From jeff at bioinformatics.org Tue May 1 00:32:13 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:20:48 2006 Subject: [Pipet Devel] poster Message-ID: <3AEE3C4D.6E6B6A0@bioinformatics.org> FYI, I'm presenting a poster about Piper at the UMass Bioinformatics Conference on Tuesday. Wish me luck. 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 jeff at bioinformatics.org Wed May 2 17:34:44 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:20:48 2006 Subject: [Pipet Devel] name Message-ID: <3AF07D74.959EDC41@bioinformatics.org> Hi guys! I've been checking Freshmeat from time to time to see if the "other Piper" was given another name, as Erik promised. Well, it's still called "Piper". Jarl, perhaps you'd like to send Erik a reminder, since he's pretty mad at me now. Here's his address: emeusel@cs.uni-magdeburg.de BTW, if we have too many problems like this, how about changing Piper's name to Pijper? I don't see ANY programs with that name, just some Dutch sirnames. Cheers. Jeff From jarl at xs4all.nl Thu May 3 12:05:22 2001 From: jarl at xs4all.nl (Jarl van Katwijk) Date: Fri Feb 10 19:20:48 2006 Subject: [Pipet Devel] name References: <3AF07D74.959EDC41@bioinformatics.org> Message-ID: <3AF181C2.3EB996B1@xs4all.nl> Hela Jeff, How did the poster presentation go? > Jarl, perhaps you'd like to send Erik a reminder, since he's pretty mad at > me now. Here's his address: > > emeusel@cs.uni-magdeburg.de Mad at you? How come, I missed something? K, I'll contact him and get that piper project registered to us. > > BTW, if we have too many problems like this, how about changing Piper's name > to Pijper? I don't see ANY programs with that name, just some Dutch > sirnames. Lucky bastards ;-) Let's not take that name, stick to Piper. jarl From jeff at bioinformatics.org Wed May 2 18:17:44 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:20:49 2006 Subject: [Pipet Devel] name References: <3AF07D74.959EDC41@bioinformatics.org> <3AF181C2.3EB996B1@xs4all.nl> Message-ID: <3AF08788.7C5A7517@bioinformatics.org> Hi Jarl! Jarl van Katwijk wrote: > > How did the poster presentation go? Fine, thanks. It was a small meeting. I wanted to give a talk, but space was reserved for faculty. The meeting was for people in the 5-campus UMass system interested in bioinformatics. I've been hoping to convince people in the system that Piper could be used to connect bioinformatics resources between campuses. > Mad at you? How come, I missed something? Cuz he kept asking for a compromise, but I wouldn't agree to one, and then he seemed kind of pithy in the end. > K, I'll contact him and get that piper project registered to us. Thanks! > > BTW, if we have too many problems like this, how about changing Piper's name > > to Pijper? I don't see ANY programs with that name, just some Dutch > > sirnames. > Lucky bastards ;-) Let's not take that name, stick to Piper. Too embarrassing, eh? ;-) Jeff From jeff at bioinformatics.org Wed May 2 23:26:33 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:20:49 2006 Subject: [Pipet Devel] natural language processing Message-ID: <3AF0CFE9.73695064@bioinformatics.org> Hi Jeff! I read on the ISMB website that you're giving a tutorial on natural language processing. It's interesting to see that, because we've been talking about NLP for Piper. Basically, I've been thinking about a text-based interface that works like a first-person adventure game, where you give text commands to traverse nodes in Piper and do things with other objects at each node. For example, you might say "connect terminal 1 to terminal 2". And Piper would respond with text-based verifications of actions. The neat thing would be to integrate this with voice recognition and speech synthesis. Then, because Piper can use multiple interfaces (GUI and non-GUI), users could talk and see the results of commands graphically. We're also collaborating with some people developing an intelligent personal agent called Narval. They've been interested in connecting it to Piper to take advantage of the networking and the information stored in nodes. Interestingly, they too have thought about using NLP. With Narval/Piper, you might tell it to get some information for you, in a very abstract way, and then have the agent find what it can, possibly even building its own networks to get results. You can imagine the uses of this in bioinformatics! :-) Anyway, I thought that I'd let you know and suggest, if you're interested in developing an NLP bioinformatics system, that you might want to consider Narval/Piper :-) 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 jarl at xs4all.nl Thu May 3 22:17:38 2001 From: jarl at xs4all.nl (Jarl van Katwijk) Date: Fri Feb 10 19:20:49 2006 Subject: [Pipet Devel] Piper progress References: <3AF07D74.959EDC41@bioinformatics.org> <3AF181C2.3EB996B1@xs4all.nl> <3AF08788.7C5A7517@bioinformatics.org> Message-ID: <3AF21142.7D598909@xs4all.nl> Hi Jeff, I send an email to Mark of the other Piper project, hope to get this over with soon. I also released a new version of my personal tool Drip (http://drip.sourceforge.net), which is nearing a status where I will let it be for what it is. Drip is almost useable out-the-box for everyone, and it has all the features I need. After that, in a few weeks, I will finish the rewrite of the BL I started long ago ;) I have been thinking about what my work will be after I've done this BL work. I'll continue to work on it, but it wont be 100% Piper oriented: I'll focus on the features I started GMS once. This will result into about 50% work that will benefit the Piper goals, simply because mine and Piper's have so much in common. Issues that will not be within the Piper scope are not yet defined as such, and maybe will turn out to enrich Piper after all ;) hope to be more precise about my road plan soon, jarl From jeff at bioinformatics.org Thu May 3 15:21:42 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:20:49 2006 Subject: [Pipet Devel] Piper progress References: <3AF07D74.959EDC41@bioinformatics.org> <3AF181C2.3EB996B1@xs4all.nl> <3AF08788.7C5A7517@bioinformatics.org> <3AF21142.7D598909@xs4all.nl> Message-ID: <3AF1AFC5.2D8F1A35@bioinformatics.org> Jarl van Katwijk