diff --git a/book/2004-01.txt b/book/2004-01.txt new file mode 100644 index 0000000..262552f --- /dev/null +++ b/book/2004-01.txt @@ -0,0 +1,4201 @@ +\start +Date: Thu, 01 Jan 2004 18:00:53 +0100 +From: David MENTRE +To: axiom-developer@nongnu.org, axiom-mail@nongnu.org +Subject: [Axiom-developer] New procedure to download Axiom source code anonymously + +Hello, + +After Savannah compromise, you should now use CVS through SSH to +download Axiom source code anonymously. Here is the new procedure: + + + * export CVS_RSH="ssh" + + * cvs -z3 -d:ext:anoncvs@savannah.nongnu.org:/cvsroot/axiom co axiom + + +The SSHv2 public key fingerprints for the machine hosting the cvs trees are: + +RSA: 1024 80:5a:b0:0c:ec:93:66:29:49:7e:04:2b:fd:ba:2c:d5 +DSA: 1024 4d:c8:dc:9a:99:96:ae:cc:ce:d3:2b:b0:a3:a4:95:a5 + + +After first download, you can type usual 'cvs -z6 update' to update your +local source tree. + + +If you are a Savannah developer, please be sure that you have Protocol 2 +for all savannah related hosts in your .ssh/config file. + + +Let us know of any issues, + +PS: I have updated the web page accordingly but I don't know when +changes will take effect. + +\start +Date: 02 Jan 2004 16:13:53 -0500 +From: Camm Maguire +To: daly@idsi.net +Subject: [Axiom-developer] 15% compile time speedup + +Greetings! If one turns on SGC and does a +(si::set-gmp-allocate-relocatable t) in the initial image save (and +any possible calls to compiler::link as in my Debian patches), one can +shave off 20 min out of the original 136 min of build time. Further +enhancements doubtlessly possible, though of what interest ... + +\start +Date: 03 Jan 2004 23:46:05 -0500 +From: Camm Maguire +To: ddj@gyxi.dk +Subject: [Axiom-developer] Re: [Gcl-devel] readline support +Cc: maxima@www.ma.utexas.edu, acl2@lists.cc.utexas.edu + +[ re: command line completion with readline enabled GCL and derivatives] + +Greetings! OK, I've got something now that works like this for GCL. + +sym completes to external syms current package + used packages +pk:sym completes to external syms of pk + used packages +:sym completes to keyword symbols +::sym completes to internal symbols of current package + used packages +pk::sym completes to internal symbols of pk + used packages + +I'll probably commit shortly. But it would be great if we could +extend a bit to get useful behavior under maxima et.al. The maxima +functions appear to correspond to internal symbols in the maxima +package with a $ prepended. Perhaps we could have a user definable +function to translate back and forth from user input text to text to +be matched against lisp symbols. We also need to figure out what to +do regarding character case. + +Suggestions for maxima/acl2/axiom appreciated. + +Take care, + +ddj@gyxi.dk writes: + +> However, I'm not sure whether this should be the +> external symbols of the current package, external +> and internal, or these and the inherited symbols. +> I'm leaning toward external symbols of the current +> package and its package use list be default, +> with a possible alternate behavior when one or +> more colons appear. Thoughts? +> +> Yes. The external symbols of the current package and its +> used packages, yes that would be nice. +> +> What if by prefixing two colons, only internal symbols +> would be expanded? +> +> If I write a package name and one colon, it should behave +> as if that package is the current package. If I use two +> colons it should only try and complete internal symbols +> (including inherited internal symbols, perhaps). If I +> use two colons without any package name in front of them +> it should complete internal symbols in the current package +> (and again perhaps also all inherited internal symbols of +> the current package). +> +> On the other hand, maybe two colons should just complete +> all symbols whether they are internal, shadowed or +> external. I have on a few occassions missed the ability +> to find an internal symbol by simple completion. It can +> be very handy. I do not know if this is general enough +> to warrant its implementation. +> +> Thanks for all your work! + +\start +Date: Sun, 4 Jan 2004 00:44:48 -0500 +From: root +To: camm@enhanced.com +Subject: Re: [Axiom-developer] Re: [Gcl-devel] readline support +Cc: maxima@www.ma.utexas.edu, acl2@lists.cc.utexas.edu, ddj@gyxi.dk + +Do you call an internal "completion" function? +If so, what parameters does it take? + +\start +Date: Sun, 04 Jan 2004 16:42:12 +0100 +From: David MENTRE +To: axiom-developer@nongnu.org +Subject: [Axiom-developer] Some links on Matroids + +Hello, + +I've googled a little to know more about Matroids that were underlined +by Bertfried Fauser in a recent email (point g in +http://mail.gnu.org/archive/html/axiom-developer/2003-12/msg00064.html). + +Apparently, the reference for Matroids is: + http://members.aol.com/matroids/ + +The most simple explanation that I have found is: + http://www.ms.uky.edu/~pagano/Matridx.htm + +The MathWorld definition: http://mathworld.wolfram.com/Matroid.html + +[ Usual disclaimer: I'm not a mathematician, blah blah ] + +>From what I have understood, Matroids has several definitions, the most +frequent being related to set theory. Apparently, Matroids can be +related to graph theory and matrices. + +However, I've not found a simple explanation on how a Matroid can help +us classify information resulting from Axiom source code. In fact, +Bertfried first email is more clear that all the explanations that I +have found. Bertfried, could you give me some more insights on the use +of Matroids to classify dependent and indepent Axiom information? + +\start +Date: Sun, 4 Jan 2004 19:19:20 -0500 +From: root +To: david.mentre@wanadoo.fr +Subject: [Axiom-developer] organization + +Thanks for the pointers on matriods. + +My current thinking runs as follows: + +For each domain contruct an "index card" that contains various fields +(the index card is the structural equivalent of a "concept"). Given a +set of these index cards each field has at least one default ordering +and a default search mechanism. + +The data for the fields in the index card can be constructed by +automated programs, for example, the compiler. + +An example of an index card might contain the fields of "the general +domain structure". A general domain structure looks something like: + +Foo(Join(args with sigs)): Cat with sigs == Join(dom(args with sigs)) + with sigs + add functions + +and the compiler can create an "index card" with the following fields: + + domainName: Foo + arglist : Join(args with sigs) + category : Cat with sigs + extends : Join(dom(args with sigs)) + with : sigs + add : functions + +Of course each of the parts contains subparts (the individual list of +functions) and references to other index cards (Cat with sigs), etc. +So in general there is a lot of implied structure given on each index +card. + +Having created this pile of index cards we can look at the implied +structures and other interesting organizations. + +Now if we look at a given FIELD in the set of index cards we can create +a program which connects the cards together. So we could classify +the cards into many networks, at least one per field. For instance, +we could create the inheritance lattice based on the way that one +domain uses another. + +There can be many kinds of index cards created by many kinds of parsers +(e.g. an index card that parses a pamphlet file into component parts +like title, list of chunk names, list of section names, etc). + +The general picture emerges of a very rich set of objects (the various +index cards) that are classified into dozens of different relationships. +You can construct predicates that walk many different paths to answer a +question. For example, find the function that will be called to +multiply a matrix of complex numbers by a vector of polynomials over +the integers and tell me what file contains it. We could compute the +coercion from one domain to another per Nicolas's thesis. + +This approach breaks the problem into three parts, +writing parsers to automatically construct the index cards, +constructing the relations, +constructing the queries. + +(The generalization which I'm exploring but not explaining is that the +index card itself is just another network. The whole thing eventually +dissolves into a very large, structured network but it's too complex +to explain that way). + +I'm essentially constructing the index cards for the algebra (by hand +at the moment) to experiement with this idea. I'll post the reduced +form of these once I've completed the first pass. + +\start +Date: Mon, 5 Jan 2004 11:31:03 +0100 (CET) +From: Bertfried Fauser +To: David MENTRE +Subject: Re: [Axiom-developer] Some links on Matroids + +On Sun, 4 Jan 2004, David MENTRE wrote: + +Dear David! + + the idea with the matroid was a far reaching one, not assumed to +be practical at the very moment. I can however explain what in my mind is +behind this idea. + +* In the 30ies (if I am right) Hassler Whithney started to think about +dependency of mathematical terms. Think of: +- Linear dependence. This needs a module and a ring the module is build + over. A dependency is a *linear* relation between elements of the the + module. This is the dependency attached to linear algebra. Its too + weak to describe more sublte dependencies. +- Algebraic dependence. Think of a cyclic group (or modulo arithmetik) + where relations like x^n=1 emerge. No scalars involved here. Similar + problems occurer in Hilberts syzygie theory, where one has an + *algebra of dependencies*. +Whitney set out to distill from these type of problems the core of the +term *dependent* and coined for that the matriod definition. + +* A most perplexing feature of matroids is that they are *cryptomorphic*. +That is, matriods can be defined in quite different terms and there are +proofs that these definitions are equivalent. Eg. matroids can be defined +on graphs or on linear spaces. It shows up, that theorems easyly proved +inthe graph picture might be quite hard theorems in the linear space +picture and vice versa. This is nicely explained in a book by Peter +Laeuchli, ETH Zurich (can`t remember the title, something like +introduction to matroids for computer scientists) + Hence there is only the abstract notion of matroid and various +*presentations* which might be more or less useful.` in a given context. + +* If you are familira with incidence geometries, you may address a matriod +as a generalization of such a geometry. This is the point where I think +matroids could help to handle the algebra depencency problem. + +Let us think of the following setting. We have atributes, totally ordered +by complexity (say most complex is the code itself, and least complex is a +user help page describing usage or even just examples). We may consider +this complexity as a sort of dimension or *grading* and attach to each +such concept a sort of hypervolume. Code is one dimesnional, examples are +high dimesnional. + A dependency of two objects may then be described via an inclusion +operation (incidence relation). If a piece of code belongs to a domain, +category or user example, then there is a dependency. One might even think +of computing the dimension of the common part to give a measure how +dependent two concepts are. + This type of dependency is nonlinear in general. + + + +I am not yet sure how, but expect the following to be possible: Matriod +techniques should help to formalize and handle dependency relations in +such a high dimensional object. (its not only a graph, since you have not +only edges and vertices, but also volumes, hypervolumes etc.) + Helpful in this quest is the exchange property of matroids. Every +matroid has a set of bases, say B={b_1,b_2,...} where b_i are bases. You +might select two bases, say b_1 and b_2 and wish to delete one basis +element from b_1, say x, then there is always one element z in b_2 such +that b_1\x \/ {z} (\/ means union here) is again a basis. Such algorithms +can be used to modify dependencies into a wishful way. + +What would be needed to implement such a matroid based nonsense. The hard +thing is that one would have to describe somehow all possible dependencies +(alternatively all independent sets) of the whole set under consideration +(might be derivable from Tims index charts). Then it should be possible to +ask questions like + what ring has attribute X do not use "has commutative" +to retrieve information about the category ring having attribute X but +eliminate (if possible) the depencency on the attribut "has commutative" + +To be frank, I cannot implement such a thing in software! + +WARNING: +* There are matroids which cannot be described as vectorsapce matroid +* Matroids which can be described by vectors, might need a quite peculiar + ring of scalars, say Z_2 or even more awkward + +hope this explains my thinking a little bit, + +\start +Date: Mon, 5 Jan 2004 07:35:08 -0500 +From: root +To: axiom-developer@nongnu.org, axiom-math@nongnu.org +Subject: [Axiom-developer] Dick Jenks + +Axiom was a project at IBM Research started in the early 70s +by Richard Jenks. He managed the project over its 23 year +lifetime at IBM. I worked for him on the project and had +the greatest respect for him and held him in high esteem. +Almost everyone involved in the field of computer algebra +knew him and was influenced by him. + +Dick died yesterday. I have no words. + +\start +Date: Mon, 05 Jan 2004 06:46:42 -0800 +From: Stefan Dirnstorfer +To: axiom-developer@nongnu.org +Subject: [Axiom-developer] Questions + +I do have some serious problems with Axiom and recursions. Here are two +problems I encountered in the latest binary release. + +1) Here Axiom seems not to correctly bind the variable c + +(2) -> cons:= c+->(x+->c) + + (2) c +-> x +-> c + Type: +AnonymousFunction +(3) -> cons 3 + + (3) x +-> G1404 + Type: +AnonymousFunction +(4) -> % 6 + + (4) G1404 + + + +2) The next one might just be a problem of recursive function +definition. How else should it be done? + +(1) -> fac:= x+-> if x=0 then 1 else fac(x-1) + + (10) + x + +-> + if x= 0 + then 1 + else fac(x - 1) + Type: +AnonymousFunction +(2) -> fac 1 + + >> System error: + Bind stack overflow. + +protected-symbol-warn called with (NIL) + +\start +Date: Mon, 5 Jan 2004 08:31:22 -0500 +From: root +To: david.mentre@wanadoo.fr +Cc: axiom-developer@nongnu.org, daly@idsi.net +Subject: [Axiom-developer] fedora iso images + +http:/download.fedora.redhat.com +www.fedora.us +download.fedora.us + +\start +Date: Mon, 05 Jan 2004 11:36:11 -0500 +From: Camm Maguire +To: root +Subject: Re: [Axiom-developer] Re: [Gcl-devel] readline support +Cc: maxima@www.ma.utexas.edu, acl2@lists.cc.utexas.edu, ddj@gyxi.dk + +Greetings! + +============================================================================= +Do you call an internal "completion" function? +If so, what parameters does it take? + +t +============================================================================= + + +Effectively, the call-back gets a char * pointer to the word of text +being entered, and a 'state' integer indicating whether the call-back +has been called with this pointer before (and if so how many times). +The user supplied 'generator' routine is supposed to return the next +partially matching word of acceptable input, and NULL in the case of +no remaining matches. We may be able to alternately configure things +to give the call-back the whole line of text, together with start and +end integers bounding the current word. + +Can axiom function names be easily mapped onto the set of symbols in +some package(s)? + +\start +Date: Mon, 5 Jan 2004 13:48:05 -0500 +From: Tim Daly +To: axiom-developer@nongnu.org, axiom-mail@nongnu.org +Subject: [Axiom-developer] Richard Jenks +Cc: gilbert@sci.ccny.cuny.edu + +Would you please send the following sad notice to the Axiom mailing list. +Thanks, and + +Best regards, +Bob Caviness +---------- + Richard Dimick Jenks + Axiom Developer and Computer Algebra Pioneer + +On December 30, 2003, Dick died at the age of 66 after an extended and +courageous battle with multiple system atrophy. He received his PhD in +mathematics from the University of Illinois at Urbana-Champaign and was +a long time employee of IBM Research until his retirement in 2002. At +IBM he was a principal in the design and implementation of the Axiom +computer algebra system and a co-author with Robert Sutor of the +"Axiom" book. He was also the manager, for many years, of the IBM +group that developed Axiom. He, along with David Chudnovsky, organized +the highly successful meetings on Computers and Mathematics that were +held at Stanford in 1986 and MIT in 1989. As a legacy of those +meetings, a Jenks Prize for outstanding contributions to software +engineering in computer algebra is being established. If you would +like to make a contribution to the prize fund, make a check to the +"Jenks Computer Algebra Prize" and send it to + + Jenks Computer Algebra Prize + c/o Prof. B. F. Caviness + Dept. of Computer & Information Sciences + 103 Smith Hall + University of Delaware + Newark, DE 19716 + USA + +A memorial service is scheduled for 11AM on Saturday, January 10, 2004 at +St. Luke's Episcopal Church +850 Wolcott Avenue +Beacon, NY + +A more extensive death notice was published in the New York Times on +Sunday, January 4, 2004. + +\start +Date: Mon, 05 Jan 2004 21:20:17 +0100 +From: David MENTRE +To: Stefan Dirnstorfer +Subject: Re: [Axiom-developer] Questions + +Stefan Dirnstorfer writes: + +> 2) The next one might just be a problem of recursive function +> definition. How else should it be done? +> +> (1) -> fac:= x+-> if x=0 then 1 else fac(x-1) +> (10) +> x +> +-> +> if x= 0 +> then 1 +> else fac(x - 1) +> Type: +> AnonymousFunction +> (2) -> fac 1 +> >> System error: +> Bind stack overflow. +> protected-symbol-warn called with (NIL) + +You should use following syntax: + +(1) -> fac n == if n = 0 then 1 else n * fac(n-1) + Type: Void +(2) -> fac 1 + Compiling function fac with type Integer -> Integer + Compiling function fac as a recurrence relation. + + (2) 1 + Type: PositiveInteger +(3) -> fac 10 + + (3) 3628800 + +In your above definition (fac:= x+-> if x=0 then 1 else fac(x-1)), I +think you