Re: Patch Tagging Guidelines: DEP-3 moved to ACCEPTED status
Le Fri, Jan 20, 2012 at 09:12:16AM +, Lars Wirzenius a écrit : You're not the DEP5 driver Hi Lars and everybody, I am driving this DEP and re-listed myself at a driver to mark that fact. To summarise: - The original idea, from Sam Hocevar, was posted on this list on August 4, 2007. - A draft was written collaboratively on the wiki until March 2009, in which I had my share of contributions. - I do not remember who was the first to suggest to make a DEP out of it, but please note revision 294: I am the one to propose it in the wiki page. - The DEP was started in private, motivated in part by Ubuntu's agenda. It was a terrible mistake for me to accept this, as it resulting in purging and demotivating most of the original contributors. Nevertheless, I did a large—or perhaps the largest—share of that work in that phase. - The DEP continued in public, and the only moment where I gave up driving this project was when you stepped in. It made tremendous progresses under your direction; unfortunately you stepped down in the last mile. I dare saying that I contributed a lot. Among other things did the conversion to DocBook which has let the DEP enter in the debian-policy package, and made sure that the DEP's license short names are compatible with SPDX. - In a further phase, I organised the work through the BTS. Consensus was reached and the DEP was updated accordingly. I also coordinated the publication of the DEP on www.debian.org. At the next upload of the debian-policy package, the DEP will be on line at its canonical URL: http://www.debian.org/doc/packaging-manuals/copyright-format/1.0/ - In this final phase, I am making sure that there is no objection anymore to the release. And I will not let the momentum slip for one more year. More importantly than the procedural details: I have followed the work from its beginning, made sure that no contribution was ignored, and that most questions were answered. To the best of my free time, I made sure that past discussions were not forgotten and taken into account when the same questions were asked over and over the years. This is what I expect from a driver: being the memory of the project, keeping momentum, and making consensus on the final document. I am driving this DEP. One can argue forever on this, but please let me suggest that the best way to close the debate is to finish that work. Cheers, -- Charles Plessy Tsurumi, Kanagawa, Japan -- To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20120201151243.ga20...@merveille.plessy.net
Re: Patch Tagging Guidelines: DEP-3 moved to ACCEPTED status
On Fri, Jan 20, 2012 at 09:58:11AM +0900, Charles Plessy wrote: judging by the current status list DEPs, the problem is more stalling than accepting them too early. I admit that as DEP admins, we have not done a good job at pinging DEP drivers. Because of the current confusion of roles on DEP 5, that I push as a driver, I am waiting for its completion before pinging other DEPs as admin. You're not the DEP5 driver, Steve is; I don't think there's any confusion about that. I've effectively stepped down from drivership a year ago, and I'm happy to make it official if the process is going to drag on a lot longer. None of that should have any effect on you asking drivers of stalled DEPs what the status and future of their DEPs is. Or anyone else for that matter. -- Freedom-based blog/wiki/web hosting: http://www.branchable.com/ signature.asc Description: Digital signature
Re: Patch Tagging Guidelines: DEP-3 moved to ACCEPTED status
Le mercredi 18 janvier 2012 à 19:13 +0100, Stefano Zacchiroli a écrit : I'm not sure I see the point. DEP was never meant to be a device that gives more power to anyone. It was just a device to keep track of a discussion that was already happening, document it in some durable form, and monitor its status. It's not like that the person with the power of marking a DEP as ACCEPTED has the power of creating the corresponding consensus. “Dictator” is probably a bad term. See this as a chairman. Someone who can judge when consensus has been reached, and mark a DEP as accepted. This would avoid the countless and boring nitpicks by people who still want to discuss after the consensus has been reached. Consensus should exist by itself, ditto for an implementation, and then the corresponding DEP could be marked ACCEPTED. I don’t buy this. There will always be a large minority, if not a majority, who will refrain from using a DEP until it is marked as accepted. -- .''`. Josselin Mouette : :' : `. `' `- -- To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/1326990960.3223.860.camel@pi0307572
Re: Patch Tagging Guidelines: DEP-3 moved to ACCEPTED status