wrote: > > I send an email to Mark of the other Piper project Erik :-) > hope to get this > over with soon. I also released a new version of my personal tool Drip > (http://drip.sourceforge.net), which is nearing a status where I will > let it be for what it is. Drip is almost useable out-the-box for > everyone, and it has all the features I need. It looks very nice. I didn't know that you were into video encoding. > I have been thinking about what my work will be after I've done this BL > work. I'll continue to work on it, but it wont be 100% Piper oriented: > I'll focus on the features I started GMS once. Are you planning on creating a separate code base -- going back to "GMS"? > This will result into > about 50% work that will benefit the Piper goals, simply because mine > and Piper's have so much in common. Issues that will not be within the > Piper scope are not yet defined as such, and maybe will turn out to > enrich Piper after all ;) As long as you stick around, I'll be happy :-) I hope we can get some more developers interested soon. It can be hard though, when there are thousands of Open Source projects already out there. People are more interested in starting projects than joining them, or even finishing them. But, we'll see. Piper is interesting enough to garner a lot of attention. I think it is far more interesting than GNOME or KDE, and I wish other people could see that it is. This is why we need to make some announcements soon. Cheers. Jeff From jarl at xs4all.nl Fri May 4 10:02:33 2001 From: jarl at xs4all.nl (Jarl van Katwijk) Date: Fri Feb 10 19:20:49 2006 Subject: [Pipet Devel] Piper progress References: <3AF07D74.959EDC41@bioinformatics.org> <3AF181C2.3EB996B1@xs4all.nl> <3AF08788.7C5A7517@bioinformatics.org> <3AF21142.7D598909@xs4all.nl> <3AF1AFC5.2D8F1A35@bioinformatics.org> Message-ID: <3AF2B679.487459AB@xs4all.nl> > > I send an email to Mark of the other Piper project > Erik :-) oops.. ;-) Checked the outgoing email box: I did use Erik in the email I send to him, pfeeew! > It looks very nice. I didn't know that you were into video encoding. Just because I bough a DVDrom drive, I found there were no backup tools for non-windows around. So I wrote it. I've put way more features in it as I planned at first: much response from the 'community', so I kept on applying patches and working on it. > Are you planning on creating a separate code base -- going back to "GMS"? No, not going back to the stand alone gms code. I'm talking about putting more of the gms goals 'into' Piper. This dont even has to be in the Piper cvs\code three. Could be plugins orso. I like the situation of Piper where everybody has to comply to the specifications and discuss the features, etc. Back to GMS would mean back to coding alone and that doesn't is a good thing. > I hope we can get some more developers interested soon. It can be hard > though, when there are thousands of Open Source projects already out there. > People are more interested in starting projects than joining them, or even > finishing them. But, we'll see. Piper is interesting enough to garner a > lot of attention. I think it is far more interesting than GNOME or KDE, and > I wish other people could see that it is. This is why we need to make some > announcements soon. That's our dilemma: to few people to work on the Project, and to few Project body to get more people ;) Hope I'll be productive for Piper soon and get that framework up and running. jarl From jeff at bioinformatics.org Thu May 3 16:20:24 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:20:49 2006 Subject: [Pipet Devel] Piper progress References: <3AF07D74.959EDC41@bioinformatics.org> <3AF181C2.3EB996B1@xs4all.nl> <3AF08788.7C5A7517@bioinformatics.org> <3AF21142.7D598909@xs4all.nl> <3AF1AFC5.2D8F1A35@bioinformatics.org> <3AF2B679.487459AB@xs4all.nl> Message-ID: <3AF1BD87.7AD9C35A@bioinformatics.org> Jarl van Katwijk wrote: > > > Are you planning on creating a separate code base -- going back to "GMS"? > No, not going back to the stand alone gms code. I'm talking about > putting more of the gms goals 'into' Piper. This dont even has to be in > the Piper cvs\code three. Could be plugins orso. I like the situation of > Piper where everybody has to comply to the specifications and discuss > the features, etc. Back to GMS would mean back to coding alone and that > doesn't is a good thing. What are some of your quick thoughts on "pluggability" on the BL level? Recall that the Narval guys wanted to plug their system in as an AI layer. And you're talking about plugging other things into the BL. I would also like to see a pluggable interface that would allow users to swap Overflow for something else, like the SRS application that I sent a link to. Is there, or will there be, a well-defined API for plugging into the BL? Would it be CORBA-based? Is there much work left to have such an API? > That's our dilemma: to few people to work on the Project, and to few > Project body to get more people ;) Hope I'll be productive for Piper > soon and get that framework up and running. Well, I'm the biggest slacker here. I hope that will change soon (that I become more productive, not that everyone else becomes less ;-)). I'll be in Denmark around July 20 for an organizational meeting. I was thinking about a Piper meeting, but it would probably be only you and I (if you could go). Brad will be there, but I don't think that he will want to work on Piper. Gary won't be there. Jean-Marc hasn't said if he would go. And other people will likely not have the funds to go :-P Of course we can just start a hacking party and hope that people will join in. Cheers. Jeff From chapmanb at arches.uga.edu Thu May 3 17:48:25 2001 From: chapmanb at arches.uga.edu (Brad Chapman) Date: Fri Feb 10 19:20:49 2006 Subject: [Pipet Devel] Piper progress In-Reply-To: <3AF1BD87.7AD9C35A@bioinformatics.org> References: <3AF07D74.959EDC41@bioinformatics.org> <3AF181C2.3EB996B1@xs4all.nl> <3AF08788.7C5A7517@bioinformatics.org> <3AF21142.7D598909@xs4all.nl> <3AF1AFC5.2D8F1A35@bioinformatics.org> <3AF2B679.487459AB@xs4all.nl> <3AF1BD87.7AD9C35A@bioinformatics.org> Message-ID: <15089.53801.552129.657331@taxus.athen1.ga.home.com> Jeff, talking