From jeff at bioinformatics.org Thu Mar 1 03:28:57 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:39:14 2006 Subject: [Pipet Users] BRAD: Re: Piper config errors.... References: <200102281306.NAA00308@schreyer.oberland.net> Message-ID: <3A9E0849.EA2FC8B@bioinformatics.org> Karl-Max Wagner wrote: > > --- ./piper_old/dl/vflow/interfaces/Makefile Sun Jan 7 00:12:39 2001 > +++ ./piper/dl/vflow/interfaces/Makefile Sat Jan 27 18:57:50 2001 > @@ -203,8 +203,8 @@ > > maintainer-clean-libtool: > > -vflowmodule$(SO): $(vflowmodule__SO__OBJECTS) $(vflowmodule__SO__DEPENDENCIES) > - @rm -f vflowmodule$(SO) > +vflowmodule0 11 10 6 4 3 2 1 0SO): $(vflowmodule__SO__OBJECTS) $(vflowmodule__ > + @rm -f vflowmodule0 11 10 6 4 3 2 1 0SO) > $(vflowmodule__SO__LINK) $(vflowmodule__SO__LDFLAGS) $(vflowmodule__SO_ > > tags: TAGS This is what I have from a build on the same day, Jan 27, dl/vflow/interfaces/Makefile, lines 203-208: ------------------------------------------------------- vflowmodule$(SO): $(vflowmodule__SO__OBJECTS) $(vflowmodule__SO__DEPENDENCIES) @rm -f vflowmodule$(SO) $(vflowmodule__SO__LINK) $(vflowmodule__SO__LDFLAGS) $(vflowmodule__SO__OBJECTS) $(vflowmodule__SO__LDADD) $(LIBS) tags: TAGS ------------------------------------------------------- I don't know what to "make" of it, pun intended. It sure looks like a problem with your version of autoconf or whatever. Jeff -- J.W. Bizzaro jeff@bioinformatics.org Director, Bioinformatics.org: The Open Lab http://bioinformatics.org/~jeff "All those scientists--they're all alike! They say they're working for us, but what they really want is to rule the world!" -- Angry Villager, Young Frankenstein -- From jeff at bioinformatics.org Thu Mar 1 03:52:02 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:39:15 2006 Subject: [Pipet Users] BRAD: Re: Piper config errors.... References: <200102281306.NAA00308@schreyer.oberland.net> <3A9E0849.EA2FC8B@bioinformatics.org> Message-ID: <3A9E0DB2.DE3A24E6@bioinformatics.org> "J.W. Bizzaro" wrote: > > This is what I have from a build on the same day, Jan 27, Woops, I meant Feb 27, so it wasn't the same day. But, you say the problem exists with the current code. Have you tried a fresh checkout, or are you doing an update in the build directory? Jeff -- J.W. Bizzaro jeff@bioinformatics.org Director, Bioinformatics.org: The Open Lab http://bioinformatics.org/~jeff "All those scientists--they're all alike! They say they're working for us, but what they really want is to rule the world!" -- Angry Villager, Young Frankenstein -- From chapmanb at arches.uga.edu Thu Mar 1 04:22:37 2001 From: chapmanb at arches.uga.edu (Brad Chapman) Date: Fri Feb 10 19:39:15 2006 Subject: [Pipet Users] Re: Piper config errors.... In-Reply-To: <3A9E0849.EA2FC8B@bioinformatics.org> References: <200102281306.NAA00308@schreyer.oberland.net> <3A9E0849.EA2FC8B@bioinformatics.org> Message-ID: <15006.5341.716309.679730@taxus.athen1.ga.home.com> Hey all; > > --- ./piper_old/dl/vflow/interfaces/Makefile Sun Jan 7 00:12:39 2001 > > +++ ./piper/dl/vflow/interfaces/Makefile Sat Jan 27 18:57:50 2001 > > @@ -203,8 +203,8 @@ > > > > maintainer-clean-libtool: > > > > -vflowmodule$(SO): > > +vflowmodule0 11 10 6 4 3 2 1 0SO): I have no clue. Looking at the CVS logs, the Makefile.am and configure.in for this haven't changed in 7 months, so I'd say if you're seeing differences within January it must be something locally that you changed. All I can say is, something is really messed up if you are getting "0 11 10 6 4 3 2 1" printed instead of "$(" :-). Anyways, this diff is in the maintainer-clean-libtool part of the makefile, so it shouldn't affect a build... All I can say is to try cleaning all of the cruft (Makefile.in's, Makefiles, etc.) and try redoing it with a clean ./autogen.sh and make, etc. Sorry I can't help more -- autoconf and friends can be very mysterious... Brad From Nicolas.Chauvat at logilab.fr Thu Mar 1 06:27:03 2001 From: Nicolas.Chauvat at logilab.fr (Nicolas Chauvat) Date: Fri Feb 10 19:39:15 2006 Subject: [Pipet Users] Re: Xirium | Tech | MUD-Shell In-Reply-To: <3A9DB156.C3DC56C@bioinformatics.org> Message-ID: > I attached a couple messages containing early discussions about an NLI for > Peep. Gary and I came up with the idea in January 2000. > > Narval guys, what do you think about using an NLI with your intelligent > assistant? I've implemented a tiny little silly "Chat" module for Narval in september 1999 (First Post!), it does things like: You say "get my news" It goes fetch web pages on the web and make your newspaper, then tell you when it's done. The implementation is nothing like NLI (it's pattern matching stuff), but that's the way to go :-) -- Nicolas Chauvat http://www.logilab.com - "Mais o? est donc Ornicar ?" - LOGILAB, Paris (France) From jeff at bioinformatics.org Thu Mar 1 07:01:04 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:39:15 2006 Subject: [Pipet Users] Re: Xirium | Tech | MUD-Shell References: Message-ID: <3A9E3A00.23EB6278@bioinformatics.org> Nicolas Chauvat wrote: > > The implementation is nothing like NLI (it's pattern matching stuff), but > that's the way to go :-) >From what I have read in