define a _variable_ that _contains an _anonymous_ +function. Apparently this function is unknown to Axiom when it looks for +the recursive call. + +\start +Date: Mon, 05 Jan 2004 21:11:10 +0100 +From: David MENTRE +To: daly@idsi.net, Tim Daly +Subject: [Axiom-developer] Re: organization + +Hello Tim, + +Your email is quite interesting. Please find below my questions on your +approach: + +root writes: + +> For each domain contruct an "index card" that contains various fields +> (the index card is the structural equivalent of a "concept"). Given a +> set of these index cards each field has at least one default ordering +> and a default search mechanism. +> +> The data for the fields in the index card can be constructed by +> automated programs, for example, the compiler. + +> Of course each of the parts contains subparts (the individual list of +> functions) and references to other index cards (Cat with sigs), etc. +> So in general there is a lot of implied structure given on each index +> card. + +How would you define the structure of each card? Using Common Lisp data +type? Specifically defined data types? With a Card Definition Language? +Using standardized tools as W3C RDF or OWL? Using ACL2 common lisp +subset? + +> Having created this pile of index cards we can look at the implied +> structures and other interesting organizations. +> +> Now if we look at a given FIELD in the set of index cards we can create +> a program which connects the cards together. So we could classify +> the cards into many networks, at least one per field. For instance, +> we could create the inheritance lattice based on the way that one +> domain uses another. + +Building new structures _dynamically_ is important there. I would like +to build a structure, examining it and find it misses something, build a +new one and finally save it as a card because I find it interesting. + +> There can be many kinds of index cards created by many kinds of parsers +> (e.g. an index card that parses a pamphlet file into component parts +> like title, list of chunk names, list of section names, etc). + +How would you store the cards? On disk? As Common Lisp data structure? +On a web server? + +> The general picture emerges of a very rich set of objects (the various +> index cards) that are classified into dozens of different relationships. +> You can construct predicates that walk many different paths to answer a +> question. For example, find the function that will be called to +> multiply a matrix of complex numbers by a vector of polynomials over +> the integers and tell me what file contains it. We could compute the +> coercion from one domain to another per Nicolas's thesis. +> +> This approach breaks the problem into three parts, +> writing parsers to automatically construct the index cards, +> constructing the relations, +> constructing the queries. + +I would add: + + define a common repository for card structures (with attached semantics + for each field, or reference to other card structures) + + +> (The generalization which I'm exploring but not explaining is that the +> index card itself is just another network. The whole thing eventually +> dissolves into a very large, structured network but it's too complex +> to explain that way). +> +> I'm essentially constructing the index cards for the algebra (by hand +> at the moment) to experiement with this idea. I'll post the reduced +> form of these once I've completed the first pass. + +I would be interested to write index cards starting from the source +file. + +For example, considering the simplest card: + +Source file : file-name-type +Subsystem : choice-in(boot database spad-compiler spad-interpreter ???) +Explanation : english-text + +source file subsystem explanation +----------- --------- ----------- +src/boot/Makefile.pamphlet boot building of the boot parser sub-system +src/boot/boothdr.lisp.pamphlet boot header for boottran lisp package +src/boot/bootload.lisp.pamphlet boot loading of various lisp files constituing the boot subsystem +src/boot/btincl2.boot.pamphlet boot parsing of included lines in boot sources +src/boot/btpile2.boot.pamphlet boot handling of piles (source indentation) in boot +src/boot/btscan2.boot.pamphlet boot parsing of boot syntax +src/boot/ccl-bootsys.lsp.pamphlet boot loading of several lisp files constituing the boot subsystem +src/boot/ccl-depsys.lsp.pamphlet boot lisp file used to build depsys on top of CCL +src/boot/exports.lisp.pamphlet boot set of lisp symbols exported by the boottran package +src/boot/npextras.lisp.pamphlet boot definition in lisp of various boot variables and elementary functions +src/boot/ptyout.boot.pamphlet boot set of lisp functions to translate boot into common lisp +src/boot/tyextra.boot.pamphlet boot ?? Handling of boot data structures ?? +src/boot/typars.boot.pamphlet boot ?? Handling of boot syntax ?? +src/boot/typrops.boot.pamphlet boot Parsing of boot keywords (tokens?) +src/boot/tytree1.boot.pamphlet boot transformation of boot into common lisp +src/doc/msgs/co-eng.msgs database list of error messages +src/doc/msgs/s2-us.msgs database list of Axiom error messages + +After that, I would like to dig into each file and build new cards, for +example describing the internal structure of s2-us.msgs. I would like to +build cards indexing each boot or lisp function, using internal +information from the compiler. + +Now, two last but probably most traitorous questions: + + - do you thing we could be quickly a tool to navigate in cards? I think + a computerized tool is of most importance to attack Axiom + complexity. I can imaging throwing some common lisp data structure + and functions, however it might become rapidly unusable. Have you any + idea of a (free software) tool that we might reused? + + - don't you think attacking this issue now could delay too much the + release of a _stable_ Axiom (with a documentation, which is probably + the most lacking currently)? What about people that have submited bug + reports? + + +In short, I like your approach and I'm eager to apply it (however not on +the algebra ;-). However, I fear it would delay more important tasks. + +\start +Date: Mon, 05 Jan 2004 21:42:25 +0100 +From: David MENTRE +To: Stefan Dirnstorfer +Subject: Re: [Axiom-developer] Questions + +Stefan Dirnstorfer writes: + +> 1) Here Axiom seems not to correctly bind the variable c +> +> (2) -> cons:= c+->(x+->c) +> (2) c +-> x +-> c +> Type: +> AnonymousFunction + +I don't understand your example. What are you trying to build? + +\start +Date: 05 Jan 2004 15:30:34 -0500 +From: Camm Maguire +To: root +Subject: Re: [Axiom-developer] Re: [Gcl-devel] readline support + +Greetings! Tim, do the axiom commands all correspond to symbol names +in the BOOT package with | prepended and appended? + +\start +Date: Mon, 5 Jan 2004 17:04:26 -0500 +From: root +To: camm@enhanced.com +Subject: Re: [Axiom-developer] Re: [Gcl-devel] readline support + +The symbol naming convention is more complicated than that. +You can see the function names in the .lsp files +(e.g. int/algebra/DHMATRIX.NRLIB/code.lsp) +The name is a combination of several things including some +type-related mangling and a unique number within the datatype. + +We do have functions that can return information about the type +and its methods. Perhaps it is possible to use these functions +to suggest possible completions. + +\start +Date: 05 Jan 2004 17:12:24 -0500 +From: Camm Maguire +To: daly@idsi.net +Subject: Re: [Axiom-developer] Re: [Gcl-devel] readline support + +Greetings, and thanks! What are these functions, and how are they +called? + +Take care, + +root writes: + +> The symbol naming convention is more complicated than that. +> You can see the function names in the .lsp files +> (e.g. int/algebra/DHMATRIX.NRLIB/code.lsp) +> The name is a combination of several things including some +> type-related mangling and a unique number within the datatype. +> +> We do have functions that can return information about the type +> and its methods. Perhaps it is possible to use these functions +> to suggest possible completions. + +\start +Date: Tue, 6 Jan 2004 13:04:49 -0500 +From: "Page, Bill" +To: "'Bertfried.Fauser@uni-konstanz.de'" +Subject: RE: [Axiom-developer] Some links on Matroids + +On Monday, January 05, 2004 5:31 AM Bertfried Fauser +[mailto:fauser@spock.physik.uni-konstanz.de] wrote: +> ... +> * A most perplexing feature of matroids is that they +> are *cryptomorphic*. That is, matriods can be defined +> in quite different terms and there are proofs that +> these definitions are equivalent. Eg. matroids can be +> defined on graphs or on linear spaces. It shows up, +> that theorems easily proved in the graph picture might +> be quite hard theorems in the linear space picture and +> vice versa. This is nicely explained in a book by Peter +> Laeuchli, ETH Zurich (can`t remember the title, something +> like introduction to matroids for computer scientists) + +I haven't found any reference on the web to a book by +Peter Laeuchli about matroids. But I do know of some +lecture notes entitled: + + Combinatorics for Computer Scientists + by Dany Breslauer and Devdatt P. Dubhashi. + +see + + http://www.brics.dk/LS/95/4/BRICS-LS-95-4.ps.gz + +in + + http://www.brics.dk/LS/95/Abs/BRICS-LS-95-Abs/ + +also containing some other interesting lectures on +polymorphic type inference and category theory. + +--------- + +Combinatorics for Computer Scientists contains useful +self-contained chapters on both graph theory and on +algorithms involving matroids. For matroids refer to +chapter 23, page 185. + +>From this discuss it is quite clear, I think, how the +concept of matroid (Example 178), matroid bases and +the problem of finding a "maximal independent set of +minimum weight" (Theorem 195) relates to the issue +that Tim and I discussed much early of optimizing the +algebra "bootstrap" for building Axiom. + +> ... +> I am not yet sure how, but expect the following to +> be possible: Matriod techniques should help to +> formalize and handle dependency relations in such +> a high dimensional object. (its not only a graph, since +> you have not only edges and vertices, but also +> volumes, hypervolumes etc.) + +I think, maybe, what Bertfried is saying is that the +graph matroid that we can construct from the usual +dependency (call) graph is not the most general matroid +possible and therefore the dependency graph does not +represent all of the types of dependencies in which +we may be interested - even though it is apparently +sufficient for the bootstrap problem. + +> Helpful in this quest is the exchange property +> of matroids. Every matroid has a set of bases, say +> B={b_1,b_2,...} where b_i are bases. You might select +> two bases, say b_1 and b_2 and wish to delete one basis +> element from b_1, say x, then there is always one +> element z in b_2 such that b_1\x \/ {z} (\/ means +> union here) is again a basis. Such algorithms +> can be used to modify dependencies into a wishful way. +> +> What would be needed to implement such a matroid based +> nonsense. The hard thing is that one would have to +> describe somehow all possible dependencies (alternatively +> all independent sets) of the whole set under consideration +> (might be derivable from Tims index charts). +> +> Then it should be possible to ask questions like +> what ring has attribute X do not use "has commutative" +> to retrieve information about the category ring having +> attribute X but eliminate (if possible) the dependency +> on the attribute "has commutative" +> +> To be frank, I cannot implement such a thing in software! + +Perhaps this might be possible using software such as + + http://www.mcs.vuw.ac.nz/research/macek/ + +"(MAroids Computed Efficiently Kit) + +by Petr Hlineny + +Overview +This project is a set of tools and routines for reasonably +efficient combinatorial computations with represent able +matroids, especially for matroid constructions and for +some structural tests. (If you do not know what a matroid +is, then this project is likely not for you.) The program +is intended both to help with everyday routines that a +matroid-researcher faces (almost) every day, and to allow +for long exhausted computations of matroid classes and +their properties. ..." + +> +> WARNING: +> * There are matroids which cannot be described as +> vectorsapce matroid +> * Matroids which can be described by vectors, might +> need a quite peculiar ring of scalars, say Z_2 or even +> more awkward +> + +------------- + +However, about all of this I remain rather sceptical. +It is not clear to me that this notion of dependency +is what is really important - at least in the context +of the algebra. + +I think I would prefer a much more "structural" approach, +where perhaps more complex things (objects) are described +functorially in terms of other objects (or categories) and +operations on those objects that are simplier - often the +way things are actually defined in Axiom. Or by embedding +them into something more, or equally, complex but more well +understood by the reader - i.e. an example. + +I think there needs to be some way for the user to +"navigate" over such functorial maps, maintaining on the +computer screen (or at least quickly retrievable) a summary +of the immediately relavant mappings as an aid to memory. + +To the extent that Axiom is built in an "object oriented" +manner, this functorial approach should be extendable to +other parts of the system system - perhaps down to the +lisp level. + +\start +Date: 07 Jan 2004 13:16:16 -0500 +From: Camm Maguire +To: "gcl-devel" +Subject: [Axiom-developer] 2.6.2 +Cc: Robert Boyer , Matt Kaufmann , Mike Thomas + +Greetings! OK, I've just committed the readline patch. I'm doing the +full test suite now, but barring the unforeseen, I'd like to tag this +as 2.6.2, release, and then move on to head. Please speak now if +there are problems. On machines to which I have no access, I'm +assuming that nothing that was working has been broken. Please +correct me if this is not the case. We might want to make the +use-fast-links nil switch on Windows automatic i.e. handled by +configure, but apart from this I don't know of anything else that +needs to go into stable. + +I'll report back on my test results when done. Basically +self-compile, ansi-tests, Paul's compiler tests, maxima, acl2, axiom, +nqthm-1992, pc-nqthm-1992, and builds on all 12 Debian arches is my +test suite. Mac-OSX, BSD, and/or Windows might want to consider the +same. It will be a few days before all this is completed. Needless +to say, please don't commit anything non-platform specific to the +stable branch in this period. + +(BTW, readline is now initialized automatically. Completion can be +customized somewhat via the variable si::*readline-prefix*. See the +info pages for details. The Debian maxima package will be setting +this var to "maxima::$" for now.) + +\start +Date: Wed, 7 Jan 2004 14:41:54 -0500 +From: dpt@exoskeleton.math.harvard.edu (Dylan Thurston) +To: axiom-developer@nongnu.org +Subject: Re: [Axiom-developer] Questions + +On Mon, Jan 05, 2004 at 09:42:25PM +0100, David MENTRE wrote: +> Stefan Dirnstorfer writes: +>=20 +> > 1) Here Axiom seems not to correctly bind the variable c +> > +> > (2) -> cons:=3D c+->(x+->c) +> > (2) c +-> x +-> c +> > Type: +> > AnonymousFunction +>=20 +> I don't understand your example. What are you trying to build? + +A better name for this function might be 'constant'. 'c +-> x +-> c' +should be a function that takes an argument, 'c', and returns a function +that takes an argument 'x', ignores x, and returns c. That is, (cons 3) +is a constant function that always returns 3. + +Alternatively, you can think of 'cons' as a function that takes two +arguments, one after the other, and returns the first. + +Another way to see the bug is by looking at, say, + + c +-> x +-> c+x + +otherwise known as addition. You'll see the same behaviour. + +I'm quite worried by this bug, since it shows some really elementary +failure of lexical scoping, which I understood Axiom to have. I'll +investigate as soon as I have a chance. + +\start +Date: Wed, 07 Jan 2004 23:34:59 +0100 +From: David MENTRE +To: dpt@math.harvard.edu +Subject: Re: [Axiom-developer] Questions + +Hello Dylan, + +dpt@exoskeleton.math.harvard.edu (Dylan Thurston) writes: + +> A better name for this function might be 'constant'. 