about coding in Denmark on July 20th: > Brad will be there, but I don't think that he will want to > work on Piper. Well, I'm definately not opposed to it, but the 20th is a bad day for me. That's the middle of BOSC, so that's planned Biopython/Biocorba coding time. If you're planning on doing it at some other time not during BOSC, I would definately be into it (sounds fun!), but I guess than we have the issue of having computers and all of that. Anyways, I also might not be the best person to work on it, as I haven't had time to play with Piper in a long while. It has even taken me way to long to get back with Jean-Marc about compiling Overflow on FreeBSD :-<. Brad From jeff at bioinformatics.org Thu May 3 18:48:26 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:20:49 2006 Subject: [Pipet Devel] Piper progress References: <3AF07D74.959EDC41@bioinformatics.org> <3AF181C2.3EB996B1@xs4all.nl> <3AF08788.7C5A7517@bioinformatics.org> <3AF21142.7D598909@xs4all.nl> <3AF1AFC5.2D8F1A35@bioinformatics.org> <3AF2B679.487459AB@xs4all.nl> <3AF1BD87.7AD9C35A@bioinformatics.org> <15089.53801.552129.657331@taxus.athen1.ga.home.com> Message-ID: <3AF1E03A.5E410531@bioinformatics.org> Hi Brad! Brad Chapman wrote: > > Well, I'm definately not opposed to it, but the 20th is a > bad day for me. That's the middle of BOSC, so that's planned > Biopython/Biocorba coding time. The 20th is the second/last day of BOSC. I was thinking about including Piper in the BOSC coding party, so that's why I mentioned that day. Are BioPython and BioCORBA being coded on both days or on different days? Does each project get one or two time-slots for coding? I suppose I could ask someone coordinating BOSC, but you're immediately available ;-) You're right, having computers would be a problem for any other time. But it would be fun if we could pull it off. The other thing that I have to contend with at the conference is a mini-conference for Bioinformatics.org. We're inviting several speakers to talk about open-access publications. We currently have someone from BioMed Central and the American Library Association coming, and we could get someone from the Public Library of Science. It'll be called "BiO BoF 2001", and it is currently unannounced. But stay tuned for it. 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 jarl at xs4all.nl Fri May 4 14:05:56 2001 From: jarl at xs4all.nl (Jarl van Katwijk) Date: Fri Feb 10 19:20:49 2006 Subject: [Pipet Devel] Piper progress References: <3AF07D74.959EDC41@bioinformatics.org> <3AF181C2.3EB996B1@xs4all.nl> <3AF08788.7C5A7517@bioinformatics.org> <3AF21142.7D598909@xs4all.nl> <3AF1AFC5.2D8F1A35@bioinformatics.org> <3AF2B679.487459AB@xs4all.nl> <3AF1BD87.7AD9C35A@bioinformatics.org> Message-ID: <3AF2EF83.10452A5E@xs4all.nl> > What are some of your quick thoughts on "pluggability" on the BL level? > Recall that the Narval guys wanted to plug their system in as an AI layer. In the current BL (the one currently in Piper cvs) there are 2 plugin API's: 1- communication plugins, like the DL and PL, both are linked to the BL by this api 2- controller plugins, like the admin GUI bundled but not very well working anymore since the migration from GMS. Both fully working and fully inhabited by bugs. PS In the rewrite there will be a similair system, only better build and with extra logic surrounding it. > And you're talking about plugging other things into the BL. I would also > like to see a pluggable interface that would allow users to swap Overflow > for something else, like the SRS application that I sent a link to. Already possible if you write little wrappers that make some external code a plugin. I wrote that Baby BL for this: took the PL, and wrapped it by around 200 lines of code. Maybe some more out the BL blackbox: SRS plugged into the BL is a BL node. And that makes a BL network a set of plugged software. > Would it be CORBA-based? Is there much work left to have such an API? Nope, no corba, the plugins are binaries. Corbatized plugins break too much: handling communications for corbatized plugins is impossible, and security is done with too. > Well, I'm the biggest slacker here. I hope that will change soon (that I > become more productive, not that everyone else becomes less ;-)). You'll get sorry about this promise ;) Anything on you mind to work on already? [kopenhagen - danmark] I could be 2-3 days around, no preferences about the date. Maybe 1 day before or after the actual BOSC? jarl From jeff at bioinformatics.org Thu May 3 20:38:27 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:20:49 2006 Subject: [Pipet Devel] Piper progress References: <3AF07D74.959EDC41@bioinformatics.org> <3AF181C2.3EB996B1@xs4all.nl> <3AF08788.7C5A7517@bioinformatics.org> <3AF21142.7D598909@xs4all.nl> <3AF1AFC5.2D8F1A35@bioinformatics.org> <3AF2B679.487459AB@xs4all.nl> <3AF1BD87.7AD9C35A@bioinformatics.org> <3AF2EF83.10452A5E@xs4all.nl> Message-ID: <3AF1FA03.DEC95F2E@bioinformatics.org> Jarl van Katwijk wrote: > > Maybe some more out the BL blackbox: SRS plugged into the BL is a BL > node. And that makes a BL network a set of plugged software. I don't know if I would represent a BL node as a node in the GUI though. What do you think? > > Would it be CORBA-based? Is there much work left to have such an API? > Nope, no corba, the plugins are binaries. Corbatized plugins break too > much: handling communications for corbatized plugins is impossible, and > security is done with too. No CORBA is used with the plugins? But you just said that the DL and PL are plugins, and the use CORBA to connect to the BL. Maybe I'm getting something wrong. > > Well, I'm the biggest slacker here. I hope that will change soon (that I > > become more productive, not that everyone else becomes less ;-)). > You'll get sorry about this promise ;) Anything on you mind to work on > already? Yeah, I know that I have a list of requests from you and Brad...that old :-( Any new requests? Actually, I've been working on an unrelated program on the side too, and that has been taking up some of my time. It's a bioinformatics tool written in Python. So, at least I'm getting some practice with the language. > [kopenhagen - danmark] > I could be 