The next Fedora Packaging Committee Meeting will be held on Tuesday, July 22, 2008. (Note: Normally, the FPC meets every other week, but this meeting was scheduled since we did not have quorum on July 15, 2008)
Meeting time is at 17:00 UTC. FPC members, please try to be on-time, as the Fedora Board meets at 18:00 UTC.
Items scheduled to be discussed:
Python virtual Provides: PackagingDrafts/Python Haskell: PackagingDrafts/Haskell Font bundles amendement to Fonts policy: PackagingDrafts/Packaging_font_bundles Lisp: PackagingDrafts/Lisp
As a reminder, the process for getting items onto the FPC's agenda is documented here: http://fedoraproject.org/wiki/Packaging/Committee#Guideline_Change_Procedure
Thanks,
~spot
On Tue, 2008-07-15 at 13:25 -0400, Tom "spot" Callaway wrote:
The next Fedora Packaging Committee Meeting will be held on Tuesday, July 22, 2008. (Note: Normally, the FPC meets every other week, but this meeting was scheduled since we did not have quorum on July 15, 2008)
In case it wasn't clear to everyone already, sometimes, I'm an idiot. I'll be in route to OLS during the time I scheduled for this meeting. You guys are welcomed to have it without me, but if not, we'll push out to next Tuesday... or we could try for Monday. FPC members, please chime in.
~spot
On Thu, 2008-07-17 at 23:20 -0400, Tom "spot" Callaway wrote:
On Tue, 2008-07-15 at 13:25 -0400, Tom "spot" Callaway wrote:
The next Fedora Packaging Committee Meeting will be held on Tuesday, July 22, 2008. (Note: Normally, the FPC meets every other week, but this meeting was scheduled since we did not have quorum on July 15, 2008)
In case it wasn't clear to everyone already, sometimes, I'm an idiot. I'll be in route to OLS during the time I scheduled for this meeting. You guys are welcomed to have it without me, but if not, we'll push out to next Tuesday... or we could try for Monday. FPC members, please chime in.
FYI: I am not sure, whether I'll be able to attend any meeting throughout the next 4 weeks, because other higher commitments are likely to interfere[1]. I.e. whether it'll be Monday or Tuesday doesn't really matter much to me.
Though I'll try to join whenever meetings will be scheduled, it's impossible for me to promise to be able to do so.
Ralf
[1] This already had happened this week. I had planned to attend, but something "urgent" demanding "immediate action" had driven me AFK shortly before the "scheduled meeting".
Tom "spot" Callaway wrote:
<snip>
Meeting time is at 17:00 UTC. FPC members, please try to be on-time, as the Fedora Board meets at 18:00 UTC.
Items scheduled to be discussed:
<snip>
Lisp: PackagingDrafts/Lisp
I'm going to be without internet connectivity today and tomorrow, but I'll likely have it tonight. If anybody has questions or comments about the Lisp packaging draft, today would be a really great day to send them to me, so I can respond tonight before the meeting.
Thanks!
AG
Anthony Green wrote:
If anybody has questions or comments about the Lisp packaging draft, today would be a really great day to send them to me, so I can respond tonight before the meeting.
I am on vacation so I haven't had all that much time to review things properly, but one thing I noticed is that, as someone not familiar with Lisp who might be reviewing packages, some of the guidelines don't really enlighten me as to how I might actually tell if a particular package meets them. For example, I see that libraries should be managed by "asdf" and that they should be able to load asdf with some specific lisp code. The guidelines, however, don't inform me as to how I might tell if the package does those things.
Also, a specfile template would really be appreciated. As it is, if these guidelines were passed, I'd have no idea how to actually review lisp packages.
- J<
Jason Tibbitts wrote:
I am on vacation so I haven't had all that much time to review things properly, but one thing I noticed is that, as someone not familiar with Lisp who might be reviewing packages, some of the guidelines don't really enlighten me as to how I might actually tell if a particular package meets them. For example, I see that libraries should be managed by "asdf" and that they should be able to load asdf with some specific lisp code. The guidelines, however, don't inform me as to how I might tell if the package does those things.
