You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@corinthia.apache.org by Peter Kelly <pm...@apache.org> on 2015/08/01 17:15:07 UTC
Re: [PROPOSAL] make release 0.1
Sounds like a good plan to me.
I volunteer to write the README - I think what we have on the website/wiki is a little unclear at the moment, and have some ideas on how to better express what the project’s about.
I suggest we do this (along with LICENSES and NOTICE) as files in the root of the Git repository, so we can each make changes via the usual means as for code.
Another thing I would like to suggest is we have a CONTRIBUTORS file (or similar name) which lists all those who have contributed to the project. I’m aware this isn’t strictly required, but I think it's a good way to give credit/recognition to everyone involved for their work.
—
Dr Peter M. Kelly
pmkelly@apache.org
PGP key: http://www.kellypmk.net/pgp-key <http://www.kellypmk.net/pgp-key>
(fingerprint 5435 6718 59F0 DD1F BFA0 5E46 2523 BAA1 44AE 2966)
> On 30 Jul 2015, at 9:21 pm, jan i <ja...@apache.org> wrote:
>
> Traditionally cutting the first apache release for a podling is one big
> PITA, therefore
> I propose we get started.
>
> I propose we make release 0.1 consisting of
> - docFormats (the library only with the OOXML filter)
> - dfTest (the main test utility)
> - dfUtil (the conversion utility)
>
> We have a lot to learn here, about the files LICENSES and NOTICE and about
> how to
> handle release candidates and the final release, therefore I propose to
> include the
> "stable" parts. It can be argued that version 0.1 is not very useful for a
> end-user, which
> might be right, but it helps us get the release framework up and running.
>
> Once we have 0.1 through the needle of IPMC, version 0.2 can focus a lot
> more on the
> technical content.
>
> The steps we have to pass (I have surely forgotten a number) are:
> - Make a branch (stable_0.1) that only contains what we intent to release
> - Make it work (adapt CMake files etc)
> - Correct LICENSES and NOTICE
> - Prepare a release candidate (upload source with SHA5 to somewhere on dist)
> - Vote on the release (min. 3 +1 NO -1)
> - Ask IPMC to check the release and have them vote on it
> this is the needle, they tend to find things we have not thought about,
> but Justin (he is one of the best release checkers in incubators) earlier
> promised
> me to do a pre-check so we avoid the normal errors.
> - When the vote finally passes, move the source to the release part of dist.
> - Congratulate each other with work well done.
>
> If we agree on the content and the idea of cutting a release, I volunteer
> to do the work and
> of course write about every step in here (and on the wiki), so we all learn
> from the experience.
> I do however need help from e.g. Dave, with some of the specialties.
>
> Following the suggestion from Dave, Let us discuss this until August 7th,
> and then see
> if we have reached consensus.
>
> Comments ?
>
> rgds
> jan i.
Re: [PROPOSAL] make release 0.1
Posted by jan i <ja...@apache.org>.
On 1 August 2015 at 18:11, Peter Kelly <pm...@apache.org> wrote:
> > On 1 Aug 2015, at 11:10 pm, Peter Kelly <pm...@apache.org> wrote:
> >
> >> On 1 Aug 2015, at 10:18 pm, jan i <ja...@apache.org> wrote:
> >>
> >>> Another thing I would like to suggest is we have a CONTRIBUTORS file
> (or
> >>> similar name) which lists all those who have contributed to the
> project.
> >>> I’m aware this isn’t strictly required, but I think it's a good way to
> give
> >>> credit/recognition to everyone involved for their work.
> >>>
> >> +1, I assume you mean the file contains the names of people who have
> >> actively done commits on what is being released ?
> >
> > Yes - everyone who’s name is present as an author or committer in the
> git log (or mentioned there as having been the author of a patch). I think
> a simple alphabetical listing of names & email addresses would be
> appropriate.
>
> BTW I volunteer to create this as well.
>
Could we agree to do it without email please and use our apache id.
If I understand it correctly our stable branch, also has all the commits,
so it is a simple matter of doing git log and extracting names.
I like the idea of using apache id alphabetical, I would not like any
indication of how much work each have done.
Question, we have had some code added (I added e.g. libraries) which are
gone again, can we easily limit the list to commits that are still in the
code ?
rgds
jan i.
>
> —
> Dr Peter M. Kelly
> pmkelly@apache.org
>
> PGP key: http://www.kellypmk.net/pgp-key <http://www.kellypmk.net/pgp-key>
> (fingerprint 5435 6718 59F0 DD1F BFA0 5E46 2523 BAA1 44AE 2966)
>
>
Re: [PROPOSAL] make release 0.1
Posted by jan i <ja...@apache.org>.