On Thu, Jan 19, 2012 at 05:36:00PM +0100, Josselin Mouette wrote: “Dictator” is probably a bad term. See this as a chairman. Someone who can judge when consensus has been reached, and mark a DEP as accepted. This would avoid the countless and boring nitpicks by people who still want to discuss after the consensus has been reached. DEP0 calls these people drivers. Their job is to determine when a rough consensus has been reached. -- Freedom-based blog/wiki/web hosting: http://www.branchable.com/ signature.asc Description: Digital signature
Re: Patch Tagging Guidelines: DEP-3 moved to ACCEPTED status
Lars Wirzenius l...@liw.fi writes: On Thu, Jan 19, 2012 at 05:36:00PM +0100, Josselin Mouette wrote: “Dictator” is probably a bad term. See this as a chairman. Someone who can judge when consensus has been reached, and mark a DEP as accepted. This would avoid the countless and boring nitpicks by people who still want to discuss after the consensus has been reached. DEP0 calls these people drivers. Their job is to determine when a rough consensus has been reached. I think the concern that people have here (and I'm not sure yet whether it is enough of a concern to warrant creating more administration) is that the DEP driver is almost certainly going to have a vested interest in the DEP reaching consensus (otherwise they wouldn't have volunteered to drive it in the first place), and therefore isn't a great choice for an impartial judge of consensus if there's some dispute. -- Russ Allbery (r...@debian.org) http://www.eyrie.org/~eagle/ -- To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/87d3affv8q@windlord.stanford.edu
Re: Patch Tagging Guidelines: DEP-3 moved to ACCEPTED status
On Thu, Jan 19, 2012 at 09:38:29AM -0800, Russ Allbery wrote: Lars Wirzenius l...@liw.fi writes: On Thu, Jan 19, 2012 at 05:36:00PM +0100, Josselin Mouette wrote: “Dictator” is probably a bad term. See this as a chairman. Someone who can judge when consensus has been reached, and mark a DEP as accepted. This would avoid the countless and boring nitpicks by people who still want to discuss after the consensus has been reached. DEP0 calls these people drivers. Their job is to determine when a rough consensus has been reached. I think the concern that people have here (and I'm not sure yet whether it is enough of a concern to warrant creating more administration) is that the DEP driver is almost certainly going to have a vested interest in the DEP reaching consensus (otherwise they wouldn't have volunteered to drive it in the first place), and therefore isn't a great choice for an impartial judge of consensus if there's some dispute. Other people can also be impartial. We don't need an appointed impartial judge: if the driver declares a consensus, I'm sure those who disagree will say so. Furthermore: YAGNI. Let's not solve problems in the DEP process until and unless we have them, particularly not by making it more bureaucratic and heavy. -- Freedom-based blog/wiki/web hosting: http://www.branchable.com/ signature.asc Description: Digital signature
Re: Patch Tagging Guidelines: DEP-3 moved to ACCEPTED status
Le Thu, Jan 19, 2012 at 09:38:29AM -0800, Russ Allbery a écrit : I think the concern that people have here (and I'm not sure yet whether it is enough of a concern to warrant creating more administration) is that the DEP driver is almost certainly going to have a vested interest in the DEP reaching consensus (otherwise they wouldn't have volunteered to drive it in the first place), and therefore isn't a great choice for an impartial judge of consensus if there's some dispute. Hi Russ, judging by the current status list DEPs, the problem is more stalling than accepting them too early. I admit that as DEP admins, we have not done a good job at pinging DEP drivers. Because of the current confusion of roles on DEP 5, that I push as a driver, I am waiting for its completion before pinging other DEPs as admin. Cheers, -- Charles Plessy Tsurumi, Kanagawa, Japan -- To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20120120005811.ga28...@merveille.plessy.net
Re: Patch Tagging Guidelines: DEP-3 moved to ACCEPTED status
Le Monday 16 January 2012 19:15:07, Jakub Wilk a écrit : Does a DEP-3 parser exist? And why not? config-edit -appli dpkg (soon to become 'cme edit dpkg') is able to parse, modify and save DEP-3 patches ( note that this command also deal with debian/copyright, debian/control and some other debian files). This command is part of libconfig-model-perl package HTH Dominique -- http://config-model.wiki.sourceforge.net/ -o- http://search.cpan.org/~ddumont/ http://www.ohloh.net/accounts/ddumont -o- http://ddumont.wordpress.com/ signature.asc Description: This is a digitally signed message part.
