You can subscribe to this list here.
2002 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
(38) |
Jul
(10) |
Aug
(2) |
Sep
|
Oct
|
Nov
(7) |
Dec
|
---|---|---|---|---|---|---|---|---|---|---|---|---|
2003 |
Jan
(47) |
Feb
(5) |
Mar
(8) |
Apr
(12) |
May
|
Jun
|
Jul
(4) |
Aug
(7) |
Sep
|
Oct
|
Nov
|
Dec
(11) |
2004 |
Jan
(1) |
Feb
(1) |
Mar
(3) |
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
(1) |
Oct
(3) |
Nov
|
Dec
|
2006 |
Jan
(1) |
Feb
|
Mar
|
Apr
(2) |
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
(1) |
Nov
|
Dec
|
2007 |
Jan
|
Feb
(1) |
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
(9) |
Sep
|
Oct
|
Nov
|
Dec
|
2008 |
Jan
|
Feb
|
Mar
|
Apr
|
May
(1) |
Jun
|
Jul
(12) |
Aug
(36) |
Sep
|
Oct
|
Nov
|
Dec
|
2009 |
Jan
|
Feb
(1) |
Mar
|
Apr
(4) |
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
2010 |
Jan
|
Feb
|
Mar
(10) |
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
2011 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
(2) |
Oct
|
Nov
|
Dec
|
2012 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
(1) |
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
From: Stephen R. <re...@cy...> - 2003-03-26 15:34:23
|
On Wed, 26 Mar 2003, Piermarco Burrafato wrote: > Hi, > > I'm not sure I have understood how to relate my own costants to the Cyc > opensource framework. To better let you understand, I am working on a > research project involving knowledge representetation and anchoring of > symbols to semantics, and so I need to create my own costants in order > to build a new KB capturing the problem domain. > > Now, if I want to make my KB a "plug in" of the OpenCyc framework, I've > got the following questions: > > 1. Do I have to set the current project to OpenCycProject? And If so, why? No, Cyc uses the current project only for bookkeeping meta assertions about new terms and assertions created for that project. You can create any number of projects for your own purposes by creating a term similar to OpenCycProject. > > 2. Shell I use implementation costants or whatever for the applications > using the KB? Shell I use them anyway? No, ordinarily you would not designate your own terms as implementation constants as they are not part of the commonsense contents of the KB, but only serve as vocabulary to support Cyc behavior such as natural language paraphrasing. -Steve > Thanks very much indeed for any help. > > Best regards > -- =========================================================== Stephen L. Reed phone: 512.342.4036 Cycorp, Suite 100 fax: 512.342.4040 3721 Executive Center Drive email: re...@cy... Austin, TX 78731 web: http://www.cyc.com download OpenCyc at http://www.opencyc.org =========================================================== |
From: Piermarco B. <pm...@cs...> - 2003-03-26 11:06:31
|
Hi, I'm not sure I have understood how to relate my own costants to the Cyc=20 opensource framework. To better let you understand, I am working on a=20 research project involving knowledge representetation and anchoring of=20 symbols to semantics, and so I need to create my own costants in order=20 to build a new KB capturing the problem domain. Now, if I want to make my KB a "plug in" of the OpenCyc framework, I've=20 got the following questions: 1. Do I have to set the current project to OpenCycProject? And If so, why= ? 2. Shell I use implementation costants or whatever for the applications=20 using the KB? Shell I use them anyway? Thanks very much indeed for any help. Best regards --=20 Piermarco Burrafato Ph.D. Student DINFO Universit=E0 degli Studi Viale delle Scienze 90128 Palermo, ITALY Phone (+39)091-6615273 Fax (+39)091-488452 EMail pm...@cs... |
From: Kevin B. <co...@co...> - 2003-03-17 23:31:57
|
Sounds good. I'm not sure what method it's complaining about, because it appears all the methods on the line in question are defined... I fell back to using CycListParser in IrcChat instead of CycListKifParser for the time being. I'll try out the changes when you update, thanks. -- Kevin "Douglas Miles (Germany)" wrote: > > > Hi, > > About IrcChat.java I have been making changes for 0.7 I didn't know if > these are the ones you used from CVS. > I re-did some of the code to make it more compatible. I wasn't sure one > last CVS commit if I committed the CycListKifParser.java but over this > weekend will straighten it out in case not. I figure by Monday/Tuesday > I'll have the updates all in sync.. Sorry if it was the fact that the > two java files where out of sync (why you might get method invocation > missing error) (The advancements being made are along the lines of > paraphrasing) > > -Douglas > > -----Original Message----- > From: ope...@li... > [mailto:ope...@li...] On Behalf Of Kevin > Brosius > Sent: Friday, March 14, 2003 3:00 PM > To: OpenCyc-devel > Subject: [OpenCyc-devel] Assertions in English / cvs IrcChat > > I've got a pair of questions... > > The FAQ from the website gives an example of a simple query (#$genls > #$Dog ?X) which works fine in OpenCyc 0.7.0. Then it mentions the > ability to turn on "show asserstions in English". Is this expected to > work for the HexalateralObject genls mentioned? When I try it, and > click on that genls in the interface, the OpenCyc instance prints a > bunch of: > > CYC(1): Warning: (PPH error level 1) Recursion limit exceeded. > Warning: (PPH error level 1) Recursion limit exceeded. > Warning: (PPH error level 1) Recursion limit exceeded. > > (14 lines total) > > Then I only see: > > English Translation : > If (#$isa ?OBJ #$Dog), then (#$isa ?OBJ #$HexalateralObject). > > In the web interface, plus some justification strings (isa's) of the > same form. > > ------ > > Second question: What's the state of IrcChat in cvs? I've tried it > out, and most queries seem to return something like: > > <Cobra> query #$genls #$Dog ?X > <Cycbot> Using microtheory: #$CobraChatMt > <Cycbot> for assertions until Cobra types "mt <something>" <Cycbot> > org.opencyc.api.CycApiException: (S#24420 :MESSAGE "A SYMBOL at > 0x44010150 is not a f_6739.") <Cycbot> request: (cyc-query 'null > #$CobraChatMt) > > If I pass a: > <Cobra> query (#$genls #$Dog ?X) > <-- Cycbot has quit (Remote closed the connection) > > Then IrcChat dies with the following trace: > > token: 'query' params: '(#$genls #$Dog ?X)' > Exception in thread "main" java.lang.NoSuchMethodError > at > org.opencyc.kif.CycListKifParser.reduceDottedPairs(CycListKifParser.java > :516) > at > org.opencyc.kif.CycListKifParser.read(CycListKifParser.java:177) > at > org.opencyc.kif.CycListKifParser.read(CycListKifParser.java:106) > at org.opencyc.chat.IrcChat.toCycListString(IrcChat.java:747) > at org.opencyc.chat.IrcChat.serviceQueryUser(IrcChat.java:662) > at org.opencyc.chat.IrcChat.serviceToken(IrcChat.java:536) > at > org.opencyc.chat.IrcChat.servicePublicMessage(IrcChat.java:440) > at > org.opencyc.chat.IrcChat.serviceIRCTransaction(IrcChat.java:418) > at org.opencyc.chat.IrcChat.serviceIRCSession(IrcChat.java:396) > at org.opencyc.chat.IrcChat.serviceIRCServer(IrcChat.java:367) > at org.opencyc.chat.IrcChat.serviceLoop(IrcChat.java:227) > at org.opencyc.chat.IrcChat.run(IrcChat.java:193) > at org.opencyc.chat.IrcChat.main(IrcChat.java:113) > > This may be my fault, related to the build. Has anyone else tried > recent cvs and queries with OpenCyc 0.7.0? > > Thanks, > -- > Kevin |
From: Douglas M. \(Germany\) <dm...@is...> - 2003-03-15 13:14:39
|
Hi, About IrcChat.java I have been making changes for 0.7 I didn't know if these are the ones you used from CVS. I re-did some of the code to make it more compatible. I wasn't sure one last CVS commit if I committed the CycListKifParser.java but over this weekend will straighten it out in case not. I figure by Monday/Tuesday I'll have the updates all in sync.. Sorry if it was the fact that the two java files where out of sync (why you might get method invocation missing error) (The advancements being made are along the lines of paraphrasing) -Douglas -----Original Message----- From: ope...@li... [mailto:ope...@li...] On Behalf Of Kevin Brosius Sent: Friday, March 14, 2003 3:00 PM To: OpenCyc-devel Subject: [OpenCyc-devel] Assertions in English / cvs IrcChat I've got a pair of questions... The FAQ from the website gives an example of a simple query (#$genls #$Dog ?X) which works fine in OpenCyc 0.7.0. Then it mentions the ability to turn on "show asserstions in English". Is this expected to work for the HexalateralObject genls mentioned? When I try it, and click on that genls in the interface, the OpenCyc instance prints a bunch of: CYC(1): Warning: (PPH error level 1) Recursion limit exceeded. Warning: (PPH error level 1) Recursion limit exceeded. Warning: (PPH error level 1) Recursion limit exceeded. (14 lines total) Then I only see: English Translation : If (#$isa ?OBJ #$Dog), then (#$isa ?OBJ #$HexalateralObject). In the web interface, plus some justification strings (isa's) of the same form. ------ Second question: What's the state of IrcChat in cvs? I've tried it out, and most queries seem to return something like: <Cobra> query #$genls #$Dog ?X <Cycbot> Using microtheory: #$CobraChatMt <Cycbot> for assertions until Cobra types "mt <something>" <Cycbot> org.opencyc.api.CycApiException: (S#24420 :MESSAGE "A SYMBOL at 0x44010150 is not a f_6739.") <Cycbot> request: (cyc-query 'null #$CobraChatMt) If I pass a: <Cobra> query (#$genls #$Dog ?X) <-- Cycbot has quit (Remote closed the connection) Then IrcChat dies with the following trace: token: 'query' params: '(#$genls #$Dog ?X)' Exception in thread "main" java.lang.NoSuchMethodError at org.opencyc.kif.CycListKifParser.reduceDottedPairs(CycListKifParser.java :516) at org.opencyc.kif.CycListKifParser.read(CycListKifParser.java:177) at org.opencyc.kif.CycListKifParser.read(CycListKifParser.java:106) at org.opencyc.chat.IrcChat.toCycListString(IrcChat.java:747) at org.opencyc.chat.IrcChat.serviceQueryUser(IrcChat.java:662) at org.opencyc.chat.IrcChat.serviceToken(IrcChat.java:536) at org.opencyc.chat.IrcChat.servicePublicMessage(IrcChat.java:440) at org.opencyc.chat.IrcChat.serviceIRCTransaction(IrcChat.java:418) at org.opencyc.chat.IrcChat.serviceIRCSession(IrcChat.java:396) at org.opencyc.chat.IrcChat.serviceIRCServer(IrcChat.java:367) at org.opencyc.chat.IrcChat.serviceLoop(IrcChat.java:227) at org.opencyc.chat.IrcChat.run(IrcChat.java:193) at org.opencyc.chat.IrcChat.main(IrcChat.java:113) This may be my fault, related to the build. Has anyone else tried recent cvs and queries with OpenCyc 0.7.0? Thanks, -- Kevin ------------------------------------------------------- This SF.net email is sponsored by:Crypto Challenge is now open! Get cracking and register here for some mind boggling fun and the chance of winning an Apple iPod: http://ads.sourceforge.net/cgi-bin/redirect.pl?thaw0031en _______________________________________________ OpenCyc-devel mailing list Ope...@li... https://lists.sourceforge.net/lists/listinfo/opencyc-devel |
From: Kevin B. <co...@co...> - 2003-03-14 23:03:03
|