various places, there are very few sophisticated NLI's. Most simply look for verb and objects and then ignore everything else. The effect is sentences like this: "Today I think that I would like to take that box and put it on that shelf." Are exactly the same as "sentences" like this: "box put (on) shelf" Think about it, when NLI was all the craze (for games like Zork), these programs were less than 256 kb. They couldn't be very complex, but they sure *SEEMED* to understand what you typed. It largely is a trick, but it works. Do you think there is some way that Narval and Piper could share an NLI "engine"? Jeff -- J.W. Bizzaro jeff@bioinformatics.org Director, Bioinformatics.org: The Open Lab http://bioinformatics.org/~jeff "All those scientists--they're all alike! They say they're working for us, but what they really want is to rule the world!" -- Angry Villager, Young Frankenstein -- From Nicolas.Chauvat at logilab.fr Thu Mar 1 07:13:18 2001 From: Nicolas.Chauvat at logilab.fr (Nicolas Chauvat) Date: Fri Feb 10 19:39:15 2006 Subject: [Pipet Users] Re: Xirium | Tech | MUD-Shell In-Reply-To: <3A9E3A00.23EB6278@bioinformatics.org> Message-ID: > > The implementation is nothing like NLI (it's pattern matching stuff), but > > that's the way to go :-) > > From what I have read in various places, there are very few sophisticated > NLI's. Most simply look for verb and objects and then ignore everything > else. The effect is sentences like this: > > "Today I think that I would like to take that box and put it on that shelf." > > Are exactly the same as "sentences" like this: > > "box put (on) shelf" > > Think about it, when NLI was all the craze (for games like Zork), these > programs were less than 256 kb. They couldn't be very complex, but they sure > *SEEMED* to understand what you typed. It largely is a trick, but it works. Sure. But if that's all it does, you don't get much and appear to do stupid things. Depends on what degree of "understanding" you want to exhibit. > Do you think there is some way that Narval and Piper could share an NLI > "engine"? Yes, it may even be called www.alicebot.org :-) It's one of the best NLI stuff available in free software I've seen so far. But I could dig other things out of my bookmarks stack (ThoughtTreasure maybe, hmm... were was that URL?) -- Nicolas Chauvat http://www.logilab.com - "Mais o? est donc Ornicar ?" - LOGILAB, Paris (France) From Nicolas.Chauvat at logilab.fr Thu Mar 1 07:41:06 2001 From: Nicolas.Chauvat at logilab.fr (Nicolas Chauvat) Date: Fri Feb 10 19:39:15 2006 Subject: [Pipet Users] Re: Xirium | Tech | MUD-Shell In-Reply-To: <3A9DB156.C3DC56C@bioinformatics.org> Message-ID: > I attached a couple messages containing early discussions about an NLI for > Peep. Gary and I came up with the idea in January 2000. > > Narval guys, what do you think about using an NLI with your intelligent > assistant? I read the attached e-mail and have of few references, as we integrated some speech recognition/synthesis stuff in Narval last summer, but didn't get much time to clean/distribute it since then. We used {c,k}voiceontrol for recognition and mbrola/festival for speech synthesis. That was fun, you could tell it your name and it would talk back saying "hello Nicolas, where do you want me to go today?", etc. That needs more work, though. How would you like to integrate the same kind of things with Piper ? -- Nicolas Chauvat http://www.logilab.com - "Mais o? est donc Ornicar ?" - LOGILAB, Paris (France) From jeff at bioinformatics.org Thu Mar 1 08:16:34 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:39:15 2006 Subject: [Pipet Users] Re: Xirium | Tech | MUD-Shell References: Message-ID: <3A9E4BB2.B8573806@bioinformatics.org> Nicolas Chauvat wrote: > > > Do you think there is some way that Narval and Piper could share an NLI > > "engine"? > > Yes, it may even be called www.alicebot.org :-) It's one of the best NLI > stuff available in free software I've seen so far. But I could dig other > things out of my bookmarks stack (ThoughtTreasure maybe, hmm... were was > that URL?) I have a ton of refs too. I'll post them to the list later. [speech rec/synth] > How would you like to integrate the same kind of things with Piper ? I'm not sure if you're asking "if I would like to" or "how I would do it". For the former, sure! :-) For the latter, Loci's original MVC design was developed largely because we wanted to do such things: NLI and speech recognition/synthesis. With MVC (Model, View, Control), the "control" is the (NLI) command from the user and the speech recognition, and the "view" is the response from the computer and speech synthesis. And every command gives a response. The neat thing is, with Piper's multiple UI's, the user can control one UI (e.g., Pied/Piper) with another (e.g., Peep/Piper). Imagine saying "connect this node to that" and then seeing it happen graphically! Now imaging connecting nodes graphically and getting a verbal confirmation! Providing we have the NLI, I think that connecting to the speech rec/synth API will be smaller task. Oh, and I do recognize the wonderful uses of speech with AI :-) Cheers. Jeff From barthgregory at