'c +-> x +-> c' +> should be a function that takes an argument, 'c', and returns a function +> that takes an argument 'x', ignores x, and returns c. That is, (cons 3) +> is a constant function that always returns 3. + +Ok, that's what I guessed but I wasn't sure. + +You can try: +(1) -> cons c == x +-> c + Type: Void +(2) -> a := cons 3 + Compiling function cons with type PositiveInteger -> + AnonymousFunction + + (2) x +-> #1 + Type: AnonymousFunction +(3) -> a 2 + + (3) #1 + Type: Variable #1 + + +> I'm quite worried by this bug, since it shows some really elementary +> failure of lexical scoping, which I understood Axiom to have. I'll +> investigate as soon as I have a chance. + +However, I'm not sure if it is really a bug or a bad bahavior of Axiom +type inference system. For example, is 'c' or 'x' interpreted as a +function variable or as a polynomial variable? + +You might want to try first to precise the type of arguments and +returned type when defining cons function. + +If you have access to Axiom book, chapter 6 would be a must read. + +\start +Date: Thu, 08 Jan 2004 05:00:05 +0100 +From: Stefan Dirnstorfer +To: axiom-developer@nongnu.org +Subject: [Axiom-developer] Variable scope (was Questions) + +I did some further investigation in my intended type of variable binding +and found out, that it does work in Maple and Haskell, but not in Axiom +and Lisp. My function 'cons' was ment to produce a +constant function. Here is the example produced with Maple: + +MAPLE +----- + +> cons:= c -> (x-> c); +> + + cons := c -> x -> c + +[ This is the constant function 6, where Axiom produced x -> G1234 +> cons (6); + + x -> 6 + +[ A linear function could be created in the same way +> lin:= m -> (x-> m*x); +> + + lin := m -> x -> m x + +[ And with slope two +> lin(2); + + x -> 2 x + +[ This one is tricky, but Maple can add functions. Support for this +feature is primitive, since the result is not simplified to x-> 2x + 3 +> f:=cons(3)+lin(2); + + f := (x -> 3) + (x -> 2 x) + +[ And, to prove it is a function: +> f(2); +> + + 7 + +This plain example might seem useless, but refers to one of the +fundamential programing techniques in functional languages. + + +HASKELL +------- + +Here is the same example in Haskell: +cons :: a -> (a -> a) +cons c = \x -> c + +cons 3 2 does evaluate to 3. + + +LISP +---- + +Emacs-Lisp deals differently with this situation: +(defun constant (c) (lambda (x) (c))) + +(constant 3) +;; (lambda (x) (c)) + +(apply (constant 3) '(2)) +;; Symbol's function definition is void: c + + +AXIOM +----- + +The Lisp evaluation is just equivalent to what Axiom does. It transforms +a->b->a into a->b->#1. I do wonder where comes the unbound variable +from, where there is initially none? + + +In my eyes the Maple binding does make more sense, but there is +certainly no bug in either algebra system. I am still curious if there +is a workaround in Axiom. + +David MENTRE wrote: +> Stefan Dirnstorfer writes: +> +> +>>1) Here Axiom seems not to correctly bind the variable c +>> +>>(2) -> cons:= c+->(x+->c) +>> (2) c +-> x +-> c +>> Type: +>>AnonymousFunction +> +> +> I don't understand your example. What are you trying to build? + +\start +Date: Sun, 11 Jan 2004 11:41:14 +0100 +From: David MENTRE +To: axiom-mail@nongnu.org +Subject: CVS tree up-to-date & web site updated (was: Re: [Axiom-developer] Savannah is back) + +Hello, + +David MENTRE writes: + +> Tim Daly, Axiom project leader, has checked the Axiom sources and +> confirmed that Axiom sources has NOT been compromised[2]. + +Moreover, I have compared a freshly downloaded savannah source tree (in +anonymous mode) against my own copy of the axiom source tree. There is +no difference, so the savannah cvs repository is up-to-date with latest +update before Savannah crack. + +As a side note, the web pages of axiom web site on savannah +(http://www.nongnu.org/axiom/) have been updated. + +\start +Date: Wed, 14 Jan 2004 09:36:59 -0500 +From: Tim Daly +To: kvoa@netlane.com +Subject: [Axiom-developer] Axiom for windows + +Sam Allen wrote +>Does anyone have Axiom for Windows? +> + +Sam, + +Axiom should build and install on SuSE. A GUI is not necessary +at the moment because Axiom only does text output. You can use +TeXmacs as a front end if you want a GUI interface as it will +format the equations better. + +The windows port is in process. Windows is a particularly hard +environment for programming and the progress is slow at best. + +Where is your review website? + +\start +Date: Wed, 14 Jan 2004 20:23:37 +0100 +From: David MENTRE +To: Tim Daly +Subject: State of case issue? (was: Re: [Axiom-developer] Axiom for windows) + +Hello Tim, + +Tim Daly writes: + +> Axiom should build and install on SuSE. + +BTW, regarding SuSE or RedHat packaging, have you integrated Juergen's +patch to fix case sensitive issues? + +And what is the current state of the Axiom Book? + +\start +Date: Thu, 15 Jan 2004 11:11:28 -0500 +From: root +To: david.mentre@wanadoo.fr +Subject: Re: State of case issue? (was: Re: [Axiom-developer] Axiom for windows) + +re: case sensitive patch. + +I haven't applied it yet. I'm testing a fix to the algebra from Renaud +and plan to do an update to the website soon. I'll include this patch. + +re: book + +Still in process. + +The big drag on my time at the moment is that we are in a "funding cycle". +My department needs me to write grant proposals. Since no-one seems to +be funding computer algebra in the U.S., and especially not Axiom, my +time is being spent writing 3 20 pages papers begging for money. + +\start +Date: Thu, 15 Jan 2004 23:29:23 -0500 +From: root +To: axiom-developer@nongnu.org, daly@idsi.net +Subject: [Axiom-developer] (no subject) + +Well, I tried to update the website but it appears I'm locked out. +The FSF certificate expired and it won't let me log in. Sigh. +I've sent a note to savannah. + +There is no such thing as a simple job. + +\start +Date: Sat, 17 Jan 2004 10:41:19 +0100 +From: David MENTRE +To: daly@idsi.net +Subject: Re: [Axiom-developer] (no subject) + +Hello Tim, + +root writes: + +> Well, I tried to update the website but it appears I'm locked out. +> The FSF certificate expired and it won't let me log in. Sigh. + +I've been able to upload another savannah project according to roughly +following procedure: + + 1. modify your .ssh/config to have: +Host subversions.gnu.org + Protocol 2 + +Host savannah.nongnu.org + Protocol 2 + +Host savannah.gnu.org + Protocol 2 + + + 2. download an new fresh GOLD tree with following command: + +export CVS_RSH=ssh +cvs -z3 -d axiom@savannah.nongnu.org:/cvsroot/axiom co axiom + ^^^^^Your savannah login, Tim + + +The new authentification keys of savannah are: +RSA: 1024 80:5a:b0:0c:ec:93:66:29:49:7e:04:2b:fd:ba:2c:d5 +DSA: 1024 4d:c8:dc:9a:99:96:ae:cc:ce:d3:2b:b0:a3:a4:95:a5 + + + 3. after that, should be able to do any commit, from that tree. + + +I hope it helps, +Let me know of any issue, + +\start +Date: Sat, 17 Jan 2004 10:44:23 +0100 +From: David MENTRE +To: daly@idsi.net +Subject: Re: [Axiom-developer] (no subject) + +David MENTRE writes: + +> root writes: +> +>> Well, I tried to update the website but it appears I'm locked out. + +Oops, sorry, I did not read correctly your previous email. Did you mean +updating Axiom _Web pages_ on savannah? + +In that case, the CVS path is completely different, so the commands are: + +export CVS_RSH=ssh +cvs -z3 -d YOURLOGIN@savannah.nongnu.org:/webcvs/axiom co axiom + +Once again, you should start from a new fresh GOLD tree. + +\start +Date: Sat, 17 Jan 2004 11:44:03 +0100 +From: David MENTRE +To: axiom-developer@nongnu.org +Subject: [Axiom-developer] [crystal] A proposed framework and associated tool: LIS (Logical Information Systems) + +Hello Axiom developers and users, + +Tim has expressed the desire to have an Information System that allows +to explore and modify Axiom internals. He called that the Crystal. This +has raised some discussions on the Axiom developer list recently +(http://mail.nongnu.org/archive/html/axiom-developer/2003-12/msg00075.html). + +I think I have found a *possible* approach to solve issues raised by +Tim. It is called LIS (Logical Information Systems, +http://www.irisa.fr/LIS/). This is a research made at IRISA, Rennes, +France, by Olivier RIDOUX et al. They propose a formal framework, +algorithms and tools (i.e. running code) to classify a system, *explore* +it and *modify* this classification, always in a coherent manner. + +Roughly speaking, they propose to attach (valuated) attributes to files +and explore a file hierarchy according to those attributes, by following +attribute paths like one would follow file paths. It is even possible to +lively modify the explored hierarchy (in a coherent manner!). Moreover, +the files itself can be seen has having attributes (per file line for +example). For example, a C or Lisp file could been seen as a directory +containing the set of defined functions in that file. + +One of the most interesting feature of their approach is that they +proposed a regular file system interface (called LISFS) to use it. In +other words, from the beginning, the system is usable with a wide range +of existing tools (bash shell, emacs & vi, Gnome and KDE graphical file +browsers, find & grep, ...). + +Another interesting approach is that one can have different taxonomy +(i.e. set of classification attributes), that can be used +simultaneously. For example, language specific attributes on C files and +system attributes (file type, file size, etc.). Through a plugin system, +one can add others attributes, to enrich the file system and exibit more +information. + +Here is an example that might make the LIS more clear. + +%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% +This is a shell session done in an emacs buffer. My comments are added +between shell commands. +%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% + + We want to explore the file through their extensions, so we go into + the ext: directory. + +/sfl/sfl1 $ cd ext:/ +/sfl/sfl1/ext: $ ls +ext:c mymusic name:big name:mp3_transducer name:staying_alive size:18 size:42 +ext:jpg mypictures name:ete_indien name:naivec_adv_transducer plugins size:3 size:512284 +ext:mp3 myprograms name:foo name:small size:118 size:39 size:514395 + + We then want to explore only the C files, so we go into the ext:c directory. + +/sfl/sfl1/ext: $ cd ext:c/ +myprograms name:foo name:naivec_adv_transducer plugins size:118 size:514395 +/sfl/sfl1/ext:/ext:c $ l +total 0 +drwxrwxrwx 0 root root 0 Jan 1 1970 . +drwxrwxrwx 0 root root 0 Jan 1 1970 .. +drwxrwxrwx 0 root root 0 Jan 1 1970 myprograms +drwxrwxrwx 0 root root 0 Jan 1 1970 name:foo +drwxrwxrwx 0 root root 0 Jan 1 1970 name:naivec_adv_transducer +drwxrwxrwx 0 root root 0 Jan 1 1970 plugins +drwxrwxrwx 0 root root 0 Jan 1 1970 size:118 +drwxrwxrwx 0 root root 0 Jan 1 1970 size:514395 + + We concentrated on the foo C file (i.e. foo.c), so we go in its + directory named foo/. + +/sfl/sfl1/ext:/ext:c $ cd name:foo/ +foo.c +/sfl/sfl1/ext:/ext:c/name:foo $ l +total 0 +drwxrwxrwx 0 root root 0 Jan 1 1970 . +drwxrwxrwx 0 root root 0 Jan 1 1970 .. +-rw-r--r-- 1 pad root 118 Jan 16 18:45 foo.c + + We can edit it if we want. foo.c appears as a regular file. + +/sfl/sfl1/ext:/ext:c/name:foo $ emacs foo.c +# + + More interestingly, we can explore the foo.c file itself, using some + information of the C language. So we code in the parts/ directory of + the foo.c file (pathname /ext:c/name:foo/). + +/sfl/sfl1/ext:/ext:c/name:foo $ cd parts + + Doing a ls, we see the functions defined in foo.c. We have f and f2, + as well as the whole file, foo.c. + +/sfl/sfl1/ext:/ext:c/name:foo/parts $ ls -l +total 977 +drwxrwxrwx 0 root root 0 Jan 1 1970 . +drwxrwxrwx 0 root root 0 Jan 1 1970 .. +drwxrwxrwx 0 root root 0 Jan 1 1970 debugging +drwxrwxrwx 0 root root 0 Jan 1 1970 error +-rw-r--r-- 1 pad root 1000000 Jan 16 18:47 foo.c +drwxrwxrwx 0 root root 0 Jan 1 1970 function:f +drwxrwxrwx 0 root root 0 Jan 1 1970 function:f2 +drwxrwxrwx 0 root root 0 Jan 1 1970 synchro +drwxrwxrwx 0 root root 0 Jan 1 1970 var: + + We can see or edit foo.c. + +/sfl/sfl1/ext:/ext:c/name:foo/parts $ cat foo.c +int f(int x) { +int z; +assert(x > 1); +z = x; +fprintf(stderr, "x = 0", x); +return z * 2 +} +int f2(int z) { +return z * 4 +} + + But we can also go into the C function called f... + +/sfl/sfl1/ext:/ext:c/name:foo/parts $ cd function:f/ + + ...and visualize or edit it. + +/sfl/sfl1/ext:/ext:c/name:foo/parts/function:f $ cat foo.c + +int f(int x) { +int z; +.........:1 +z = x; +fprintf(stderr, "x = 0", x); +return z * 2 +} +%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% + +One could imagine to define attributes set for Pamhplet, SPAD, BOOT, +Lisp, etc. + +As I said previously, LIS researchers have running code (under GNU GPL +license, written in Obbective Caml), although not yet released to the +public. + +If you want to know more on LIS, they have a full range of papers on +there web site (http://www.irisa.fr/LIS/). The USENIX 2003 paper is +probably the most closed to what I have shown. + + +So, now, what about LIS and Axiom? I am personnaly convinced that LIS is +a viable approach to solve Tim issues. I'll try in the mid term to make +a first demonstrator, focused on the algebra and the exploration of the +type hierarchy. From that demonstrator, I hope to convince Axiom +developer that the LIS approach is indeed interesting. + +If you have questions, do not hesitate to ask me. + +\start +Date: Sat, 17 Jan 2004 12:00:19 +0100 +From: David MENTRE +To: Bertfried.Fauser@uni-konstanz.de +Subject: More on type browser (was: Re: [Axiom-developer] Re: Axiom interactive input syntax) + +Hello Bertfried, + +Regarding your intent to have an Axiom type browser, I think I have +found a framework and a tool to make it (see my last email on +LIS). However, I would need further information on what you expect to do +it properly. + +Bertfried Fauser writes: + +> perhaps I am describing what the AXIOM browser will do nayway. +> However, for me AXIOM is still lacking a f`type browser which can +> graphically (like a file system tree) show the dependencies of typs. +> +> Eg (fictious I have currently no AXIOM available) +> +> browse Ring +> +> Ring +> |___ Euclidean Ring +> | |____ Division Ring +> |___ Artinear Ring +> |... +> ... +> +> which shows which types are subtypes, (downsearch) + +Ok, you want (1) to show the type hierarchy using type/subtypes +relationship. + +> A second way of working should diplay the upbraces of teh type tree +> (graph?) so that you see all datastructures which contain the type +> Ring, that may be modules, groups, abelian monoids, etc.... + +Can you elaborate more on that. Do you want to search all operators +having the type Ring as argument? Would you like to search for a +specific operator? According to which criteria? a type of function? An +input or ouput type? Operator name? + + +> This type of knowledge is importand to program AXIOM packages. I spent +> currently the most time with AXIOM by searching correct types (in fact the +> most general type which allows me to do teh calculation I want to perform) + +Could you give me a concrete example, i.e. with real Axiom types and +operators, of what you would like to do in a perfect world. + +It might even be possible to realize a system that proposes the most +general type doing a computation. But I need a more concrete example to +really understand what you want to supply to the search system. + +A commented example of Axiom computation with usage example of the +(currently hypothetical) type browser and type searcher would be +perfect. + +\start +Date: Sat, 17 Jan 2004 20:27:44 +0100 +From: David MENTRE +To: axiom-developer@nongnu.org +Subject: [Axiom-developer] [axiom journal] Mizar Project and Journal of Formalized Mathematics + +Hello, + +Regarding the Axiom Journal, a friend told me about the Mizar project +(http://mizar.org/): + + The Mizar project started about 1973 with an attempt to reconstruct + mathematical vernacular. + + Since 1989, the most important activity in the Mizar project, apart + from continual improvement of the Mizar System, is the development of + a data base for mathematics. International cooperation (the main + partners: Shinshu University in Nagano and University of Alberta in + Edmonton) resulted in creating a data base which includes more than 2 + thousands of definitions of mathematical concepts and more than 30 + thousands of theorems. + + +One interesting point about the project is that they have a journal +(http://mizar.org/JFM/) which seems similar to what Tim wanted for his +Axiom Journal, i.e. with machine readable description of paper content +(http://mizar.org/language/pages/mizar-article.html). + +It might be of some interest to look at it before crafting the specs of +the Axiom Journal. + +\start +Date: Sat, 17 Jan 2004 16:16:37 -0500 +From: root +To: david.mentre@wanadoo.fr +Subject: [Axiom-developer] mizar + +David, + +Mizar looks like an interesting idea. Carlo Traverso is working +on a literate programming journal. I'll pass the reference on to +him. + +This also fits a second goal, the CATS (Computer Algebra Test Suite). + +Good work. Thanks. +Tim + +============================================================== +Regarding the Axiom Journal, a friend told me about the Mizar project +(http://mizar.org/): + + The Mizar project started about 1973 with an attempt to reconstruct + mathematical vernacular. + + Since 1989, the most important activity in the Mizar project, apart + from continual improvement of the Mizar System, is the development of + a data base for mathematics. International cooperation (the main + partners: Shinshu University in Nagano and University of Alberta in + Edmonton) resulted in creating a data base which includes more than 2 + thousands of definitions of mathematical concepts and more than 30 + thousands of theorems. + + +One interesting point about the project is that they have a journal +(http://mizar.org/JFM/) which seems similar to what Tim wanted for his +Axiom Journal, i.e. with machine readable description of paper content +(http://mizar.org/language/pages/mizar-article.html). + +It might be of some interest to look at it before crafting the specs of +the Axiom Journal. + +\start +Date: Sat, 17 Jan 2004 17:05:04 -0500 +From: root +To: mus@mizar.uwb.edu.pl, mml@mizar.uwb.edu.pl +Subject: [Axiom-developer] mizar +Cc: gilbert@sci.ccny.cuny.edu + +My name is Tim Daly. I'm the lead developer on the Axiom project. + +Axiom is a free, open source, general purpose computer algebra system +similar to Mathematica and Maple. It was originally developed at IBM +Research starting in 1973. It was later sold to NAG (Numerical Algorithms +Group). Axiom was a commercial product until 2001 when it was withdrawn +from the market and released as free software under the Modified BSD license. +The current home page is: +http://savannah.nongnu.org/projects/axiom + +Axiom consists of many hundreds of algorithms, many of which are +implementations of research level mathematics. Many of these algorithms +cannot be understood without reading the associated research papers. +Currently the research papers are in libraries, journals, or proceedings. + +We have been discussing ways of combining the research with the executable +code. Knuth has recommended a style called "literate programming" which +combines TeX files with special tags that embed the source code. This +would allow us to keep the research paper with the implementation code. +Dr Carlo Traverso is looking at developing a specialized journal in +computational mathematics that would accept articles written as literate +programs. + +As computers begin to have a larger impact on mathemtatics there will +be a subset of mathematics, which I call computational mathematics, +which can only be performed by computer. We need to develop formal +ways of handling the mathematics and its associated computational +aspects. In particular, we need to develop machinery (similar to +your Mizar syntax) that allows the computer to access and manipulate +the