I've got a pair of questions... The FAQ from the website gives an example of a simple query (#$genls #$Dog ?X) which works fine in OpenCyc 0.7.0. Then it mentions the ability to turn on "show asserstions in English". Is this expected to work for the HexalateralObject genls mentioned? When I try it, and click on that genls in the interface, the OpenCyc instance prints a bunch of: CYC(1): Warning: (PPH error level 1) Recursion limit exceeded. Warning: (PPH error level 1) Recursion limit exceeded. Warning: (PPH error level 1) Recursion limit exceeded. (14 lines total) Then I only see: English Translation : If (#$isa ?OBJ #$Dog), then (#$isa ?OBJ #$HexalateralObject). In the web interface, plus some justification strings (isa's) of the same form. ------ Second question: What's the state of IrcChat in cvs? I've tried it out, and most queries seem to return something like: <Cobra> query #$genls #$Dog ?X <Cycbot> Using microtheory: #$CobraChatMt <Cycbot> for assertions until Cobra types "mt <something>" <Cycbot> org.opencyc.api.CycApiException: (S#24420 :MESSAGE "A SYMBOL at 0x44010150 is not a f_6739.") <Cycbot> request: (cyc-query 'null #$CobraChatMt) If I pass a: <Cobra> query (#$genls #$Dog ?X) <-- Cycbot has quit (Remote closed the connection) Then IrcChat dies with the following trace: token: 'query' params: '(#$genls #$Dog ?X)' Exception in thread "main" java.lang.NoSuchMethodError at org.opencyc.kif.CycListKifParser.reduceDottedPairs(CycListKifParser.java:516) at org.opencyc.kif.CycListKifParser.read(CycListKifParser.java:177) at org.opencyc.kif.CycListKifParser.read(CycListKifParser.java:106) at org.opencyc.chat.IrcChat.toCycListString(IrcChat.java:747) at org.opencyc.chat.IrcChat.serviceQueryUser(IrcChat.java:662) at org.opencyc.chat.IrcChat.serviceToken(IrcChat.java:536) at org.opencyc.chat.IrcChat.servicePublicMessage(IrcChat.java:440) at org.opencyc.chat.IrcChat.serviceIRCTransaction(IrcChat.java:418) at org.opencyc.chat.IrcChat.serviceIRCSession(IrcChat.java:396) at org.opencyc.chat.IrcChat.serviceIRCServer(IrcChat.java:367) at org.opencyc.chat.IrcChat.serviceLoop(IrcChat.java:227) at org.opencyc.chat.IrcChat.run(IrcChat.java:193) at org.opencyc.chat.IrcChat.main(IrcChat.java:113) This may be my fault, related to the build. Has anyone else tried recent cvs and queries with OpenCyc 0.7.0? Thanks, -- Kevin |
From: Rich M. <rd...@cf...> - 2003-02-26 00:40:41
|
>GA generation? Sure. But the *test* is the thing. How does one >'automate' that? The *test* of the assertions is a hard problem. By way of a response, I'm going to give more background on what I'm hoping to do, eventually working my way around to some possibilities for specific tests. This may help to answer your questions, but it should also be interesting as background. Actually, I'd love to hear about things that other OpenCyc developers are doing (or planning :-). My current project is an OS browser (initially aimed at Mac OS X) that is supported by a set of local (task-scheduling, etc.) daemons and a remote server. The idea is that the local machines collect and report OS data to the server; the server aggregates information, analyzes patterns, and (ultimately) models OS behavior. I should note, in passing, that I'm very pragmatic about the technology; human and mechanized input are both acceptable; I just want to build a model that will be useful to administrators, programmers, etc. Consider the problem of analyzing/documenting the relationships between the files and directories in Mac OS X. Including the Developer Tools, this comes to ~170K items. Neither human nor mechanized effort alone is likely to get much traction on this, but a combination should work: * Static information can be gleaned from the file tree (e.g., aliases, hard links, symbolic links, ownership information) and documentation (e.g., FILES and SEE ALSO references in manual pages). * Monitoring of process and file activity can yield a "genealogical graph" that ties process ancestry to file access, creation, etc. * Mechanical analysis can highlight "common practice" (e.g., which permission modes are the most common) and "typical events" (e.g., what is the normal range of activities of the printer subsystem). * Human specification of relationships (e.g., this lock file is part of that subsystem) and rules (e.g., section 1 man pages tend to document files in these "bin" directories) can fill in the gaps. The FreeBSD Browser explores a subset of these ideas. It combines static OS information with human-supplied relationships and rules in a "virtual graph" that describes OS files and manual pages. A CGI script can then do a (constrained, breadth-first) search of the local neighborhood of a specified item, finding sets of "closely-related items". The prototype does about as well as I might wish. It finds most of the things that I (a 20-year Unix programmer) might find, with few "false positives", in FAR less time. However, it has some deficiencies: * The (all-Perl) implementation is inflexible: It can only do the searches that are hard-wired into the CGI script. * Information on OS variations are ignored: the presence of a FILES reference in a Linux or OpenBSD man page does nothing to assist a FreeBSD search. * No dynamic information (e.g., file and process history) is used, let alone information from large numbers of similar machines. The version I'm currently developing should address these issues. I'm hopeful that OpenCyc can provide the server with a framework for posing questions, testing assertions, and generally modeling the OS. Let's say that a researcher (or a program :-) asserts that "files in this directory sub-tree are only created by a specified set of user/ program combinations". The information needed to test this statement may not be available in the server's knowledge base, but that's not a real problem; by issuing a request, it can get a bevy of machines to collect the needed information. As the data comes in, the assertion (excuse me; micro-theory :-) can be tuned to fit it more closely. Obviously, there's a lot of hand-waving above, but I'm confident that something along these lines is feasible. And, although I'm quite busy with infrastructure and user-interface issues, it isn't too early to start looking at how to approach the higher-level analysis... -r -- email: rd...@cf...; phone: +1 650-873-7841 http://www.cfcl.com/rdm - my home page, resume, etc. http://www.cfcl.com/Meta - The FreeBSD Browser, Meta Project, etc. http://www.ptf.com/dossier - Prime Time Freeware's DOSSIER series http://www.ptf.com/tdc - Prime Time Freeware's Darwin Collection |
