Difference between revisions of "PMC proposals/Debian package mantainer"
(cat) |
|||
(13 intermediate revisions by 5 users not shown) | |||
Line 2: | Line 2: | ||
=2011/ |
=2011/11/27 #1: Debian package maintainer= |
||
==Summary== |
==Summary== |
||
Line 30: | Line 30: | ||
==Comments== |
==Comments== |
||
I'm a little surprised that its €1000/year. How many hours is that? 30? |
|||
I prefer that we start out offering €1000/year, but when then see how much there really is to update every year and negociate salary accordingly. -- [[User:Jacob Nordfalk|Jacob Nordfalk]] |
|||
I don't want to assume that Jacob means the same thing I do, but I would assume that there would be a lot more work involved the first time around, as the majority of the packages in Debian - if not all - are out of date. On the issue of money, our only income to date has been from GSoC, and I think that it would be risky to make commitments based on the assumption that Google will continue to select Apertium as a mentoring organisation. Overall, there is a lack of specific detail in this proposal, and I can't help but wonder - as it is not at all clear from the proposal - if this is a proposal to both consider paying someone ''and'' to pay them, or if the details will be voted on, later? If the latter, I would vote yes; if the former, no. |
|||
I would greatly prefer, when dealing with predominantly volunteer-based organisations, that some attempt to enlist volunteers were made. I recall that Fran had difficulties with Debian, but they have made significant strides in enabling a more casual level of contribution in the last few years, and enlisting interested Apertiumers is an available option. |
|||
In any event, there are a few things that we can do to make life easier for the Debian packaging effort. I've committed some fixes to Debian-related bugs, but a new release of Apertium will be required. Also, there are some missing licence headers in the source, which should be fixed, to not fall foul of Debian's licence checks. First, the win32 directory - which is unused - contains some code with missing licence information, and some with GPL-incompatible terms. I propose to remove this directory entirely. Second, the DTDs have no licence header - can someone confirm that they are GPLed? -- [[User:Jimregan|Jimregan]] 04:35, 18 February 2013 (UTC) |
|||
I should probably rephrase that as a set of questions, to be more clear: |
|||
# As written, it seems to me that it is anticipated that the workload will be more or less the same, year after year. Is this the case? Can someone point me to some documentation that backs this up? (Mikel: "I think that Fran estimated that workload. But we can say this: how much is Apertium willing to pay to be up-to-date in Debian?") |
|||
# What provisions are in place if we do not continue to be accepted as a GSoC mentoring organisation? (Mikel: "Well, that's an easy one I think. The contract would be renewed yearly, provided that we have at least as much money in the bank as it is needed and we're willing to spend it in that. If we have no money or we believe we have to spend it elsewhere, then the job would not be renewed for an additional year.") |
|||
# Is this proposal intended primarily for consideration? i.e., will there be a second, more detailed proposal when the details are available? |
|||
On the matters unrelated to this vote: |
|||
# Are there any objections to removing the win32 directory? |
|||
# Can someone (Mikel? Juan Antonio?) who speaks on behalf of UA confirm that the DTDs were intended to also be GPLd? (Mikel confirmed that they are, the SVN versions now contain a licence header) |
|||
Hopefully, that's a little more clear. -- [[User:Jimregan|Jimregan]] 15:54, 18 February 2013 (UTC) |
|||
To make life easier, I've made a Debian copyright file in the machine readable format, for whoever ends up doing this: [https://gist.github.com/jimregan/4980312 here] -- [[User:Jimregan|Jimregan]] 20:05, 18 February 2013 (UTC) |
|||
==Voting== |
==Voting== |
||
=== Agree === |
=== Agree === |
||
# [[User:Francis Tyers|Francis Tyers]] 21:27, 27 November 2012 (UTC) |
|||
# [[User:Fsanchez|Felipe Sánchez-Martínez]] 11:24, 28 November 2012 (UTC) |
|||
# [[User:Japerez|Juan Antonio Pérez]] 15:10, 28 November 2012 (UTC) |
|||
# --[[User:Jacob Nordfalk|Jacob Nordfalk]] 16:56, 17 February 2013 (UTC) |
|||
===Disagree=== |
===Disagree=== |
||
===Abstain=== |
===Abstain=== |
||
[[Category:Project Management Committee]] |
Latest revision as of 22:26, 3 August 2013
2011/11/27 #1: Debian package maintainer[edit]
Summary[edit]
Recruiting someone to make sure that Apertium packages in Debian are reasonably up-to-date.