2-3 days around, no preferences about the date. Maybe 1 day > before or after the actual BOSC? Cool. Would you expect several people to commit to participating before you make the trip, or would you be willing to go just for little old me? ;-) 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 valj01 at gel.usherb.ca Thu May 3 23:05:41 2001 From: valj01 at gel.usherb.ca (Jean-Marc Valin) Date: Fri Feb 10 19:20:49 2006 Subject: [Pipet Devel] Piper progress References: <3AF07D74.959EDC41@bioinformatics.org> <3AF181C2.3EB996B1@xs4all.nl> <3AF08788.7C5A7517@bioinformatics.org> <3AF21142.7D598909@xs4all.nl> <3AF1AFC5.2D8F1A35@bioinformatics.org> <3AF2B679.487459AB@xs4all.nl> <3AF1BD87.7AD9C35A@bioinformatics.org> <15089.53801.552129.657331@taxus.athen1.ga.home.com> Message-ID: <3AF21C85.DEA1D554@gel.usherb.ca> > Anyways, I also might not be the best person to work on it, as I > haven't had time to play with Piper in a long while. It has even taken > me way to long to get back with Jean-Marc about compiling Overflow on > FreeBSD :-<. You might want to try the 0.4.2 version I just released. Also, something I forgot to mention in the e-mail is that I've added "wrappers" for Overflow. I have a C++ API (in wrapper.h) and C API (in cwrpper.h). As for Kopenhagen, I don't think I'll be there. However, we could probably arrange a coding session with me in Montreal. BTW, I'll be unavailable for the next week, since I'll be at ICASSP in Salt-Lake City (wanna meet there?!!). Jean-Marc -- Jean-Marc Valin Universite de Sherbrooke - Genie Electrique valj01@gel.usherb.ca From chapmanb at arches.uga.edu Fri May 4 15:36:24 2001 From: chapmanb at arches.uga.edu (Brad Chapman) Date: Fri Feb 10 19:20:49 2006 Subject: [Pipet Devel] Piper progress In-Reply-To: <3AF1E03A.5E410531@bioinformatics.org> References: <3AF07D74.959EDC41@bioinformatics.org> <3AF181C2.3EB996B1@xs4all.nl> <3AF08788.7C5A7517@bioinformatics.org> <3AF21142.7D598909@xs4all.nl> <3AF1AFC5.2D8F1A35@bioinformatics.org> <3AF2B679.487459AB@xs4all.nl> <3AF1BD87.7AD9C35A@bioinformatics.org> <15089.53801.552129.657331@taxus.athen1.ga.home.com> <3AF1E03A.5E410531@bioinformatics.org> Message-ID: <15091.1208.398131.10162@taxus.athen1.ga.home.com> Hi Jeff! > The 20th is the second/last day of BOSC. I was thinking about > including Piper in the BOSC coding party, so that's why I mentioned > that day. Are BioPython and BioCORBA being coded on both days or on > different days? Does each project get one or two time-slots for > coding? I have no idea how it will work -- I'm not even sure if there is a plan yet. The only thing I heard was that mornings were for talks and afternoons/evenings were for coding. I imagine it'll be pretty informal with the people who are into the projects doing most of the coding. You'd probably have to ask Ewan or someone in the know about the plans, but I don't even know if anything is set yet. > You're right, having computers would be a problem for any other > time. But it would be fun if we could pull it off. You might want to ask about how many computers they'll have before planning it -- there can only be a finite number of computers (hey, Suns are expensive you know!), so I don't know they'll be split up or whatever. Just something to think about. > It'll be called "BiO BoF 2001", and it is > currently unannounced. But stay tuned for it. Sounds fun, although with all of those alternating upper and lowercase letters, people might think you're talking about some new LaTeX project :-). Brad From jeff at bioinformatics.org Mon May 7 15:58:28 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:20:49 2006 Subject: [Pipet Devel] Juno papers Message-ID: <3AF6FE64.7A8A1ED3@bioinformatics.org> Yury sent me some papers about Juno's supercomputer plans (you may recall earlier messages about this; basically they want to collaborate with Bioinformatics.org on something, perhaps at some point involving Piper). It's over 1 MB, so if you want to see them, just ask, and I'll send them directly to you instead of the list. He didn't mention non-disclosure, so I'll assume that there are no secrets to worry about. Jeff From jarl at casema.net Tue May 8 07:04:18 2001 From: jarl at casema.net (Jarl van Katwijk) Date: Fri Feb 10 19:20:49 2006 Subject: [Pipet Devel] Juno papers References: <3AF6FE64.7A8A1ED3@bioinformatics.org> Message-ID: <3AF7D2B1.60EF6350@casema.net> Gimme one bartender ;-) Non disclosere <- I wont distribute it any futher, jarl > Yury sent me some papers about Juno's supercomputer plans (you may > recall earlier messages about this; basically they want to collaborate > with Bioinformatics.org on something, perhaps at some point involving > Piper). It's over 1 MB, so if you want to see them, just ask, and I'll > send them directly to you instead of the list. He didn't mention > non-disclosure, so I'll assume that there are no secrets to worry about. From gvd at penguin.pharmacy.ualberta.ca Tue May 8 18:33:34 2001 From: gvd at penguin.pharmacy.ualberta.ca (Gary Van Domselaar) Date: Fri Feb 10 19:20:49 2006 Subject: [Pipet Devel] Juno papers In-Reply-To: <3AF7D2B1.60EF6350@casema.net> Message-ID: Jeff, Could you put a copy in my home directory on the server? I wouldnt mind taking a look. Thnaks! g. -- -- Gary Van Domselaar Ph.D. Candidate, Associate Director, Faculty of Pharmacy, Bioinformatics.org: The Open Lab University of Alberta gary@bioinformatics.org gary@penguin.pharmacy.ualberta.ca http://bioinformatics.org/gary -- -- From jeff at bioinformatics.org Tue May 8 18:55:06 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:20:49 2006 Subject: [Pipet Devel] Juno papers References: Message-ID: <3AF8794A.B41DF687@bioinformatics.org> Gary Van Domselaar wrote: > > Could you put a copy in my home directory on the server? I wouldnt mind > taking a look. Okey dokey! Jeff From jchang at SMI.Stanford.EDU Thu May 10 18:08:59 2001 From: jchang at SMI.Stanford.EDU (Jeffrey Chang) Date: Fri Feb 10 19:20:49 2006 Subject: [Pipet Devel] Re: natural language processing In-Reply-To: <3AF0CFE9.73695064@bioinformatics.org> Message-ID: Hi Jeff, Sounds like Piper's moving right along. Congratulations! The plans you have in terms of NLP are really exciting, and I'm