Well, it does say where asdf system definition (.asd) files need to be installed. I guess I'm assuming that the packager would realize that the library isn't managed by asdf if no .asd file is present upstream. I've small changes to clarify this. I've also clarified that you should type "(require 'asdf)" at the REPL in order to test if the Lisp implementation is capable of loading asdf. I think that if a package reviewer doesn't know what a REPL is, then they aren't qualified to review a Common Lisp implementation package (although Lisp libraries should be easily reviewable without any Lisp domain knowledge).
Also, a specfile template would really be appreciated. As it is, if these guidelines were passed, I'd have no idea how to actually review lisp packages.
I'll post a spec template for Lisp libraries. I agree that this would be helpful. But the current draft as written already says where things need to be installed and what the required dependencies are. How can you say it gives you no idea how to review lisp packages?
Thanks,
AG
- J<
-- Fedora-packaging mailing list Fedora-packaging@redhat.com https://www.redhat.com/mailman/listinfo/fedora-packaging
Anthony Green wrote:
I think that if a package reviewer doesn't know what a REPL is, then they aren't qualified to review a Common Lisp implementation package (although Lisp libraries should be easily reviewable without any Lisp domain knowledge).
Wow, really? I guess that in that case all I can do is wish you luck in getting any packages reviewed at all. Well, that and that I'm inclined to vote against guidelines which don't explain things sufficiently so that the available reviewer pool is capable of understanding them.
- J<
Jason Tibbitts wrote:
Anthony Green wrote:
I think that if a package reviewer doesn't know what a REPL is, then they aren't qualified to review a Common Lisp implementation package (although Lisp libraries should be easily reviewable without any Lisp domain knowledge).
Wow, really? I guess that in that case all I can do is wish you luck in getting any packages reviewed at all.
Remember that I'm making a distinction between Lisp implementations and Lisp libraries. The Common Lisp situation is similar to that of Java, in that there are multiple implementations of the language runtime, and many library packages that are designed to work with any of the implementations. Fedora's Java packaging guidelines only cover java libraries, and ignore the implementation side. I think that packaging a Java implementation, much like packaging a Lisp implementation, requires a certain amount of specialized domain knowledge. That's probably why it was omitted from the Java packaging guidelines.
Well, that and that I'm inclined to vote against guidelines which don't explain things sufficiently so that the available reviewer pool is capable of understanding them.
I could simply drop the Lisp implementation packaging guidelines from my draft (to mirror our Java guidelines), but I'd rather not since what I have is better than nothing. All of the major FOSS Common Lisp implementations are already in Fedora, so it's not like we'll need many reviewers with the rudimentary Lisp knowledge necessary to deal with them.
AG
Jason Tibbitts wrote:
Also, a specfile template would really be appreciated. As it is, if these guidelines were passed, I'd have no idea how to actually review lisp packages.
Ok, done. https://fedoraproject.org/wiki/PackagingDrafts/Lisp
When is the next meeting on this subject?
Thanks,
AG
On Tue, Jul 15, 2008 at 01:25:09PM -0400, Tom spot Callaway wrote:
Items scheduled to be discussed:
[list not including MinGW]
As a reminder, the process for getting items onto the FPC's agenda is documented here: http://fedoraproject.org/wiki/Packaging/Committee#Guideline_Change_Procedure
I edited this page a couple of week ago: http://fedoraproject.org/wiki/PackagingDrafts/DraftsTodo but the MinGW stuff doesn't appear in the list?
Rich.
On Mon, Jul 21, 2008 at 11:05:34AM -0400, Tom spot Callaway wrote:
On Mon, 2008-07-21 at 15:54 +0100, Richard W.M. Jones wrote:
[list not including MinGW]
MinGW is blocked at FESCo at the moment (I think).
Don't things go to FPC first? How do I find out its status at FESCo & how to unblock it?
Rich.