netcourrier.com Thu Mar 1 06:29:03 2001 From: barthgregory at netcourrier.com (Barth Gregory) Date: Fri Feb 10 19:39:15 2006 Subject: [Pipet Users] End of year degree investigation Message-ID: <002d01c0a242$d2273520$719282c3@pandora.be> Skipped content of type multipart/related-------------- next part -------------- A non-text attachment was scrubbed... Name: Investigation.doc Type: application/msword Size: 96768 bytes Desc: not available Url : http://bioinformatics.org/pipermail/pipet-users/attachments/20010301/6248cd49/Investigation.doc From jeff at bioinformatics.org Thu Mar 1 17:32:37 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:39:15 2006 Subject: [Pipet Users] Re: End of year degree investigation References: <002d01c0a242$d2273520$719282c3@pandora.be> Message-ID: <3A9ECE05.DE7E1416@bioinformatics.org> Hi Barth. I wasn't able to open your attachment since some of that funky Microsoft formatting doesn't cut it in my browser. Anyway, Piper is *NOT* being developed by a company for commercial purposes. It is a Free Software project licensed with the GNU GPL license and developed by volunteers around the world. Please visit http://gnu.org for more information on the nature of such programs. The P2P field or "marketplace" is growing rapidly, but most programs are now commercial and non-Free. Piper is one of the few that are not. Cheers. Jeff From jarl at xs4all.nl Sat Mar 3 19:47:20 2001 From: jarl at xs4all.nl (Jarl van Katwijk) Date: Fri Feb 10 19:39:15 2006 Subject: [Pipet Users] End of year degree investigation References: <002d01c0a242$d2273520$719282c3@pandora.be> Message-ID: <3AA19098.9C08AFBC@xs4all.nl> Hi Barth Gregory, could you please re-post your questions as html or plain text? I'm afraid .doc attachments are not readable by most of the Piper team. jarl From jarl at xs4all.nl Sat Mar 3 20:10:09 2001 From: jarl at xs4all.nl (Jarl van Katwijk) Date: Fri Feb 10 19:39:15 2006 Subject: [Pipet Users] UI Piper layer WAS: Xirium | Tech | MUD-Shell References: <3A9E4BB2.B8573806@bioinformatics.org> Message-ID: <3AA195F1.8A3A6404@xs4all.nl> Hi Pipers, Untill the 7th post in this thread I kinda acted passively, scared of the huge amount of text I guess ;) I'm a bit puzzled about this thread, I'm missing the knowledge about the scope of Pied and that of the DL. I know the DL should offer 'an API for UI support', but it aint clear where the border lies. I also know that the 'official DL api' is non exsistent, but propably Jeff can explain me what is the current scope of Pied and the DL. What is done where for the frontend. I'm asking this because it seems a NMI is close to be a complete UI at its own. If this is true, the name UI could be generalized into CL, being Control Layer. Apart from being more clear it would make stuff like batch scripting be more @ home. 'UI' with some more explaination would be fine to me too though. jarl From jarl at xs4all.nl Sat Mar 3 21:40:19 2001 From: jarl at xs4all.nl (Jarl van Katwijk) Date: Fri Feb 10 19:39:15 2006 Subject: [Pipet Users] Overflow (PL) updates References: <3A9D309D.E40DFB9E@locusdialogue.com> Message-ID: <3AA1AB13.F6C922A@xs4all.nl> Hi Jean-Marc, nice work. Compiled Overflow (linux\gcc-2.95.2\glib2.2) and used it to compile Piper. Seems to work fine. I'll commit few Makefile-stuff changes to Piper cvs to sync that with Overflow cvs. I got these new though, any thoughs JM? ----- error: net LP already existed error: net HP already existed error: net LPC_DECOMP already existed ----- jarl From valj01 at gel.usherb.ca Sun Mar 4 23:26:45 2001 From: valj01 at gel.usherb.ca (Jean-Marc Valin) Date: Fri Feb 10 19:39:15 2006 Subject: [Pipet Users] Overflow (PL) updates References: <3A9D309D.E40DFB9E@locusdialogue.com> <3AA1AB13.F6C922A@xs4all.nl> Message-ID: <3AA31585.9A9B78D9@gel.usherb.ca> > nice work. Compiled Overflow (linux\gcc-2.95.2\glib2.2) and used it to > compile Piper. Seems to work fine. I'll commit few Makefile-stuff > changes to Piper cvs to sync that with Overflow cvs. Oh, and I forgot the last thing item in the changelog: There is now 3D Now! support in Overflow (for basic vector primitives), all you need to do is add -D_USE_3DNOW to your CXXFLAGS. I might also add SSE support eventually. > I got these new though, any thoughs JM? > ----- > error: net LP already existed > error: net HP already existed > error: net LPC_DECOMP already existed I see two (more likely) causes: 1) Conflict with a previously installed version of Overflow 2) You don't need to include $(prefix)/lib and $(prefix)/toolbox/... in your VFLOW_PATH environment variable, as of the last updates. In fact, if you didn't add any new (.n) node, you don't even need to set VFLOW_PATH. Jean-Marc -- Jean-Marc Valin Universite de Sherbrooke - Genie Electrique valj01@gel.usherb.ca From jarl at xs4all.nl Mon Mar 5 07:45:46 2001 From: jarl at xs4all.nl (Jarl van Katwijk) Date: Fri Feb 10 19:39:15 2006 Subject: [Pipet Users] Overflow (PL) updates References: <3A9D309D.E40DFB9E@locusdialogue.com> <3AA1AB13.F6C922A@xs4all.nl> <3AA31585.9A9B78D9@gel.usherb.ca> Message-ID: <3AA38A7A.200B3B80@xs4all.nl> > Oh, and I forgot the last thing item in the changelog: There is now 3D