sure it's going to be an exciting problem. It sounds like the right thing to do to start, is to compile a list of "templates" for actions that people can type in. For example: add [named ] connect to ... A good start would be to require people to type or say things that exactly match a template. Then, you can expand it, so that it could learn different ways of expressing the information in the templates. As for the Narval stuff, to be able to handle more abstract requests, the system will have to understand the interrelationships between each of the objects. Do you have that functionality already in Piper? Hmmm... I'm starting to get carried away now. Unfortunately, I'm not going to have time to work more on that project for a while -- I'm trying to put together my thesis now. However, integrating with NLP with Piper sounds exciting, and I'll check in from time to time to see what's going on. Good luck! Jeff > From: "J.W. Bizzaro" > Date: Thu, 03 May 2001 03:26:33 +0000 > To: Jeffrey Chang > Cc: pipet-devel@bioinformatics.org > Subject: natural language processing > > Hi Jeff! > > I read on the ISMB website that you're giving a tutorial on natural language > processing. It's interesting to see that, because we've been talking about > NLP for Piper. Basically, I've been thinking about a text-based interface > that works like a first-person adventure game, where you give text commands to > traverse nodes in Piper and do things with other objects at each node. For > example, you might say "connect terminal 1 to terminal 2". And Piper would > respond with text-based verifications of actions. > > The neat thing would be to integrate this with voice recognition and speech > synthesis. Then, because Piper can use multiple interfaces (GUI and non-GUI), > users could talk and see the results of commands graphically. > > We're also collaborating with some people developing an intelligent personal > agent called Narval. They've been interested in connecting it to Piper to > take advantage of the networking and the information stored in nodes. > Interestingly, they too have thought about using NLP. With Narval/Piper, you > might tell it to get some information for you, in a very abstract way, and > then have the agent find what it can, possibly even building its own networks > to get results. You can imagine the uses of this in bioinformatics! :-) > > Anyway, I thought that I'd let you know and suggest, if you're interested in > developing an NLP bioinformatics system, that you might want to consider > Narval/Piper :-) > > 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 jeff at bioinformatics.org Sat May 12 00:39:27 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:20:49 2006 Subject: [Pipet Devel] name References: <3AF07D74.959EDC41@bioinformatics.org> <3AF181C2.3EB996B1@xs4all.nl> Message-ID: <3AFCBE7F.764A1990@bioinformatics.org> Jarl van Katwijk wrote: > > K, I'll contact him and get that piper project registered to us. Any luck? 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 jarl at xs4all.nl Sat May 12 05:24:10 2001 From: jarl at xs4all.nl (Jarl van Katwijk) Date: Fri Feb 10 19:20:49 2006 Subject: [Pipet Devel] name References: <3AF07D74.959EDC41@bioinformatics.org> <3AF181C2.3EB996B1@xs4all.nl> <3AFCBE7F.764A1990@bioinformatics.org> Message-ID: <3AFD013A.65C8C977@xs4all.nl> "J.W. Bizzaro" wrote: > > Jarl van Katwijk wrote: > > > > K, I'll contact him and get that piper project registered to us. > > Any luck? Not that much, got an email Erik had gad a buzy time, and he would do it as soon as possible, I send a reminder to him. jarl From jeff at bioinformatics.org Tue May 22 10:16:11 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:20:49 2006 Subject: [Pipet Devel] [Fwd: OiBC-2001] Message-ID: <3B0A74AB.B0B53809@bioinformatics.org> -------------- next part -------------- An embedded message was scrubbed... From: W_David_Benton@sbphrd.com Subject: OiBC-2001 Date: Tue, 22 May 2001 09:55:48 -0400 Size: 2569 Url: http://bioinformatics.org/pipermail/pipet-devel/attachments/20010522/86bf8cd4/attachment.mht From jeff at bioinformatics.org Mon May 28 10:45:20 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:20:49 2006 Subject: [Pipet Devel] ideas for a smaller piper core Message-ID: <3B126480.EB2B1F67@bioinformatics.org> Pipers, I've been thinking a bit about the direction that I'd like to see Piper head, particularly in the short-term where our problem is making a usable system right away (this will help attract users and more developers). I think to solve the usability problem, we should look at ways to get Piper compiled and running with less requirements. And the requirements to consider are the ones that are not for the core of Piper (the DL and the BL). These are the more elaborate UI's and PL's, such as Pied and Overflow. NOW, I'M NOT SAYING THAT WE SHOULD NOT SUPPORT THESE! I'm just saying that they should not be required to compile and run Piper. Regarding Pied, I would like to distribute it as a separate package. I've mentioned this before. I think that Peep, our console UI, should be the only one distributed with the core of Piper. This would remove the requirement of the GTK/GNOME and PyGTK libraries (although Jarl still uses GTK in the BL). I personally would like to see Piper used as a "better UNIX GUI", allowing CLI progs to be connected via GUI. It should be no surprise, since most of my announcements about Piper have mentioned this. Piper is, of course, a dataflow system, but there aren't as many systems that do what I'd like Piper to do with CLI progs, and I think that we'd get more users this way. In light of this, I'd like to rename Pied as "Piper Desktop". (There are some other Pied (Desktop) issues that I will address later.) Regarding Overflow and the "better UNIX GUI" idea, I'd like to have a very small PL distributed with the core of Piper that simply wraps CLI prorgams, per my discourse: http://www.bioinformatics.org/piper/documentation/command-compilation.html YES, I know that Overflow does this and has a better way of doing it, and I'd like to do it Overflow's way. I simply want this feature by default in Piper without requiring the use of a sophisticated data flow system. As for myself, I WOULD WANT to use Overflow, but perhaps many Piper users will not find the need for it. Also, I would like to see systems like SRS http://www.lionbioscience.com/eng/index_c_1_1_1.htm and Narval plug directly into the BL as PL's. In other words, we could have any number of PL's. We should produce some API docs for that. As Jarl mentioned to me, each PL, if not custom made for Piper, would require a small wrapper. (So, the code we now have in Piper for wrapping Overflow can also be separated.) So, what are your thoughts on this? Yes, no? 