On Mon, 2008-07-21 at 16:39 +0100, Richard W.M. Jones wrote:
On Mon, Jul 21, 2008 at 11:05:34AM -0400, Tom spot Callaway wrote:
On Mon, 2008-07-21 at 15:54 +0100, Richard W.M. Jones wrote:
[list not including MinGW]
MinGW is blocked at FESCo at the moment (I think).
Don't things go to FPC first? How do I find out its status at FESCo & how to unblock it?
Well, in this case, I think the workflow is:
The Fedora Board was asked to determine whether including MinGW bits was a good idea. They said that it was, but that it should be separated from the main Fedora repository, and that FESCo should determine the specifics. (see: https://fedoraproject.org/wiki/Board/Meetings/2008-07-15)
Once FESCo does this, the FPC will review the specific packaging bits and approve the guidelines.
~spot
On Mon, Jul 21, 2008 at 12:01:36PM -0400, Tom spot Callaway wrote:
The Fedora Board was asked to determine whether including MinGW bits was a good idea. They said that it was, but that it should be separated from the main Fedora repository, and that FESCo should determine the specifics. (see: https://fedoraproject.org/wiki/Board/Meetings/2008-07-15)
This meeting wasn't announced in the normal way (by a posting "Plan for tomorrows (DATE) ..." on fedora-devel-list), there are no IRC logs anywhere, no one asked anyone in the MinGW SIG to attend (and so they weren't there) and all we have is this fait-accompli message after the fact ...
Rich.
On Mon, 2008-07-21 at 17:45 +0100, Richard W.M. Jones wrote:
On Mon, Jul 21, 2008 at 12:01:36PM -0400, Tom spot Callaway wrote:
The Fedora Board was asked to determine whether including MinGW bits was a good idea. They said that it was, but that it should be separated from the main Fedora repository, and that FESCo should determine the specifics. (see: https://fedoraproject.org/wiki/Board/Meetings/2008-07-15)
This meeting wasn't announced in the normal way (by a posting "Plan for tomorrows (DATE) ..." on fedora-devel-list), there are no IRC logs anywhere, no one asked anyone in the MinGW SIG to attend (and so they weren't there) and all we have is this fait-accompli message after the fact ...
The Board meets (usually) every week on Tuesday. One meeting a month, that meeting is public (IRC):
https://fedoraproject.org/wiki/Board/
This particular meeting was not public, thus, no one other than Board members were invited.
~spot
On Mon, Jul 21, 2008 at 12:58:39PM -0400, Tom spot Callaway wrote:
On Mon, 2008-07-21 at 17:45 +0100, Richard W.M. Jones wrote:
On Mon, Jul 21, 2008 at 12:01:36PM -0400, Tom spot Callaway wrote:
The Fedora Board was asked to determine whether including MinGW bits was a good idea. They said that it was, but that it should be separated from the main Fedora repository, and that FESCo should determine the specifics. (see: https://fedoraproject.org/wiki/Board/Meetings/2008-07-15)
This meeting wasn't announced in the normal way (by a posting "Plan for tomorrows (DATE) ..." on fedora-devel-list), there are no IRC logs anywhere, no one asked anyone in the MinGW SIG to attend (and so they weren't there) and all we have is this fait-accompli message after the fact ...
The Board meets (usually) every week on Tuesday. One meeting a month, that meeting is public (IRC):
https://fedoraproject.org/wiki/Board/
This particular meeting was not public, thus, no one other than Board members were invited.
And IRC logs of this meeting ...?
I notice people were at the meeting who aren't in FESCo. OK, maybe technically they just happened to be there and weren't "invited", but I think it's in the interests of everyone to find out who said what, in the open.
Rich.