On 1 August 2015 at 19:49, Dennis E. Hamilton <de...@acm.org>
wrote:
> Some contributors may object to their email addresses being put in
> plaintext files.
>
> I don't have a workaround, and the ASF uses email addresses as identifiers
> and as part of code provenance.
>
> So long as there is a place where we can recover email addresses from the
> names we list, I think we can demonstrate provenance without publishing the
> email addresses directly.
>
Do you also publishing the apache id (which is a mail address as well) is a
problem, and would you prefer just their full name ?
>
> In general, we do not accept anonymous contributions, so there needs to be
> a way to handle the objections, whenever one arrives.
>
correct.
>
> It is probably better to figure this out now rather than wait until a
> contributor complains.
>
+1
rgds
jan i
>
> - Dennis
>
> -----Original Message-----
> From: Peter Kelly [mailto:pmkelly@apache.org]
> Sent: Saturday, August 1, 2015 09:12
> To: dev@corinthia.incubator.apache.org
> Subject: Re: [PROPOSAL] make release 0.1
>
> > On 1 Aug 2015, at 11:10 pm, Peter Kelly <pm...@apache.org> wrote:
> >
> >> On 1 Aug 2015, at 10:18 pm, jan i <ja...@apache.org> wrote:
> >>
> >>> Another thing I would like to suggest is we have a CONTRIBUTORS file
> (or
> >>> similar name) which lists all those who have contributed to the
> project.
> >>> I’m aware this isn’t strictly required, but I think it's a good way to
> give
> >>> credit/recognition to everyone involved for their work.
> >>>
> >> +1, I assume you mean the file contains the names of people who have
> >> actively done commits on what is being released ?
> >
> > Yes - everyone who’s name is present as an author or committer in the
> git log (or mentioned there as having been the author of a patch). I think
> a simple alphabetical listing of names & email addresses would be
> appropriate.
>
> BTW I volunteer to create this as well.
>
> —
> Dr Peter M. Kelly
> pmkelly@apache.org
>
> PGP key: http://www.kellypmk.net/pgp-key <http://www.kellypmk.net/pgp-key>
> (fingerprint 5435 6718 59F0 DD1F BFA0 5E46 2523 BAA1 44AE 2966)
>
>
>
Re: [PROPOSAL] make release 0.1
Posted by Peter Kelly <pm...@apache.org>.
I’m ok with just using names. The email addresses can be recovered anyway via git log.
—
Dr Peter M. Kelly
pmkelly@apache.org
PGP key: http://www.kellypmk.net/pgp-key <http://www.kellypmk.net/pgp-key>
(fingerprint 5435 6718 59F0 DD1F BFA0 5E46 2523 BAA1 44AE 2966)
> On 2 Aug 2015, at 12:49 am, Dennis E. Hamilton <de...@acm.org> wrote:
>
> Some contributors may object to their email addresses being put in plaintext files.
>
> I don't have a workaround, and the ASF uses email addresses as identifiers and as part of code provenance.
>
> So long as there is a place where we can recover email addresses from the names we list, I think we can demonstrate provenance without publishing the email addresses directly.
>
> In general, we do not accept anonymous contributions, so there needs to be a way to handle the objections, whenever one arrives.
>
> It is probably better to figure this out now rather than wait until a contributor complains.
>
> - Dennis
>
> -----Original Message-----
> From: Peter Kelly [mailto:pmkelly@apache.org]
> Sent: Saturday, August 1, 2015 09:12
> To: dev@corinthia.incubator.apache.org
> Subject: Re: [PROPOSAL] make release 0.1
>
>> On 1 Aug 2015, at 11:10 pm, Peter Kelly <pm...@apache.org> wrote:
>>
>>> On 1 Aug 2015, at 10:18 pm, jan i <ja...@apache.org> wrote:
>>>
>>>> Another thing I would like to suggest is we have a CONTRIBUTORS file (or
>>>> similar name) which lists all those who have contributed to the project.
>>>> I’m aware this isn’t strictly required, but I think it's a good way to give
>>>> credit/recognition to everyone involved for their work.
>>>>
>>> +1, I assume you mean the file contains the names of people who have
>>> actively done commits on what is being released ?
>>
>> Yes - everyone who’s name is present as an author or committer in the git log (or mentioned there as having been the author of a patch). I think a simple alphabetical listing of names & email addresses would be appropriate.
