Re: [darcs-users] new darcs maintainer: Guillaume Hoffmann

[Guillaume Hoffmann]

Hi David,

> At the moment I don't really know what to expect from Darcs.  The Darcs
> development team is passionate and highly competent but very small.  I also
> don't know what the direction will be.  What are the major improvements
> you'd all like to work on?

Right now, I'm not working on any major improvement but rather seeing
what should go into the next release, which I'd like to do next
February. It will probably not have any new killer feature and will be
mostly code improvements with some visible improvements (hopefully
we'll solve the "couldn't fetch" bugs this time!). Also it is good to
regularly keep up with new GHC releases.

In parallel I'd like to have students work on projects. I put online
the following Projects page: http://darcs.net/Projects , which would
be for any kind of student projects, not only GSoC.

> From the top of my head the perceived shortcomings of previous Git users are

> - switching branches is not possible within the same directory [...]

+1

See http://bugs.darcs.net/issue555 for an old discussion about this
and a shell script that "implements" this feature, while we wait for
Ganesh's stash implementation :)

> - it's not as easy to refer to a specific state of the repository using a
> hash

I'd like to see progress on this issue, within the possibilities of
darcs' current repo format.

See http://bugs.darcs.net/issue992 and http://darcs.net/Ideas/ShortSecureId

Or maybe can we use darcsden somehow to store more information about
repositories?

> - when merging conflicts it's difficult to understand which part of the
> marked conflict belongs to which patch

See http://bugs.darcs.net/issue833

Ganesh did some work on this:

http://hub.darcs.net/ganesh/darcs-conflict-marking

I remember Ganesh said it was inefficient, but anyway I wonder if we
can have this into 2.12 ?

> - I'd like to amend-record a patch that other patches depend on (without
> having to suspend, amend-record, unsuspend)

I wonder if we can have such detection of cases where rebasing can be
automatic? It would definitely be great to have!

thanks for the feedback David!

Guillaume
___
darcs-users mailing list
darcs-users@darcs.net
http://lists.osuosl.org/mailman/listinfo/darcs-users

Re: [darcs-users] new darcs maintainer: Guillaume Hoffmann

[Ganesh Sittampalam]

My main priorities, in no particular order:

 - fix the conflict problem once and for all
- I'm very excited by the recent Pijul announcement and hope to be
able to adopt the same approach for Darcs.

 - develop darcsden as a good local UI as well as hosting platform
- We've had some GSoC projects to help this along in the last few years.

 - develop my stash prototype and release it
- with some UI work, I think this addresses the "switching branches"
point below.

 - easy two-way darcs-git interop
- we learnt a bunch of stuff when writing the original bridge, but
need to start the imlementation again to get this right.

 - Patch groups (as below)

 - Improve rebase (including the items below)

I tend to work on things in no particular order as and when the mood
takes me, which doesn't really make for a great roadmap, I'm afraid!


On 26/10/2015 15:04, David Leuschner wrote:
> Hi Guillaume,
> 
> thank you very much for dedicating so much time to Darcs!   As you
> probably know, we at factis research use Darcs for all of our projects
> since about 2006.  Stefan and I are very happy with Darcs and since we
> have rebase there are no major shortcomings.  Still it becomes
> constantly harder for us to justify the decision to use Darcs in front
> of our younger and newer colleagues, who have been using Git previously.
> At the moment I don't really know what to expect from Darcs.  The Darcs
> development team is passionate and highly competent but very small.  I
> also don't know what the direction will be.  What are the major
> improvements you'd all like to work on?
> 
> From the top of my head the perceived shortcomings of previous Git users are
> - switching branches is not possible within the same directory (keeping
> the directory is important because build products can be reused)
> - there's no way to group patches ("feature branches") to be able to
> treat them as a unit (for pulling/applying/suspending)
> - there's no way to see at which time and by whom a patch has been
> applied to a repository
> - it's not as easy to refer to a specific state of the repository using
> a hash
> - when merging conflicts it's difficult to understand which part of the
> marked conflict belongs to which patch
> 
> Other frequent questions and wishes include
> - Why can't I push to and pull from a repository with a rebase in progress?
> - Why do I have to record all my changes before unsuspending a
> (unrelated) patch?
> - I'd like to explicitly swap to patches without having to suspend one,
> obliterate the other, unsupend the first, reapply the other
> - I'd like to undo unsuspending a patch
> - I'd like to amend-record a patch that other patches depend on (without
> having to suspend, amend-record, unsuspend)
> 
> Thank's for taking on maintainership!  We're always looking forward to
> the next release!
> 
> Cheers,
> 
>  David
> 
> 2015-10-08 20:36 GMT+02:00 Guillaume Hoffmann  >:
> 
> Hi everyone,
> 
> I guess I should say a few words about this :-)
> 
> First I'd like to thank Eric for his work as a maintainer for 7 years,
> and hope he'll stick around!
> 
> For the observers that might wonder what is going to change for the
> Darcs project, the answer is: not much.
> 
> Ever since I started participating to Darcs (2009, my first sprint
> attendance :,-) ), I found the project to be very collaborative and
> consensus-lead. Anyone is welcome to send patches, code reviews are
> public and are a great opportunity to learn about Darcs' codebase;
> release management duties have been shared between several members over
> time; coding sprints are open for participation, and indeed, every time
> we have a few non-darcsers visiting us and helping. I am very happy
> with these aspects of Darcs as a free software project, and Eric ensured
> all of this happened over the years.
> 
> So, things will continue as before!
> 
> As for Darcs the software, how is it going to evolve? I cannot answer
> this question alone for the long term. What I can say, is that in the
> past I sometimes found Darcs' evolution as being a little too slow and
> conservative. We lost valuable energy and focus in decisions which did
> not pay off eventually. On some topics we did not reach any conclusion
> and sticked to the status quo. My guess is that this drove a few hackers
> to stop contributing, while the users also left anyway!
> 
> I don't mean that we're going to break everything just to make us
> developers happy. The very nature of Darcs leads us to be conservative:
> Darcs is not just a software but also a data format (Darcs repositories)
> which we don't want to change unless we have a *really* good reason
> to do so. Also, Darcs handles other people's data, so we want it to be
> safe and predictable. On the other hand, I think we should really commit
> to 