On Mon, 2008-07-21 at 17:54 +0100, Richard W.M. Jones wrote:
On Mon, Jul 21, 2008 at 12:58:39PM -0400, Tom spot Callaway wrote:
On Mon, 2008-07-21 at 17:45 +0100, Richard W.M. Jones wrote:
On Mon, Jul 21, 2008 at 12:01:36PM -0400, Tom spot Callaway wrote:
The Fedora Board was asked to determine whether including MinGW bits was a good idea. They said that it was, but that it should be separated from the main Fedora repository, and that FESCo should determine the specifics. (see: https://fedoraproject.org/wiki/Board/Meetings/2008-07-15)
This meeting wasn't announced in the normal way (by a posting "Plan for tomorrows (DATE) ..." on fedora-devel-list), there are no IRC logs anywhere, no one asked anyone in the MinGW SIG to attend (and so they weren't there) and all we have is this fait-accompli message after the fact ...
The Board meets (usually) every week on Tuesday. One meeting a month, that meeting is public (IRC):
https://fedoraproject.org/wiki/Board/
This particular meeting was not public, thus, no one other than Board members were invited.
And IRC logs of this meeting ...?
I notice people were at the meeting who aren't in FESCo. OK, maybe technically they just happened to be there and weren't "invited", but I think it's in the interests of everyone to find out who said what, in the open.
The Fedora Board is different from FESCo. The Fedora Board holds its meetings over the telephone (except for the one monthly open IRC meeting).
Again, to reiterate, the Board simply said that it was in support of MinGW in Fedora, but it should be separated, and that FESCo should handle the technical specifics.
~spot
On Mon, Jul 21, 2008 at 01:22:09PM -0400, Tom spot Callaway wrote:
Again, to reiterate, the Board simply said that it was in support of MinGW in Fedora, but it should be separated, and that FESCo should handle the technical specifics.
What is the board's rationale for putting MinGW packages in a separate repository, when other cross-compiler toolchain (eg ARM) are in the main Fedora repository. Seems to me like we're penalizing MinGW just because it happens to be related to Windows, even though MinGW's code is still just as open source as anything else in our repos.
Daniel
On Tue, 2008-07-22 at 10:29 +0100, Daniel P. Berrange wrote:
What is the board's rationale for putting MinGW packages in a separate repository, when other cross-compiler toolchain (eg ARM) are in the main Fedora repository. Seems to me like we're penalizing MinGW just because it happens to be related to Windows, even though MinGW's code is still just as open source as anything else in our repos.
Actually I think the prevailing thought that the Board has (although it's up to FESCo to really nail it down) is that the mingw tools themselves are absolutely suitable for Fedora. The libraries compiled against it for windows use are what should be in another repo.
My personal opinion is that if you're going to need to munge spec files in order to produce packages built against mingw, those munges need to be done outside our cvs repo as well.
However that's just my opinion, and since the board has asked FESCo to sort out the technical details, and I'm not in FESCo anymore, that opinion doesn't amount to much (:
On Tue, Jul 22, 2008 at 08:06:59AM -0400, Jesse Keating wrote:
On Tue, 2008-07-22 at 10:29 +0100, Daniel P. Berrange wrote:
What is the board's rationale for putting MinGW packages in a separate repository, when other cross-compiler toolchain (eg ARM) are in the main Fedora repository. Seems to me like we're penalizing MinGW just because it happens to be related to Windows, even though MinGW's code is still just as open source as anything else in our repos.
Actually I think the prevailing thought that the Board has (although it's up to FESCo to really nail it down) is that the mingw tools themselves are absolutely suitable for Fedora. The libraries compiled against it for windows use are what should be in another repo.
[I'm going to prepare something more detailed, hopefully integrating efforts with the cross-compiler folks, but just on these two points ...]
If we ship only the four base packages (mingw-gcc, mingw-binutils, mingw-w32api and mingw-runtime) then the only software that can be compiled is software which doesn't use any libraries. That's pretty restrictive.
To compile, for example, libvirt, one needs six other libraries. As with Linux, you need the library around (foo-0.dll) in order to link. Anyone compiling libvirt would need to download the source for each of these six libraries and './configure --host=i686-pc-mingw32 ; make ; make install' before they could start on libvirt, and of course it isn't really that simple since those libraries don't all just cross-compile without needing tweaks and patches. Tweaks and patches are what spec files are for. This is why we'd like to ship pre-compiled DLLs (only) of those six libs.