mathematics as well as the algorithms. + +I've been looking at your Journal of Formalized Mathematics. It appears +that we have areas where our goals overlap. For instance, in the paper +"Introduction to Turing Machines" there is a description of turing machines +which can easily be reduced to executable code. I notice, however, that +it does not include the code. Traverso's Journal would extend the article +with such code. + +If you have any interest in considering cooperative efforts (such as +extending the Mizar syntax to allow executable code) please let me know. + +\start +Date: Sun, 18 Jan 2004 12:20:26 -0500 +From: root +To: Dacian.Tudor@Comneon.com +Subject: [Axiom-developer] Re: Axiom help needed + +Dacian, + +I know that Axiom has OpenMath support. NAG, especially Mike Dewar, +was deeply involved in this effort. I'm unfamiliar with the details +of OpenMath so I don't how much help I can be. + +The best place to get help is to send email to axiom-developer@nongnu.org +This is the Axiom mailing list. Some of the developers know more about it +than I do. + +Do you have CVS? You can get the code from the website by typing: + +cvs -d:pserver:anoncvs@subversions.gnu.org:/projects/axiom login +cvs -d:pserver:anoncvs@subversions.gnu.org:/projects/axiom co axiom + +The first command will log you in as an anonymous user. It will +prompt you for a password. Just hit enter. + +The second command will create a new directory called axiom in +your current directory. If you current directory is + +/home/dacian + +you would type: + +cd axiom +export AXIOM=/home/dacian/axiom/mnt/linux +export PATH=$AXIOM/bin:$PATH +make + +Let me know if you have any problems. + +================================================================== + +Dear Sir, + +I am doing some project research that is involving AXIOM and I am writing to +ask for your help in this matter. + +One of the goals of the project is to make use of AXIOM and provide an +interface to it based on OpenMath. I'd like to know if there is built in +support for OpenMath in AXIOM 2.3 and if you could help me getting AXIOM +application. I noticed that the internet download section is still down for +several weeks and I have no possibility to get the code from somewhere else. + +It is very important for me that AXIOM should have OpenMath support. It +would be nice if AXIOM could accept OpenMath input and return OpenMath +output as well. Please let me know if there is such release and if not what +other solutions there are for this problem. + +I am looking forward to receiving your answer. + +\start +Date: Sun, 18 Jan 2004 19:18:08 +0100 +From: David MENTRE +To: daly@idsi.net +Subject: Re: [Axiom-developer] Re: Axiom help needed +Cc: Dacian.Tudor@Comneon.com + +Hello Tim and Dacian, + +root writes: + +> Do you have CVS? You can get the code from the website by typing: +> +> cvs -d:pserver:anoncvs@subversions.gnu.org:/projects/axiom login +> cvs -d:pserver:anoncvs@subversions.gnu.org:/projects/axiom co axiom +> +> The first command will log you in as an anonymous user. It will +> prompt you for a password. Just hit enter. +> +> The second command will create a new directory called axiom in +> your current directory. If you current directory is + +The procedure to download the Axiom sources from CVS has changed after +savannah crack. Dacian, you should do: + +export CVS_RSH="ssh" ## depending on your shell, in this case for bash +cvs -z3 -d:ext:anoncvs@savannah.nongnu.org:/cvsroot/axiom co axiom + +The SSHv2 public key fingerprints for the machine hosting the cvs trees are: + +RSA: 1024 80:5a:b0:0c:ec:93:66:29:49:7e:04:2b:fd:ba:2c:d5 +DSA: 1024 4d:c8:dc:9a:99:96:ae:cc:ce:d3:2b:b0:a3:a4:95:a5 + + +Of course, you should have ssh available on your machine. + +The remaining procedure of Tim to compile Axiom is correct. + +Let us know of any issues. + + + +Regarding OpenMath, I'm afraid but I think all NAG work to support +OpenMath is no longer available in current free Axiom. However, Bill +Page, Camm Maguire (lead developer of Gnu Common Lisp) and of course +Mike Dewar where also interested by OpenMath and they might help +you. They are on this list. + + +\start +Date: 19 Jan 2004 10:04:26 -0500 +From: Camm Maguire +To: David MENTRE +Subject: Re: [Axiom-developer] Re: Axiom help needed +Cc: Dacian.Tudor@Comneon.com + +Greetings! + +David MENTRE writes: + +> Regarding OpenMath, I'm afraid but I think all NAG work to support +> OpenMath is no longer available in current free Axiom. However, Bill +> Page, Camm Maguire (lead developer of Gnu Common Lisp) and of course +> Mike Dewar where also interested by OpenMath and they might help +> you. They are on this list. +> + +If this is a priority item, I might find time later this week/weekend +to write a GCL interface to the OpenMath library. + +\start +Date: 19 Jan 2004 09:59:13 -0500 +From: Camm Maguire +To: David MENTRE +Subject: Re: [Axiom-developer] [crystal] A proposed framework and associated tool: LIS (Logical Information Systems) + +Greetings! This looks very interesting! Just thought I'd also +mention another related though less sophisticated approach from the C +world -- cxref. Basically, it is a hook into the compiler which +captures the variable, function, call tree, and comment structure of +the source and outputs to latex/html/.... Whether the output is in +this form or in a browseable filesystem (very neat), it might be +logical for the lisp compiler to parse and cross reference this +information from the input source, rather like GCL already does to +some extent with the .fn files. Time permitting, I'd be interested in +extending GCL to process source files in this manner. + +\start +Date: Mon, 19 Jan 2004 16:39:10 +0100 +From: Dacian.Tudor@Comneon.com +To: camm@enhanced.com, david.mentre@wanadoo.fr +Subject: RE: [Axiom-developer] Re: Axiom help needed + +Hi all, + +I need a solution to have AXIOM running with OpenMath. Can you tell me if +such a solution is available at the moment and if not what would be the +steps I should follow to have it. I read that there is a AXIOM 2.3 delivery +with OpenMath support but I don't know if it is true and where I can get it +from. + +-----Original Message----- +From: Camm Maguire [mailto:camm@enhanced.com] +Sent: Montag, 19. Januar 2004 16:04 +To: David MENTRE +Cc: daly@idsi.net; Tudor Dacian (CN L External); axiom-developer@nongnu.org +Subject: Re: [Axiom-developer] Re: Axiom help needed + + +Greetings! + +David MENTRE writes: + +> Regarding OpenMath, I'm afraid but I think all NAG work to support +> OpenMath is no longer available in current free Axiom. However, Bill +> Page, Camm Maguire (lead developer of Gnu Common Lisp) and of course +> Mike Dewar where also interested by OpenMath and they might help you. +> They are on this list. +> + +If this is a priority item, I might find time later this week/weekend to +write a GCL interface to the OpenMath library. + +\start +Date: Mon, 19 Jan 2004 17:03:32 +0100 +From: Dacian.Tudor@Comneon.com +To: camm@enhanced.com +Subject: RE: [Axiom-developer] Re: Axiom help needed + +Actually this would not be the focus of the project and I believe that you +will do a better job than me. From what you are telling, you will implement +OpenMath support for AXIOM (the same version available from cvs I hope). +When do you thing it will be ready ? + +Will there be a built in support for OpenMath ? '(e.g input - a file +containing OpenMath and output the result also in OpenMah). + +Regards, +Dacian + +-----Original Message----- +From: Camm Maguire [mailto:camm@enhanced.com] +Sent: Montag, 19. Januar 2004 16:56 +To: Tudor Dacian (CN L External) +Cc: david.mentre@wanadoo.fr; daly@idsi.net; axiom-developer@nongnu.org +Subject: Re: [Axiom-developer] Re: Axiom help needed + + +Greetings! I don't know about AXIOM 2.3 -- is this the last commercial +release? The open source version available via cvs does not yet have the +openmath bindings to my understanding. These do not appear difficult to +implement. If you are in a hurry and want to try your hand before I get to +it, I can describe briefly how it should be done. + +Take care, + +Dacian.Tudor@Comneon.com writes: + +> Hi all, +> +> I need a solution to have AXIOM running with OpenMath. Can you tell me +> if such a solution is available at the moment and if not what would be +> the steps I should follow to have it. I read that there is a AXIOM 2.3 +> delivery with OpenMath support but I don't know if it is true and +> where I can get it from. +> +> Best wishes, +> Dacian +> +> -----Original Message----- +> From: Camm Maguire [mailto:camm@enhanced.com] +> Sent: Montag, 19. Januar 2004 16:04 +> To: David MENTRE +> Cc: daly@idsi.net; Tudor Dacian (CN L External); +axiom-developer@nongnu.org +> Subject: Re: [Axiom-developer] Re: Axiom help needed +> +> +> Greetings! +> +> David MENTRE writes: +> +> > Regarding OpenMath, I'm afraid but I think all NAG work to support +> > OpenMath is no longer available in current free Axiom. However, Bill +> > Page, Camm Maguire (lead developer of Gnu Common Lisp) and of course +> > Mike Dewar where also interested by OpenMath and they might help you. +> > They are on this list. +> > +> +> If this is a priority item, I might find time later this week/weekend +> to write a GCL interface to the OpenMath library. + +\start +Date: Mon, 19 Jan 2004 12:18:45 -0500 +From: root +To: Dacian.Tudor@Comneon.com +Subject: Re: [Axiom-developer] Re: Axiom help needed +Cc: camm@enhanced.com + +Dacian, Camm, + +I believe the zips directory in the axiom CVS contains a file called +OMCv1.4a.tgz which is the library used to support Axiom's Openmath +implementation. I'll look at the details of this library and what +it requires. + +\start +Date: 19 Jan 2004 10:56:15 -0500 +From: Camm Maguire +To: Dacian.Tudor@Comneon.com +Subject: Re: [Axiom-developer] Re: Axiom help needed + +Greetings! I don't know about AXIOM 2.3 -- is this the last +commercial release? The open source version available via cvs does +not yet have the openmath bindings to my understanding. These do not +appear difficult to implement. If you are in a hurry and want to try +your hand before I get to it, I can describe briefly how it should be +done. + +Take care, + +Dacian.Tudor@Comneon.com writes: + +> Hi all, +> +> I need a solution to have AXIOM running with OpenMath. Can you tell me if +> such a solution is available at the moment and if not what would be the +> steps I should follow to have it. I read that there is a AXIOM 2.3 delivery +> with OpenMath support but I don't know if it is true and where I can get it +> from. +> +> Best wishes, +> Dacian +> +> -----Original Message----- +> From: Camm Maguire [mailto:camm@enhanced.com] +> Sent: Montag, 19. Januar 2004 16:04 +> To: David MENTRE +> Cc: daly@idsi.net; Tudor Dacian (CN L External); axiom-developer@nongnu.org +> Subject: Re: [Axiom-developer] Re: Axiom help needed +> +> +> Greetings! +> +> David MENTRE writes: +> +> > Regarding OpenMath, I'm afraid but I think all NAG work to support +> > OpenMath is no longer available in current free Axiom. However, Bill +> > Page, Camm Maguire (lead developer of Gnu Common Lisp) and of course +> > Mike Dewar where also interested by OpenMath and they might help you. +> > They are on this list. +> > +> +> If this is a priority item, I might find time later this week/weekend to +> write a GCL interface to the OpenMath library. + +\start +Date: Mon, 19 Jan 2004 12:32:46 -0500 +From: root +To: Renaud.Rioboo@lip6.fr +Subject: [Axiom-developer] Re: reclos.spad + +Renaud, + +Your algebra fixes for reclos.spad have (finally) been applied. +I've also updated the input file with the test case you supplied. + +There was a considerable delay as I have not been able to update +the Axiom CVS until this morning due to a number of errors (mostly +mine). + +Is there any chance of getting copies of the two papers you reference? +(Real Algebraic Closure of an ordered Field: Implementation in Axiom +and Generic computation of a real closure of an ordered field) +I'd really like to get a literate program version of the reclos domain. + +\start +Date: Mon, 19 Jan 2004 12:34:09 -0500 +From: root +To: fauser@spock.physik.uni-konstanz.de +Subject: [Axiom-developer] Re: reclos.spad + +Bertfried, + +Have you made any progress on the Clifford or Hopf algebra work? +Is there something I can do to help? + +\start +Date: Mon, 19 Jan 2004 14:46:30 -0500 +From: root +To: lecerf@math.uvsq.fr +Subject: [Axiom-developer] Dynamic Real Closure and Axiom +Cc: J.H.Davenport@bath.ac.uk + +Gregoire, + +I've come across your work on dynamic real closures. +Your paper is very close to a literate program which is a method +of combining research results with executable code. + +Axiom has been released as free and open source software. +I'm the lead developer on the project. +(http://savannah.nongnu.org/projects/axiom) + +I'd like your permission (and assistance if you can) in converting +your paper to working code within Axiom. In particular this would +involve modifying the TeX document into a "pamphlet" file which is +Axiom's literate programming format (It is basically TeX with two +additional tags that allow program source to be extracted). + +Is it possible to get a TeX version of your paper and permission +to modify it to include it as part of Axiom? + +\start +Date: Mon, 19 Jan 2004 23:02:22 +0100 (CET) +From: Bertfried Fauser +To: root +Subject: [Axiom-developer] Crystal -> doxygen + +Dear All, + + since there was a discussion on helper tools, my brother (he _is_ +programmer) recommended the doxygen package. Perhaps its worth a look, +even if it (obviously) does not (yet) supports AXIOM. + +\start +Date: Mon, 19 Jan 2004 15:53:01 -0500 +From: root +To: caruso@posso.dm.unipi.it +Subject: [Axiom-developer] symbolic summation beta + +Fabrizio, + +>My previous work on Symbolic summation was related +>to hypergeometric symbolic summation and Zeilberger's algorithm, +>which I have implemented in Ma(cs/x)yma. + +Can you send me a copy of the Maxima code for this? + +> +>One of the people working on the book is +>Prof. Laureano Gonzalez-Vega (Santander, Spain). +>I will ask him about the status and let you know as soon as possible. + +Have you been able to contact Laureano? + +> +>I also mentioned the work of Carsten Schneider +>(Risc-Linz), who has improved and implemented Karr's algorithm +>in his Mathematica package (Sigma), which should now be available for +>download at the RISC site after writing an email to him or his boss (P.Paule). +>His improved and generalized version of Karr's algorithm is +>the most sophisticated summation tool. It resembles Risk's algorithm +>for symbolic integration. + +I have copies of his papers but no code. +It is interesting work and we need to make a start on symbolic +summation in Axiom. + +\start +Date: Mon, 19 Jan 2004 20:36:14 +0100 +From: David MENTRE +To: Bertfried.Fauser@uni-konstanz.de +Subject: Re: [Axiom-developer] Crystal -> doxygen + +Hello Bertfried, + +Bertfried Fauser writes: + +> since there was a discussion on helper tools, my brother (he _is_ +> programmer) recommended the doxygen package. Perhaps its worth a look, +> even if it (obviously) does not (yet) supports AXIOM. + +While I've not used Doxygen extensively, I know at least its +capabilities. Doxygen is good to document C++ code, draw class diagram +and add documentation to pieces of code, however it is very very far +from Tim's ideas of multiple exploration axes for the Crystal. That's +why tim's ideas are so interesting. :) + +Yours, +d. + +PS : For reference, the doxygen web site: + http://www.stack.nl/~dimitri/doxygen/ + +\start +Date: Mon, 19 Jan 2004 16:08:53 -0500 +From: root +To: wyscc@cunyvm.cuny.edu +Subject: [Axiom-developer] pleqn.spad + +Bill, + +I've updated the pleqn.spad with the latest version you've given me. +It cleanly compiles and has been uploaded into the distribution. +I'm going to look at the paper you sent me to see what needs to be +done to form a literate program. Once I get the basic dog-work done +I'll need you to review it for content. + +\start +Date: Mon, 19 Jan 2004 17:32:59 -0500 +From: root +To: gilbert@sci.ccny.cuny.edu, cfm@groups.sci.ccny.cuny.edu +Subject: [Axiom-developer] group theory classification + +Gilbert, Chuck, + +I've been looking at the classification scheme of finitely presented +and finitely generated groups so I can implement the proper category +hierarchy in Axiom. I'm now looking for references that will give me +the axioms which define each group. Any help would be appreciated. +I'd like to state the axioms that are added at each point in the +lattice. + +The "finitely presented simple group (+WP)" is hanging out unclassified. + +Where do nilpotent groups of order 2 fit? + +There are a list of groups that need classification. From a discussion +with Gilbert I find a very bushy tree of the form: + +layer 1 + FPG finitely presented group + +layer 2 + FN free nilpotent + HNN HNN group + OR one relator + AUTO automatic + AMAL amalgamated + SC small cancellation + F free + +layer 3 + HYPER hyperbolic + NIL nilpotent + +layer 4 + ABEL abelian + +4 ABEL + | + | +3 NIL HYPER + | | + | | +2 FN HNN OR AUTO AMAL SC F + | | | | | | | + | | | | | | | + -------------------------------------------------- + | + | +1 FPG + + + +Among my notes I found the attached diagram: + +layer 1 + FGA finitely generated abelian + (+WP, +CP, +GWP, +IsoP) + FPRF finitely presented residually free + (+WP, ?CP, -GWP, ?IsoP) + +layer 2 + FPM finitely presented metabelian + (+WP, +CP, +GWP, ?IsoP) + FGN finitely generated nilpotent + (+WP, +CP, +GWP, +IsoP) + +layer 3 + FPSDL3 finitely presented solvable derived length 3 + (-WP, -CP, -GWP, -IsoP) + FGM finitely generated metabelian + (+WP, +CP, +GWP, ?IsoP) + FPRN finitely presented residually nilpotent + (+WP, -CP, -GWP, -IsoP) + P polycyclic + (+WP, +CP, +GWP, +IsoP) + A arithmetic + (+WP, +CP, -GWP, ?IsoP) + +layer 4 + FGABN finitely generated abelian-by-nilpotent + (+WP, ?CP, +GWP, ?IsoP) + SA S-arithmetic + (+WP, +CP, -GWP, ?IsoP) + FPS finitely presented subgroups + (+WP, ?CP, -GWP, ?IsoP) + +layer 5 + FGABP finitely generated abelian-by-polycyclic + (+WP, ?CP, ?GWP, ?IsoP) + FGSGL finitely generated subgroups of GL(n,Z) + (+WP, -CP, -GWP, -IsoP) + FPRF