Re: [darcs-users] new darcs maintainer: Guillaume Hoffmann

[David Leuschner]

Hi Guillaume,

thank you very much for dedicating so much time to Darcs!   As you probably
know, we at factis research use Darcs for all of our projects since about
2006.  Stefan and I are very happy with Darcs and since we have rebase
there are no major shortcomings.  Still it becomes constantly harder for us
to justify the decision to use Darcs in front of our younger and newer
colleagues, who have been using Git previously.
At the moment I don't really know what to expect from Darcs.  The Darcs
development team is passionate and highly competent but very small.  I also
don't know what the direction will be.  What are the major improvements
you'd all like to work on?

>From the top of my head the perceived shortcomings of previous Git users are
- switching branches is not possible within the same directory (keeping the
directory is important because build products can be reused)
- there's no way to group patches ("feature branches") to be able to treat
them as a unit (for pulling/applying/suspending)
- there's no way to see at which time and by whom a patch has been applied
to a repository
- it's not as easy to refer to a specific state of the repository using a
hash
- when merging conflicts it's difficult to understand which part of the
marked conflict belongs to which patch

Other frequent questions and wishes include
- Why can't I push to and pull from a repository with a rebase in progress?
- Why do I have to record all my changes before unsuspending a (unrelated)
patch?
- I'd like to explicitly swap to patches without having to suspend one,
obliterate the other, unsupend the first, reapply the other
- I'd like to undo unsuspending a patch
- I'd like to amend-record a patch that other patches depend on (without
having to suspend, amend-record, unsuspend)

Thank's for taking on maintainership!  We're always looking forward to the
next release!

Cheers,

 David

2015-10-08 20:36 GMT+02:00 Guillaume Hoffmann :

> Hi everyone,
>
> I guess I should say a few words about this :-)
>
> First I'd like to thank Eric for his work as a maintainer for 7 years,
> and hope he'll stick around!
>
> For the observers that might wonder what is going to change for the
> Darcs project, the answer is: not much.
>
> Ever since I started participating to Darcs (2009, my first sprint
> attendance :,-) ), I found the project to be very collaborative and
> consensus-lead. Anyone is welcome to send patches, code reviews are
> public and are a great opportunity to learn about Darcs' codebase;
> release management duties have been shared between several members over
> time; coding sprints are open for participation, and indeed, every time
> we have a few non-darcsers visiting us and helping. I am very happy
> with these aspects of Darcs as a free software project, and Eric ensured
> all of this happened over the years.
>
> So, things will continue as before!
>
> As for Darcs the software, how is it going to evolve? I cannot answer
> this question alone for the long term. What I can say, is that in the
> past I sometimes found Darcs' evolution as being a little too slow and
> conservative. We lost valuable energy and focus in decisions which did
> not pay off eventually. On some topics we did not reach any conclusion
> and sticked to the status quo. My guess is that this drove a few hackers
> to stop contributing, while the users also left anyway!
>
> I don't mean that we're going to break everything just to make us
> developers happy. The very nature of Darcs leads us to be conservative:
> Darcs is not just a software but also a data format (Darcs repositories)
> which we don't want to change unless we have a *really* good reason
> to do so. Also, Darcs handles other people's data, so we want it to be
> safe and predictable. On the other hand, I think we should really commit
> to the transitions we set ourselves to do, and remember that developer
> energy is a scarce resource.
>
> Now for the short term, what is going to happen? We had a big 2.10
> release last April, 3 years after 2.8. It included a lot of new stuff,
> which made us very happy; but the release process was consequently
> quite long. I'd like the next release to happen sooner and I propose
> myself to take care of it for a second time. The release process will
> start after the GHC 8 release and the next sprint, which means February.
>
> Happy hacking!
>
> Guillaume
> ___
> darcs-users mailing list
> darcs-users@darcs.net
> http://lists.osuosl.org/mailman/listinfo/darcs-users
>
___
darcs-users mailing list
darcs-users@darcs.net
http://lists.osuosl.org/mailman/listinfo/darcs-users