From: Michael R. A. <mic...@ya...> - 2003-02-25 22:55:55
|
Rich, GA generation? Sure. But the *test* is the thing. How does one 'automate' that? The *test* of the assertions is a hard problem. Michael Roy Ames ----- Original Message ----- From: "Rich Morin" <rd...@cf...> To: <ope...@li...> Sent: Monday, February 24, 2003 9:09 PM Subject: [OpenCyc-devel] PBI: using GAs to generate (YA)CycL assertions? > I have a Partly-Baked Idea about using Genetic Algorithms to generate > CycL assertions. I could then test these assertions against my data > set and evaluate the validity of the statements (rinse; repeat :-). > > Comments, questions, and suggestions are all welcome... > > -r > -- > email: rd...@cf...; phone: +1 650-873-7841 > http://www.cfcl.com/rdm - my home page, resume, etc. > http://www.cfcl.com/Meta - The FreeBSD Browser, Meta Project, etc. > http://www.ptf.com/dossier - Prime Time Freeware's DOSSIER series > http://www.ptf.com/tdc - Prime Time Freeware's Darwin Collection > > > ------------------------------------------------------- > This sf.net email is sponsored by:ThinkGeek > Welcome to geek heaven. > http://thinkgeek.com/sf > _______________________________________________ > OpenCyc-devel mailing list > Ope...@li... > https://lists.sourceforge.net/lists/listinfo/opencyc-devel > |
From: Rich M. <rd...@cf...> - 2003-02-25 02:18:06
|
I have a Partly-Baked Idea about using Genetic Algorithms to generate CycL assertions. I could then test these assertions against my data set and evaluate the validity of the statements (rinse; repeat :-). Comments, questions, and suggestions are all welcome... -r -- email: rd...@cf...; phone: +1 650-873-7841 http://www.cfcl.com/rdm - my home page, resume, etc. http://www.cfcl.com/Meta - The FreeBSD Browser, Meta Project, etc. http://www.ptf.com/dossier - Prime Time Freeware's DOSSIER series http://www.ptf.com/tdc - Prime Time Freeware's Darwin Collection |
From: Yeb H. <yeb...@po...> - 2003-02-24 10:24:05
|
> The FAQ states: > > Yes, it is compatible with apache but also has its own built-in http > server (this is the default) -- it can use either > > Has anyone done this yet? Is there documentation available? Or is it in > the opencyc docs and I've just missed it? To be honest, I am not sure how to interpret "compatible with apache" (compatible with what standard? CGI? HTTP? Apache module API?). Since there are references to 'your www server and name of the cgi program', I assume 'compatible with apache' means 'compatible with the cgi interface'. But there is no cgi program anywhere in the distribution currently. > I'm running FreeBSD and would really like to avoid running linux emulation > for the built-in server. I'm not Java literate yet, but I'm more than > familiar with Apache. Does anyone have any suggestions? Still, if there would be a program like 'cg' (the default name of the cyc cgi program' in the distribution, it would be distributed as a linux binary, independent of the type of architecture your apache was compiled on. No programs run 'under Apache', since it is not an operating system. Also, I think, there would not only be a cgi program but also a server, like in the current distribution. So linux emulation is your only option, currently.. Have you tried: http://sourceforge.net/forum/message.php?msg_id=1539242 ? |
From: Stephen C. <cl...@ww...> - 2003-02-20 06:47:21
|
The FAQ states: Yes, it is compatible with apache but also has its own built-in http server (this is the default) -- it can use either Has anyone done this yet? Is there documentation available? Or is it in the opencyc docs and I've just missed it? I'm running FreeBSD and would really like to avoid running linux emulation for the built-in server. I'm not Java literate yet, but I'm more than familiar with Apache. Does anyone have any suggestions? -Stephen |
From: Stephen R. <re...@cy...> - 2003-01-23 18:12:29
|
Yeb, I opened a telnet session with OpenCyc and began a long request: > telnet localhost 3601 > (do-assertions (assertion)) Note that the above form iteratives over all the assertions in the OpenCyc KB taking no other action on them. Then in the OpenCyc console: CYC(10): (show-processes) Run 3601 server I/O Wait Port 3614 server dispatcher I/O Wait Port 3601 server dispatcher I/O Wait Port 3600 server dispatcher Sleep Cyc Agenda Run Background Run Initial NIL CYC(11): (kill-process-named "3601 server") The above form kills the api session. If you are using the java api in the default binary (cfasl) mode, then the process should be named "3614 server". At this time there is no distinction between several threads all having the same "3601 server" name. In a future OpenCyc release we will turn on the api task processor by default and it will at that time provide a way to kill a given api request without killing the api session. -Steve On Thu, 23 Jan 2003, Yeb Havinga wrote: > Hello all, > > How can I do the what cb-abort-inference does in the KB browser on the SubL > or the Java API? > <long motivation deleted> > > Thanks > Yeb > > > > > > ------------------------------------------------------- > This SF.NET email is sponsored by: > SourceForge Enterprise Edition + IBM + LinuxWorld = Something 2 See! > http://www.vasoftware.com > _______________________________________________ > OpenCyc-devel mailing list > Ope...@li... > https://lists.sourceforge.net/lists/listinfo/opencyc-devel > -- =========================================================== Stephen L. Reed phone: 512.342.4036 Cycorp, Suite 100 fax: 512.342.4040 3721 Executive Center Drive email: re...@cy... Austin, TX 78731 web: http://www.cyc.com download OpenCyc at http://www.opencyc.org =========================================================== |