>
> BTW I volunteer to create this as well.
>
> —
> Dr Peter M. Kelly
> pmkelly@apache.org
>
> PGP key: http://www.kellypmk.net/pgp-key <http://www.kellypmk.net/pgp-key>
> (fingerprint 5435 6718 59F0 DD1F BFA0 5E46 2523 BAA1 44AE 2966)
>
>
RE: [PROPOSAL] make release 0.1
Posted by "Dennis E. Hamilton" <de...@acm.org>.
Some contributors may object to their email addresses being put in plaintext files.
I don't have a workaround, and the ASF uses email addresses as identifiers and as part of code provenance.
So long as there is a place where we can recover email addresses from the names we list, I think we can demonstrate provenance without publishing the email addresses directly.
In general, we do not accept anonymous contributions, so there needs to be a way to handle the objections, whenever one arrives.
It is probably better to figure this out now rather than wait until a contributor complains.
- Dennis
-----Original Message-----
From: Peter Kelly [mailto:pmkelly@apache.org]
Sent: Saturday, August 1, 2015 09:12
To: dev@corinthia.incubator.apache.org
Subject: Re: [PROPOSAL] make release 0.1
> On 1 Aug 2015, at 11:10 pm, Peter Kelly <pm...@apache.org> wrote:
>
>> On 1 Aug 2015, at 10:18 pm, jan i <ja...@apache.org> wrote:
>>
>>> Another thing I would like to suggest is we have a CONTRIBUTORS file (or
>>> similar name) which lists all those who have contributed to the project.
>>> I’m aware this isn’t strictly required, but I think it's a good way to give
>>> credit/recognition to everyone involved for their work.
>>>
>> +1, I assume you mean the file contains the names of people who have
>> actively done commits on what is being released ?
>
> Yes - everyone who’s name is present as an author or committer in the git log (or mentioned there as having been the author of a patch). I think a simple alphabetical listing of names & email addresses would be appropriate.
BTW I volunteer to create this as well.
—
Dr Peter M. Kelly
pmkelly@apache.org
PGP key: http://www.kellypmk.net/pgp-key <http://www.kellypmk.net/pgp-key>
(fingerprint 5435 6718 59F0 DD1F BFA0 5E46 2523 BAA1 44AE 2966)
Re: [PROPOSAL] make release 0.1
Posted by Peter Kelly <pm...@apache.org>.
> On 1 Aug 2015, at 11:10 pm, Peter Kelly <pm...@apache.org> wrote:
>
>> On 1 Aug 2015, at 10:18 pm, jan i <ja...@apache.org> wrote:
>>
>>> Another thing I would like to suggest is we have a CONTRIBUTORS file (or
>>> similar name) which lists all those who have contributed to the project.
>>> I’m aware this isn’t strictly required, but I think it's a good way to give
>>> credit/recognition to everyone involved for their work.
>>>
>> +1, I assume you mean the file contains the names of people who have
>> actively done commits on what is being released ?
>
> Yes - everyone who’s name is present as an author or committer in the git log (or mentioned there as having been the author of a patch). I think a simple alphabetical listing of names & email addresses would be appropriate.
BTW I volunteer to create this as well.
—
Dr Peter M. Kelly
pmkelly@apache.org
PGP key: http://www.kellypmk.net/pgp-key <http://www.kellypmk.net/pgp-key>
(fingerprint 5435 6718 59F0 DD1F BFA0 5E46 2523 BAA1 44AE 2966)
Re: [PROPOSAL] make release 0.1
Posted by jan i <ja...@apache.org>.
On 1 August 2015 at 20:33, Dennis E. Hamilton <de...@acm.org>
wrote:
> Isn't this limiting contributors to those mentioned in a commit log entry
> or who push files to the repository?
>
it is, but that is not the key. The key is that the file is not used for
tracking, but are generated from the logs.
rgds
jan i.
>
> - Dennis
>
> -----Original Message-----
> From: jan i [mailto:jani@apache.org]
> Sent: Saturday, August 1, 2015 11:06
> To: dev@corinthia.incubator.apache.org
> Subject: Re: [PROPOSAL] make release 0.1
>
> On 1 August 2015 at 19:59, Andrea Pescetti <pe...@apache.org> wrote:
>
> > Peter Kelly wrote:
> >
> >> On 1 Aug 2015, at 10:18 pm, jan i wrote:
> >>> +1, I assume you mean the file contains the names of people who have
> >>> actively done commits on what is being released ?