Re: Patch Tagging Guidelines: DEP-3 moved to ACCEPTED status
* Dominique Dumont d...@debian.org, 2012-01-18, 10:41: Does a DEP-3 parser exist? And why not? config-edit -appli dpkg (soon to become 'cme edit dpkg') is able to parse, modify and save DEP-3 patches ( note that this command also deal with debian/copyright, debian/control and some other debian files). Huh? What has dpkg to do with DEP-3? And how do I use this parser? I want something as simple as: for a given patch, check if the header complies to DEP-3 and if it does, dump it in some machine-readable format. -- Jakub Wilk -- To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20120118174144.ga9...@jwilk.net
Re: Patch Tagging Guidelines: DEP-3 moved to ACCEPTED status
On Tue, Jan 17, 2012 at 10:05:06AM +0100, Josselin Mouette wrote:
Le lundi 16 janvier 2012 à 18:07 +, Ian Jackson a écrit :
I think the DPL should appoint a dictator who will rule on when
consensus has been achieved on a DEP.
(I originally interpreted this as being enclosed within sarcasm tags)
Seconded. The DEP process is missing a clear way to make a DEP change
state. With a single-person (or small team) responsibility, everything
should be clearer.
I'm not sure I see the point. DEP was never meant to be a device that
gives more power to anyone. It was just a device to keep track of a
discussion that was already happening, document it in some durable form,
and monitor its status. It's not like that the person with the power of
marking a DEP as ACCEPTED has the power of creating the corresponding
consensus.
Consensus should exist by itself, ditto for an implementation, and then
the corresponding DEP could be marked ACCEPTED. If that happens too
soon, no big deal, it's in a VCS, we can revert the commit. Some people
might be fooled in the interim in believing something is more standard
than how much it actual is, but the same could have happened to anyone
looking at the archive of some discussion (i.e. without DEP).
If a DEP is in strict need of a formal rubber stamp of standardization,
then its implementation should probably correspond to formal
integration into policy (as, IIRC, it is the case for DEP-5).
The above notwithstanding, we can probably learn from this thread that,
for the future, it would help to first announce I'm about to mark
DEP-$x as ACCEPTED and then doing that. I personally don't think it is
a big deal, but given that others disagree, ... why not.
Cheers.
--
Stefano Zacchiroli zack@{upsilon.cc,pps.jussieu.fr,debian.org} . o .
Maître de conférences .. http://upsilon.cc/zack .. . . o
Debian Project Leader... @zack on identi.ca ...o o o
« the first rule of tautology club is the first rule of tautology club »
signature.asc
Description: Digital signature
Re: Patch Tagging Guidelines: DEP-3 moved to ACCEPTED status
Le Wednesday 18 January 2012 18:41:44, Jakub Wilk a écrit : config-edit -appli dpkg (soon to become 'cme edit dpkg') is able to parse, modify and save DEP-3 patches ( note that this command also deal with debian/copyright, debian/control and some other debian files). Huh? What has dpkg to do with DEP-3? This command is designed to help debian packager do their job, i.e editing debian package files, including debian/patches in DEP-3 format. And how do I use this parser? I want something as simple as: for a given patch, check if the header complies to DEP-3 and if it does, dump it in some machine-readable format. Currently, it cannot be used outside of the more general debian package files editor. I guess that config-edit could be slightly modified to be applied to individual package files. In check only mode, it should be able to validate the DEP-3 patches and issues error or warning in case of trouble. Even if Config::Model does not really qualify as simple, would that interest you ? ( then we could work out the dump it in some machine-readable format part ) If not, feel free to reuse the parser code [1] All the best Dominique [1] https://metacpan.org/source/DDUMONT/Config-Model-1.265/lib/Config/Model/Backend/Debian/Dpkg/Patch.pm -- http://config-model.wiki.sourceforge.net/ -o- http://search.cpan.org/~ddumont/ http://www.ohloh.net/accounts/ddumont -o- http://ddumont.wordpress.com/ signature.asc Description: This is a digitally signed message part.