From: Yeb H. <yeb...@po...> - 2003-01-23 17:23:47
|
Hello all, How can I do the what cb-abort-inference does in the KB browser on the SubL or the Java API? <long motivation deleted> Thanks Yeb |
From: Richard H. M. <rh...@cd...> - 2003-01-18 16:53:37
|
You might be interested in my Knowledge Explorer program http://rhm.cdepot.net/ which implements my knowledge representation language, MKR. Its knowledge base includes the definitions of its parameters & = commands. =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=20 Dick McCullough=20 knowledge :=3D man do identify od existent done knowledge haspart proposition list ----- Original Message -----=20 From: James Michael DuPont=20 To: ope...@li...=20 Sent: Friday, January 17, 2003 7:05 AM Subject: Re: Is this DAML export? (was Re: [OpenCyc-devel] new OpenCyc = FAQ) --- Yeb Havinga <yeb...@po...> wrote: > > Imagine if you could get all the opencyc java code inside of > opencyc > > then it would be a "reflective" system, or self aware? >=20 > It would only contain information about the java api, not the > inference > engine. I think it is very confusing to talk about 'self awareness' > this > way. yes. But it is a start.=20 The opencyc project does not want to release all code, but we have to start somewhere.=20 > > then you can refactor itself... the possibilites are endless. >=20 > There is always something funny about having a compiler compile > itself. But > the end result is still a compiler ;-) it is still a compiler. The trick is if you can add in features on the way. =3D=3D=3D=3D=3D James Michael DuPont http://introspector.sourceforge.net/ __________________________________________________ Do you Yahoo!? Yahoo! Mail Plus - Powerful. Affordable. Sign up now. http://mailplus.yahoo.com ------------------------------------------------------- This SF.NET email is sponsored by: Thawte.com - A 128-bit supercerts = will allow you to extend the highest allowed 128 bit encryption to all your = clients even if they use browsers that are limited to 40 bit = encryption.=20 http://ads.sourceforge.net/cgi-bin/redirect.pl?thaw0030en _______________________________________________ OpenCyc-devel mailing list Ope...@li... https://lists.sourceforge.net/lists/listinfo/opencyc-devel |
From: James M. D. <mdu...@ya...> - 2003-01-17 16:20:50
|
Sounds like a good plan. I have had very good experience with the usage of RDF for the storage : eg : http://introspector.sourceforge.net/2002/12/16/tmp_rdf-core.h_.tu__global__Mon_Dec_16_23_39_52_2002_38-dump.ntriple.gz --- John De Oliveira <jo...@cy...> wrote: > Mike, > > From what you're say about reflective systems, you might be > interested in > something David Whitten and I would like to do. We want to reify > (store in > the KB) all SubL API functions (from there, representing Java API > methods > would be a logical next step). I'm going to work with Steve Reed to > figure > out what we'd want to say about each function, and we'll together > propose > some standard vocabulary. I have been building a vocab based on the gcc, but it is stored currently in the introspector meta-model. I would love to define a OWL/RDFS based ontology for it. > After representing a few functions, we were > going > to turn it over to David to begin representing hundreds of them. I'm > sure > he could use some help, if you're interested. > > -- John D. I am interested. Lisp like languages and SUBL would be a great addition. mike ===== James Michael DuPont http://introspector.sourceforge.net/ __________________________________________________ Do you Yahoo!? Yahoo! Mail Plus - Powerful. Affordable. Sign up now. http://mailplus.yahoo.com |
From: John De O. <jo...@cy...> - 2003-01-17 15:57:49
|
Mike, From what you're say about reflective systems, you might be interested in something David Whitten and I would like to do. We want to reify (store in the KB) all SubL API functions (from there, representing Java API methods would be a logical next step). I'm going to work with Steve Reed to figure out what we'd want to say about each function, and we'll together propose some standard vocabulary. After representing a few functions, we were going to turn it over to David to begin representing hundreds of them. I'm sure he could use some help, if you're interested. -- John D. At 07:05 AM 1/17/2003 -0800, James Michael DuPont wrote: >--- Yeb Havinga <yeb...@po...> wrote: > > > Imagine if you could get all the opencyc java code inside of > > opencyc > > > then it would be a "reflective" system, or self aware? > > > > It would only contain information about the java api, not the > > inference > > engine. I think it is very confusing to talk about 'self awareness' > > this > > way. > >yes. But it is a start. >The opencyc project does not want to release all code, >but we have to start somewhere. > > > > then you can refactor itself... the possibilites are endless. > > > > There is always something funny about having a compiler compile > > itself. But > > the end result is still a compiler ;-) > >it is still a compiler. The trick is if you can add in features on the >way. > > >===== >James Michael DuPont >http://introspector.sourceforge.net/ > >__________________________________________________ >Do you Yahoo!? >Yahoo! Mail Plus - Powerful. Affordable. Sign up now. >http://mailplus.yahoo.com > > >------------------------------------------------------- >This SF.NET email is sponsored by: Thawte.com - A 128-bit supercerts will >allow you to extend the highest allowed 128 bit encryption to all your >clients even if they use browsers that are limited to 40 bit encryption. >http://ads.sourceforge.net/cgi-bin/redirect.pl?thaw0030en >_______________________________________________ >OpenCyc-devel mailing list >Ope...@li... >https://lists.sourceforge.net/lists/listinfo/opencyc-devel |