finitely presented residually finite + (+WP, -CP, -GWP, -IsoP) + +layer 6 + FGL finitely generated linear + (+WP, -CP, -GWP, -IsoP) + FPH finitely generated hopfian + (-WP, -CP, -GWP, -IsoP) + + + + +6 FGL FPH + | | + | | +5 FGABP | FGS FPRF + | | | | + | | | | +4 FGABN | SA FPS | + | | | | | | + | | | | | | + ----------- ------- | | + | | | | + | | | | +3 FPDSL3 FGM A P FPRN + | | | | | | + | | | | | | + ---------- -------------- | + | | | + | | | +2 FPM FGN | + | | | + ------------------ | + | | + | | +1 FGA FPRF + +\start +Date: Mon, 19 Jan 2004 17:43:05 -0500 +From: root +To: kvoa@netlane.com +Subject: [Axiom-developer] Axiom for Windows + +Sam, + +"Building Axiom" requires no skill (that's why I could do it :-) ). +Essentially you download the sources, type make, and wait for a +really long time. At the end of the make you should have a running Axiom. +I don't have a SuSE binary available as I don't have a running SuSE. +My compile farm is currently hosed and awaiting my time and attention. + +I'll look at your review website as soon as I get a chance. + +\start +Date: Mon, 19 Jan 2004 15:12:54 -0500 +From: Tim Daly +To: research@ijs.co.nz +Subject: [Axiom-developer] Re: Scintilla GUI + +Craig, + +> +>I produced a GUI for my new 'Tope' symbolic algebra solver that solves +>linear logic problems, e.g (Exists x)(a (aI use the Scintilla DLL which provides the cursor and text features +>that their text editor needs. I used the GWindows Ada 95 bindings to +>Scintilla. That binding would be eliminated since GWindows won't +>run in Linux. It is a thin binding and not much is used. +> +>http://www.scintilla.org/ +>http://www.adapower.com/ +> +>A Windows .EXE file "Tope" implementing my Scintilla GUI: +>http://www.ijs.co.nz/code/ada95_symbolic_algebra_polytope_logic.zip +> +>(If wishing to see how to compile it then get this: +>http://www.ijs.co.nz/code/ada95_adagoop_parser.zip ) +> +>The code is for n-D polytopes so there may never be a time when I +>need more than a GUI-ized line by line interface. +> +>My Tope's GUI is a bit like this: http://www.jsoftware.com/ +>(pressing ENTER on text would copy it to the bottom, and Scintilla +>provides read-only text marking the ends.) +> +>I suppose that this message allows a much better estimate on how much +>work is involved if Scintilla is used. A main problem was that I +>didn't see how to intercept ENTER, Ctrl-UpArrow, LeftArrow, before +>Scintilla found out about them. + +I've never heard of Scintilla but I'll look at your references. + +> +>Sometimes people show up and recommend Java and Tcl/Tk. I am sure +>that Scintilla would be a better choice. +> +>I am not subscribed to your mailing lists. I partly write to +>discourage use of Tcl/Tk and Java(TM) and other inferior options. + +We are looking at Tcl/TK at the moment as we have a user interface +from the Magnus project available (it was done in Tcl/TK). I personally +don't care for it but it has the advantage of being running code. + +...[snip]... +> +>Craig Carey +>Auckland, New Zealand + +\start +Date: Mon, 19 Jan 2004 23:21:39 -0500 +From: dpt@exoskeleton.math.harvard.edu (Dylan Thurston) +To: axiom-developer@nongnu.org +Subject: Re: [Axiom-developer] group theory classification + +On Mon, Jan 19, 2004 at 05:32:59PM -0500, root wrote: +> Where do nilpotent groups of order 2 fit? + +What is a nilpotent group of order 2? Do you mean the length of the +lower central series is 2? Why make a special case for order 2, and not +order N? + +> layer 2 +> FN free nilpotent +> HNN HNN group +> OR one relator=20 +> AUTO automatic +> AMAL amalgamated +> SC small cancellation +> F free + +What's a free nilpotent group? It sounds more special than a nilpotent +group, so I'm confused by your hierarchy. + +This seems like a list of properties of groups, rather than +constructions of groups, so 'free' doesn't seem to quite belong. But +maybe it does, since subgroups of free groups are automatically free. +But then why don't you have free abelian on this list? + +Amalgamated and HNN groups are both best understood as special cases of +graphs of groups. It would be nice to do the general case and unify +these two. + +Aren't small cancellation groups necessarily automatic? + +By the way, I object to putting finitely presented groups at the base of +the hierarchy (if I understand the diagram correctly). There +interesting groups which are not finitely presented, which you can still +work with in practice. Maybe it has a countable set of relations, or +maybe it's a subgroup of a finitely presented group, or maybe it's +infinitely generated. + +Two infinitely generated groups that I know well is the infinite +derangement group (with only a finite number of objects displaced) and +the infinite braid group (likewise). Both have solvable word problem +and (I think) conjugacy problem. (The word problem, at least, is +quadratic time for the braid group.) + +Several of the properties you list (like automatic) do specifically +refer to finitely presented groups, but others (like nilpotent) do not. + +I think I must be misunderstanding something. Please help clear up my +confusion! + +> Among my notes I found the attached diagram: + +I didn't understand the diagram. What do the lines mean? It doesn't +seem to be subtyping like in your earlier diagram. + +> layer 1 +> FGA finitely generated abelian=20 +>=20 +> (+WP, +CP, +GWP, +IsoP) + +What is GWP? The others must relate to solvability of + +WP =3D word problem +CP =3D conjugacy problem +IsoP =3D isomorphism problem + +\start +Date: Tue, 20 Jan 2004 01:00:48 -0500 +From: root +To: dpt@math.harvard.edu +Subject: Re: [Axiom-developer] group theory classification + +Dylan, + +>On Mon, Jan 19, 2004 at 05:32:59PM -0500, root wrote: +>> Where do nilpotent groups of order 2 fit? +> +>What is a nilpotent group of order 2? Do you mean the length of the +>lower central series is 2? Why make a special case for order 2, and not +>order N? +> +>> layer 2 +>> FN free nilpotent +>> HNN HNN group +>> OR one relator=20 +>> AUTO automatic +>> AMAL amalgamated +>> SC small cancellation +>> F free +> +>What's a free nilpotent group? It sounds more special than a nilpotent +>group, so I'm confused by your hierarchy. + +free nilpotent groups of order 2 are special groups that we study +locally. I expect that you'll find the hierarchy confusing as I'm +confused by it myself. I'm trying to sort out the various infinite +group classifications into some sort of a category hierarchy so I +can properly encode them in Axiom. It appears that this hasn't been +done before. + +Ideally I'd like some sort of a Venn diagram showing the containment +and overlap of the groups, their axioms and their properties. + +(In fact, I once saw such a detailed Venn diagram for groups, rings, +fields, etc in a book but have not been able to lay hands on +it again. If you happen to see it please give me the reference.) + +>This seems like a list of properties of groups, rather than +>constructions of groups, so 'free' doesn't seem to quite belong. But +>maybe it does, since subgroups of free groups are automatically free. +>But then why don't you have free abelian on this list? + +I've been struggling with the properties vs axioms of the groups. +I'm trolling the literature to construct the tree of axioms associated +with the various groups. + +>Amalgamated and HNN groups are both best understood as special cases of +>graphs of groups. It would be nice to do the general case and unify +>these two. + +Do you have a reference I could look at? I'm unfamiliar with the +more general case but I'm willing to learn. + +>Aren't small cancellation groups necessarily automatic? + +I don't know yet. I haven't seen a theorem to that effect. + +>By the way, I object to putting finitely presented groups at the base of +>the hierarchy (if I understand the diagram correctly). There +>interesting groups which are not finitely presented, which you can still +>work with in practice. Maybe it has a countable set of relations, or +>maybe it's a subgroup of a finitely presented group, or maybe it's +>infinitely generated. + +Magnus (my other computer algebra project) specializes in finitely +presented groups. Thus finitely presented groups represent a base +assumption for the kinds of groups I want to classify. I want to +enable Axiom to perform the computations currently done by Magnus. +To do that properly I need to figure out a category hierarchy. + +>Two infinitely generated groups that I know well is the infinite +>derangement group (with only a finite number of objects displaced) and +>the infinite braid group (likewise). Both have solvable word problem +>and (I think) conjugacy problem. (The word problem, at least, is +>quadratic time for the braid group.) + +I don't have any machinery to handle infinitely generated groups. +Once I get the category classification straight I need to organize +the algorithms in Magnus (and Sim's book "Computation with Finitely +Presented Groups" Cambridge Univ. Press 1994) into the hierarchy. +I don't have any algorithms on infinitely generated groups. + +Ah, braid groups. I missed that one. In fact, there is another +hierarchy on my list of semi-random lattice diagrams namely one in +topology (where braid groups are used in knots). + +>Several of the properties you list (like automatic) do specifically +>refer to finitely presented groups, but others (like nilpotent) do not. +> +>I think I must be misunderstanding something. Please help clear up my +>confusion! + +No, the misunderstanding is mine. As I said, this apparently hasn't +been done before so I'm trying to derive the category hierarchy. +What appears to be confusion on your part is a lack of understanding +on my part. + +> +>> Among my notes I found the attached diagram: + +>I didn't understand the diagram. What do the lines mean? It doesn't +>seem to be subtyping like in your earlier diagram. + +Indeed the two diagrams are describing two different kinds of +classification. The second diagram looks at what kinds of problems +are known to be solvable (+), unsolvable (-), or unknown (?). + +> +>> layer 1 +>> FGA finitely generated abelian +>> +>> (+WP, +CP, +GWP, +IsoP) +> +>What is GWP? The others must relate to solvability of + +GWP is the generalized word problem. + +> +>WP word problem +>CP conjugacy problem +>IsoP isomorphism problem +> +>Peace, +> Dylan + +Sorry for the confusion. + +\start +Date: Tue, 20 Jan 2004 01:54:43 -0500 +From: root +To: dpt@math.harvard.edu +Subject: Re: [Axiom-developer] group theory classification + +Dylan, + +If you can find a copy of +Steen, Lynn Arthur and Seebach, J. Arthur "Counterexamples in Topology" +Springer-Verlag (1978) or Dover (1995) see page 16. The diagram given +there for topology is essentially what I'm trying to achieve for the +group theory work. (It is an astonishing diagram and I don't hope to +achieve nearly so complete an understanding.) I'd also like a similar +diagram describing rings, unique factorization domains, etc for algebra. +This kind of exposition would put our understanding of the fundamentals +of Axiom on a solid footing. + +I need to achieve this kind of understanding (or some shadow of it) in +order to properly add categories to Axiom and to construct the crystal. + +\start +Date: Tue, 20 Jan 2004 08:14:14 -0500 +From: root +To: C.Miller@ms.unimelb.edu.au +Subject: [Axiom-developer] group theory ordering +Cc: gilbert@sci.ccny.cuny.edu + +Chuck, + +Thanks for the notes. I realize that there are several different ways +to order the lattice of group based on different metrics. I have an +example from topology (which I hope to keyboard later today). It comes +from an excellent book called "Counterexamples in Topology" by Steen +and Seebach (Springer-Verlag, 1978; Dover 1995) on page 16. They +order the groups based on axioms. They also have order diagrams +based on properties such as connectedness and compactness which are +entirely different. I'm trying to derive similar information for our +area and it appears nobody has done so yet. As Dylan Thurston pointed +out I still have several points of confusion. + +One point of confusion appears to be that some terms are defined as +as axioms and some are defined as properties. Sometimes I see writers +use abelian as an axiom and sometimes I see it used as a property. +I'm beginning to see that Axiom has this confusion also. + +I'm looking to attack the problem of ordering these things for 3 reasons: + +1) I need to understand (and classify) these groups in some systematic ways + so I can get a better handle on the pile of results and algorithms. +2) I need to understand (and classify) these groups so I can figure out + a category structure that Axiom can use to construct these groups and + order the algorithms. +3) It seems like a good domain to build this beast I'm calling a "crystal". + Picture a huge ball of string in space with many knots (the + ball being a graph of the many relationships and the knots being + clusters of axioms and properties related to a concept. Now "wrap" the + knotted string with a "crystal" with many facets. Each facet extracts + a different structural relationship of the concepts in the ball. This + whole thing (viewer and network; aka facets and string) I call crystal. + Some facets show the math, some the lattice, some the underlying code. + +Virtually every area Axiom has touched has the same problem (though less +intense) as Infinite Group Theory. Everything is ordered by everything +else. How you want to order it depends on how you want to think about it. +I want to look at Infinite Groups ordered in many different ways. + +So crystal is an attempt to let you string things together in all of the +ways you want to think about them (the ball) and view them in all of the +ways you want to see them (the facets). Hard problems within crystal are +things like automatically classifying concepts (by using a metric like +subsumption) so (a) they fit in the "appropriate" place and (b) you can +find them again. You want it to be automatic because you want the machine +to do the work. Axiom already automatically extracts a great deal of +information when it compiles things and even more when you contruct +type towers of domains. At the moment this information is exported as +"databases". The idea of a "database" is too limiting to order the kind +of math we need to play with. Semantic networks are closer but, like the +ball itself, they aren't the only way to build the structure. + +Of course, none of this is magic which means that I have to understand, +extract, categorize, code, compile, classify, and view the concepts. +I've done a complete reduction of Axiom's categories and domains and +I'm looking at the compiler output to see what I can automatically +extract. + +I'll look at the information on your website. + +\start +Date: Fri, 23 Jan 2004 02:35:34 -0500 +From: William Sit +To: Tim Daly +Subject: [Axiom-developer] self-repair Axiom? + +Dear Tim: + +Since you have this 30 year horizon for Axiom, would it make sense to +incorporate the idea of self-repair? See article below: + +http://www.trnmag.com/Stories/2004/011404/Software_repairs_itself_on_the_go_011404.html + +This may be combined with the test-suite, but is much more general. I am not +sure how the robustness of a self-repair system relates to the (mathematical) +correctness of computation. Anyone familiar with the MIT project? + +\start +Date: Fri, 23 Jan 2004 04:04:21 -0500 +From: root +To: wyscc@cunyvm.cuny.edu +Subject: [Axiom-developer] Re: self-repair Axiom? + +self-repair is hot news these days. IBM calls it "autonomic computing" +because they make an analogy to the human autonomic nervous system which +adjusts your body functions (e.g. you sweat when it is too hot). The +system tries to return to a known state when disturbed. For example, +your windows box will tend to a blue screen (the normal state) :-) + +I'll look at your reference later this morning. + +\start +Date: Sun, 25 Jan 2004 11:53:05 -0500 +From: "Bill Page" +To: "'David MENTRE'" +Subject: RE: [Axiom-developer] Re: Axiom help needed + +Tim and David; + +What is required of us in order to restore the contents +of the savannah files section for AXIOM? + + http://savannah.nongnu.org/download/axiom/ + +\start +Date: Sun, 25 Jan 2004 18:16:48 +0100 +From: David MENTRE +To: "Bill Page" +Subject: Re: [Axiom-developer] Re: Axiom help needed + +Hello Bill, + +"Bill Page" writes: + +> What is required of us in order to restore the contents +> of the savannah files section for AXIOM? +> +> http://savannah.nongnu.org/download/axiom/ + +To follow a fairly complicated upload procedure: + http://savannah.nongnu.org/forum/forum.php?forum_id=2838 + +Apparently, we need to sign packages before uploading them with +FTP. Seems complicated for me. I wonder if we can't use another FTP/HTTP +server for download area? + +\start +Date: Sun, 25 Jan 2004 14:04:21 -0500 +From: root +To: bill.page1@sympatico.ca +Subject: Re: [Axiom-developer] Re: Axiom help needed + +Bill, + +I don't know. I'll send a note to Jaime and ask. + +\start +Date: Sun, 25 Jan 2004 14:49:46 -0500 +From: root +To: villate@debian.org +Subject: [Axiom-developer] upload/download area on savannah + +Jaime, + +What is the magic necessary to restore the upload/download files +area in Axiom? Is there a FAQ or doc item someplace? + +\start +Date: Mon, 26 Jan 2004 01:41:08 -0500 +From: root +To: Renaud.Rioboo@lip6.fr +Subject: [Axiom-developer] Re: Axiom Installation + +Renaud, + + +>I finally resigned installing Axiom at my office machines because I have no +>root priviledges and this is a mess. Among the problems I had, I remember +>noweb installation which requires root proviledges to run texhash in the +>install part of noweb. You could change the line +> +> -texhash || echo "Program texhash not found or failed" +> +>in the obj/noweb/src/Makefile to +> +> -texhash $(TEXINPUTS) || echo "Program texhash not found or failed" +> +>in order to only change the directory which is under axiom's control. +> +>I thus decided to install my new laptop which has some disk space and, as of +>now, I have no major problem. I run RedHat 9 on my laptop. + +I'll have to look at what texhash does. I