Re: Patch Tagging Guidelines: DEP-3 moved to ACCEPTED status
* Dominique Dumont domi.dum...@free.fr, 2012-01-18, 19:37: https://metacpan.org/source/DDUMONT/Config-Model-1.265/lib/Config/Model/Backend/Debian/Dpkg/Patch.pm Judging by a quick look, it doesn't support dpatch patches[0] or pseudo-headers[0][1]. [0] Don't ask what are these features good for. But they are in the specification. [1] Also don't ask me how to unambiguously tell where the pseudo-header starts. -- Jakub Wilk -- To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20120118191501.ga4...@jwilk.net
Re: Patch Tagging Guidelines: DEP-3 moved to ACCEPTED status
On Wed, Jan 18, 2012 at 4:37 PM, Dominique Dumont domi.dum...@free.fr wrote: Le Wednesday 18 January 2012 18:41:44, Jakub Wilk a écrit : And how do I use this parser? I want something as simple as: for a given patch, check if the header complies to DEP-3 and if it does, dump it in some machine-readable format. Currently, it cannot be used outside of the more general debian package files editor. You can use dpkg-copyright instead of dpkg, though: http://search.cpan.org/~ddumont/Config-Model-1.265/lib/Config/Model/models/Debian/Dpkg/Copyright.pod Regards, -- To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/CANVYNa-ivQQYdbEcno5di=+yv8f_iu6chnef4ddqwxvb0fw...@mail.gmail.com
Re: Patch Tagging Guidelines: DEP-3 moved to ACCEPTED status
Le lundi 16 janvier 2012 à 18:07 +, Ian Jackson a écrit : I think the DPL should appoint a dictator who will rule on when consensus has been achieved on a DEP. Seconded. The DEP process is missing a clear way to make a DEP change state. With a single-person (or small team) responsibility, everything should be clearer. -- .''`. Josselin Mouette : :' : `. `' `- -- To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/1326791106.3223.346.camel@pi0307572
Re: Patch Tagging Guidelines: DEP-3 moved to ACCEPTED status
On 16/01/12 16:01, Jonathan Wiltshire wrote: A CVE field, mandatory if a CVE has been published for this patch and is the major component of this patch, would allow easy tracing of patches back to CVE publications later (for review perhaps, or by other distributions). I wonder whether CVE IDs are close enough to being a (limited-scope) bug tracking system to treat them as such, analogous to Bug-Debian, Bug-Fedora etc.; I've previously used Bug-CVE: CVE-2011- in ioquake3, although I haven't been completely consistent about that. (Also, a Bug-* line would ideally have a URI - is there a canonical URI corresponding to each CVE ID, preferably one that doesn't still just say RESERVED long after the embargo date?) S -- To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/4f155d83.4030...@debian.org
Re: Patch Tagging Guidelines: DEP-3 moved to ACCEPTED status
On 2012-01-17 11:37, Simon McVittie wrote: On 16/01/12 16:01, Jonathan Wiltshire wrote: A CVE field, mandatory if a CVE has been published for this patch and is the major component of this patch, would allow easy tracing of patches back to CVE publications later (for review perhaps, or by other distributions). I wonder whether CVE IDs are close enough to being a (limited-scope) bug tracking system to treat them as such, analogous to Bug-Debian, Bug-Fedora etc.; I've previously used Bug-CVE: CVE-2011- in ioquake3, although I haven't been completely consistent about that. It *should* be the case that each CVE identifiers is unique to a problem; occasionally they get revoked if a duplicate becomes apparent. In rare cases they are disputed and marked as such. (Also, a Bug-* line would ideally have a URI - is there a canonical URI corresponding to each CVE ID, preferably one that doesn't still just say RESERVED long after the embargo date?) Useful: http://security-tracker.debian.org/tracker/CVEID https://bugzilla.redhat.com/show_bug.cgi?id=CVEID Generally not so useful: http://cve.mitre.org/cgi-bin/cvename.cgi?name=CVEID (the official CVE database) http://web.nvd.nist.gov/view/vuln/detail?vulnId=CVEID -- Jonathan Wiltshire j...@debian.org Debian Developer http://people.debian.org/~jmw 4096R: 0xD3524C51 / 0A55 B7C5 1223 3942 86EC 74C3 5394 479D D352 4C51 -- To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/823bcc7d8ef2bcd0a17e814917f98...@hogwarts.powdarrmonkey.net
Re: Patch Tagging Guidelines: DEP-3 moved to ACCEPTED status
Hi, FTR given that I got no reports of problems with DEP-3, that it's already well established, I just changed the state of the DEP-3 from CANDIDATE to ACCEPTED. just because that you didn't get any reports you should not set a status to ACCEPTED. IMHO the driver of a DEP should not do that at all, at least not without asking on common lists first. No reaction on your DEP could just mean that people consider it as a waste of time or don't like your format. -- Bernd ZeimetzDebian GNU/Linux Developer http://bzed.dehttp://www.debian.org GPG Fingerprints: ECA1 E3F2 8E11 2432 D485 DD95 EB36 171A 6FF9 435F -- To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/4f142f0c.40...@bzed.de