From: James M. D. <mdu...@ya...> - 2003-01-17 15:05:17
|
--- Yeb Havinga <yeb...@po...> wrote: > > Imagine if you could get all the opencyc java code inside of > opencyc > > then it would be a "reflective" system, or self aware? > > It would only contain information about the java api, not the > inference > engine. I think it is very confusing to talk about 'self awareness' > this > way. yes. But it is a start. The opencyc project does not want to release all code, but we have to start somewhere. > > then you can refactor itself... the possibilites are endless. > > There is always something funny about having a compiler compile > itself. But > the end result is still a compiler ;-) it is still a compiler. The trick is if you can add in features on the way. ===== James Michael DuPont http://introspector.sourceforge.net/ __________________________________________________ Do you Yahoo!? Yahoo! Mail Plus - Powerful. Affordable. Sign up now. http://mailplus.yahoo.com |
From: Yeb H. <yeb...@po...> - 2003-01-17 14:57:48
|
> Imagine if you could get all the opencyc java code inside of opencyc > then it would be a "reflective" system, or self aware? It would only contain information about the java api, not the inference engine. I think it is very confusing to talk about 'self awareness' this way. > then you can refactor itself... the possibilites are endless. There is always something funny about having a compiler compile itself. But the end result is still a compiler ;-) |
From: James M. D. <mdu...@ya...> - 2003-01-17 14:11:58
|
--- Yeb Havinga <yeb...@po...> wrote: > > I am someone who places a greate value on the ability to compile. > > i see your point, but for me it would be a good place to start > learning > > the java code in compiling it, and extending it. > > Did you miss that I showed how to compile all files in > org/opencyc/xml ? no i did not, but that is not ant. i know that you can compile one file, but... that is not why you have an ant system. > > As for building OpenCyc.jar yourself, did you already finish building > > FIPA_OSv2_1_0.jar commons-collections.jar jdom.jar > rdf-api-2001-01-19.jar > dynamicjava.jar jena.jar xerces.jar > ViolinStrings.jar icu4j.jar jug.jar > jakarta-oro-2.0.3.jar junit.jar > > ? (I want screen dumps!) good point. But that is coming. if i want to use gcj then i need to compile it all. *everything* Take a look at my efforts to port gtk, from the bottom up. http://introspector.sourceforge.net/dia_win32.htm it might take a year or two, but gcj will benefit. I have a new tool that will extract the data out of the java code via the gcc. Imagine if you could get all the opencyc java code inside of opencyc then it would be a "reflective" system, or self aware? then you can refactor itself... the possibilites are endless. don't say, "but java reflection", because we are also talking here about the method bodies, and java does not reflect on that. mike ===== James Michael DuPont http://introspector.sourceforge.net/ __________________________________________________ Do you Yahoo!? Yahoo! Mail Plus - Powerful. Affordable. Sign up now. http://mailplus.yahoo.com |
From: Yeb H. <yeb...@po...> - 2003-01-17 13:59:57
|
> I am someone who places a greate value on the ability to compile. > i see your point, but for me it would be a good place to start learning > the java code in compiling it, and extending it. Did you miss that I showed how to compile all files in org/opencyc/xml ? As for building OpenCyc.jar yourself, did you already finish building FIPA_OSv2_1_0.jar commons-collections.jar jdom.jar rdf-api-2001-01-19.jar dynamicjava.jar jena.jar xerces.jar ViolinStrings.jar icu4j.jar jug.jar jakarta-oro-2.0.3.jar junit.jar ? (I want screen dumps!) Yeb |
From: James M. D. <mdu...@ya...> - 2003-01-17 13:48:06
|
--- Yeb Havinga <yeb...@po...> wrote: > > Thank you for the help with the export, > > that is what i wanted! > Did you get the file opencyc.daml? > > > can you please update the shell scripts that call export to pass > the > > right parameters? > > No, I don't have access to those files; the releases are built by > Cycorp. > > About compiling code; sure, the ant build script gives some errors > because > it's trying to compile the whole tree. > But when I enter: > :~/opencyc-0.7.0/src/org/opencyc/xml$ javac ExportDaml.java > security properties not found. using defaults. > ~/opencyc-0.7.0/src/org/opencyc/xml$ ls *.class > ExportDaml.class > > it works. (with the same classpath setting I mentioned earlier) And > the same > actually holds for every other source in .../opencyc.xml. There is no > need > to compile other stuff if you want to change things to the xml / daml > code. > Use the jars that are provided, it's for your own convenience ;-) > > greetings > Yeb > I am someone who places a greate value on the ability to compile. i see your point, but for me it would be a good place to start learning the java code in compiling it, and extending it. mike ===== James Michael DuPont http://introspector.sourceforge.net/ __________________________________________________ Do you Yahoo!? Yahoo! Mail Plus - Powerful. Affordable. Sign up now. http://mailplus.yahoo.com |
From: Yeb H. <yeb...@po...> - 2003-01-17 13:44:51
|