> >>>
> >> Yes - everyone who’s name is present as an author or committer in the
> git
> >> log
> >>
> >
> > I've seen this debated somewhere. I don't remember the answer, but maybe
> > it's good to check with the Incubator PMC and ask. For sure it's stupid
> to
> > annotate the source files directly, and nobody is proposing this here. As
> > for a separate file with contributors, I find it totally reasonable but I
> > seem to recall a discussion saying that attribution at the ASF is tracked
> > exclusively in version control logs... I don't remember seeing a
> compelling
> > reason for that anyway.
> >
> I remember the debate, and the keyword here is "tracked". We cannot use
> CONTRIBUTORS to track e.g. people who do patches, that must in the version
> control itself.
>
> But we can extract from the version control and generate CONTRIBUTORS. In
> order to avoid discussions later (with IPMC), I would add a couple of
> sentences in the top of the file, explaining how it is generated.
>
> rgds
> jan i.
>
>
> >
> > Regards,
> > Andrea.
> >
>
>
RE: [PROPOSAL] make release 0.1
Posted by "Dennis E. Hamilton" <de...@acm.org>.
Isn't this limiting contributors to those mentioned in a commit log entry or who push files to the repository?
- Dennis
-----Original Message-----
From: jan i [mailto:jani@apache.org]
Sent: Saturday, August 1, 2015 11:06
To: dev@corinthia.incubator.apache.org
Subject: Re: [PROPOSAL] make release 0.1
On 1 August 2015 at 19:59, Andrea Pescetti <pe...@apache.org> wrote:
> Peter Kelly wrote:
>
>> On 1 Aug 2015, at 10:18 pm, jan i wrote:
>>> +1, I assume you mean the file contains the names of people who have
>>> actively done commits on what is being released ?
>>>
>> Yes - everyone who’s name is present as an author or committer in the git
>> log
>>
>
> I've seen this debated somewhere. I don't remember the answer, but maybe
> it's good to check with the Incubator PMC and ask. For sure it's stupid to
> annotate the source files directly, and nobody is proposing this here. As
> for a separate file with contributors, I find it totally reasonable but I
> seem to recall a discussion saying that attribution at the ASF is tracked
> exclusively in version control logs... I don't remember seeing a compelling
> reason for that anyway.
>
I remember the debate, and the keyword here is "tracked". We cannot use
CONTRIBUTORS to track e.g. people who do patches, that must in the version
control itself.
But we can extract from the version control and generate CONTRIBUTORS. In
order to avoid discussions later (with IPMC), I would add a couple of
sentences in the top of the file, explaining how it is generated.
rgds
jan i.
>
> Regards,
> Andrea.
>
Re: [PROPOSAL] make release 0.1
Posted by jan i <ja...@apache.org>.
On 1 August 2015 at 19:59, Andrea Pescetti <pe...@apache.org> wrote:
> Peter Kelly wrote:
>
>> On 1 Aug 2015, at 10:18 pm, jan i wrote:
>>> +1, I assume you mean the file contains the names of people who have
>>> actively done commits on what is being released ?
>>>
>> Yes - everyone who’s name is present as an author or committer in the git
>> log
>>
>
> I've seen this debated somewhere. I don't remember the answer, but maybe
> it's good to check with the Incubator PMC and ask. For sure it's stupid to
> annotate the source files directly, and nobody is proposing this here. As
> for a separate file with contributors, I find it totally reasonable but I
> seem to recall a discussion saying that attribution at the ASF is tracked
> exclusively in version control logs... I don't remember seeing a compelling
> reason for that anyway.
>
I remember the debate, and the keyword here is "tracked". We cannot use
CONTRIBUTORS to track e.g. people who do patches, that must in the version
control itself.
But we can extract from the version control and generate CONTRIBUTORS. In
order to avoid discussions later (with IPMC), I would add a couple of
sentences in the top of the file, explaining how it is generated.
rgds
jan i.
>
> Regards,
> Andrea.
>
Re: [PROPOSAL] make release 0.1
Posted by Andrea Pescetti <pe...@apache.org>.
Peter Kelly wrote:
>> On 1 Aug 2015, at 10:18 pm, jan i wrote:
>> +1, I assume you mean the file contains the names of people who have
>> actively done commits on what is being released ?
> Yes - everyone who’s name is present as an author or committer in the git log
I've seen this debated somewhere. I don't remember the answer, but maybe
it's good to check with the Incubator PMC and ask. For sure it's stupid
to annotate the source files directly, and nobody is proposing this
here. As for a separate file with contributors, I find it totally
reasonable but I seem to recall a discussion saying that attribution at
the ASF is tracked exclusively in version control logs... I don't
remember seeing a compelling reason for that anyway.