I think people have somehow got the impression we want to (a) ship FIREFOX.EXE and/or (b) cross-compile every library in Fedora. I'd like to say that (a) is not our intention, ever, and (b) isn't even technically possible, nevermind that it is completely undesirable.
My personal opinion is that if you're going to need to munge spec files in order to produce packages built against mingw, those munges need to be done outside our cvs repo as well.
There are two ways that we've proposed that one could build 'mingw-gnutls'. One is as a completely separate package, another is as a subpackage of the ordinary gnutls. I investigated and built packages both ways (see links below) just to see what was technically feasible. It turns out that both methods are *technically* feasible. Which is better from technical, organizational or political points of view is a completely different question.
http://hg.et.redhat.com/misc/fedora-mingw--devel/?cmd=manifest;manifest=91a8... http://hg.et.redhat.com/misc/fedora-mingw--devel/?cmd=manifest;manifest=91a8...
Rich.
Richard W.M. Jones wrote:
On Tue, Jul 22, 2008 at 08:06:59AM -0400, Jesse Keating wrote:
On Tue, 2008-07-22 at 10:29 +0100, Daniel P. Berrange wrote:
What is the board's rationale for putting MinGW packages in a separate repository, when other cross-compiler toolchain (eg ARM) are in the main Fedora repository. Seems to me like we're penalizing MinGW just because it happens to be related to Windows, even though MinGW's code is still just as open source as anything else in our repos.
Actually I think the prevailing thought that the Board has (although it's up to FESCo to really nail it down) is that the mingw tools themselves are absolutely suitable for Fedora. The libraries compiled against it for windows use are what should be in another repo.
[I'm going to prepare something more detailed, hopefully integrating efforts with the cross-compiler folks, but just on these two points ...]
If we ship only the four base packages (mingw-gcc, mingw-binutils, mingw-w32api and mingw-runtime) then the only software that can be compiled is software which doesn't use any libraries. That's pretty restrictive.
To compile, for example, libvirt, one needs six other libraries. As with Linux, you need the library around (foo-0.dll) in order to link. Anyone compiling libvirt would need to download the source for each of these six libraries and './configure --host=i686-pc-mingw32 ; make ; make install' before they could start on libvirt, and of course it isn't really that simple since those libraries don't all just cross-compile without needing tweaks and patches. Tweaks and patches are what spec files are for. This is why we'd like to ship pre-compiled DLLs (only) of those six libs.
When people talk about a separate repo, it's something that would still allow this workflow to happen. The separate repo exists on the Fedora master mirror but mirrors of us have the option to include or exclude these other repos depending on their ability to carry the extra packages.
This separate repo will have a yum configuration file that I think should be shipped by default. I think it should also be turned on by default. This would make the fact that there is a different repo for the packages transparent to end users. (However, this portion is something that FESCo decides, not FPC... this case would need to be argued in front of FESCo).
I think people have somehow got the impression we want to (a) ship FIREFOX.EXE and/or (b) cross-compile every library in Fedora. I'd like to say that (a) is not our intention, ever, and (b) isn't even technically possible, nevermind that it is completely undesirable.
Who is "we"? That is the crux of your statements. If a group of Fedora contributors who are not the libvirt team decide that they want to have a complete cross-compilation environment to be able to build firefox.exe for windows under Fedora at some point in the future, I'd like us to not stand in their way. OTOH, even if that never happens, there is still the issue that MingW is not the only crosscompilation system that we want in Fedora. To scale across architectures as well as in depth on one os-architecture also has an impact on mirrors which can be mitigated by having a separate repo.
My personal opinion is that if you're going to need to munge spec files in order to produce packages built against mingw, those munges need to be done outside our cvs repo as well.