Now! > support in Overflow (for basic vector primitives), all you need to do is add > -D_USE_3DNOW to your CXXFLAGS. I might also add SSE support eventually. Sweet, AMD's first ;) > > I got these new though, any thoughs JM? > > ----- > > error: net LP already existed > > error: net HP already existed > > error: net LPC_DECOMP already existed > > I see two (more likely) causes: > > 1) Conflict with a previously installed version of Overflow > 2) You don't need to include $(prefix)/lib and $(prefix)/toolbox/... in your > VFLOW_PATH environment variable, as of the last updates. In fact, if you didn't > add any new (.n) node, you don't even need to set VFLOW_PATH. Did you tried latest Piper cvs code JM? Did you got the same 'errors' ? jarl From jeff at bioinformatics.org Sun Mar 11 11:25:52 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:39:15 2006 Subject: [Pipet Users] Medialogic - ML-View Message-ID: <3AABA710.7C324469@bioinformatics.org> This isn't exactly how I planned to run GUIs in Piper, but the possibilities are interesting: http://www.medialogic.it/projects/mlview/ Jeff From jeff at bioinformatics.org Sun Mar 11 11:41:53 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:39:15 2006 Subject: [Pipet Users] Open-source firm dips into services SOUP - Tech News - CNET.com Message-ID: <3AABAAD1.B37C32E1@bioinformatics.org> It looks like Ximian/Gnome will be integrating SOAP and .NET capabilities into the desktop: http://news.cnet.com/news/0-1003-200-5079895.html Jeff From jarl at xs4all.nl Sun Mar 11 15:53:43 2001 From: jarl at xs4all.nl (Jarl van Katwijk) Date: Fri Feb 10 19:39:15 2006 Subject: [Pipet Users] Open-source firm dips into services SOUP - Tech News - CNET.com References: <3AABAAD1.B37C32E1@bioinformatics.org> Message-ID: <3AABE5D7.25134005@xs4all.nl> "J.W. Bizzaro" wrote: > > It looks like Ximian/Gnome will be integrating SOAP and .NET capabilities into > the desktop: Anybody experienced with SOAP (http://www.w3.org/TR/SOAP/)? jarl From jeff at bioinformatics.org Sun Mar 11 16:53:58 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:39:15 2006 Subject: [Pipet Users] Open-source firm dips into services SOUP - Tech News - CNET.com References: <3AABAAD1.B37C32E1@bioinformatics.org> <3AABE5D7.25134005@xs4all.nl> Message-ID: <3AABF3F6.B2BAD9D0@bioinformatics.org> Jarl van Katwijk wrote: > > Anybody experienced with SOAP (http://www.w3.org/TR/SOAP/)? So, we're still looking for an RPC for Internet communications, assuming that CORBA isn't good at that, right? Is that why you're asking, Jarl? Jeff -- J.W. Bizzaro jeff@bioinformatics.org Director, Bioinformatics.org: The Open Lab http://bioinformatics.org/~jeff "All those scientists--they're all alike! They say they're working for us, but what they really want is to rule the world!" -- Angry Villager, Young Frankenstein -- From jarl at xs4all.nl Mon Mar 12 05:07:53 2001 From: jarl at xs4all.nl (Jarl van Katwijk) Date: Fri Feb 10 19:39:15 2006 Subject: [Pipet Users] Open-source firm dips into services SOUP - Tech News - CNET.com References: <3AABAAD1.B37C32E1@bioinformatics.org> <3AABE5D7.25134005@xs4all.nl> <3AABF3F6.B2BAD9D0@bioinformatics.org> Message-ID: <3AAC9FF9.33C05E31@xs4all.nl> > > Anybody experienced with SOAP (http://www.w3.org/TR/SOAP/)? > > So, we're still looking for an RPC for Internet communications, assuming that > CORBA isn't good at that, right? Is that why you're asking, Jarl? Right. The Fipa specs and SOAP seem to be two candidates. Just using Corba wont be sufficient, it might be the technology to implement the communications, but we still need the api\design on top of that. jarl From Nicolas.Chauvat at logilab.fr Mon Mar 12 09:05:18 2001 From: Nicolas.Chauvat at logilab.fr (Nicolas Chauvat) Date: Fri Feb 10 19:39:15 2006 Subject: [Pipet Users] Open-source firm dips into services SOUP - Tech News - CNET.com In-Reply-To: <3AABE5D7.25134005@xs4all.nl> Message-ID: > Anybody experienced with SOAP (http://www.w3.org/TR/SOAP/)? Guess what... Narval uses it for its UI/engine communication :-) We use www.pythonware.com's implementation. SOAP is slow... because parsing XML messages is slow and using a generic format for marshalling objects is verbose (ie much slower than Corba and its stub mechanism). But it works and looks like a good open-standard candidate. -- Nicolas Chauvat http://www.logilab.com - "Mais o? est donc Ornicar ?" - LOGILAB, Paris (France) From jeff at bioinformatics.org Mon Mar 12 09:37:01 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:39:16 2006 Subject: [Pipet Users] Open-source firm dips into services SOUP - Tech News -CNET.com References: Message-ID: <3AACDF0D.922BC877@bioinformatics.org> Nicolas Chauvat wrote: > > SOAP is slow... because parsing XML messages is slow and using a generic > format for marshalling objects is verbose (ie much slower than Corba and > its stub mechanism). But it works and looks like a good open-standard > candidate. We don't have to choose one or the other for everything. There's nothing wrong with: Local: CORBA (UIL <--> DL <--> BL <--> PL) Remote: SOAP or Fipa (BL <--> BL or DL <--> DL) Or is there? Jeff -- J.W. Bizzaro jeff@bioinformatics.org