Re: [darcs-users] new darcs maintainer: Guillaume Hoffmann

[David Leuschner]

Hi Eric,

thank you very much for the great work you've done!  We hope that you'll
stick around!


Cheers,

  David

2015-10-08 16:58 GMT+02:00 Eric Kow :

> Hello Darcs users!
>
> So last month I had a chance to reunite with the active members of the
> Darcs team (Ganesh and Guillaume) at the Paris hacking sprint.  It was
> a very nice time and I hope to see everyone again soon. During the
> sprint (in between hack sessions), we also spent some time thinking
> about the general future of Darcs, what sorts of directions we could
> go in and how we can go about getting there. For example, what if we
> thought a little outside the version control box? Could Darcs find a
> niche for for making patch reordering work for other version control
> systems? Possibly. It's worth exploring.
>
> It's a different world out there now (but ever changing) than seven
> years ago. In particular I think we have done an absolutely *splendid*
> job adhering to the old Haskell slogan “avoid success at all costs!”
> ;-).  On the one hand the smaller community makes it harder to find
> help and get effort directed at things. On the other hand, it's also a
> blessing of sorts. With a little less juggling on our hands, we can
> now direct more attention to the core work of getting Darcs right
> (that's why we're pretty excited to see Pijul; this sort of work is
> very helpful to us).
>
> We still love Darcs; we still think it has something unique to offer
> to the world and we still think it's interesting enough to work on.
> There is a lot of work to do: hammering out of the core theory,
> cleaning up of the code base, improving the basic engineering, the UI,
> the ecosystem… In the meantime, I am glad to have been able to spend
> some more time with Darcs folks. I get a lot of happiness out seeing
> that people are still interested in working on Darcs and taking us
> further along.
>
> That said, as much as I may love Darcs personally, I also recognise
> that for the past three years and perhaps many more I have not been in
> a place where I knew how to fit free/open source software hacking into
> my life. In fact, I hadn't been doing any such fitting at all, having
> wandered off except for manning the spam filter, writing the odd HCAR
> report. I've even — gasp! — gotten much much better at using Git.
>
> This is where Guillaume come in. Luckily for Darcs, some active
> members of the team have been merrily going about the business of not
> just hacking on Darcs, but running summer of code projects, recruiting
> new developers, making links with the academic world. So how about we
> make it official? How about we swap Guillaume in as the new face of
> Darcs since he's already been effectively doing the job anyway?
>
> I hope I didn't overdo the “you'll be fine!” downplaying when I was in
> Paris.  The role is a bit more than a nominal one and does come with
> some responsibility. But I've seen Guillaume at work, seen the energy
> he's putting into organising things, into chasing up on people, seen
> the way he runs a sprint. I think he'll be more than fine.
>
> So without further ado… I'd like to say, welcome to our new
> maintainer, Guillaume Hoffmann!
>
> Eric
>
> PS: Good luck! No really, you will just be fine! :-)
>
> PPS Unfortunately, I'm not likely to make it to Seville for the next
> one  but hope you have a great time,
> and that I can see you in the future.
>
> --
> Eric Kow 
> ___
> darcs-users mailing list
> darcs-users@darcs.net
> http://lists.osuosl.org/mailman/listinfo/darcs-users
>
___
darcs-users mailing list
darcs-users@darcs.net
http://lists.osuosl.org/mailman/listinfo/darcs-users

[darcs-users] new darcs maintainer: Guillaume Hoffmann

[Eric Kow]

Hello Darcs users!