There are two ways that we've proposed that one could build 'mingw-gnutls'. One is as a completely separate package, another is as a subpackage of the ordinary gnutls. I investigated and built packages both ways (see links below) just to see what was technically feasible. It turns out that both methods are *technically* feasible. Which is better from technical, organizational or political points of view is a completely different question.
This is partially a FPC issue and partially a FESCo/Board issue. The four people present for the FPC meeting last week discussed this informally and there was consensus that separate packging made more sense. However, FESCo will need to decide how having a separate download repository maps to our cvs repository. The two options I see are separate packages (as discussed by FPC) and separate branches within CVS. Which one is decided will have some influence over any eventual Guidelines that the FPC writes and/or approves.
-Toshio
On Tue, 2008-07-22 at 17:22 +0100, Richard W.M. Jones wrote:
On Tue, Jul 22, 2008 at 08:06:59AM -0400, Jesse Keating wrote:
On Tue, 2008-07-22 at 10:29 +0100, Daniel P. Berrange wrote:
What is the board's rationale for putting MinGW packages in a separate repository, when other cross-compiler toolchain (eg ARM) are in the main Fedora repository. Seems to me like we're penalizing MinGW just because it happens to be related to Windows, even though MinGW's code is still just as open source as anything else in our repos.
Actually I think the prevailing thought that the Board has (although it's up to FESCo to really nail it down) is that the mingw tools themselves are absolutely suitable for Fedora. The libraries compiled against it for windows use are what should be in another repo.
[I'm going to prepare something more detailed, hopefully integrating efforts with the cross-compiler folks, but just on these two points ...]
If we ship only the four base packages (mingw-gcc, mingw-binutils, mingw-w32api and mingw-runtime) then the only software that can be compiled is software which doesn't use any libraries. That's pretty restrictive.
This is way too restrictive. In fact, such a restriction closes out any cross-toolchain from Fedora.
My personal opinion is that if you're going to need to munge spec files in order to produce packages built against mingw, those munges need to be done outside our cvs repo as well.
Building cross-toolchains inevitably needs some target-libraries. If you want to see cross-toolchain packages in Fedora, these target-libraries must be shipped as part of Fedora.
Ralf
Tom "spot" Callaway wrote:
The next Fedora Packaging Committee Meeting will be held on Tuesday, July 22, 2008. (Note: Normally, the FPC meets every other week, but this meeting was scheduled since we did not have quorum on July 15, 2008)
Meeting time is at 17:00 UTC. FPC members, please try to be on-time, as the Fedora Board meets at 18:00 UTC.
Items scheduled to be discussed:
Python virtual Provides: PackagingDrafts/Python Haskell: PackagingDrafts/Haskell Font bundles amendement to Fonts policy: PackagingDrafts/Packaging_font_bundles Lisp: PackagingDrafts/Lisp
As a reminder, the process for getting items onto the FPC's agenda is documented here: http://fedoraproject.org/wiki/Packaging/Committee#Guideline_Change_Procedure
Wat is the plan with this meeting, is it going through ? I'm asking because there have been some cancellations so I wonder how usefull it is to get together. I'll be there this time, sorry for missing the last few times.
Regards,
Hans
On Tuesday, 15 July 2008 at 19:25, Tom spot Callaway wrote:
The next Fedora Packaging Committee Meeting will be held on Tuesday, July 22, 2008. (Note: Normally, the FPC meets every other week, but this meeting was scheduled since we did not have quorum on July 15, 2008)
Meeting time is at 17:00 UTC. FPC members, please try to be on-time, as the Fedora Board meets at 18:00 UTC.
I couldn't make it, again, and I'm sorry. The current meeting time is consistently overlapping with my sport activities which can't be rescheduled. That time might change after the vacations are over, but I won't know until then. I'd very much like to change that meeting time or step down from FPC if that's not possible. I don't see any point in being on the Committee if I can't attend the meetings.
Moreover, I'm on vacation for the most of August and will only have Internet connectivity via a cellphone, if at all, and that's expensive, so I'll be limiting it as much as possible.
Regards, R.
packaging@lists.fedoraproject.org