Director, Bioinformatics.org: The Open Lab http://bioinformatics.org/~jeff "All those scientists--they're all alike! They say they're working for us, but what they really want is to rule the world!" -- Angry Villager, Young Frankenstein -- From jmvalin at locusdialogue.com Mon Mar 12 10:02:23 2001 From: jmvalin at locusdialogue.com (Jean-Marc Valin) Date: Fri Feb 10 19:39:16 2006 Subject: [Pipet Users] Open-source firm dips into services SOUP - Tech News -CNET.com References: <3AACDF0D.922BC877@bioinformatics.org> Message-ID: <3AACE4FF.C0745EA1@locusdialogue.com> > We don't have to choose one or the other for everything. There's nothing > wrong with: > > Local: CORBA > (UIL <--> DL <--> BL <--> PL) > > Remote: SOAP or Fipa > (BL <--> BL or DL <--> DL) > > Or is there? If CORBA becomes only for local use, then why not use something like SysV IPC (shm, messages, ...) which is much faster. Jean-Marc From Nicolas.Chauvat at logilab.fr Mon Mar 12 12:08:41 2001 From: Nicolas.Chauvat at logilab.fr (Nicolas Chauvat) Date: Fri Feb 10 19:39:16 2006 Subject: [Pipet Users] Open-source firm dips into services SOUP - Tech News -CNET.com In-Reply-To: <3AACE4FF.C0745EA1@locusdialogue.com> Message-ID: > > Local: CORBA > > (UIL <--> DL <--> BL <--> PL) > > > > Remote: SOAP or Fipa > > (BL <--> BL or DL <--> DL) > > If CORBA becomes only for local use, then why not use something like > SysV IPC (shm, messages, ...) which is much faster. I'd vote for the former with the following modification: Local LAN = CORBA / Internet = SOAP CORBA's natural distribution is a plus. You can have several CORBA servers on the same host or on "close" hosts and it still works fine. That's harder to achieve with SysV IPC mechanisms. OTOH, SOAP is a HTTP-based protocol, that is easier to control and secure with common tools (ssl, etc.) and to let thru firewalls etc. Typically what you need over the internet. -- Nicolas Chauvat http://www.logilab.com - "Mais o? est donc Ornicar ?" - LOGILAB, Paris (France) From bizzaro at mercury.capeonramp.com Mon Mar 12 14:02:57 2001 From: bizzaro at mercury.capeonramp.com (bizzaro) Date: Fri Feb 10 19:39:16 2006 Subject: [Pipet Users] Open-source firm dips into services SOUP - Tech News Message-ID: <200103121402.AA1089077586@mercury.capeonramp.com> >I'd vote for the former with the following modification: > >Local LAN = CORBA / Internet = SOAP Are you saying that Piper <--> Piper communication should switch between CORBA and SOAP, depending on the type of network? Or, are you saying that CORBA and SOAP be used in separate parts of Piper, as I mentioned earlier? Jeff From jarl at xs4all.nl Mon Mar 12 16:51:03 2001 From: jarl at xs4all.nl (Jarl van Katwijk) Date: Fri Feb 10 19:39:16 2006 Subject: [Pipet Users] Open-source firm dips into services SOUP - Tech News -CNET.com References: <3AACDF0D.922BC877@bioinformatics.org> Message-ID: <3AAD44C7.CC2C015E@xs4all.nl> "J.W. Bizzaro" wrote: > > Nicolas Chauvat wrote: > > > > SOAP is slow... because parsing XML messages is slow and using a generic > > format for marshalling objects is verbose (ie much slower than Corba and > > its stub mechanism). But it works and looks like a good open-standard > > candidate. > > We don't have to choose one or the other for everything. There's nothing > wrong with: > > Local: CORBA > (UIL <--> DL <--> BL <--> PL) > > Remote: SOAP or Fipa > (BL <--> BL or DL <--> DL) > > Or is there? There always is, but very little compared to other designs. This is the setup I had in mind too. jarl From jarl at xs4all.nl Mon Mar 12 16:58:51 2001 From: jarl at xs4all.nl (Jarl van Katwijk) Date: Fri Feb 10 19:39:16 2006 Subject: [Pipet Users] Open-source firm dips into services SOUP - Tech News -CNET.com References: <3AACDF0D.922BC877@bioinformatics.org> <3AACE4FF.C0745EA1@locusdialogue.com> Message-ID: <3AAD469B.7F57B823@xs4all.nl> Jean-Marc Valin wrote: > > > We don't have to choose one or the other for everything. There's nothing > > wrong with: > > > > Local: CORBA > > (UIL <--> DL <--> BL <--> PL) > > > > Remote: SOAP or Fipa > > (BL <--> BL or DL <--> DL) > > > > Or is there? > > If CORBA becomes only for local use, then why not use something like > SysV IPC (shm, messages, ...) which is much faster. > Because corba offers that we are able to create the layers in any language and hosting hardware. Good corba implementations use shm e.a. for it's ipc. jarl From jeff at bioinformatics.org Mon Mar 12 17:06:07 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:39:16 2006 Subject: [Pipet Users] Open-source firm dips into services SOUP - Tech News -CNET.com References: <3AACDF0D.922BC877@bioinformatics.org> <3AACE4FF.C0745EA1@locusdialogue.com> <3AAD469B.7F57B823@xs4all.nl> Message-ID: <3AAD484F.A328615C@bioinformatics.org> Jarl van Katwijk wrote: > > Because corba offers that we are able to create the layers in any > language and hosting hardware. Good corba implementations use shm e.a. > for it's ipc. To echo Jarl's sentiment, we're concerned with pluggability. We lose that and modularity with some approaches. Jeff From Nicolas.Chauvat at logilab.fr Tue Mar 13 08:16:48 2001 From: Nicolas.Chauvat at logilab.fr (Nicolas Chauvat) Date: Fri