know there is another lurking +issue (how to put the axiom.sty file someplace where tex can find it) +that needs to be addressed and I believe that texhash will solve the problem. +I'll look into this and get back to you. + +>Just one point, I took the habit (for readabilty) to name directories with +>uppercases first letters designating words as in Axiom, this yield to some +>problems when compiling axiom. I had to make symbolic links in order to have +>things work. + +The case sensitive issue was addressed by Juergen Weiss a few months ago. +I've been remiss in adding it for no obvious reason. Anyway it's been +applied, tested, and added to the CVS (thanks, Juergen). . So now Axiom's +pathnames should not be case sensitive. + +Is it possible to get an electronic version of your reclos paper? +I'd like to read it. + +\start +Date: Mon, 26 Jan 2004 01:55:31 -0500 +From: root +To: axiom-developer@nongnu.org +Subject: [Axiom-developer] bitkeeper + +Has anyone used bitkeeper? I've been trolling the linux kernel +mailing list and it seems that a fair number of the developers +have moved off CVS and onto bitkeeper (www.bitkeeper.com). + +\start +Date: 26 Jan 2004 14:09:03 +0100 +From: Peter Simons +To: axiom-developer@nongnu.org +Subject: [Axiom-developer] Building error + +I've tried to build the latest Axiom version from CVS, but +got an error during the build. I'll attach the complete log +output to this e-mail. + +Does anyone know what's causing this? And, more importantly, +how to fix it? + +Peter + + +0 SPAD=/usr/local/src/axiom-current/mnt/linux SYS=linux SPD=/usr/local/src/axiom-current LSP=/usr/local/src/axiom-current/lsp GCLDIR=/usr/local/src/axiom-current/lsp/gcl-2.6.1 SRC=/usr/local/src/axiom-current/src INT=/usr/local/src/axiom-current/int OBJ=/usr/local/src/axiom-current/obj MNT=/usr/local/src/axiom-current/mnt ZIPS=/usr/local/src/axiom-current/zips TMP=/usr/local/src/axiom-current/obj/tmp SPADBIN=/usr/local/src/axiom-current/mnt/linux/bin INC=/usr/local/src/axiom-current/src/include CCLBASE=/usr/local/src/axiom-current/obj/linux/ccl/ccllisp PART=cprogs SUBPART=everything NOISE=-o /usr/local/src/axiom-current/obj/tmp/trace GCLVERSION=gcl-2.6.1 TANGLE=/usr/local/src/axiom-current/mnt/linux/bin/lib/notangle +10 copying /usr/local/src/axiom-current/src/scripts to /usr/local/src/axiom-current/mnt/linux/bin +1 making a linux system, PART=cprogs SUBPART=everything +2 Environment SPAD=/usr/local/src/axiom-current/mnt/linux SYS=linux SPD=/usr/local/src/axiom-current LSP=/usr/local/src/axiom-current/lsp GCLDIR=/usr/local/src/axiom-current/lsp/gcl-2.6.1 SRC=/usr/local/src/axiom-current/src INT=/usr/local/src/axiom-current/int OBJ=/usr/local/src/axiom-current/obj MNT=/usr/local/src/axiom-current/mnt ZIPS=/usr/local/src/axiom-current/zips TMP=/usr/local/src/axiom-current/obj/tmp SPADBIN=/usr/local/src/axiom-current/mnt/linux/bin INC=/usr/local/src/axiom-current/src/include CCLBASE=/usr/local/src/axiom-current/obj/linux/ccl/ccllisp PART=cprogs SUBPART=everything NOISE=-o /usr/local/src/axiom-current/obj/tmp/trace GCLVERSION=gcl-2.6.1 TANGLE=/usr/local/src/axiom-current/mnt/linux/bin/lib/notangle +make[1]: Entering directory `/usr/local/src/axiom-current' +11 checking directory structure +12 Environment: PLF=LINUXplatform CCF=-O2 -fno-strength-reduce -Wall -D_GNU_SOURCE -DLINUXplatform -I/usr/X11/include LDF=-L/usr/X11R6/lib CC=gcc AWK=gawk RANLIB=ranlib TOUCH=touch TAR=tar AXIOMXLROOT=/usr/local/src/axiom-current/mnt/linux/compiler O=o BYE=bye LISP=lsp DAASE=/usr/local/src/axiom-current/src/share XLIB=/usr/X11R6/lib +18 making /usr/local/src/axiom-current/src +make[2]: Entering directory `/usr/local/src/axiom-current/src' +1 making /usr/local/src/axiom-current/src/scripts +make[3]: Entering directory `/usr/local/src/axiom-current/src/scripts' +1 making /usr/local/src/axiom-current/src/scripts +make[3]: Leaving directory `/usr/local/src/axiom-current/src/scripts' +17 making /usr/local/src/axiom-current/src/lib +make[3]: Entering directory `/usr/local/src/axiom-current/src/lib' +72 finished making /usr/local/src/axiom-current/obj/linux/lib/libspad.a /usr/local/src/axiom-current/obj/linux/lib/cfuns-c.o /usr/local/src/axiom-current/obj/linux/lib/hash.o +make[3]: Leaving directory `/usr/local/src/axiom-current/src/lib' +make[2]: Leaving directory `/usr/local/src/axiom-current/src' +0 PLF=LINUXplatform CCF=-O2 -fno-strength-reduce -Wall -D_GNU_SOURCE -DLINUXplatform -I/usr/X11/include LDF=-L/usr/X11R6/lib CC=gcc AWK=gawk RANLIB=ranlib TOUCH=touch TAR=tar AXIOMXLROOT=/usr/local/src/axiom-current/mnt/linux/compiler O=o BYE=bye LISP=lsp DAASE=/usr/local/src/axiom-current/src/share XLIB=/usr/X11R6/lib +10 copying /usr/local/src/axiom-current/src/scripts to /usr/local/src/axiom-current/mnt/linux/bin +19 making /usr/local/src/axiom-current/lsp +make[2]: Entering directory `/usr/local/src/axiom-current/lsp' +make[2]: `gcldir' is up to date. +make[2]: Leaving directory `/usr/local/src/axiom-current/lsp' +15 making /usr/local/src/axiom-current/src +make[2]: Entering directory `/usr/local/src/axiom-current/src' +21 making /usr/local/src/axiom-current/src/boot +make[3]: Entering directory `/usr/local/src/axiom-current/src/boot' +make[3]: `/usr/local/src/axiom-current/obj/linux/bin/bootsys' is up to date. +make[3]: Leaving directory `/usr/local/src/axiom-current/src/boot' +25 making /usr/local/src/axiom-current/src/interp +make[3]: Entering directory `/usr/local/src/axiom-current/src/interp' +3 making /usr/local/src/axiom-current/obj/linux/bin/depsys + +> +Loading /usr/local/src/axiom-current/obj/linux/interp/makedep.lisp +Loading /usr/local/src/axiom-current/obj/linux/interp/sys-pkg.lsp +Finished loading /usr/local/src/axiom-current/obj/linux/interp/sys-pkg.lsp +Loading /usr/local/src/axiom-current/obj/linux/interp/nocompil.lsp +Finished loading /usr/local/src/axiom-current/obj/linux/interp/nocompil.lsp +Loading /usr/local/src/axiom-current/obj/linux/interp/util.lsp +Finished loading /usr/local/src/axiom-current/obj/linux/interp/util.lsp +Loading /usr/local/src/axiom-current/int/interp/vmlisp.lisp +Finished loading /usr/local/src/axiom-current/int/interp/vmlisp.lisp +Loading /usr/local/src/axiom-current/int/interp/ggreater.lisp +Finished loading /usr/local/src/axiom-current/int/interp/ggreater.lisp +Loading /usr/local/src/axiom-current/int/interp/hash.lisp +Finished loading /usr/local/src/axiom-current/int/interp/hash.lisp +Loading /usr/local/src/axiom-current/int/interp/bootfuns.lisp +Finished loading /usr/local/src/axiom-current/int/interp/bootfuns.lisp +Loading /usr/local/src/axiom-current/int/interp/union.lisp +Finished loading /usr/local/src/axiom-current/int/interp/union.lisp +Loading /usr/local/src/axiom-current/int/interp/nlib.lisp +Finished loading /usr/local/src/axiom-current/int/interp/nlib.lisp +Loading /usr/local/src/axiom-current/int/interp/macros.lisp +Finished loading /usr/local/src/axiom-current/int/interp/macros.lisp +Loading /usr/local/src/axiom-current/int/interp/comp.lisp +Finished loading /usr/local/src/axiom-current/int/interp/comp.lisp +Loading /usr/local/src/axiom-current/int/interp/spaderror.lisp +Finished loading /usr/local/src/axiom-current/int/interp/spaderror.lisp +Loading /usr/local/src/axiom-current/int/interp/debug.lisp +Finished loading /usr/local/src/axiom-current/int/interp/debug.lisp +Loading /usr/local/src/axiom-current/int/interp/spad.lisp +Finished loading /usr/local/src/axiom-current/int/interp/spad.lisp +Loading /usr/local/src/axiom-current/int/interp/bits.lisp +Finished loading /usr/local/src/axiom-current/int/interp/bits.lisp +Loading /usr/local/src/axiom-current/int/interp/setq.lisp +Finished loading /usr/local/src/axiom-current/int/interp/setq.lisp +Loading /usr/local/src/axiom-current/int/interp/property.lisp +Finished loading /usr/local/src/axiom-current/int/interp/property.lisp +Loading /usr/local/src/axiom-current/int/interp/unlisp.lisp +Finished loading /usr/local/src/axiom-current/int/interp/unlisp.lisp +Loading /usr/local/src/axiom-current/int/interp/foam_l.lisp +Finished loading /usr/local/src/axiom-current/int/interp/foam_l.lisp +Loading /usr/local/src/axiom-current/int/interp/axext_l.lisp +Finished loading /usr/local/src/axiom-current/int/interp/axext_l.lisp +Loading /usr/local/src/axiom-current/lsp/gcl-2.6.1/cmpnew/collectfn.o +start address -T 0x875d000 Finished loading /usr/local/src/axiom-current/lsp/gcl-2.6.1/cmpnew/collectfn.o +Loading /usr/local/src/axiom-current/obj/linux/interp/postpar.fn +Finished loading /usr/local/src/axiom-current/obj/linux/interp/postpar.fn +Loading /usr/local/src/axiom-current/obj/linux/interp/proclaims.lisp +Finished loading /usr/local/src/axiom-current/obj/linux/interp/proclaims.lisp +Compiling /usr/local/src/axiom-current/obj/linux/interp/postpar.lsp. +; (DEFUN |postError| ...) is being compiled. +;; The variable |$defOp| is undefined. +;; The compiler will assume this variable is a global. +; (DEFUN |postCategory| ...) is being compiled. +;; The variable |$insidePostCategoryIfTrue| is undefined. +;; The compiler will assume this variable is a global. +; (DEFUN |postDef| ...) is being compiled. +;; The variable |$maxSignatureLineNumber| is undefined. +;; The compiler will assume this variable is a global. +;; The variable |$docList| is undefined. +;; The compiler will assume this variable is a global. +;; The variable |$headerDocumentation| is undefined. +;; The compiler will assume this variable is a global. +; (DEFUN |postIf| ...) is being compiled. +;; Warning: The variable |x| is not used. +End of Pass 1. +End of Pass 2. + +Error: Caught fatal error [memory may be damaged] +Fast links are on: do (si::use-fast-links nil) for debugging +Error signalled by COND. +Broken at APPLY. Type :H for Help. +COMPILER>>4 /usr/local/src/axiom-current/obj/linux/bin/depsys created +136 making /usr/local/src/axiom-current/obj/linux/interp/vmlisp.o from /usr/local/src/axiom-current/int/interp/vmlisp.lisp +/bin/sh: line 1: /usr/local/src/axiom-current/obj/linux/bin/depsys: No such file or directory +make[3]: *** [/usr/local/src/axiom-current/obj/linux/interp/vmlisp.o] Error 127 +make[3]: Leaving directory `/usr/local/src/axiom-current/src/interp' +make[2]: *** [interpdir] Error 2 +make[2]: Leaving directory `/usr/local/src/axiom-current/src' +make[1]: *** [srcdir] Error 2 +make[1]: Leaving directory `/usr/local/src/axiom-current' +make: *** [all] Error 2 + +--=-=-=-- + +\start +Date: Mon, 26 Jan 2004 08:57:12 -0500 +From: Tim Daly +To: simons@cryp.to +Subject: [Axiom-developer] build errors + +Two questions: + +1) Are you running Fedora (the free version redhat)? + If you are you need to turn off exec-shield first. + +2) Did you just update from last night? + I did an update last night and might have screwed something up. + I'll download from CVS and do a rebuild to check it. + +\start +Date: Mon, 26 Jan 2004 09:39:31 -0500 +From: Tim Daly +To: simons@cryp.to +Subject: [Axiom-developer] Building error + +Peter, + +You mentioned that this is a complete log of the build. +The lisp appears to have already been built. +The error occurs while compiling lisp code. +Which lisp are you using? + +\start +Date: 26 Jan 2004 10:52:13 -0500 +From: Camm Maguire +To: Tim Daly +Subject: Re: [Axiom-developer] build errors +Cc: simons@cryp.to + +Greetings! + +Tim Daly writes: + +> Two questions: +> +> 1) Are you running Fedora (the free version redhat)? +> If you are you need to turn off exec-shield first. +> + +Tim, is this still necessary? I though I put in automatic addition of +-Wa,--execstack when needed. + +Take care, + +> 2) Did you just update from last night? +> I did an update last night and might have screwed something up. +> I'll download from CVS and do a rebuild to check it. +> + +\start +Date: 26 Jan 2004 10:51:07 -0500 +From: Camm Maguire +To: Peter Simons +Subject: Re: [Axiom-developer] Building error + +Greetings! I think you have a bad GCL build. Can you post the full +configure and build output? Also, what machine are you using? + +Take care, + +Peter Simons writes: + +> I've tried to build the latest Axiom version from CVS, but +> got an error during the build. I'll attach the complete log +> output to this e-mail. +> +> Does anyone know what's causing this? And, more importantly, +> how to fix it? + +\start +Date: 26 Jan 2004 10:58:15 -0500 +From: Camm Maguire +To: daly@idsi.net +Subject: Re: [Axiom-developer] bitkeeper + +Greetings! + +root writes: + +> Has anyone used bitkeeper? I've been trolling the linux kernel +> mailing list and it seems that a fair number of the developers +> have moved off CVS and onto bitkeeper (www.bitkeeper.com). +> + +Must confess I've never used it, but understand that although it has +some nice features, there can be nasty strings attached. I've seen +reports of the bitkeeper folk revoking certain individuals' 'license' +to use their software, and also understand that it is forbidden, for +example, to use bitkeeper to develop an alternate source revision +control system. My impression is that it is quite restricted, and can +lead to trouble down the road. + +I've heard of a number of other truly open alternatives, though have +not tried any (e.g. aegis?, subversion?). Would be interested in +reading a knowledgeable comparison. + +\start +Date: Mon, 26 Jan 2004 10:13:37 -0500 +From: Tim Daly +To: Camm@enhanced.com +Subject: [Axiom-developer] exec-shield + +I don't know if it is still necessary. I meta-hosed the fedora box while +trying to get some extra space and I haven't gotten it back to working +order yet. + +\start +Date: Mon, 26 Jan 2004 10:20:16 -0500 +From: Tim Daly +To: Camm@enhanced.com +Subject: [Axiom-developer] bitkeeper + +I have about 6 "development" trees open locally and I'm trying to find +a way to manage them. It would be nice if other developers on a particular +subproject could just "pull" one of the trees. In particular, now that I +have the compile farm up and broken it would be nice if I could pull an +individual tree. + +re: yanking licenses. they are quite sensitive to people trying to +develop clones of bitkeeper since they actually sell it. however the +only requirement for using the free version is that you host your +meta-information on their site (not your code). they use the meta +info to generate test cases and debug reports I guess. + +\start +Date: Mon, 26 Jan 2004 18:56:54 +0100 +From: David MENTRE +To: daly@idsi.net +Subject: Re: [Axiom-developer] bitkeeper +Cc: Camm@enhanced.com + +Hello Tim, + +Tim Daly writes: + +> I have about 6 "development" trees open locally and I'm trying to find +> a way to manage them. It would be nice if other developers on a particular +> subproject could just "pull" one of the trees. In particular, now that I +> have the compile farm up and broken it would be nice if I could pull an +> individual tree. + +As Camm, I would be quite reluctant at using bitkeeper, as it is not +free software. + +Tim, have you considered GNU Arch +(http://www.gnu.org/software/gnu-arch/)? It is very similar to +bitkeeper in spirit. It has a big developer community (well maintained?) +and, more importantly, it keeps modifications as change sets. It is also +free software. + +Regarding your particular request, it works without central +repository. It allows developers to pull source tree from other +developers' source trees[1]. + +I've heard of very positive feedback from one of TeXmacs developer +(David Allouche) who uses it for daily work (no pun intented). + +[1] http://www.gnu.org/software/gnu-arch/tutorial/shared-and-public-archives.html#Shared_and_Public_Archives + +\start +Date: Mon, 26 Jan 2004 12:23:17 -0500 +From: Tim Daly +To: david.mentre@wanadoo.fr +Subject: [Axiom-developer] gnu-arch + +ok. I'll try that rather than bitkeeper. +Do you run gnu-arch? + +\start +Date: 26 Jan 2004 19:19:53 +0100 +From: Peter Simons +To: axiom-developer@nongnu.org +Subject: [Axiom-developer] Re: Building error + +Tim Daly writes: + + > The lisp appears to have already been built ... + +You're right. The full log file is available at + + http://peti.cryp.to/axiom-log.gz + +BTW, all stack-protection features _are_ disabled while +compiling Axiom. + +\start +Date: 26 Jan 2004 19:21:54 +0100 +From: Peter Simons +To: Camm Maguire +Subject: Re: [Axiom-developer] Building error + +Camm Maguire writes: + + > Also, what machine are you using? + +Linux/x86 2.6.1, gcc-3.2.3, glibc 2.3.2 + +\start +Date: Mon, 26 Jan 2004 19:26:09 +0100 +From: David MENTRE +To: Tim Daly +Subject: [Axiom-developer] Re: gnu-arch + +Tim Daly writes: + +> Do you run gnu-arch? + +Myself no. But I'm ready to try. + +\start +Date: Mon, 26 Jan 2004 12:41:18 -0500 +From: Tim Daly +To: david.mentre@wanadoo.fr +Subject: [Axiom-developer] Re: gnu-arch + +well i'm building it at the moment. if i get it to run you can +build it and try a pull from my source tree. -- t + +\start +Date: Mon, 26 Jan 2004 13:32:53 -0500 +From: "Bill Page" +To: "'Tim Daly'" , +Subject: RE: [Axiom-developer] gnu-arch + +Tim, David; + +I would also be very interested in experimenting with +the use of gnu-arch. + +Is it hosted somewhere like Savannah? Or, who does it +work when you (David) say that "It allows developers to +pull source tree from other developers' source trees."? + +Bill Page. + +> -----Original Message----- +> From: +> axiom-developer-bounces+bill.page1=sympatico.ca@nongnu.org +> [mailto:axiom-developer-bounces+bill.page1=sympatico.ca@nongnu +> .org] On Behalf Of Tim Daly +> Sent: Monday, January 26, 2004 12:23 PM +> To: david.mentre@wanadoo.fr +> Cc: axiom-developer@nongnu.org; daly@idsi.net +> Subject: [Axiom-developer] gnu-arch +> +> +> ok. I'll try that rather than bitkeeper. +> Do you run gnu-arch? +> + +\start +Date: Mon, 26 Jan 2004 12:55:53 -0500 +From: Tim Daly +To: simons@cryp.to +Subject: [Axiom-developer] build bug + +well, piffle. the CVS version is broken as I failed to upload a +change to one Makefile.pamphlet. I'll do that once I get home. + +That isn't Peter's problem, however. + +I don't see any reason for the build to fail at that point. + +lisp appears to have been built successfully. + +bootsys appears to have been built successfully. (bootsys is an +interpreted image that is used to compile the src/boot files). + +now the system is building depsys. this involves compiling the +files in src/boot. the first file you try to compile fails. +this means that bootsys is broken somehow. try the following +and let me know the result: + +export AXIOM=/usr/local/src/axiom-current/mnt/linux +export PATH=/usr/local/src/axiom-current/obj/linux/bin:$PATH +cd /usr/local/src/axiom-current/obj/linux/interp +lisp +> (load "makedep.lisp") + +if this fails please send me the makedep.lisp file. + +if all else fails start again: + +export AXIOM=/usr/local/src/axiom-current/mnt/linux +export PATH=/usr/local/src/axiom-current/obj/linux/bin:$PATH +cd /usr/local/src/axiom-current +make clean +make + +\start +Date: Mon, 26 Jan 2004 19:45:52 +0100 +From: David MENTRE +To: "Bill Page" +Subject: Re: [Axiom-developer] gnu-arch + +"Bill Page" writes: + +> Is it hosted somewhere like Savannah? + +No. Or this is not an obligation. + +> Or, who does it work when you (David) say that "It allows developers +> to pull source tree from other developers' source trees."