> Thank you for the help with the export, > that is what i wanted! Did you get the file opencyc.daml? > can you please update the shell scripts that call export to pass the > right parameters? No, I don't have access to those files; the releases are built by Cycorp. About compiling code; sure, the ant build script gives some errors because it's trying to compile the whole tree. But when I enter: :~/opencyc-0.7.0/src/org/opencyc/xml$ javac ExportDaml.java security properties not found. using defaults. ~/opencyc-0.7.0/src/org/opencyc/xml$ ls *.class ExportDaml.class it works. (with the same classpath setting I mentioned earlier) And the same actually holds for every other source in .../opencyc.xml. There is no need to compile other stuff if you want to change things to the xml / daml code. Use the jars that are provided, it's for your own convenience ;-) greetings Yeb |
From: James M. D. <mdu...@ya...> - 2003-01-17 12:01:42
|
> Question: why do you want to compile? I want to compile so that I can make changes to the opencyc java code. i would like to try and port the java code to use the gnu java compiler. i would like to be able to recompile the code. Thank you for the help with the export, that is what i wanted! can you please update the shell scripts that call export to pass the right parameters? thanks again for the help mike --- Yeb Havinga <yeb...@po...> wrote: > ----- Original Message ----- > From: "James Michael DuPont" <mdu...@ya...> > To: <ope...@li...> > Sent: Friday, January 17, 2003 12:21 PM > Subject: Re: [OpenCyc-devel] new OpenCyc FAQ > > > > thanks, > > I have tried a couple of times. Even got it to stare. > > there are two export classes, the batch files dont work, > > all of this has been reported. > > > > I will wait for the next release before testing again, > > i have deleted all the opencyc stuff for now. > > > > i am not able to compile the java code under linux. > > if you clean up the ant files, and test it under linux, then I will > > cross check it, but untill then, I dont have the time or experience > to > > fix the ant. > > > > mike > > Question: why do you want to compile? > My opencyc is in /home/httpd/opencyc-0.7.0 > > $ java -version > java version "1.3.1" > Java(TM) 2 Runtime Environment, Standard Edition (build > Blackdown-1.3.1-FCS) > Java HotSpot(TM) Client VM (build Blackdown-1.3.1-FCS, mixed mode) > > make CLASSPATH env variable > $for i in *.jar ; do export > CLASSPATH=$CLASSPATH:/home/httpd/opencyc-0.7.0/lib/$i ; done > > $java org/opencyc/xml/ExportEeldDaml all > security properties not found. using defaults. > Choosing KB selection: all > Exporting Entire KB subset > Getting terms from Cyc > Selected 13234 CycFort terms > Sorting 10398 CycConstant terms > ... > etc > > I don't know if this is what you wanted. > > Yeb ===== James Michael DuPont http://introspector.sourceforge.net/ __________________________________________________ Do you Yahoo!? Yahoo! Mail Plus - Powerful. Affordable. Sign up now. http://mailplus.yahoo.com |
From: Yeb H. <yeb...@po...> - 2003-01-17 11:50:40
|
----- Original Message ----- From: "James Michael DuPont" <mdu...@ya...> To: <ope...@li...> Sent: Friday, January 17, 2003 12:21 PM Subject: Re: [OpenCyc-devel] new OpenCyc FAQ > thanks, > I have tried a couple of times. Even got it to stare. > there are two export classes, the batch files dont work, > all of this has been reported. > > I will wait for the next release before testing again, > i have deleted all the opencyc stuff for now. > > i am not able to compile the java code under linux. > if you clean up the ant files, and test it under linux, then I will > cross check it, but untill then, I dont have the time or experience to > fix the ant. > > mike Question: why do you want to compile? My opencyc is in /home/httpd/opencyc-0.7.0 $ java -version java version "1.3.1" Java(TM) 2 Runtime Environment, Standard Edition (build Blackdown-1.3.1-FCS) Java HotSpot(TM) Client VM (build Blackdown-1.3.1-FCS, mixed mode) make CLASSPATH env variable $for i in *.jar ; do export CLASSPATH=$CLASSPATH:/home/httpd/opencyc-0.7.0/lib/$i ; done $java org/opencyc/xml/ExportEeldDaml all security properties not found. using defaults. Choosing KB selection: all Exporting Entire KB subset Getting terms from Cyc Selected 13234 CycFort terms Sorting 10398 CycConstant terms ... etc I don't know if this is what you wanted. Yeb > > > --- Yeb Havinga <yeb...@po...> wrote: > > A piece of the irc log you may find interesting: > > > > 08:06:44 <bediako> ping > > 08:06:59 <bediako> anybody home? > > 08:09:56 <bediako> I was wondering if anyone coudl outline the best > > method > > to affect a DAML export of the OpenCyc ontology > > 08:10:20 <bediako> I have successfully compiled the OpenCyc java API > > 08:10:51 <bediako> however I have had little success running the > > Export > > functionality > > 08:31:13 <bediako> ok never mind > > 08:31:27 <bediako> I realize now that "ALL" is required > > 08:31:50 <bediako> perhaps the java code could throw an exception if > > no > > parameter is specified? > > 08:47:21 <yepster> hello > > 08:47:51 <yepster> never tried to export daml, I'm afraid I can't > > help you > > 09:41:38 <bediako> thanks yepster > > 09:41:45 <bediako> I actually succeeded > > 09:42:21 <bediako> the problem I experience was due to the fact that > > I never > > specified the type of export I wanted to do > > 09:42:35 <bediako> when I specified all it worked well > > 09:43:00 <bediako> I would like to see a more infomative error thrown > > though > > 09:43:07 <bediako> :) > > 09:43:13 <yepster> yeah > > 09:43:46 <bediako> ok well off to lunch. see ya > > > > ----- Original Message ----- > > From: "James Michael DuPont" <mdu...@ya...> > > To: <ope...@li...> > > Sent: Friday, January 17, 2003 12:53 AM > > Subject: Re: [OpenCyc-devel] new OpenCyc FAQ > > > > > > > > > > --- John De Oliveira <jo...@cy...