Re: Patch Tagging Guidelines: DEP-3 moved to ACCEPTED status
On Mon, Jan 16, 2012 at 03:07:08PM +0100, Bernd Zeimetz wrote: just because that you didn't get any reports you should not set a status to ACCEPTED. IMHO the driver of a DEP should not do that at all, at least not without asking on common lists first. No reaction on your DEP could just mean that people consider it as a waste of time or don't like your format. Who should have that authority, then? The DEP-0 proposers? Since the whole DEP process itself is still in CANDIDATE, we could end up in an interesting situation if/when it comes to migrate *that* to ACCEPTED ☺ DEP-0 merely says consensus exists that the implementation has been a success Perhaps that needs unpacking. -- Jon Dowland -- To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20120116144947.GB9047@pris
Re: Patch Tagging Guidelines: DEP-3 moved to ACCEPTED status
Hi, On Mon, 16 Jan 2012, Bernd Zeimetz wrote: just because that you didn't get any reports you should not set a status to ACCEPTED. IMHO the driver of a DEP should not do that at all, at least not without asking on common lists first. No reaction on your DEP could just mean that people consider it as a waste of time or don't like your format. We did have lots of discussion when we were designing it. People commented and reacted. Remember that this format is there to help and is not mandatory (although it's likely to be considered as a best practice in terms of packaging). So if you find it a waste of time, ignore it. But it's already widely used (I have used it for my own packages, Ubuntu is recommending it too), it has been designed following an open process to let everybody participate and ensure it fits as many scenario as possible. It's lightweight and compatible with many of Git's convention. And I have been asked about moving it to ACCEPTED by someone else already (I think it was zack but I no longer remember). And the reason why I post here is precisely so that people can object _if needed_. So do you have a reason to object to the ACCEPTED status of this DEP or was this pure rhetoric ? Cheers, -- Raphaël Hertzog ◈ Debian Developer Pre-order a copy of the Debian Administrator's Handbook and help liberate it: http://debian-handbook.info/liberation/ -- To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20120116145824.ga2...@rivendell.home.ouaza.com
Re: Patch Tagging Guidelines: DEP-3 moved to ACCEPTED status
On 2012-01-16 15:02, Stefano Zacchiroli wrote: Does anyone have further comments about DEP-3? If so, please state them. Otherwise, let's forget about the process details (no matter if they could have been better or not) and rejoice for a nice standard way of adding useful metadata to patches in the Debian archive. It is only a small thing but I did not realise DEP-3 was still a candidate or I would have spoken earlier. A CVE field, mandatory if a CVE has been published for this patch and is the major component of this patch, would allow easy tracing of patches back to CVE publications later (for review perhaps, or by other distributions). Such a field should probably be comma-separated if more than one CVE identifier is relevant to the patch. -- Jonathan Wiltshire j...@debian.org Debian Developer http://people.debian.org/~jmw 4096R: 0xD3524C51 / 0A55 B7C5 1223 3942 86EC 74C3 5394 479D D352 4C51 -- To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/e9846bdb29accb9445e617d2fa272...@hogwarts.powdarrmonkey.net
Re: Patch Tagging Guidelines: DEP-3 moved to ACCEPTED status
Jonathan Wiltshire, 2012-01-16 17:01+0100: It is only a small thing but I did not realise DEP-3 was still a candidate or I would have spoken earlier. A CVE field, mandatory if a CVE has been published for this patch and is the major component of this patch, would allow easy tracing of patches back to CVE publications later (for review perhaps, or by other distributions). Then it would be better to make it independant from CVE, since they are not the only security vulnerability database. -- ,--. : /` ) Tanguy Ortolo xmpp:tan...@ortolo.eu irc://irc.oftc.net/Tanguy | `-'Debian Developer \_ -- To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/jf1k4a$27a$1...@dough.gmane.org
Re: Patch Tagging Guidelines: DEP-3 moved to ACCEPTED status