? + +Each developer has a copy of the repository with his own additions. He +can make parts of it public (probably through a web server). + +But do not take this for granted. I've not read yet the Arch doc. :) + +\start +Date: Mon, 26 Jan 2004 13:08:19 -0500 +From: Tim Daly +To: bill.page1@sympatico.ca, david.mentre@wanadoo.fr +Subject: [Axiom-developer] gnu-arch + +If it works anything like bitkeeper (and it appears to do just that) +then it works as follows: + +I create a tree (or several trees) locally. You can ask to "pull" a +copy of my tree. Now my tree is the parent and yours is the child. +You can make changes locally and send me "changesets". A changeset +is just a collection of patches, hopefully with a common goal. This +solves the problem of changing multiple files in sync (so you don't +make the mistake I just made last night and forget a file change). + +You can pull the latest parent tree and it gets automatically merged +with your local changes. + +The main advantage is that I can create several trees and share them. +Right now the only shared resource is the savannah CVS and I have to +be very careful what I put up there because I can (and do) break the +build. + +Local trees mean that we can create sub-projects to explore ideas (which +is how I work locally, albeit by hand) without breaking the build for +the world. For instance, I want to work on the automatic coercion +code from Nicolas's PhD thesis which will deeply break the world for +a while. So all you would normally see is silence from me while I +get it working. It would be better if you could see and change the +latest broken code without waiting for CVS. + +So hopefully gnu-arch will give us the ability to share experimental +trees. + +\start +Date: Mon, 26 Jan 2004 13:09:55 -0500 +From: Tim Daly +To: bill.page1@sympatico.ca, david.mentre@wanadoo.fr +Subject: [Axiom-developer] gnu-arch + +well, gnu-arch built and installed cleanly. +i'll try to make a local axiom tree and see if you can reach it. + +\start +Date: Wed, 28 Jan 2004 13:58:10 +1100 +From: Jason White +To: David MENTRE +Subject: Re: [Axiom-developer] gnu-arch +Cc: Bill Page + +David MENTRE writes: + > + > Each developer has a copy of the repository with his own additions. He + > can make parts of it public (probably through a web server). + +or more precisely, through ftp, http or sftp. + > + > But do not take this for granted. I've not read yet the Arch doc. :) + > + > Yours, + > d. + > -- + +There are tools available for sharing projects between arch and cvs: +http://wiki.sourcecontrol.net/ + +\start +Date: Tue, 27 Jan 2004 22:53:23 -0500 +From: root +To: jasonjgw@pacific.net.au +Subject: Re: [Axiom-developer] gnu-arch +Cc: bill.page1@sympatico.ca + +I'm setting up a first version of gnu-arch now. +I'll let you know when it can be (hopefully) reached. + +\start +Date: Wed, 28 Jan 2004 11:32:09 +0100 +From: Dacian.Tudor@Comneon.com +To: camm@enhanced.com +Subject: RE: [Axiom-developer] Re: Axiom help needed + +Hi everybody, + +I am sorry to disturb you again with an AXIOM question. I have AXIOM running +and when it is starting it displays the following text : + +(AXIOM Sockets) The AXIOM server number is undefined. +---------------------------------------------------------------------------- +- + Issue )copyright to view copyright notices. + Issue )summary for a summary of useful system commands. + Issue )quit to leave AXIOM and return to shell. +Saturday January 10, 2004 at 20:14:29 +---------------------------------------------------------------------------- +- + +I have 3 questions: + +1. How can I connect to axiom ? Can I use some socket connection ? How can I +configure the port in Axiom ? + +2. How can I find the version of axiom ? If I use )copyright or )summary I +don't get any message. + +3. This is for Camm Maguire. Can you please give me some hints on +implementing a phrasebook for AXIOM ? Are there some encoders/decoders from +OpenMath to AXIOM and vice-versa available ? + +Thank you very much, +Dacian + +-----Original Message----- +From: Camm Maguire [mailto:camm@enhanced.com] +Sent: Montag, 19. Januar 2004 16:56 +To: Tudor Dacian (CN L External) +Cc: david.mentre@wanadoo.fr; daly@idsi.net; axiom-developer@nongnu.org +Subject: Re: [Axiom-developer] Re: Axiom help needed + + +Greetings! I don't know about AXIOM 2.3 -- is this the last commercial +release? The open source version available via cvs does not yet have the +openmath bindings to my understanding. These do not appear difficult to +implement. If you are in a hurry and want to try your hand before I get to +it, I can describe briefly how it should be done. + +Take care, + +Dacian.Tudor@Comneon.com writes: + +> Hi all, +> +> I need a solution to have AXIOM running with OpenMath. Can you tell me +> if such a solution is available at the moment and if not what would be +> the steps I should follow to have it. I read that there is a AXIOM 2.3 +> delivery with OpenMath support but I don't know if it is true and +> where I can get it from. +> +> Best wishes, +> Dacian +> +> -----Original Message----- +> From: Camm Maguire [mailto:camm@enhanced.com] +> Sent: Montag, 19. Januar 2004 16:04 +> To: David MENTRE +> Cc: daly@idsi.net; Tudor Dacian (CN L External); +axiom-developer@nongnu.org +> Subject: Re: [Axiom-developer] Re: Axiom help needed +> +> +> Greetings! +> +> David MENTRE writes: +> +> > Regarding OpenMath, I'm afraid but I think all NAG work to support +> > OpenMath is no longer available in current free Axiom. However, Bill +> > Page, Camm Maguire (lead developer of Gnu Common Lisp) and of course +> > Mike Dewar where also interested by OpenMath and they might help you. +> > They are on this list. +> > +> +> If this is a priority item, I might find time later this week/weekend +> to write a GCL interface to the OpenMath library. + +\start +Date: Wed, 28 Jan 2004 07:21:41 -0500 +From: root +To: Dacian.Tudor@Comneon.com +Subject: Re: [Axiom-developer] Re: Axiom help needed +Cc: camm@enhanced.com + +Axiom has a higher level program (not yet running) called sman (superman) +which manages the sockets. + +Axiom 2.3 is the last commercial version. +Versions are now listed by date and time of the build. +Yours was last built on: +Saturday January 10, 2004 at 20:14:29 + +The encoders/decoders for openmath are in the C code +zips/OMCv1.4a.tgz +and the code to use OpenMath is in the +src/algebra/*.spad.pamphlet +files. + +\start +Date: Wed, 28 Jan 2004 12:51:48 +0100 +From: Dacian.Tudor@Comneon.com +To: daly@idsi.net, Dacian.Tudor@Comneon.com +Subject: RE: [Axiom-developer] Re: Axiom help needed +Cc: camm@enhanced.com + +Hi Tim, + +Are you saying that there is no way to connect to AXIOM from other programs +and to send commands to AXIOM ? (e.g ports) + +I downloaded the sources from cvs (is this AXIOM 2.3 you are talking about +?) and I have some error messages when trying to compile. Here is the +message I receive : + +[tudor@lnzsx01c(sauron):~]$ cd axiom +[tudor@lnzsx01c(sauron):~/axiom]$ export AXIOM=/home/tudor/axiom/mnt/linux +[tudor@lnzsx01c(sauron):~/axiom]$ export PATH=$AXIOM/bin:$PATH +[tudor@lnzsx01c(sauron):~/axiom]$ make +13 making noweb +patching file modules.c +mnt.o(.text+0x365): In function `emitfile': +: the use of `tmpnam' is dangerous, better use `mkstemp' +make[1]: [install-shell] Error 1 (ignored) +make[1]: [install-code] Error 1 (ignored) +/bin/sh: line 1: texhash: command not found +make[1]: [install-elisp] Error 1 (ignored) +0 SPAD=/home/tudor/axiom/mnt/linux SYS=linux SPD=/home/tudor/axiom +LSP=/home/tud +or/axiom/lsp GCLDIR=/home/tudor/axiom/lsp/gcl-2.6.1 +SRC=/home/tudor/axiom/src IN +T=/home/tudor/axiom/int OBJ=/home/tudor/axiom/obj MNT=/home/tudor/axiom/mnt +ZIPS +=/home/tudor/axiom/zips TMP=/home/tudor/axiom/obj/tmp +SPADBIN=/home/tudor/axiom/ +mnt/linux/bin INC=/home/tudor/axiom/src/include +CCLBASE=/home/tudor/axiom/obj/li +nux/ccl/ccllisp PART=cprogs SUBPART=everything NOISE=-o +/home/tudor/axiom/obj/tm +p/trace GCLVERSION=gcl-2.6.1 +TANGLE=/home/tudor/axiom/mnt/linux/bin/lib/notangle +10 copying /home/tudor/axiom/src/scripts to /home/tudor/axiom/mnt/linux/bin +1 making a linux system, PART=cprogs SUBPART=everything +2 Environment SPAD=/home/tudor/axiom/mnt/linux SYS=linux +SPD=/home/tudor/axiom L +SP=/home/tudor/axiom/lsp GCLDIR=/home/tudor/axiom/lsp/gcl-2.6.1 +SRC=/home/tudor/ +axiom/src INT=/home/tudor/axiom/int OBJ=/home/tudor/axiom/obj +MNT=/home/tudor/ax +iom/mnt ZIPS=/home/tudor/axiom/zips TMP=/home/tudor/axiom/obj/tmp +SPADBIN=/home/ +tudor/axiom/mnt/linux/bin INC=/home/tudor/axiom/src/include +CCLBASE=/home/tudor/ +axiom/obj/linux/ccl/ccllisp PART=cprogs SUBPART=everything NOISE=-o +/home/tudor/ +axiom/obj/tmp/trace GCLVERSION=gcl-2.6.1 +TANGLE=/home/tudor/axiom/mnt/linux/bin/ +lib/notangle +make[1]: Entering directory `/home/tudor/axiom' +11 checking directory structure +12 Environment: PLF=LINUXplatform CCF=-O2 -fno-strength-reduce -Wall +-D_GNU_SOUR +CE -DLINUXplatform -I/usr/X11/include LDF=-L/usr/X11R6/lib CC=gcc AWK=gawk +RANLI +B=ranlib TOUCH=touch TAR=tar +AXIOMXLROOT=/home/tudor/axiom/mnt/linux/compiler O= +o BYE=bye LISP=lsp DAASE=/home/tudor/axiom/src/share XLIB=/usr/X11R6/lib +16 making /home/tudor/axiom/src/Makefile from +/home/tudor/axiom/src/Makefile.pam +phlet +which: no latex in +(/home/tudor/axiom/mnt/linux/bin:/usr/local/bin:/bin:/usr/bin +:/usr/X11R6/bin:/usr/atria/bin:/home/tudor/bin:./) +document ERROR You must install latex first +18 making /home/tudor/axiom/src +make[2]: Entering directory `/home/tudor/axiom/src' +make[2]: *** No rule to make target `setup'. Stop. +make[2]: Leaving directory `/home/tudor/axiom/src' +make[1]: *** [srcsetup] Error 2 +make[1]: Leaving directory `/home/tudor/axiom' +make: *** [all] Error 2 + +I am wondering if this has to do with latex or not. + +I am looking forward to receiving your answer. + +Dacian + +-----Original Message----- +From: root [mailto:daly@idsi.net] +Sent: Mittwoch, 28. Januar 2004 13:22 +To: Dacian.Tudor@Comneon.com +Cc: camm@enhanced.com; david.mentre@wanadoo.fr; daly@idsi.net; +axiom-developer@nongnu.org +Subject: Re: [Axiom-developer] Re: Axiom help needed + + + +Axiom has a higher level program (not yet running) called sman (superman) +which manages the sockets. + +Axiom 2.3 is the last commercial version. +Versions are now listed by date and time of the build. +Yours was last built on: +Saturday January 10, 2004 at 20:14:29 + +The encoders/decoders for openmath are in the C code zips/OMCv1.4a.tgz +and the code to use OpenMath is in the +src/algebra/*.spad.pamphlet +files. + +\start +Date: Wed, 28 Jan 2004 07:59:07 -0500 +From: root +To: Dacian.Tudor@Comneon.com +Subject: Re: [Axiom-developer] Re: Axiom help needed +Cc: Dacian.Tudor@Comneon.com, camm@enhanced.com + +The failure of the build is due to the fact that latex is not installed. + +yes, it is possible to connect to Axiom but it requires a bit of +low level knowledge to do it. Axiom adds a socket package to GCL +and you can access the Lisp code directly. + +For details see src/lib/sockio.c.pamphlet line 1264 +and you can see that Axiom adds code to GCL to do things like +redirect standard I/O + +This code is used by src/interp/sockio.lisp.pamphlet (around line 93). +So it is possible to use sockets from lisp. + +It is even possible to use sockets directly from the spad algebra +language. Function calls to lisp functions (e.g. car) can be reached by +CAR(x)$LISP +that is, the function name must be uppercased and $LISP appended. + +\start +Date: 28 Jan 2004 12:05:33 -0500 +From: Camm Maguire +To: Dacian.Tudor@Comneon.com +Subject: Re: [Axiom-developer] Re: Axiom help needed + +Greetings! + +Dacian.Tudor@Comneon.com writes: + +> 3. This is for Camm Maguire. Can you please give me some hints on +> implementing a phrasebook for AXIOM ? Are there some encoders/decoders from +> OpenMath to AXIOM and vice-versa available ? + +I'm afraid I don't know what a 'phrasebook' is in this context. Once +I write the common lisp bindings in GCL to the (C) openmath library, +doesn't axiom already know about that api? + +Take care, + +> +> Thank you very much, +> Dacian +> +> -----Original Message----- +> From: Camm Maguire [mailto:camm@enhanced.com] +> Sent: Montag, 19. Januar 2004 16:56 +> To: Tudor Dacian (CN L External) +> Cc: david.mentre@wanadoo.fr; daly@idsi.net; axiom-developer@nongnu.org +> Subject: Re: [Axiom-developer] Re: Axiom help needed +> +> +> Greetings! I don't know about AXIOM 2.3 -- is this the last commercial +> release? The open source version available via cvs does not yet have the +> openmath bindings to my understanding. These do not appear difficult to +> implement. If you are in a hurry and want to try your hand before I get to +> it, I can describe briefly how it should be done. +> +> Take care, +> +> Dacian.Tudor@Comneon.com writes: +> +> > Hi all, +> > +> > I need a solution to have AXIOM running with OpenMath. Can you tell me +> > if such a solution is available at the moment and if not what would be +> > the steps I should follow to have it. I read that there is a AXIOM 2.3 +> > delivery with OpenMath support but I don't know if it is true and +> > where I can get it from. +> > +> > Best wishes, +> > Dacian +> > +> > -----Original Message----- +> > From: Camm Maguire [mailto:camm@enhanced.com] +> > Sent: Montag, 19. Januar 2004 16:04 +> > To: David MENTRE +> > Cc: daly@idsi.net; Tudor Dacian (CN L External); +> axiom-developer@nongnu.org +> > Subject: Re: [Axiom-developer] Re: Axiom help needed +> > +> > +> > Greetings! +> > +> > David MENTRE writes: +> > +> > > Regarding OpenMath, I'm afraid but I think all NAG work to support +> > > OpenMath is no longer available in current free Axiom. However, Bill +> > > Page, Camm Maguire (lead developer of Gnu Common Lisp) and of course +> > > Mike Dewar where also interested by OpenMath and they might help you. +> > > They are on this list. +> > > +> > +> > If this is a priority item, I might find time later this week/weekend +> > to write a GCL interface to the OpenMath library. + +\start +Date: 28 Jan 2004 12:11:23 -0500 +From: Camm Maguire +To: daly@idsi.net +Subject: Re: [Axiom-developer] Re: Axiom help needed +Cc: Dacian.Tudor@Comneon.com + +Greetings! Just a quick note here -- all should work below as Tim +says, but if anyone would like to use GCL's native socket support +instead/in addition and could provide a quick example of a socket +session to be so converted, I could take a look at it (at some point +in the moderately near future :-)). + +Take care, + +root writes: + +> The failure of the build is due to the fact that latex is not installed. +> +> yes, it is possible to connect to Axiom but it requires a bit of +> low level knowledge to do it. Axiom adds a socket package to GCL +> and you can access the Lisp code directly. +> +> For details see src/lib/sockio.c.pamphlet line 1264 +> and you can see that Axiom adds code to GCL to do things like +> redirect standard I/O +> +> This code is used by src/interp/sockio.lisp.pamphlet (around line 93). +> So it is possible to use sockets from lisp. +> +> It is even possible to use sockets directly from the spad algebra +> language. Function calls to lisp functions (e.g. car) can be reached by +> CAR(x)$LISP +> that is, the function name must be uppercased and $LISP appended. + +\start +Date: Wed, 28 Jan 2004 13:05:09 -0500 +From: root +To: camm@enhanced.com +Subject: Re: [Axiom-developer] Re: Axiom help needed +Cc: Dacian.Tudor@Comneon.com + +Camm, + +Yes, we do need to start using GCL-native socket support. +We can #+ it into place. +Please submit a bug report for this. + +\start +Date: Wed, 28 Jan 2004 19:54:38 +0100 +From: David MENTRE +To: axiom-developer@nongnu.org +Subject: [Axiom-developer] Possible switch from CVS to GNU Arch + +Hello, + +For demexp development, I'm seriously considering switching from CVS to +GNU Arch (http://www.gnu.org/software/gnu-arch/). + +Pros: + - excellent support for renaming, handling of patchset, etc. + - excellent support for branches + - support of non-centralized trees + - apparently well maintained + +Cons: + - not available as debian package in woody (isa impacted, however it + can be compiled from sources) + - no windows port (windows port of demexp) + - not tested yet (I've just read the docs ;) + + +In any case, if I switch to Arch, I won't force other demexp developers +to also switch to Arch. I'll update the CVS tree on a regular basis. We +should also do regular snaposhots of the development tree. + +Any opinion ? Any news ? + +\start +Date: Wed, 28 Jan 2004 15:02:26 -0500 +From: root +To: dmentre@linux-france.org +Subject: Re: [Axiom-developer] Possible switch from CVS to GNU Arch + +Well, the big news seems to be that you switched email addresses. +Or somebody is posing as you. + +No news yet. Gnu-arch compiled cleanly. I put it on several machines +and it installed without a problem. I'm slowly reading the docs and +trying each of the commands. I have an archive up on one machine +which I'm populating at the moment. + +Opinion? Well, CVS works to keep things archived, it is standard on +savannah so I'm stuck with it there. However it forces a single +point-of-development model and requires me to do detailed bookkeeping. +I've already hosed the master site twice