> wrote: > > > > I read through them, but I'm sure there are still > > > > some > > > > mistakes. So, if you let me know what you find, I'll make > > > > corrections. > > > > > > please add : > > > > > > How can i compile this java code under linux? > > > How can i export this daml under linux? > > > > > > > > ------------------------------------------------------- > > This SF.NET email is sponsored by: Thawte.com > > Understand how to protect your customers personal information by > > implementing > > SSL on your Apache Web Server. Click here to get our FREE Thawte > > Apache > > Guide: http://ads.sourceforge.net/cgi-bin/redirect.pl?thaw0029en > > _______________________________________________ > > OpenCyc-devel mailing list > > Ope...@li... > > https://lists.sourceforge.net/lists/listinfo/opencyc-devel > > > ===== > James Michael DuPont > http://introspector.sourceforge.net/ > > __________________________________________________ > Do you Yahoo!? > Yahoo! Mail Plus - Powerful. Affordable. Sign up now. > http://mailplus.yahoo.com > > > ------------------------------------------------------- > This SF.NET email is sponsored by: Thawte.com > Understand how to protect your customers personal information by implementing > SSL on your Apache Web Server. Click here to get our FREE Thawte Apache > Guide: http://ads.sourceforge.net/cgi-bin/redirect.pl?thaw0029en > _______________________________________________ > OpenCyc-devel mailing list > Ope...@li... > https://lists.sourceforge.net/lists/listinfo/opencyc-devel > |
From: James M. D. <mdu...@ya...> - 2003-01-17 11:21:35
|
thanks, I have tried a couple of times. Even got it to stare. there are two export classes, the batch files dont work, all of this has been reported. I will wait for the next release before testing again, i have deleted all the opencyc stuff for now. i am not able to compile the java code under linux. if you clean up the ant files, and test it under linux, then I will cross check it, but untill then, I dont have the time or experience to fix the ant. mike --- Yeb Havinga <yeb...@po...> wrote: > A piece of the irc log you may find interesting: > > 08:06:44 <bediako> ping > 08:06:59 <bediako> anybody home? > 08:09:56 <bediako> I was wondering if anyone coudl outline the best > method > to affect a DAML export of the OpenCyc ontology > 08:10:20 <bediako> I have successfully compiled the OpenCyc java API > 08:10:51 <bediako> however I have had little success running the > Export > functionality > 08:31:13 <bediako> ok never mind > 08:31:27 <bediako> I realize now that "ALL" is required > 08:31:50 <bediako> perhaps the java code could throw an exception if > no > parameter is specified? > 08:47:21 <yepster> hello > 08:47:51 <yepster> never tried to export daml, I'm afraid I can't > help you > 09:41:38 <bediako> thanks yepster > 09:41:45 <bediako> I actually succeeded > 09:42:21 <bediako> the problem I experience was due to the fact that > I never > specified the type of export I wanted to do > 09:42:35 <bediako> when I specified all it worked well > 09:43:00 <bediako> I would like to see a more infomative error thrown > though > 09:43:07 <bediako> :) > 09:43:13 <yepster> yeah > 09:43:46 <bediako> ok well off to lunch. see ya > > ----- Original Message ----- > From: "James Michael DuPont" <mdu...@ya...> > To: <ope...@li...> > Sent: Friday, January 17, 2003 12:53 AM > Subject: Re: [OpenCyc-devel] new OpenCyc FAQ > > > > > > --- John De Oliveira <jo...@cy...> wrote: > > > I read through them, but I'm sure there are still > > > some > > > mistakes. So, if you let me know what you find, I'll make > > > corrections. > > > > please add : > > > > How can i compile this java code under linux? > > How can i export this daml under linux? > > > > ------------------------------------------------------- > This SF.NET email is sponsored by: Thawte.com > Understand how to protect your customers personal information by > implementing > SSL on your Apache Web Server. Click here to get our FREE Thawte > Apache > Guide: http://ads.sourceforge.net/cgi-bin/redirect.pl?thaw0029en > _______________________________________________ > OpenCyc-devel mailing list > Ope...@li... > https://lists.sourceforge.net/lists/listinfo/opencyc-devel ===== James Michael DuPont http://introspector.sourceforge.net/ __________________________________________________ Do you Yahoo!? Yahoo! Mail Plus - Powerful. Affordable. Sign up now. http://mailplus.yahoo.com |
From: Yeb H. <yeb...@po...> - 2003-01-17 11:09:21
|
A piece of the irc log you may find interesting: 08:06:44 <bediako> ping 08:06:59 <bediako> anybody home? 08:09:56 <bediako> I was wondering if anyone coudl outline the best method to affect a DAML export of the OpenCyc ontology 08:10:20 <bediako> I have successfully compiled the OpenCyc java API 08:10:51 <bediako> however I have had little success running the Export functionality 08:31:13 <bediako> ok never mind 08:31:27 <bediako> I realize now that "ALL" is required 08:31:50 <bediako> perhaps the java code could throw an exception if no parameter is specified? 08:47:21 <yepster> hello 08:47:51 <yepster> never tried to export daml, I'm afraid I can't help you 09:41:38 <bediako> thanks yepster 09:41:45 <bediako> I actually succeeded 09:42:21 <bediako> the problem I experience was due to the fact that I never specified the type of export I wanted to do 09:42:35 <bediako> when I specified all it worked well 09:43:00 <bediako> I would like to see a more infomative error thrown though 09:43:07 <bediako> :) 09:43:13 <yepster> yeah 09:43:46 <bediako> ok well off to lunch. see ya ----- Original Message ----- From: "James Michael DuPont" <mdu...@ya...> To: <ope...@li...> Sent: Friday, January 17, 2003 12:53 AM Subject: Re: [OpenCyc-devel] new OpenCyc FAQ > > --- John De Oliveira <jo...@cy...> wrote: > > I read through them, but I'm sure there are still > > some > > mistakes. So, if you let me know what you find, I'll make > > corrections. > > please add : > > How can i compile this java code under linux? > How can i export this daml under linux? |