From andym@ActiveState.com Thu Aug 23 17:47:53 2001 From: andym@ActiveState.com (Andy McKay) Date: Thu, 23 Aug 2001 09:47:53 -0700 Subject: [Catalog-sig] Status Message-ID: <057001c12bf3$5c653b30$ae03a8c0@activestate.com> Hi there, Just wondering what the status of the catalog-sig, whats the plan for moving forward, and what can I do to help? Looking at the status it looks like we had two prototypes one in Zope (link not working) and another that I havent looked at installing yet. But after that activity seems to have stopped... Cheers. -- Andy McKay. From guido@python.org Thu Aug 23 20:18:22 2001 From: guido@python.org (Guido van Rossum) Date: Thu, 23 Aug 2001 15:18:22 -0400 Subject: [Catalog-sig] Status In-Reply-To: Your message of "Thu, 23 Aug 2001 09:47:53 PDT." <057001c12bf3$5c653b30$ae03a8c0@activestate.com> References: <057001c12bf3$5c653b30$ae03a8c0@activestate.com> Message-ID: <200108231918.f7NJIMs07607@odiug.digicool.com> > Hi there, > > Just wondering what the status of the catalog-sig, whats the plan for moving > forward, and what can I do to help? > > Looking at the status it looks like we had two prototypes one in Zope (link > not working) and another that I havent looked at installing yet. But after > that activity seems to have stopped... > > Cheers. I would *love* to see this move forward. What kind of help do you have to offer? Note that there's also discussion about this in the distutils-sig, and there is a PEP about metadata. --Guido van Rossum (home page: http://www.python.org/~guido/) From andym@ActiveState.com Fri Aug 24 00:04:51 2001 From: andym@ActiveState.com (Andy McKay) Date: Thu, 23 Aug 2001 16:04:51 -0700 Subject: [Catalog-sig] Status References: <057001c12bf3$5c653b30$ae03a8c0@activestate.com> <200108231918.f7NJIMs07607@odiug.digicool.com> Message-ID: <00ce01c12c28$04485dd0$ae03a8c0@activestate.com> > I would *love* to see this move forward. What kind of help do you > have to offer? Note that there's also discussion about this in the > distutils-sig, and there is a PEP about metadata. Some time, lots of interest, all that wishy washy stuff. This is one of the biggest problems with Python (and Zope to a certain extent) at the moment and want to see it fixed. Id also like to make sure any work ActiveState does on this is kept in sync with the community and vice versa. Cheers. -- Andy McKay. From guido@python.org Fri Aug 24 02:02:51 2001 From: guido@python.org (Guido van Rossum) Date: Thu, 23 Aug 2001 21:02:51 -0400 Subject: [Catalog-sig] Status In-Reply-To: Your message of "Thu, 23 Aug 2001 16:04:51 PDT." <00ce01c12c28$04485dd0$ae03a8c0@activestate.com> References: <057001c12bf3$5c653b30$ae03a8c0@activestate.com> <200108231918.f7NJIMs07607@odiug.digicool.com> <00ce01c12c28$04485dd0$ae03a8c0@activestate.com> Message-ID: <200108240102.VAA02194@cj20424-a.reston1.va.home.com> > > I would *love* to see this move forward. What kind of help do you > > have to offer? Note that there's also discussion about this in the > > distutils-sig, and there is a PEP about metadata. > > Some time, lots of interest, all that wishy washy stuff. This is one of the > biggest problems with Python (and Zope to a certain extent) at the moment > and want to see it fixed. Id also like to make sure any work ActiveState > does on this is kept in sync with the community and vice versa. What's most needed seemt to be a "BDFL" for this project who has a vision and won't stop until the vision is implemented. Are you strong enough? --Guido van Rossum (home page: http://www.python.org/~guido/) From andym@ActiveState.com Fri Aug 24 19:37:56 2001 From: andym@ActiveState.com (Andy McKay) Date: Fri, 24 Aug 2001 11:37:56 -0700 Subject: [Catalog-sig] Status References: <057001c12bf3$5c653b30$ae03a8c0@activestate.com> <200108231918.f7NJIMs07607@odiug.digicool.com> <00ce01c12c28$04485dd0$ae03a8c0@activestate.com> <200108240102.VAA02194@cj20424-a.reston1.va.home.com> Message-ID: <012c01c12ccb$e7128b60$ae03a8c0@activestate.com> > > Some time, lots of interest, all that wishy washy stuff. This is one of the > > biggest problems with Python (and Zope to a certain extent) at the moment > > and want to see it fixed. Id also like to make sure any work ActiveState > > does on this is kept in sync with the community and vice versa. > > What's most needed seemt to be a "BDFL" for this project who has a > vision and won't stop until the vision is implemented. Wow, thats quite a difference from just helping out here and there. Isnt AMK the BDFL? I'd hate to tread on his toes. > Are you strong enough? Probably, what I need is a vision. I havent got one yet. Cheers. -- Andy McKay. From guido@python.org Fri Aug 24 19:49:14 2001 From: guido@python.org (Guido van Rossum) Date: Fri, 24 Aug 2001 14:49:14 -0400 Subject: [Catalog-sig] Status In-Reply-To: Your message of "Fri, 24 Aug 2001 11:37:56 PDT." <012c01c12ccb$e7128b60$ae03a8c0@activestate.com> References: <057001c12bf3$5c653b30$ae03a8c0@activestate.com> <200108231918.f7NJIMs07607@odiug.digicool.com> <00ce01c12c28$04485dd0$ae03a8c0@activestate.com> <200108240102.VAA02194@cj20424-a.reston1.va.home.com> <012c01c12ccb$e7128b60$ae03a8c0@activestate.com> Message-ID: <200108241849.OAA15173@cj20424-a.reston1.va.home.com> > > What's most needed seemt to be a "BDFL" for this project who has a > > vision and won't stop until the vision is implemented. > > Wow, thats quite a difference from just helping out here and > there. Isnt AMK the BDFL? I'd hate to tread on his toes. You should ask him, but I bet he'd be delighted to let someone else take the lead. > > Are you strong enough? > > Probably, what I need is a vision. I havent got one yet. That's the easy part. You need an archive that you can set up quickly and that can change over time to adapt to user needs. :-) --Guido van Rossum (home page: http://www.python.org/~guido/) From pete@shinners.org Fri Aug 24 20:14:20 2001 From: pete@shinners.org (Pete Shinners) Date: Fri, 24 Aug 2001 12:14:20 -0700 Subject: [Catalog-sig] Status References: <057001c12bf3$5c653b30$ae03a8c0@activestate.com> <200108231918.f7NJIMs07607@odiug.digicool.com> <00ce01c12c28$04485dd0$ae03a8c0@activestate.com> <200108240102.VAA02194@cj20424-a.reston1.va.home.com> Message-ID: <3B86A78C.2090109@shinners.org> Guido van Rossum wrote: > What's most needed seemt to be a "BDFL" for this project who has a > vision and won't stop until the vision is implemented. Are you strong > enough? i think that's right. at some point there just needs to be a lot of decisions made. i don't think there'd be any trouble finding people who want to help work on a good implementation. someone just needs to come up with that good implementation and give it a little kick in the pants to get it rolling. after that i think plenty of people would be glad to help out. wouldn't that be "BDFP"? ... For Project? :] From akuchlin@mems-exchange.org Fri Aug 24 20:49:59 2001 From: akuchlin@mems-exchange.org (Andrew Kuchling) Date: Fri, 24 Aug 2001 15:49:59 -0400 Subject: [Catalog-sig] Status In-Reply-To: <012c01c12ccb$e7128b60$ae03a8c0@activestate.com>; from andym@ActiveState.com on Fri, Aug 24, 2001 at 11:37:56AM -0700 References: <057001c12bf3$5c653b30$ae03a8c0@activestate.com> <200108231918.f7NJIMs07607@odiug.digicool.com> <00ce01c12c28$04485dd0$ae03a8c0@activestate.com> <200108240102.VAA02194@cj20424-a.reston1.va.home.com> <012c01c12ccb$e7128b60$ae03a8c0@activestate.com> Message-ID: <20010824154959.B779@ute.mems-exchange.org> On Fri, Aug 24, 2001 at 11:37:56AM -0700, Andy McKay wrote: >Wow, thats quite a difference from just helping out here and there. Isnt AMK >the BDFL? I'd hate to tread on his toes. Not really. I've written up a draft PEP for an installed package database, but never find the time to focus on implementing it or an interface. If you have more time and can actually get something *done*, great! (The most annoying thing about this, to me, is that I think just a month's work -- heck, perhaps a week's work -- would be enough for a decent and fairly simple implementation that wouldn't solve every problem, but would still be useful. This problem just isn't that hard.) --amk From faassen@vet.uu.nl Wed Aug 29 13:16:20 2001 From: faassen@vet.uu.nl (Martijn Faassen) Date: Wed, 29 Aug 2001 14:16:20 +0200 Subject: [Catalog-sig] Status In-Reply-To: <057001c12bf3$5c653b30$ae03a8c0@activestate.com> References: <057001c12bf3$5c653b30$ae03a8c0@activestate.com> Message-ID: <20010829141619.A11065@vet.uu.nl> Andy McKay wrote: > Just wondering what the status of the catalog-sig, whats the plan for moving > forward, and what can I do to help? > > Looking at the status it looks like we had two prototypes one in Zope (link > not working) and another that I havent looked at installing yet. But after > that activity seems to have stopped... Coming into this thread a bit late, I kept hearing messages about a BoF or something of the kind going to take place at the Open Source conference to kick this project forward again. Wasn't Paul Prescod (also at ActiveState) organizing something like that? Did nothing happen? The PSU must have put a curse on the Catalog project. :) It's-the-only-explanation-that-makes-sense-ly yours, Martijn From andym@ActiveState.com Wed Aug 29 15:35:27 2001 From: andym@ActiveState.com (Andy McKay) Date: Wed, 29 Aug 2001 07:35:27 -0700 Subject: [Catalog-sig] Status References: <057001c12bf3$5c653b30$ae03a8c0@activestate.com> <20010829141619.A11065@vet.uu.nl> Message-ID: <000501c13097$dc4f6430$abcc4e18@cr582427a> > Coming into this thread a bit late, I kept hearing messages about > a BoF or something of the kind going to take place at the Open Source > conference to kick this project forward again. Wasn't Paul Prescod > (also at ActiveState) organizing something like that? Paul did a demo of some work ActiveState put into the catalog-sig in attempt to kick start the project. However its not totally integrated with