Regards,
Andrea.
Re: [PROPOSAL] make release 0.1
Posted by Peter Kelly <pm...@apache.org>.
> On 1 Aug 2015, at 10:18 pm, jan i <ja...@apache.org> wrote:
>
>> Another thing I would like to suggest is we have a CONTRIBUTORS file (or
>> similar name) which lists all those who have contributed to the project.
>> I’m aware this isn’t strictly required, but I think it's a good way to give
>> credit/recognition to everyone involved for their work.
>>
> +1, I assume you mean the file contains the names of people who have
> actively done commits on what is being released ?
Yes - everyone who’s name is present as an author or committer in the git log (or mentioned there as having been the author of a patch). I think a simple alphabetical listing of names & email addresses would be appropriate.
—
Dr Peter M. Kelly
pmkelly@apache.org
PGP key: http://www.kellypmk.net/pgp-key <http://www.kellypmk.net/pgp-key>
(fingerprint 5435 6718 59F0 DD1F BFA0 5E46 2523 BAA1 44AE 2966)
Re: [PROPOSAL] make release 0.1
Posted by jan i <ja...@apache.org>.
On 1 August 2015 at 17:15, Peter Kelly <pm...@apache.org> wrote:
> Sounds like a good plan to me.
>
> I volunteer to write the README - I think what we have on the website/wiki
> is a little unclear at the moment, and have some ideas on how to better
> express what the project’s about.
>
super, then we can later update the website/wiki or you do it the other way
round, whatever you prefer.
>
> I suggest we do this (along with LICENSES and NOTICE) as files in the root
> of the Git repository, so we can each make changes via the usual means as
> for code.
>
+1
>
> Another thing I would like to suggest is we have a CONTRIBUTORS file (or
> similar name) which lists all those who have contributed to the project.
> I’m aware this isn’t strictly required, but I think it's a good way to give
> credit/recognition to everyone involved for their work.
>
+1, I assume you mean the file contains the names of people who have
actively done commits on what is being released ?
rgds
jan i.
>
> —
> Dr Peter M. Kelly
> pmkelly@apache.org
>
> PGP key: http://www.kellypmk.net/pgp-key <http://www.kellypmk.net/pgp-key>
> (fingerprint 5435 6718 59F0 DD1F BFA0 5E46 2523 BAA1 44AE 2966)
>
> > On 30 Jul 2015, at 9:21 pm, jan i <ja...@apache.org> wrote:
> >
> > Traditionally cutting the first apache release for a podling is one big
> > PITA, therefore
> > I propose we get started.
> >
> > I propose we make release 0.1 consisting of
> > - docFormats (the library only with the OOXML filter)
> > - dfTest (the main test utility)
> > - dfUtil (the conversion utility)
> >
> > We have a lot to learn here, about the files LICENSES and NOTICE and
> about
> > how to
> > handle release candidates and the final release, therefore I propose to
> > include the
> > "stable" parts. It can be argued that version 0.1 is not very useful for
> a
> > end-user, which
> > might be right, but it helps us get the release framework up and running.
> >
> > Once we have 0.1 through the needle of IPMC, version 0.2 can focus a lot
> > more on the
> > technical content.
> >
> > The steps we have to pass (I have surely forgotten a number) are:
> > - Make a branch (stable_0.1) that only contains what we intent to release
> > - Make it work (adapt CMake files etc)
> > - Correct LICENSES and NOTICE
> > - Prepare a release candidate (upload source with SHA5 to somewhere on
> dist)
> > - Vote on the release (min. 3 +1 NO -1)
> > - Ask IPMC to check the release and have them vote on it
> > this is the needle, they tend to find things we have not thought about,
> > but Justin (he is one of the best release checkers in incubators)
> earlier
> > promised
> > me to do a pre-check so we avoid the normal errors.
> > - When the vote finally passes, move the source to the release part of
> dist.
> > - Congratulate each other with work well done.
> >
> > If we agree on the content and the idea of cutting a release, I volunteer
> > to do the work and
> > of course write about every step in here (and on the wiki), so we all
> learn
> > from the experience.
> > I do however need help from e.g. Dave, with some of the specialties.
> >
> > Following the suggestion from Dave, Let us discuss this until August 7th,
> > and then see
> > if we have reached consensus.
> >
> > Comments ?
> >
> > rgds
> > jan i.
>
>