On 2012-01-16 16:43, Tanguy Ortolo wrote: Jonathan Wiltshire, 2012-01-16 17:01+0100: It is only a small thing but I did not realise DEP-3 was still a candidate or I would have spoken earlier. A CVE field, mandatory if a CVE has been published for this patch and is the major component of this patch, would allow easy tracing of patches back to CVE publications later (for review perhaps, or by other distributions). Then it would be better to make it independant from CVE, since they are not the only security vulnerability database. Ack; but we (in the security team) only track CVE really. The Debian bug number is useful but only within Debian, the CVE identifier is cross-distribution. -- Jonathan Wiltshire j...@debian.org Debian Developer http://people.debian.org/~jmw 4096R: 0xD3524C51 / 0A55 B7C5 1223 3942 86EC 74C3 5394 479D D352 4C51 -- To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/2e6c5e69749b3b03b3cf91bcdc007...@hogwarts.powdarrmonkey.net
Re: Patch Tagging Guidelines: DEP-3 moved to ACCEPTED status
On Mon, Jan 16, 2012 at 12:14:26PM +0100, Raphael Hertzog wrote: FTR given that I got no reports of problems with DEP-3, that it's already well established, I just changed the state of the DEP-3 from CANDIDATE to ACCEPTED. Of course this does not mean that the DEP-3 can't be extended or improved (in particular when it doesn't break backwards compatibility) but it does mean that this format is ready for widespread usage. Use it to document the patches that you add to Debian packages: http://dep.debian.net/deps/dep3/ +1 for moving this to accepted. -- Steve Langasek Give me a lever long enough and a Free OS Debian Developer to set it on, and I can move the world. Ubuntu Developerhttp://www.debian.org/ slanga...@ubuntu.com vor...@debian.org signature.asc Description: Digital signature
Re: Patch Tagging Guidelines: DEP-3 moved to ACCEPTED status
On 01/16/2012 07:14 PM, Raphael Hertzog wrote: Hello, FTR given that I got no reports of problems with DEP-3, that it's already well established, I just changed the state of the DEP-3 from CANDIDATE to ACCEPTED. Of course this does not mean that the DEP-3 can't be extended or improved (in particular when it doesn't break backwards compatibility) but it does mean that this format is ready for widespread usage. Use it to document the patches that you add to Debian packages: http://dep.debian.net/deps/dep3/ Cheers, IMHO, that's a very good thing if we can improve Debian, and don't hold back proposals indefinitively, especially when most of us are already implementing such DEP. I'm really not sure what makes you authoritative for it though, and I'd like to understand (which doesn't conflict with the fact I'm happy dep3 is in state ACCEPTED, and that you decided to do it!). Also, does this mean that you've patched the policy, that lintian would soon more aggressively complain about lacks of patch comments, and that we'll have a new Standard-Version? BTW, what's the status of DEP5? I'm already always uploading DEP5 compliant copyright files myself, and I'd be happy to see it go in the policy. Having them parsable is, IMHO, a very good thing, so that we can make statistics about what license we have, and do all sorts of incompatibility checks (like, who's using GPL and badly mixing it with MPL or OpenSSL for example...). Cheers, Thomas -- To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/4f145f83.5050...@debian.org
Re: Patch Tagging Guidelines: DEP-3 moved to ACCEPTED status
On 16/01/12 18:33, Thomas Goirand wrote: Also, does this mean that you've patched the policy, that lintian would soon more aggressively complain about lacks of patch comments, and that we'll have a new Standard-Version? Lintian already complains when a quilt patch doesn't contain a description, fwiw. See http://lintian.debian.org/tags/quilt-patch-missing-description.html Regards, -- Mehdi -- To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/4f1464df.5000...@dogguy.org
Re: Patch Tagging Guidelines: DEP-3 moved to ACCEPTED status
* Raphael Hertzog hert...@debian.org, 2012-01-16, 12:14: FTR given that I got no reports of problems with DEP-3, that it's already well established, I just changed the state of the DEP-3 from CANDIDATE to ACCEPTED. Does a DEP-3 parser exist? And why not? -- Jakub Wilk -- To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20120116181507.ga2...@jwilk.net
Re: Patch Tagging Guidelines: DEP-3 moved to ACCEPTED status
Jon Dowland writes (Re: Patch Tagging Guidelines: DEP-3 moved to ACCEPTED status): Who should have that authority, then? The DEP-0 proposers? Since the whole DEP process itself is still in CANDIDATE, we could end up in an interesting situation if/when it comes to migrate *that* to ACCEPTED ☺ I think the DPL should appoint a dictator who will rule on when consensus has been achieved on a DEP. If the dictator gets it wrong then insofar as a DEP is a technical policy for Debian (which DEP-3 definitely is) the Technical Committee could overrule the decision, as could a GR of course. Ian. -- To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20244.26459.605817.973...@chiark.greenend.org.uk