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 jeff at bioinformatics.org Mon May 28 10:50:43 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:20:49 2006 Subject: [Pipet Devel] ideas for a smaller piper core References: <3B126480.EB2B1F67@bioinformatics.org> Message-ID: <3B1265C3.D5DC7536@bioinformatics.org> "J.W. Bizzaro" wrote: > > We should produce some API docs for that. As Jarl > mentioned to me, each PL, if not custom made for Piper, would require a small > wrapper. Brad starting writing about this: http://www.bioinformatics.org/piper/documentation/piper_plugin.pdf We need to expand upon it. What has changed since then? 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 jmvalin at locusdialogue.com Mon May 28 11:07:43 2001 From: jmvalin at locusdialogue.com (Jean-Marc Valin) Date: Fri Feb 10 19:20:50 2006 Subject: [Pipet Devel] ideas for a smaller piper core References: <3B126480.EB2B1F67@bioinformatics.org> Message-ID: <3B1269BF.70EA4A1B@locusdialogue.com> > So, what are your thoughts on this? Yes, no? No. This will make matters even more complicated, since we'd have to support Piper without Overflow and Piper with Overflow. Besides that, I managed to get rid of most of Overflow dependencies. The only required library is now libxml, which is used by the DL anyway. It would also be easy to put the Overflow tree into the Piper tree for releases, so there's no big deal... Last thing I was not sure I understood from your previous mail: were you thinking of allowing another PL than Overflow or releasing Piper without a PL at all. Either way, I don't think I see the point... Jean-Marc From jeff at bioinformatics.org Mon May 28 13:14:23 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:20:50 2006 Subject: [Pipet Devel] ideas for a smaller piper core References: <3B126480.EB2B1F67@bioinformatics.org> <3B1269BF.70EA4A1B@locusdialogue.com> Message-ID: <3B12876F.4C5539B1@bioinformatics.org> Jean-Marc Valin wrote: > > No. This will make matters even more complicated, since we'd have to > support Piper without Overflow and Piper with Overflow. I don't think that it would mean supporting two code bases, as you're saying. If Overflow is a plugin to Piper, then, like all plugins, there should be no major modification required for Piper. > Besides that, I > managed to get rid of most of Overflow dependencies. The only required > library is now libxml, which is used by the DL anyway. Great! > It would also be > easy to put the Overflow tree into the Piper tree for releases, so > there's no big deal... Hmmm. I suppose we could pop Overflow in there for major releases. I guess we were thinking that people would have to go fetch it as a separate package. But it would require some changes to the build scripts. Would you like to help us change the build scripts to support Overflow options? > Last thing I was not sure I understood from your previous mail: were you > thinking of allowing another PL than Overflow or releasing Piper without > a PL at all. Either way, I don't think I see the point... I was thinking about a very small PL that would simply allow command-line options to be strung together as nodes in a network, like Overflow does, but that would be the only thing that this PL would do. The point? To let users use Piper for just command-line wrapping without the overhead of a dataflow system, also to make it more obvious how command-line wrapping is done. In any case, I want to emphasize that we don't want to drop support for Overflow or make it more difficult to run Piper with Overflow. Talking to some people about Piper over the past year, I've been frequently asked if Piper could support a certain existing system (e.g., some collaboratory), which would work in place of Overflow. And the question that I have to ask (of us) is "can Piper even work without Overflow?" Again, Jean-Marc, I'm not trying to ditch Overflow, so please don't take it the wrong way. Overflow can work without Piper, and I simply think that Piper users may want the same flexibility. Perhaps I'm just a raving lunatic. 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 jeff at bioinformatics.org Mon May 28 13:20:26 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:20:50 2006 Subject: [Pipet Devel] ideas for a smaller piper core References: <3B126480.EB2B1F67@bioinformatics.org> <3B1269BF.70EA4A1B@locusdialogue.com> <3B12876F.4C5539B1@bioinformatics.org> Message-ID: <3B1288DA.97B3649F@bioinformatics.org> Anyway, these changes would require everyone's full support. I won't make the mistake (again) of shoving my ideas down your throats. So, we'll get comments and take some votes. Jarl? Brad? Gary? Deanne? Dominic? 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 jeff at bioinformatics.org Mon May 28 13:38:16 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:20:50 2006 Subject: [Pipet Devel] [Fwd: OiBC-2001] References: <3B0A74AB.B0B53809@bioinformatics.org> Message-ID: <3B128D08.72FCC6F0@bioinformatics.org> I don't think that I sent the abstract to this list. It's attached. Jeff -------------- next part -------------- A non-text attachment was scrubbed... Name: bizzaro-abstract.pdf Type: application/pdf Size: 2444 bytes Desc: not available Url : http://bioinformatics.org/pipermail/pipet-devel/attachments/20010528/a0444ed5/bizzaro-abstract.pdf From jeff at bioinformatics.org Mon May 28 14:05:52 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:20:50 2006 Subject: [Pipet Devel] ideas for a smaller piper core References: <3B126480.EB2B1F67@bioinformatics.org> <3B1269BF.70EA4A1B@locusdialogue.com> <3B12876F.4C5539B1@bioinformatics.org> <3B1288DA.97B3649F@bioinformatics.org> Message-ID: <3B129380.BD2E7608@bioinformatics.org> Just yammering... This has nothing to do with liking or disliking any one part of Piper, and it has everything to do with modularity. Jarl once said that someone somewhere may want to swap out his BL for another one. Should Piper be modular enough to allow that? I think so. And I really don't care if something thinks that my Pied (Desktop) is sh*t and wants to use something else. I would even support putting the UIL, DL, BL and PL into separate CVS modules and distributing them as separate packages. In fact, let's just put that on the table. Some may argue that it becomes too confusing for users, but our users won't be complete novices anyway. 