Feb 10 19:39:16 2006 Subject: [Pipet Users] Open-source firm dips into services SOUP - Tech News In-Reply-To: <200103121402.AA1089077586@mercury.capeonramp.com> Message-ID: > >Local LAN = CORBA / Internet = SOAP > > Are you saying that Piper <--> Piper communication should switch > between CORBA and SOAP, depending on the type of network? Or, are you > saying that CORBA and SOAP be used in separate parts of Piper, as I > mentioned earlier? I hadn't thought about the former. I was agreeing with you on the latter. But now that you mention it... maybe it could be interesting to consider the former. BUT, it all depends on the expected message trafic. Our experience here is that SOAP tends to slow things down, so using as much CORBA as possible would be good for performance. OTOH, we'd better stick with a simple design at first (use CORBA and SOAP in separate parts) and improve efficiency later when needed. Piper's design is thought to be modular from the ground up, replacing a communication mechanism at some point in the future shouldn't be a big deal... don't you think? -- Nicolas Chauvat http://www.logilab.com - "Mais o? est donc Ornicar ?" - LOGILAB, Paris (France) From jeff at bioinformatics.org Tue Mar 13 08:39:23 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:39:16 2006 Subject: [Pipet Users] Open-source firm dips into services SOUP - Tech News References: Message-ID: <3AAE230B.F0303050@bioinformatics.org> Nicolas Chauvat wrote: > > OTOH, we'd better stick with a simple design at first (use CORBA and SOAP > in separate parts) and improve efficiency later when needed. I agree. > Piper's > design is thought to be modular from the ground up, replacing a > communication mechanism at some point in the future shouldn't be a big > deal... don't you think? The idea of "pluggable communications", in addition to pluggable applications, is very appealing. Jeff -- J.W. Bizzaro jeff@bioinformatics.org Director, Bioinformatics.org: The Open Lab http://bioinformatics.org/~jeff "All those scientists--they're all alike! They say they're working for us, but what they really want is to rule the world!" -- Angry Villager, Young Frankenstein -- From valj01 at gel.usherb.ca Wed Mar 21 15:27:47 2001 From: valj01 at gel.usherb.ca (Jean-Marc Valin) Date: Fri Feb 10 19:39:16 2006 Subject: [Pipet Users] Overflow 0.4 released Message-ID: <3AB90EC3.F310D42E@gel.usherb.ca> Hello, We're pleased to announce version 0.4.0 of Overflow (aka the Open Mind Speech environment, aka Piper PL). This new release brings a number of major improvements over version 0.3.0. Some of the changes are: * The VFLOW_PATH environment variable is no longer required (it's determined from the --prefix configure option) if you don't add your own nodes. * The FFTW library is no longer required to build Overflow. If you don't have it, the build will continue without FFT support. * It's now possible to have Feedback loops in a flow. This is done by using the "Feedback" node and a demo is available in the examples directory. * Overflow also supports throwing and catching exceptions within the flow using the "Throw" and "Catch" node. This is still undocumented. * Support for AMD 3D Now! instructions in vector primitives (use -D_USE_3DNOW) * Rewrite of the Neural network toolbox to make it faster * More stuff in the "examples" directory. * Updated html documentation * Better layout the links on the GUI (links can be broken lines) * As usual, a whole bunch of new node Overflow is a visual development environment, using a GNOME UI, that allows easy development of applications using toolboxes. Current toolboxes include signal processing, image processing, neural networks, vector quantization and more... Download, compile, enjoy, report bugs! From jarl at xs4all.nl Wed Mar 21 18:40:25 2001 From: jarl at xs4all.nl (Jarl van Katwijk) Date: Fri Feb 10 19:39:16 2006 Subject: [Pipet Users] Overflow 0.4 released References: <3AB90EC3.F310D42E@gel.usherb.ca> Message-ID: <3AB93BE9.54288D86@xs4all.nl> > * Overflow also supports throwing and catching exceptions within the > flow using the "Throw" and "Catch" node. This is still undocumented. What are the possibilities if these nodes, are they capable of reporting to other layers in piper? > * Support for AMD 3D Now! instructions in vector primitives (use -D_USE_3DNOW) How did you enable this, just a gcc flag or asm mmx code? > * Rewrite of the Neural network toolbox to make it faster Can I be tolded what the toolbox is ... or do I have to see it for myself? > * Updated html documentation ;-) Let me guess: no api changes since the cvs version you annouced? (the one where the naming of the libraries was done) jarl From valj01 at gel.usherb.ca Wed Mar 21 20:45:06 2001 From: valj01 at gel.usherb.ca (Jean-Marc Valin) Date: Fri Feb 10 19:39:16 2006 Subject: [Pipet Users] Overflow 0.4 released References: <3AB90EC3.F310D42E@gel.usherb.ca> <3AB93BE9.54288D86@xs4all.nl> Message-ID: <3AB95922.6A6305E7@gel.usherb.ca> > > * Overflow also supports throwing and catching exceptions within the > > flow using the "Throw" and "Catch" node. This is still undocumented. > What are the possibilities if these nodes, are they capable