Re: Patch Tagging Guidelines: DEP-3 moved to ACCEPTED status
On Tue, 17 Jan 2012, Thomas Goirand wrote: I'm really not sure what makes you authoritative for it though, and I'd like to understand (which doesn't conflict with the fact I'm happy dep3 is in state ACCEPTED, and that you decided to do it!). I just did it as the DEP driver because I believe that there's a consensus that the implementation has been a success and that's the criteria set in DEP-0. Since the goal was only to provide a format to standardize the meta-information and that many people are successfully using this format to document their patch, I think we can assert that the DEP has been successful. I have not counted how many patches embed those standardized fields so I can't say how widely it is used but I know from the interaction with various DD / teams that it's relatively well accepted (the quilt maintainer even recently added a --dep3 option to quilt header). Also, does this mean that you've patched the policy, that lintian would soon more aggressively complain about lacks of patch comments, and that we'll have a new Standard-Version? No, the policy is not the proper place for this, but I believe that a recommendation in the developers-reference would be appropriate. Lintian already recommends the usage of DEP3 in the long description of the relevant informative tags it has: http://lintian.debian.org/tags/dpatch-missing-description.html http://lintian.debian.org/tags/quilt-patch-missing-description.html Cheers, -- Raphaël Hertzog ◈ Debian Developer Pre-order a copy of the Debian Administrator's Handbook and help liberate it: http://debian-handbook.info/liberation/ -- To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20120116194919.ge15...@rivendell.home.ouaza.com
Re: Patch Tagging Guidelines: DEP-3 moved to ACCEPTED status
On 01/17/2012 01:56 AM, Mehdi Dogguy wrote: On 16/01/12 18:33, Thomas Goirand wrote: Also, does this mean that you've patched the policy, that lintian would soon more aggressively complain about lacks of patch comments, and that we'll have a new Standard-Version? Lintian already complains when a quilt patch doesn't contain a description, fwiw. I know that, but it's currently just warnings, not hard errors. Thomas -- To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/4f1489d4.5080...@debian.org
Re: Patch Tagging Guidelines: DEP-3 moved to ACCEPTED status
Le Mon, Jan 16, 2012 at 12:14:26PM +0100, Raphael Hertzog a écrit : FTR given that I got no reports of problems with DEP-3, that it's already well established, I just changed the state of the DEP-3 from CANDIDATE to ACCEPTED. Of course this does not mean that the DEP-3 can't be extended or improved (in particular when it doesn't break backwards compatibility) but it does mean that this format is ready for widespread usage. Use it to document the patches that you add to Debian packages: http://dep.debian.net/deps/dep3/ Bonjour Raphaël, with my DEP admin hat on, I congratulate and thank you for homing this DEP to completion. In my understanding of the process, DEP 3 will not be changed anymore. The format it defines has been implemented in different tools, and this is the achievement of DEP 3. Modifications of the format are of course possible, as a new DEP (taking as inspiration the RFCs 822, 2822 and then 5322), or outside the DEP process. Have a nice day, -- Charles Plessy Debian Enhancement Process team, http://dep.debian.net Tsurumi, Kanagawa, Japan -- To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20120116235059.gc9...@merveille.plessy.net
Re: Patch Tagging Guidelines: DEP-3 moved to ACCEPTED status
Thomas Goirand z...@debian.org writes: Also, does this mean that you've patched the policy, that lintian would soon more aggressively complain about lacks of patch comments, and that we'll have a new Standard-Version? No. DEP-3 is an optional standard. I'm not sure if it should be incorporated into Policy or not. It's probably not a bad idea, although we should deal with DEP-5 first and see if that provides a reasonable precedent for how to do so. BTW, what's the status of DEP5? One of the DEP drivers is not yet happy with the level of specificity and detail provided to ensure that the results are interoperable, and I'd like to see those concerns resolved before including it in Policy. (Which is currently being worked on, as I understand it.) -- Russ Allbery (r...@debian.org) http://www.eyrie.org/~eagle/ -- To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/87ty3vqffk@windlord.stanford.edu