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 jmvalin at locusdialogue.com Mon May 28 14:21:25 2001 From: jmvalin at locusdialogue.com (Jean-Marc Valin) Date: Fri Feb 10 19:20:50 2006 Subject: [Pipet Devel] ideas for a smaller piper core References: <3B126480.EB2B1F67@bioinformatics.org> <3B1269BF.70EA4A1B@locusdialogue.com> <3B12876F.4C5539B1@bioinformatics.org> <3B1288DA.97B3649F@bioinformatics.org> <3B129380.BD2E7608@bioinformatics.org> Message-ID: <3B129725.489C2847@locusdialogue.com> > This has nothing to do with liking or disliking any one part of Piper, and it > has everything to do with modularity. Jarl once said that someone somewhere > may want to swap out his BL for another one. Should Piper be modular enough > to allow that? I think so. And I really don't care if something thinks that > my Pied (Desktop) is sh*t and wants to use something else. I've got one problem with that: interoperability. Since the idea of Piper is about distributed computing, you want (almost) the same Piper to run on all the machines. If you have 4 different DL's, 3 BL's and 5 PL's, you get 60 different Piper configurations and a total mess. It's impossible to do QA, and you can only use Piper on 1/60th of the machines... Modularity is nice, but to much of a good thing is bad. BTW, correct me if I'm wrong but your initial post was about a non-distributed Piper desktop... well for that, all you'd need is Overflow with (maybe) a more user-friendly GUI. ...and you don't even need the DL/BL. Those are for distributed computing. Jean-Marc From jarl at xs4all.nl Mon May 28 14:43:11 2001 From: jarl at xs4all.nl (Jarl van Katwijk) Date: Fri Feb 10 19:20:50 2006 Subject: [Pipet Devel] ideas for a smaller piper core References: <3B126480.EB2B1F67@bioinformatics.org> Message-ID: <3B129C3F.7D215796@xs4all.nl> Hi Jeff, Just a short inpression of me, I'll write a long detailed one later this evening, have to go out now. Basically I agree fully with you Jeff that Piper should get less dependancies, or put otherwise: Piper should requiere less external packages needed to install. A solution might be to bundle everything that's needed into Piper releases and make and install procedure that handles already availeble dependacies by not installing them. About the modulairity: to me Piper is just a set of specifications implemented by packages of software labeled 'Piper compliant' by us. There has to be an 'official' Piper package containing the default software for every layer (a GUI,DL,BL and PL). For now using Overflow++ for the PL seems most logical. Could change, like it can for every layer, when there are more implementations and we chose the 'best' whatever that might be. more on this later ;) Jarl From jeff at bioinformatics.org Mon May 28 14:54:43 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:20:50 2006 Subject: [Pipet Devel] ideas for a smaller piper core References: <3B126480.EB2B1F67@bioinformatics.org> <3B1269BF.70EA4A1B@locusdialogue.com> <3B12876F.4C5539B1@bioinformatics.org> <3B1288DA.97B3649F@bioinformatics.org> <3B129380.BD2E7608@bioinformatics.org> <3B129725.489C2847@locusdialogue.com> Message-ID: <3B129EF3.84188490@bioinformatics.org> Jean-Marc Valin wrote: > > If you have 4 different DL's, 3 BL's and 5 > PL's, you get 60 different Piper configurations and a total mess. That's a good point, and a good argument for standards. The WWW also runs off of numerous servers and clients, and yes there are a lot of compatibility problems, but there are some standards (e.g., HTML X.X) that make the Web very usable. > BTW, correct me if I'm wrong but your initial post was about a > non-distributed Piper desktop... well for that, all you'd need is > Overflow with (maybe) a more user-friendly GUI. ...and you don't even > need the DL/BL. Those are for distributed computing. True, I wouldn't need the DL and BL (or BL and PL, depending on how you look at it) if I didn't want to make a distributed desktop. But I do :-) 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 jeff at bioinformatics.org Mon May 28 15:25:42 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:20:50 2006 Subject: [Pipet Devel] ideas for a smaller piper core References: <3B126480.EB2B1F67@bioinformatics.org> <3B129C3F.7D215796@xs4all.nl> Message-ID: <3B12A636.C924CAA8@bioinformatics.org> Jarl van Katwijk wrote: > > About the modulairity: to me Piper is just a set of specifications > implemented by packages of software labeled 'Piper compliant' by us. I agree that "Piper" can be constructed using components that comply with a standard. I also understand that our packaging of UIL+DL+BL+PL together forms a standard (or reference), something like the X Windowing System reference distributed by the Open Group and how the Linux Standards Base might use Debian as a reference. But, separate packaging of the UIL+DL+BL+PL will not prevent us from calling them a standard/reference, providing we specify which components should be used for a standard/reference. I think that we should also have a documented standard. > Could change, like it can for every > layer, when there are more implementations and we chose the 'best' > whatever that might be. What is my motivation for wanting separate packages? What would the standard gain by it? As you said, separate packaging will facilitate a quick change in the standard, if needed. What would users gain by it? As I mentioned before, it would allow high-end users (those wanting to incorporate Piper into their systems, like Narval or BlueBox) to quickly swap components. Additionally, low-end users won't have to upgrade the whole Piper system when there is a change in only one component. What would the developers and code base gain by it? It would help ensure that changes in one layer will not have deleterious effects on the other layers. This is good practice for any modular system, as it contributes to the robustness of any one component in the face of the failure of others. If I screw something up in the Desktop (Pied), for example, and commit it to CVS (as I've been known to do), it won't keep everyone else from using Piper, since you will have an older version of the Desktop available, separately. 