by failing to upload a changed +file. Hopefully Gnu-arch will cure two of the three problems. I'll still +have to export to CVS for savannah but Camm should be able to make +changes in the gnu-arch tree for GCL-native socket support and send it +in as one "changeset" (fixing single-point-of-development). And it +should cure the missing file issue since everything gets bundled as a +changeset (fixing the missing file upload issue). + +Further, I have a few "parallel" projects with various people (for +example I want to explore the coerce code with Nicolas) so I want +separate trees with experimental code without breaking the master tree. +Gnu-arch appears to have a better merge model. + +So I'm on the learning curve at the moment. Linus uses bitkeeper for +Linux because he faced the same issues so if it works for him it should +work for me. But, as you point out, bitkeeper has a non-free license +and, if free works, why bother? + +\start +Date: Wed, 28 Jan 2004 20:39:33 +0100 +From: David MENTRE +To: daly@idsi.net +Subject: Re: [Axiom-developer] Possible switch from CVS to GNU Arch +Cc: dmentre@linux-france.org + +root writes: + +> Well, the big news seems to be that you switched email addresses. + +Oops. Sorry, my message was not intended for this mailing list. + +\start +Date: Thu, 29 Jan 2004 11:54:02 -0600 +From: Michal Lijowski +To: axiom-developer@nongnu.org +Subject: [Axiom-developer] Installing AXIOM on RedHat Fedora Core 1 crashes + +Hello, + +I am trying to install AXIOM on RedHat Fedora Core 1 with +gcc 3.3.2 compiler. The installation crashes with following +message + +Loading ./../gcl-tk/tk-package.lsp +Finished loading ./../gcl-tk/tk-package.lsp +Loading ./../cmpnew/cmpmain.lsp +Warning: COMPILE-FILE is being redefined. +Warning: COMPILE is being redefined. +Warning: DISASSEMBLE is being redefined. +Finished loading ./../cmpnew/cmpmain.lsp +Loading ./../cmpnew/lfun_list.lsp +Finished loading ./../cmpnew/lfun_list.lsp +Loading ./../cmpnew/cmpopt.lsp +Finished loading ./../cmpnew/cmpopt.lsp +Loading ./../lsp/auto_new.lsp +Finished loading ./../lsp/auto_new.lsp +Warning: LISP-IMPLEMENTATION-VERSION is being redefined. +T + +> +#<"USER" package> + +> +Unrecoverable error: Segmentation violation.. +make[4]: *** [saved_gcl] Error 134 +rm init_gcl.lsp.tmp raw_gcl +make[4]: Leaving directory +`/home/michal/tmp/axiom/lsp/gcl-2.5.2/unixport' +make[3]: *** [unixport/saved_gcl] Error 2 +make[3]: Leaving directory `/home/michal/tmp/axiom/lsp/gcl-2.5.2' +make[2]: Leaving directory `/home/michal/tmp/axiom/lsp' +15 making /home/michal/tmp/axiom/src +make[2]: Entering directory `/home/michal/tmp/axiom/src' +9 making /home/michal/tmp/axiom/src/boot +make[3]: Entering directory `/home/michal/tmp/axiom/src/boot' +2 making /home/michal/tmp/axiom/int/boot/boothdr.lisp from +/home/michal/tmp/axiom/src/boot/boothdr.lisp.pamphlet +1 making /home/michal/tmp/axiom/obj/linux/boot/boothdr.o from +/home/michal/tmp/axiom/int/boot/boothdr.lisp +/bin/sh: line 1: /home/michal/tmp/axiom/obj/linux/bin/lisp: Permission +denied +make[3]: *** [/home/michal/tmp/axiom/obj/linux/boot/boothdr.o] Error 126 +make[3]: Leaving directory `/home/michal/tmp/axiom/src/boot' +make[2]: *** [bootdir] Error 2 +make[2]: Leaving directory `/home/michal/tmp/axiom/src' +make[1]: *** [srcdir] Error 2 +make[1]: Leaving directory `/home/michal/tmp/axiom' +make: *** [all] Error 2 + +The file /home/michal/tmp/axiom/obj/linux/bin/lisp is not permitted +to be executed. + +Michal Lijowski + +\start +Date: Thu, 29 Jan 2004 14:12:23 -0500 +From: root +To: michal@cvu.wustl.edu +Subject: Re: [Axiom-developer] Installing AXIOM on RedHat Fedora Core 1 crashes + +try the following... (as root) + + echo 0 >/proc/sys/kernel/exec-shield + +then, + +cd home/michal/tmp/axiom +make clean +make + +let me know if you succeed or fail. + +\start +Date: 29 Jan 2004 15:55:13 -0500 +From: Camm Maguire +To: daly@idsi.net +Subject: Re: [Axiom-developer] Installing AXIOM on RedHat Fedora Core 1 crashes +Cc: michal@cvu.wustl.edu + +Greetings! + +root writes: + +> try the following... (as root) +> +> echo 0 >/proc/sys/kernel/exec-shield +> +> then, +> +> cd home/michal/tmp/axiom +> make clean +> make +> +> +> let me know if you succeed or fail. +> + +Or try latest stable cvs GCL via + +export CSH_RSH=ssh +export CVSROOT=:ext:anoncvs@subversions.gnu.org:/cvsroot/gcl +cvs -z9 -q co -r Version_2_6_1 -d gcl-2.6.1 gcl +cd gcl-2.6.1 +./configure && make + +Should let you build on Fedora without being root. I'd appreciate +knowing if not. + +\start +Date: Thu, 29 Jan 2004 17:14:56 -0500 +From: root +To: michal@cvu.wustl.edu +Subject: Re: [Axiom-developer] Installing AXIOM on RedHat Fedora Core 1 crashes +Cc: camm@enhanced.com + +It should take approx 3 hours on a 2GHz machine -- t + +\start +Date: Thu, 29 Jan 2004 17:17:38 -0500 +From: root +To: camm@enhanced.com +Subject: Re: [Axiom-developer] Installing AXIOM on RedHat Fedora Core 1 crashes +Cc: michal@cvu.wustl.edu + +Camm, + +I will update the version in the cvs to use the one you reference +in the command. I don't have a working fedora but I'll try to build +one here again. I've been off on other parts of the compile farm. + +btw, did you see the 1 terabyte disk? USB enabled, external. $1200. + +\start +Date: Thu, 29 Jan 2004 15:18:09 -0600 +From: Michal Lijowski +To: Camm Maguire +Subject: Re: [Axiom-developer] Installing AXIOM on RedHat Fedora Core 1 crashes + +After removing untared axiom directory, untaring axiom.20031211.tgz, +and restarting installation, axiom make is still running over one +hour on 2GHz P4 with 1GB RAM. + + Michal + +On Thu, 2004-01-29 at 14:55, Camm Maguire wrote: +> Greetings! +> +> root writes: +> +> > try the following... (as root) +> > +> > echo 0 >/proc/sys/kernel/exec-shield +> > +> > then, +> > +> > cd home/michal/tmp/axiom +> > make clean +> > make +> > +> > +> > let me know if you succeed or fail. +> > +> +> Or try latest stable cvs GCL via +> +> export CSH_RSH=ssh +> export CVSROOT=:ext:anoncvs@subversions.gnu.org:/cvsroot/gcl +> cvs -z9 -q co -r Version_2_6_1 -d gcl-2.6.1 gcl +> cd gcl-2.6.1 +> ./configure && make +> +> Should let you build on Fedora without being root. I'd appreciate +> knowing if not. + +\start +Date: 29 Jan 2004 17:12:50 -0500 +From: Camm Maguire +To: daly@idsi.net +Subject: Re: [Axiom-developer] Installing AXIOM on RedHat Fedora Core 1 crashes +Cc: michal@cvu.wustl.edu + +Greetings! + +root writes: + +> Camm, +> +> I will update the version in the cvs to use the one you reference +> in the command. I don't have a working fedora but I'll try to build +> one here again. I've been off on other parts of the compile farm. + +No prob/need for hurry. + +> +> btw, did you see the 1 terabyte disk? USB enabled, external. $1200. +> + +!!!! No I did not. What an age we live in. + +\start +Date: Thu, 29 Jan 2004 16:50:52 -0600 +From: Michal Lijowski +To: daly@idsi.net +Subject: Re: [Axiom-developer] Installing AXIOM on RedHat Fedora Core 1 crashes +Cc: camm@enhanced.com + +Hello, + +I installed successfully axiom20031211 on RedHat Fedora Core 1. +Next, I have to find out how to use it. + +Thank you all, + +Michal + +On Thu, 2004-01-29 at 16:17, root wrote: +> Camm, +> +> I will update the version in the cvs to use the one you reference +> in the command. I don't have a working fedora but I'll try to build +> one here again. I've been off on other parts of the compile farm. +> +> btw, did you see the 1 terabyte disk? USB enabled, external. $1200. + +\start +Date: Thu, 29 Jan 2004 23:48:14 -0500 +From: root +To: axiom-developer@nongnu.org +Subject: [Axiom-developer] gnu-arch + +I have several efforts on Axiom in process at the same time. +None of these efforts can show up in the savannah tree because +they break the build for our normal "customers". + +We need a development tool so we can work on new ideas and allow +them to proceed in parallel. To that end I've been investigating +Gnu-arch. Gnu-arch allows you to "pull" a copy of a particular +subproject, work on it locally, and submit a "changeset" which +is a series of patches collected into one. + +I've succeeded in setting up the main project. I'm still working +on the details of how to create subversions of this project. You +can get Gnu-arch and play with it if you like. The latest build +I have is tla-1.1.tar.gz from the gnu-arch website. You can get +up to speed by the following steps: + +(download tla-1.1.tar.gz) +tar -zcf tla-1.1.tar.gz +cd tla-1.1/src +mkdir =build +cd =build +../configure --prefix /usr/local +make +make install + +tla my-id "Your Name " +tla register-archive axiom@tenkan.org--axiom-SOURCE http://axiom.tenkan.org/current +tla make-archive --mirror-from axiom@tenkan.org--axiom-SOURCE you@place--axiom +tla archive-mirror axiom@tenkan.org--axiom +tla my-default-archive axiom@tenkan.org--axiom +tla get axiom--release--1 axiom +cd axiom + +Notes: + +tla is the only executable. All commands are parameters to tla. + +tla my-id has a standard format "Your Name " + +tla register-archive says that the name "axiom@tenkan.org--axiom-SOURCE" + should resolve to "http://axiom.tenkan.org/current" which is where + the current version of Axiom resides (for developers). The current + stable customer version will be on savannah. + +tla make-archive says that when you ask for "axiom@tenkan.org--axiom" + you want it copy the Gnu-arch project tree to "you@place--axiom" + +tla archive-mirror does a network copy from "axiom@tenkan.org--axiom" to + "you@place--axiom" + +tla my-default-archive "axiom@tenkan.org--axiom" says that all commands + are assumed to refer to this archive by default. + +tla get reaches into "axiom--release--1" (which is hiding in + "you@place--axiom" in gnu-arch form) and expands it into the + directory "axiom". + +cd axiom will get you to where the source resides. + +As for further details, well, I'm still reading the docs. +If you're interested in tracking one of the various subprojects +you might want to try this example. + +Let me know if you succeed or fail. + +\start +Date: Fri, 30 Jan 2004 23:48:47 +0100 +From: David MENTRE +To: daly@idsi.net +Subject: Re: [Axiom-developer] gnu-arch + +Hello Tim, + +root writes: + +> (download tla-1.1.tar.gz) +> tar -zcf tla-1.1.tar.gz +> cd tla-1.1/src +> mkdir =build +> cd =build +> ../configure --prefix /usr/local +> make +> make install + +or "apt-get install tla" (as root) on a Debian sarge/unstable/testing +system. :) + + +> tla my-id "Your Name " + +done. + +> tla register-archive axiom@tenkan.org--axiom-SOURCE http://axiom.tenkan.org/current + +done. + +> tla make-archive --mirror-from axiom@tenkan.org--axiom-SOURCE you@place--axiom + +done. By the way, "you@place--axiom" is the name of the local directory +where the archive is stored so I prefer to use a directory name like +"~/pub/axiom-libre/Tim-axiom-archive". + +> tla archive-mirror axiom@tenkan.org--axiom + +Be aware, it takes some time without any indication from tla. The base +file is about 27 Mbytes. + +> tla my-default-archive axiom@tenkan.org--axiom + +Not done. I don't want to do that (I'm working with my own default +archive) so... + +> tla get axiom--release--1 axiom + +... I have done: + +tla get -A axiom@tenkan.org--axiom axiom--release--1 axiom + + +> cd axiom + +And I have the sources. Great! :) + + + +I also need to reread the Arch tutorial: + http://www.gnu.org/software/gnu-arch/tutorial/arch.html + + +By the way, there is a tool called cscvs to transform CVS history into a +proper Arch history. It takes a long time but it works. I've used it on +another project. + http://wiki.sourcecontrol.net/moin.cgi/cscvs + + +BTW, a question, is your axiom--release--1 tree the latest CVS tree? +Looking at the CHANGELOG, it seems so, but just to be sure. I also +noticed that most of files have executable rights (like +src/Makefile.pamphlet). I think it is not needed. + + +\start +Date: Sat, 31 Jan 2004 09:29:03 +0100 +From: David MENTRE +To: daly@idsi.net, Tim Daly +Subject: Re: [Axiom-developer] gnu-arch + +David MENTRE writes: + +> By the way, there is a tool called cscvs to transform CVS history into a +> proper Arch history. It takes a long time but it works. I've used it on +> another project. +> http://wiki.sourcecontrol.net/moin.cgi/cscvs + +Do not forget to use the -S option when doing the totla +command. Otherwise the Summary lines of tla logs are not very +meaningful (you would have patchset-84, patchset-85, ...). + +cscvs totla -S -i 1 /path/to/arch/dir +cscvs totla -S -c 2: /path/to/arch/dir + +\start +Date: Sat, 31 Jan 2004 04:47:00 -0500 +From: root +To: david_mentre@wanadoo.fr +Subject: [Axiom-developer] axiom from tla + +Yes, the sources are the latest CVS copy. +I'm trying to figure out how to create a second project +that uses the first project. +There is no such thing as a simple job :-) + +\start +Date: Sat, 31 Jan 2004 09:58:23 +0100 +From: David MENTRE +To: Tim Daly +Subject: [Axiom-developer] Clef does not work, patch proposed + +Hello Tim, + +Last time I checked, Clef was not working. I've open bug #7527 for this +issue as a remainder. + +I proposed a patch in november: +http://mail.nongnu.org/archive/html/axiom-developer/2003-11/msg00290.html + +Let me know if you want me to rework the patch with current Arch tree. + +\start +Date: Sat, 31 Jan 2004 04:53:50 -0500 +From: root +To: david.mentre@wanadoo.fr +Subject: [Axiom-developer] Re: Clef does not work, patch proposed + +David, + +Try to send me a changeset of the patch and I will try to apply it. + +\start +Date: Sat, 31 Jan 2004 05:18:32 -0500 +From: root +To: david.mentre@wanadoo.fr +Subject: [Axiom-developer] Re: Clef does not work, patch proposed + +David, + +If you do a changeset please write down the exact sequence of +tla commands you used. + +\start +Date: Sat, 31 Jan 2004 10:53:29 +0100 +From: David MENTRE +To: axiom-developer@nongnu.org +Subject: [Axiom-developer] Error when compiling axiom, gcl issue? + +Hello, + +I tried to compile latest Arch Axiom and it failed with following error: + +Finished compiling /home/david/00-poubelle/Axiom/axiom/int/algebra/./NFINTBAS.NRLIB/code.lsp. +------------------------------------------------------------------------ + NumberFieldIntegralBasis is now explicitly exposed in frame initial + NumberFieldIntegralBasis will be automatically loaded when needed + from + /home/david/00-poubelle/Axiom/axiom/int/algebra/NFINTBAS.NRLIB/code + + +Unrecoverable error: mark botch. +make[3]: *** [/home/david/00-poubelle/Axiom/axiom/int/algebra/NFINTBAS.NRLIB] Error 134 + + +Unfortunatly, I have stored the full log. I'll remake a compilation to +have it. + +I'm compiling on an Athlon 2500+, Debian sarge system. + +\start +Date: Sat, 31 Jan 2004 05:52:32 -0500 +From: root +To: david.mentre@wanadoo.fr, camm@enhanced.com +Subject: Re: [Axiom-developer] Error when compiling axiom, gcl issue? + +mark botch is a failure in the garbage collection routine. +this is a gcl issue. + +Tim + +================================================================= + + +I tried to compile latest Arch Axiom and it failed with following error: + +Finished compiling /home/david/00-poubelle/Axiom/axiom/int/algebra/./NFINTBAS.NRLIB/code.lsp. +------------------------------------------------------------------------ + NumberFieldIntegralBasis is now explicitly exposed in frame initial + NumberFieldIntegralBasis will be automatically loaded when needed + from + /home/david/00-poubelle/Axiom/axiom/int/algebra/NFINTBAS.NRLIB/code + + +Unrecoverable error: mark botch. +make[3]: *** [/home/david/00-poubelle/Axiom/axiom/int/algebra/NFINTBAS.NRLIB] Error 134 + + +Unfortunatly, I have stored the full log. I'll remake a compilation to +have it. + +I'm compiling on an Athlon 2500+, Debian sarge system. + +\start +Date: Sat, 31 Jan 2004 15:35:56 +0100 +From: David MENTRE +To: camm@enhanced.com, axiom-developer@nongnu.org, gcl-devel@gnu.org +Subject: Re: [Axiom-developer] Error when compiling axiom, gcl issue? + +Camm, + +me: + +> Unfortunatly, I have stored the full log. I'll remake a compilation to +> have it. +> +> I'm compiling on an Athlon 2500+, Debian sarge system. + +GCL version used in Axiom sources is 2.6.1. + +\start +Date: 31 Jan 2004 12:55:21 -0500 +From: Camm Maguire +To: David MENTRE +Subject: Re: [Gcl-devel] Re: [Axiom-developer] Error when compiling axiom, gcl issue? + +Hi David! Will I be able to reproduce this with latest cvs? I'm +still a page behind with gnu-arch. BTW, I'm getting bounces from your +mail address. Another BTW, just released a new axiom Debian package, +0.20040128-1, which built without a hitch for me and should eventually +make its way across all the other Debian platforms when the +autobuilders take hold. This build used Debian's externally +installed gcl, 2.6.1-23, but I can't think of any changes since axiom +took a snapshot of the 2.6.1 tree which would explain this. + +Take care, + +David MENTRE writes: + +> Camm, +> +> me: +> +> > Unfortunatly, I have stored the full log. I'll remake a compilation to +> > have it. +> > +> > I'm compiling on an Athlon 2500+, Debian sarge system. +> +> GCL version used in Axiom sources is 2.6.1. + +\start +Date: Sat, 31 Jan 2004 17:10:25 -0500 +From: root +To: camm@enhanced.com +Subject: [Axiom-developer] savannah CVS vs tenkan arch + +The arch server is where I plan to "externalize" my projects +(like writing the book) so others can help. + +The main version of Axiom will first appear on arch and if +everyone likes it then I'll post the working version to CVS +on savannah. + + + + diff --git a/changelog b/changelog index 47c8129..e5a8ffa 100644 --- a/changelog +++ b/changelog @@ -1,3 +1,5 @@ +20140420 tpd src/axiom-website/patches.html 20140421.01.tpd.patch +20140420 tpd book/2004-01.txt regularize 20140420 tpd src/axiom-website/patches.html 20140420.03.tpd.patch 20140420 tpd book/2003-12.txt regularize 20140420 tpd src/axiom-website/patches.html 20140420.02.tpd.patch diff --git a/src/axiom-website/patches.html b/src/axiom-website/patches.html index a65768f..2604279 100644 --- a/src/axiom-website/patches.html +++ b/src/axiom-website/patches.html @@ -4286,6 +4286,8 @@ book/2003-10.txt regularize book/2003-11.txt regularize 20140420.03.tpd.patch book/2003-12.txt regularize +20140421.01.tpd.patch +book/2004-01.txt regularize