So last month I had a chance to reunite with the active members of the
Darcs team (Ganesh and Guillaume) at the Paris hacking sprint.  It was
a very nice time and I hope to see everyone again soon. During the
sprint (in between hack sessions), we also spent some time thinking
about the general future of Darcs, what sorts of directions we could
go in and how we can go about getting there. For example, what if we
thought a little outside the version control box? Could Darcs find a
niche for for making patch reordering work for other version control
systems? Possibly. It's worth exploring.

It's a different world out there now (but ever changing) than seven
years ago. In particular I think we have done an absolutely *splendid*
job adhering to the old Haskell slogan “avoid success at all costs!”
;-).  On the one hand the smaller community makes it harder to find
help and get effort directed at things. On the other hand, it's also a
blessing of sorts. With a little less juggling on our hands, we can
now direct more attention to the core work of getting Darcs right
(that's why we're pretty excited to see Pijul; this sort of work is
very helpful to us).

We still love Darcs; we still think it has something unique to offer
to the world and we still think it's interesting enough to work on.
There is a lot of work to do: hammering out of the core theory,
cleaning up of the code base, improving the basic engineering, the UI,
the ecosystem… In the meantime, I am glad to have been able to spend
some more time with Darcs folks. I get a lot of happiness out seeing
that people are still interested in working on Darcs and taking us
further along.

That said, as much as I may love Darcs personally, I also recognise
that for the past three years and perhaps many more I have not been in
a place where I knew how to fit free/open source software hacking into
my life. In fact, I hadn't been doing any such fitting at all, having
wandered off except for manning the spam filter, writing the odd HCAR
report. I've even — gasp! — gotten much much better at using Git.

This is where Guillaume come in. Luckily for Darcs, some active
members of the team have been merrily going about the business of not
just hacking on Darcs, but running summer of code projects, recruiting
new developers, making links with the academic world. So how about we
make it official? How about we swap Guillaume in as the new face of
Darcs since he's already been effectively doing the job anyway?

I hope I didn't overdo the “you'll be fine!” downplaying when I was in
Paris.  The role is a bit more than a nominal one and does come with
some responsibility. But I've seen Guillaume at work, seen the energy
he's putting into organising things, into chasing up on people, seen
the way he runs a sprint. I think he'll be more than fine.

So without further ado… I'd like to say, welcome to our new
maintainer, Guillaume Hoffmann!

Eric

PS: Good luck! No really, you will just be fine! :-)

PPS Unfortunately, I'm not likely to make it to Seville for the next
one  but hope you have a great time,
and that I can see you in the future.

-- 
Eric Kow 
___
darcs-users mailing list
darcs-users@darcs.net
http://lists.osuosl.org/mailman/listinfo/darcs-users

Re: [darcs-users] new darcs maintainer: Guillaume Hoffmann

[Guillaume Hoffmann]

Hi everyone,

I guess I should say a few words about this :-)

First I'd like to thank Eric for his work as a maintainer for 7 years,
and hope he'll stick around!

For the observers that might wonder what is going to change for the
Darcs project, the answer is: not much.

Ever since I started participating to Darcs (2009, my first sprint
attendance :,-) ), I found the project to be very collaborative and
consensus-lead. Anyone is welcome to send patches, code reviews are
public and are a great opportunity to learn about Darcs' codebase;
release management duties have been shared between several members over
time; coding sprints are open for participation, and indeed, every time
we have a few non-darcsers visiting us and helping. I am very happy
with these aspects of Darcs as a free software project, and Eric ensured
all of this happened over the years.

So, things will continue as before!

As for Darcs the software, how is it going to evolve? I cannot answer
this question alone for the long term. What I can say, is that in the
past I sometimes found Darcs' evolution as being a little too slow and
conservative. We lost valuable energy and focus in decisions which did
not pay off eventually. On some topics we did not reach any conclusion
and sticked to the status quo. My guess is that this drove a few hackers
to stop contributing, while the users also left anyway!

I don't mean that we're going to break everything just to make us
developers happy. The very nature of Darcs leads us to be conservative:
Darcs is not just a software but also a data format (Darcs repositories)
which we don't want to change unless we have a *really* good reason
to do so. Also, Darcs handles other people's data, so we want it to be
safe and predictable. On the other hand, I think we should really commit
to the transitions we set ourselves to do, and remember that developer
energy is a scarce resource.

Now for the short term, what is going to happen? We had a big 2.10
release last April, 3 years after 2.8. It included a lot of new stuff,
which made us very happy; but the release process was consequently
quite long. I'd like the next release to happen sooner and I propose
myself to take care of it for a second time. The release process will
start after the GHC 8 release and the next sprint, which means February.

Happy hacking!

Guillaume
___
darcs-users mailing list
darcs-users@darcs.net
http://lists.osuosl.org/mailman/listinfo/darcs-users