of reporting > to other layers in piper? These are the "Throw" and "Catch" nodes. "Throw" throws an exception, so there has to be a Catch node somewhere further in the flow. When "Catch" catches an exception, it "pulls" from the "CATCH" input... There's a demo of that in the "examples" directory. > > * Support for AMD 3D Now! instructions in vector primitives (use -D_USE_3DNOW) > How did you enable this, just a gcc flag or asm mmx code? Eventually, there should be a configure flag. For now, you need to add -D_USE_3DNOW to the CXXFLAGS environment variable, e.g. export CXXFLAGS='-O3 -D_USE_3DNOW -mpentiumpro' ./configure ..... If you want to see the actual 3D Now! code, look in data-flow/include/vec.h > > * Rewrite of the Neural network toolbox to make it faster > Can I be tolded what the toolbox is ... or do I have to see it for > myself? What do you mean? Are you asking what are neural networks? (It's an artificial intelligence thing that roughly approximates the way neurons work). > > * Updated html documentation > Let me guess: no api changes since the cvs version you annouced? (the > one where the naming of the libraries was done) No API changes... Jean-Marc -- Jean-Marc Valin Universite de Sherbrooke - Genie Electrique valj01@gel.usherb.ca From jeff at bioinformatics.org Thu Mar 22 17:10:14 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:39:16 2006 Subject: [Pipet Users] compiling with gcc 2.96 Message-ID: <3ABA7846.DE09AF6@bioinformatics.org> Pipers, We've tried compiling Piper and Overflow with GCC 2.96, the most recent GNU compiler. Unfortunately, we've encountered a number of problems that cannot be fixed at the moment. GCC 2.96 is an "experimental" compiler that (1) has some bugs and (2) is not fully ANSI C++ compatible. It is also the standard compiler for some Linux distributions. When it reaches 3.0, we believe that we will be able to once again compile with it, but, in the meantime, GCC 2.95 should continue to work. If anyone has successfully compiled Piper/Overflow with another compiler, please let us know. Cheers. Jeff -- J.W. Bizzaro jeff@bioinformatics.org Director, Bioinformatics.org: The Open Lab http://bioinformatics.org/~jeff "All those scientists--they're all alike! They say they're working for us, but what they really want is to rule the world!" -- Angry Villager, Young Frankenstein -- From jmvalin at locusdialogue.com Thu Mar 22 17:56:54 2001 From: jmvalin at locusdialogue.com (Jean-Marc Valin) Date: Fri Feb 10 19:39:16 2006 Subject: [Pipet Users] compiling with gcc 2.96 References: <3ABA7846.DE09AF6@bioinformatics.org> Message-ID: <3ABA8336.27C46CE8@locusdialogue.com> > We've tried compiling Piper and Overflow with GCC 2.96, the most recent GNU > compiler. Unfortunately, we've encountered a number of problems that cannot > be fixed at the moment. GCC 2.96 is an "experimental" compiler that (1) has > some bugs and (2) is not fully ANSI C++ compatible. It is also the standard > compiler for some Linux distributions. When it reaches 3.0, we believe that > we will be able to once again compile with it, but, in the meantime, GCC 2.95 > should continue to work. > > If anyone has successfully compiled Piper/Overflow with another compiler, > please let us know. Note: gcc 2.96 *almost* works, that is, it compiles and it only breaks one feature (AFAIK). I think egcs 1.1.2 does work too, although I don't have a machine to test. The last time I tried, gcc 3.0 only breaks the GUI (vflow) and it's my fault. Since Piper doesn't use vflow, it might be OK with 3.0. So, if you try and there's a compilation problem, let me know (BTW, always use -O2 or -O3). Jean-Marc From jeff at bioinformatics.org Thu Mar 22 21:57:36 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:39:16 2006 Subject: [Pipet Users] Overflow 0.4 released References: <3AB90EC3.F310D42E@gel.usherb.ca> Message-ID: <3ABABBA0.F14D62D3@bioinformatics.org> Jean-Marc Valin wrote: > > We're pleased to announce version 0.4.0 of Overflow (aka the Open Mind Speech > environment, aka Piper PL). FYI, Overflow packages are kept in FTP at SourceForge.net: ftp://ftp1.sourceforge.net/pub/sourceforge/freespeech/ There are RPM's made for version 0.4.0! We also keep a mirror of Overflow packages at Bioinformatics.org: ftp://bioinformatics.org/pub/Overflow/ (new directory) Plus, as mentioned before, we make daily snapshots of Overflow from CVS and put them in FTP: ftp://bioinformatics.org/pub/Overflow/CVS/ Cheers. Jeff -- J.W. Bizzaro jeff@bioinformatics.org Director, Bioinformatics.org: The Open Lab http://bioinformatics.org/~jeff "All those scientists--they're all alike! They say they're working for us, but what they really want is to rule the world!" -- Angry Villager, Young Frankenstein -- From jeff at bioinformatics.org Fri Mar 23 01:32:18 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:39:16 2006 Subject: [Pipet Users] Natural Programming and Integration Message-ID: <3ABAEDF2.6DC6ED9E@bioinformatics.org> It looks like our friends at dLoo are still alive and well: http://bluebox.sourceforge.net/technology/index.html We were at one point considering using the BlueBox API for creating GUI's. It's something like