the PEP and part of the feedback from that was a feeling that we should be involved in more with the community catalog-sig. Hence.. still-trying-to-read-the-peps-on-the-subject-ly-yours -- Andy McKay From k_vertigo@yahoo.com Fri Aug 31 19:39:38 2001 From: k_vertigo@yahoo.com (Kapil Thangavelu) Date: Fri, 31 Aug 2001 11:39:38 -0700 (PDT) Subject: [Catalog-sig] metadata Message-ID: <20010831183938.51696.qmail@web11608.mail.yahoo.com> I've started work on a fourth catalog implementation. I've had a few questions/comments, i wanted to send to the list to resolve regarding metadata. I spent some time looking at the metadata of some other packaging systems namely Debian's apt, ACS's apm, and the OSD format. i see the overall goal of this project as creating a pythonic version of cpan/apt, to automate installation of new packages with depedency resolution. i think this goal is best solved by three separate software packages, the catalog server, the distutils, and the catalog client sitting on top of the distutils. the key for interoperability among the three is having package metadata. looking over pep 241, i can note several deficiencies that i would like to address. While the use of rfc822 for metadata definition does lower the author burden is unextensible and creates the opportunity for ambiguity in the metadata, i'd like to change this to an xml based format. probably the biggest problem with adoption of pep241 is the lack of dependency info. Dependency info should be both version specific and capable of being os dependent. following along with debian's apt, i'd like to add in multiple types of dependencies basically mirroring the apt, with the addition of additional type 'EXTERNAL' to denote non python dependencies. one concept i've been debating is the introduction of logical package keys for dependency tracking. this would basically allow a single package to register itself as a provider of multiple logical keys, say zope does zpublisher, dtml, zodb, extension class,etc. a package would always have its own name as a logical key making the definition of other logical keys optional. i'm not sure about the utility for resolving dependencies though since outside of some standards driven packages, packages with the same logical keys will likely have different interfaces and underlying implementation concerns. however it would allow for aggregate distributions like the current egenix system. ? another issue is the ability to define multiple authors and a vendor. for example the pyxml package is has multiple authors and a single vendor (xml-sig). the current description in pep241 is single author biased and introduces ambiguity regarding an authors contact info. the use of keywords as classification doesn't due an adequate job IMO, to allow for automated classification within the catalog. while i'd still like to keep the use of keywords, i'd like to add in the addition of heirarchical categories. there is also an assumption within the pep241 and 243 i'd like to address. namely that the author of a package will be the person to upload a package. at least initially this is likely to be unlikely, especially during an initial rush to fill up the repository via some semi-automated extraction from the vaults. something else i was considering is a some type of global unique identifier to allow for replication of information to different repositories. i was thinking of something along the lines of a new uri protocol that identified a package on the basis of its classification with the catalog... i'm a little fuzzy on this. to facilitate software updates, i'd like to add in a release date to the metadata info. also the addition of a required python version for a given package version. i'll try and write up an xml schema which defines this package-metadata xml format. cheers kapil thangavelu __________________________________________________ Do You Yahoo!? Get email alerts & NEW webcam video instant messaging with Yahoo! Messenger http://im.yahoo.com From mats@laplaza.org Fri Aug 31 21:08:04 2001 From: mats@laplaza.org (Mats Wichmann) Date: Fri, 31 Aug 2001 16:08:04 -0400 Subject: [Catalog-sig] metadata In-Reply-To: <20010831183938.51696.qmail@web11608.mail.yahoo.com> Message-ID: <5.1.0.14.1.20010831160217.02032d40@204.151.72.2> At 11:39 AM 8/31/2001 -0700, Kapil Thangavelu wrote: >I've started work on a fourth catalog implementation. >I've had a few questions/comments, i wanted to send to >the list to resolve regarding metadata. I have a sort of meta-comment which is that recent experiences with "rpm" packages suggests caution is needed. I have have a whole host of problems arising from rpm or rpm authors getting too ambitious with putting information into the packages, and thus ending up with dependencies that are not needed: MUST have zot-1.30 except that if you were to try against 1.29, 1.28 or 1.27 it would work fine: zot may be a dependency but the listed version is not.