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 jarl at xs4all.nl Tue May 29 17:50:47 2001 From: jarl at xs4all.nl (Jarl van Katwijk) Date: Fri Feb 10 19:20:50 2006 Subject: [Pipet Devel] ideas for a smaller piper core References: <3B126480.EB2B1F67@bioinformatics.org> <3B129C3F.7D215796@xs4all.nl> <3B12A636.C924CAA8@bioinformatics.org> Message-ID: <3B1419B7.A42B7B36@xs4all.nl> > > About the modulairity: to me Piper is just a set of specifications > > implemented by packages of software labeled 'Piper compliant' by us. > separate packaging of the UIL+DL+BL+PL will not > prevent us from calling them a standard/reference, providing we specify which > components should be used for a standard/reference. right. Maybe I'm missing the point here: are you looking for an easy way to install Piper? Practically bundling all needed components into on package is the most simple installation. > What is my motivation for wanting separate packages? What would the standard > gain by it? As you said, separate packaging will facilitate a quick change in > the standard, if needed. I'm not sure I got this one: are you saying separate packages allow a more flexible upgrade procedure? Also I'm not saying we should use separate packages for the distribution of Piper: use a complete tarball containing the best implementations we got so for. Untill we got a nice installation procedure based on a fully documented Piper environment ;-) In short: supply as much functionality in one package untill we're no longer in alpha state. Something that could be very close to what you actaully want and did 1st post I guess. You seem to be lookin for the quickest way to get a simple working piece of code that's working? > What would users gain by it? As I mentioned before, it would allow high-end > users (those wanting to incorporate Piper into their systems, like Narval or > BlueBox) to quickly swap components. Additionally, low-end users won't have > to upgrade the whole Piper system when there is a change in only one > component. This could be, and to my view should be, done in the configure script. Maybe have a try Jeff at chancing that script. It's very confusing systax at 1st, but you manage ;) There's is another PRO for an installation system that's capable of installing separate packages Jeff did not mention. An Piper that always installs the 4 default layers will overwrite or de-configure non-official parts during an upgrade. So lets package 4 default layers into Piper, and make (some?) of them optional to install. Jarl From jeff at bioinformatics.org Tue May 29 18:12:58 2001 From: jeff at bioinformatics.org (J.W. Bizzaro) Date: Fri Feb 10 19:20:50 2006 Subject: [Pipet Devel] ideas for a smaller piper core References: <3B126480.EB2B1F67@bioinformatics.org> <3B129C3F.7D215796@xs4all.nl> <3B12A636.C924CAA8@bioinformatics.org> <3B1419B7.A42B7B36@xs4all.nl> Message-ID: <3B141EEA.9443CEA2@bioinformatics.org> Jarl van Katwijk wrote: > > right. Maybe I'm missing the point here: are you looking for an easy way > to install Piper? Practically bundling all needed components into on > package is the most simple installation. The point that I made in my first message was EASY = FEW EXTERNAL DEPENDENCIES True, making the external dependencies internal components will eliminate them ;-) I subsequently made the point EASY = MODULARITY > > As you said, separate packaging will facilitate a quick change in > > the standard, if needed. > I'm not sure I got this one: are you saying separate packages allow a > more flexible upgrade procedure? That too. I meant that when we want to change the "reference Piper", we won't have to rip apart and reconfigure a directory tree and start supporting serveral different trees. > Also I'm not saying we should use separate packages for the distribution > of Piper: use a complete tarball containing the best implementations we > got so for. Untill we got a nice installation procedure based on a fully > documented Piper environment ;-) > In short: supply as much functionality in one package untill we're no > longer in alpha state. Something that could be very close to what you > actaully want and did 1st post I guess. You seem to be lookin for the > quickest way to get a simple working piece of code that's working? In my first message, I was looking for the quickest way to get to beta, and that could include folding dependencies into the source code. Later on I was looking for the best solution for the long-run. And I think that would involve modularity to the extent that each component is developed and distributed independently. I would agree that we can keep one massive source tree until Piper is near 1.0, but I would like to split it up after that. > This could be, and to my view should be, done in the configure script. > Maybe have a try Jeff at chancing that script. It's very confusing > systax at 1st, but you manage ;) I'm much better with shell scripting than autoconf. I'm tempted to write a single install script for novices. > There's is another PRO for an installation system that's capable of > installing separate packages Jeff did not mention. An Piper that always > installs the 4 default layers will overwrite or de-configure > non-official parts during an upgrade. Good point! > So lets package 4 default layers into Piper, and make (some?) of them > optional to install. Would you agree that later on the components should be developed and packaged separately? I think we have 5 or 6 good reasons for doing so. Cheers. Jeff