GLADE XML. Unfortunately, they wanted to part ways for now. Jeff From nile at nest.dloo.com Sun Mar 25 21:44:35 2001 From: nile at nest.dloo.com (nile@nest.dloo.com) Date: Fri Feb 10 19:39:16 2006 Subject: [Pipet Users] Natural Programming and Integration In-Reply-To: <3ABAEDF2.6DC6ED9E@bioinformatics.org> Message-ID: Hi Jeff, Not at all! We think Piper is cool! We don't want to part ways. Back in October, we just really needed to focus on coding BlueBox itself and realized that adding people (Mythical ManMonth style) would slow down the project. It was a hard decision, but we really needed to focus on the project. Please feel free to use BlueBox when it is released. It uses the word model -as opposed to the object model - which I think you'll find very useful for Piper. We're going to be releasing the C++ source - which can only create non-scripted interfaces) tommorow and putting the Python version in CVS. A *useful* developer version should be available at the end of April. It is being released under the BSD. Anyway, we were planning on emailing you tommorow when we released the source and thanking you for your patience as we got this together. Best of luck on Piper: I think you guys are working on something great. Nile P.S. I will be working hard on the source code the next week and responding to people, so there may be a small delay in replies. On Fri, 23 Mar 2001, J.W. Bizzaro wrote: > It looks like our friends at dLoo are still alive and well: > > http://bluebox.sourceforge.net/technology/index.html > > We were at one point considering using the BlueBox API for creating GUI's. > It's something like GLADE XML. Unfortunately, they wanted to part ways for > now. > > Jeff > > _______________________________________________ > pipet-users maillist - pipet-users@bioinformatics.org > http://bioinformatics.org/mailman/listinfo/pipet-users > From jeff at bioinformatics.org Mon Mar 26 01:09:20 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:39:16 2006 Subject: [Pipet Users] Natural Programming and Integration References: Message-ID: <3ABEDD10.BB4F459A@bioinformatics.org> Hi Nile! nile@nest.dloo.com wrote: > > Not at all! We think Piper is cool! We don't want to part ways. Back in > October, we just really needed to focus on coding BlueBox itself > and realized that adding people (Mythical ManMonth style) would slow down > the project. It was a hard decision, but we really needed to focus on the > project. Sehr gut! I was under the impression that you thought the Piper model might not be compatible with BlueBox's, not currently and possibly not ever. > Please feel free to use BlueBox when it is released. Documentation on your API (how GUI's are described and generated) would be most helpful. You didn't have it ready last fall. Nodes in Piper can describe GUI's (as meta XML), and, as you know, the act of combining nodes builds GUI's. Thus, in effect, Pied/Piper can function as a BlueBox SDK. Cheers. Jeff -- J.W. Bizzaro jeff@bioinformatics.org Director, Bioinformatics.org http://bioinformatics.org/~jeff "As we enjoy great advantages from the inventions of others, we should be glad of an opportunity to serve others by any invention of ours; and this we should do freely and generously." -- Benjamin Franklin -- From nile at nest.dloo.com Mon Mar 26 13:13:22 2001 From: nile at nest.dloo.com (nile@nest.dloo.com) Date: Fri Feb 10 19:39:16 2006 Subject: [Pipet Users] Natural Programming and Integration In-Reply-To: <3ABEDD10.BB4F459A@bioinformatics.org> Message-ID: Hi Jeff, Yep. Piper can be definitely be an SDK. As a quick warning, the BlueBox XMLGUI is not finished. It does not have Python scripting at the moment, not till the end of April. cheers, Nile On Mon, 26 Mar 2001, J.W. Bizzaro wrote: > Hi Nile! > > nile@nest.dloo.com wrote: > > > > Not at all! We think Piper is cool! We don't want to part ways. Back in > > October, we just really needed to focus on coding BlueBox itself > > and realized that adding people (Mythical ManMonth style) would slow down > > the project. It was a hard decision, but we really needed to focus on the > > project. > > Sehr gut! I was under the impression that you thought the Piper model might > not be compatible with BlueBox's, not currently and possibly not ever. > > > Please feel free to use BlueBox when it is released. > > Documentation on your API (how GUI's are described and generated) would be > most helpful. You didn't have it ready last fall. Nodes in Piper can > describe GUI's (as meta XML), and, as you know, the act of combining nodes > builds GUI's. Thus, in effect, Pied/Piper can function as a BlueBox SDK. > > Cheers. > Jeff > From frogerie at ilog.fr Tue Mar 27 08:37:02 2001 From: frogerie at ilog.fr (Flavio Rogerie) Date: Fri Feb 10 19:39:16 2006 Subject: [Pipet Users] More Informations Message-ID: <62EBE1BD35CBD311B93B009027936AF801270DE5@montrouge.ilog.fr> Hi, I am currently looking for informations on peer-to-peer. I have just watched your website and I would like to know the contact names & e-mails of Piper developers and project managers. I will appreciate if you could send me these informations. Thanks for your help. Kind regards Flavio ROGERIE