The recruitee would get a yearly stipend, for making a Debian release of all the stable packages with changes every 6 months with a maintenance release if and when libpcre becomes incompatible.
The stipend would be €1000/year, based on Fran's experience of how much work it is.
If we continue in GSOC in the same way, this would represent 1/6 of the annual GSOC-earnings.
This could be a good place to advertise it:
Proposed by: User:Mlforcada
Seconded by: User:Francis Tyers
In detail[edit]
Caveats[edit]
Comments[edit]
I'm a little surprised that its €1000/year. How many hours is that? 30?
I prefer that we start out offering €1000/year, but when then see how much there really is to update every year and negociate salary accordingly. -- Jacob Nordfalk
I don't want to assume that Jacob means the same thing I do, but I would assume that there would be a lot more work involved the first time around, as the majority of the packages in Debian - if not all - are out of date. On the issue of money, our only income to date has been from GSoC, and I think that it would be risky to make commitments based on the assumption that Google will continue to select Apertium as a mentoring organisation. Overall, there is a lack of specific detail in this proposal, and I can't help but wonder - as it is not at all clear from the proposal - if this is a proposal to both consider paying someone and to pay them, or if the details will be voted on, later? If the latter, I would vote yes; if the former, no.
I would greatly prefer, when dealing with predominantly volunteer-based organisations, that some attempt to enlist volunteers were made. I recall that Fran had difficulties with Debian, but they have made significant strides in enabling a more casual level of contribution in the last few years, and enlisting interested Apertiumers is an available option.
In any event, there are a few things that we can do to make life easier for the Debian packaging effort. I've committed some fixes to Debian-related bugs, but a new release of Apertium will be required. Also, there are some missing licence headers in the source, which should be fixed, to not fall foul of Debian's licence checks. First, the win32 directory - which is unused - contains some code with missing licence information, and some with GPL-incompatible terms. I propose to remove this directory entirely. Second, the DTDs have no licence header - can someone confirm that they are GPLed? -- Jimregan 04:35, 18 February 2013 (UTC)
I should probably rephrase that as a set of questions, to be more clear:
- As written, it seems to me that it is anticipated that the workload will be more or less the same, year after year. Is this the case? Can someone point me to some documentation that backs this up? (Mikel: "I think that Fran estimated that workload. But we can say this: how much is Apertium willing to pay to be up-to-date in Debian?")
- What provisions are in place if we do not continue to be accepted as a GSoC mentoring organisation? (Mikel: "Well, that's an easy one I think. The contract would be renewed yearly, provided that we have at least as much money in the bank as it is needed and we're willing to spend it in that. If we have no money or we believe we have to spend it elsewhere, then the job would not be renewed for an additional year.")
- Is this proposal intended primarily for consideration? i.e., will there be a second, more detailed proposal when the details are available?
On the matters unrelated to this vote:
- Are there any objections to removing the win32 directory?
- Can someone (Mikel? Juan Antonio?) who speaks on behalf of UA confirm that the DTDs were intended to also be GPLd? (Mikel confirmed that they are, the SVN versions now contain a licence header)
Hopefully, that's a little more clear. -- Jimregan 15:54, 18 February 2013 (UTC)
To make life easier, I've made a Debian copyright file in the machine readable format, for whoever ends up doing this: here -- Jimregan 20:05, 18 February 2013 (UTC)
Voting[edit]
Agree[edit]
- Francis Tyers 21:27, 27 November 2012 (UTC)
- Felipe Sánchez-Martínez 11:24, 28 November 2012 (UTC)
- Juan Antonio Pérez 15:10, 28 November 2012 (UTC)
- --Jacob Nordfalk 16:56, 17 February 2013 (UTC)