I feel guilty for letting Rawstudio rotten.  Sorry Anders!  I started
hunting a bug with the colour of my SONY a850 but things got awfully
busy.  The little time I can spare for FLOSS is devoted to hunting a bug
in recent Linux kernels that makes Ubuntu 14.04 (and 14.10) nearly
useless on my laptop.  I hope to be done with the bisecting before
Christmas, and to have some time during the Holidays to hunt the colour
bug in Rawstudio.

On 14-11-29 05:20 PM, Tuomas Haarala wrote:
> On Sat, Nov 29, 2014 at 07:33:51PM +0000, Sérgio Basto wrote:
>> About plugins at least put it as optional, could be better , if breaks
>> compilation, we don't compile with it , but if someone want it, can fix
>> it . 
> 
> Ability to define own external program call as an upload plugin would
> be nice. Everyone could roll their own or use something that is
> generally available.

A Phyton scripting interface has the potential of playing wonders in
this area.  A few years back Kay Jahnke added a Python scripting
interface to Hugin and it has set some creative talent free.


>>> 1. Move the source repository to github.
>>> 2. Start fresh.

+1

>>> 3. Remove the following plugins from Rawstudio:

+1.  Expose a generic Python interface and let users of facebook,
flickr, picasa, etc. deal with that.  Focus on what Rawstudio was meant
to do: quick processing of Raw camera output.

>>> 4. Drop support for older versions

+1.  I would not bother Ubuntu 12.04.  Or any other version other than
the most recent one, and maybe the most recent LTS (14.04).  An (older)
version of Rawstudio is in the repos, users who want the bleeding edge
will take care of themselves, until the project is rebooted and has
sufficient capacity to provide backports.

>>> 5. Replace all deprecated API's

I don't know enough about this (i.e. how much effort? for what benefit?)
but if an API is deprecated, keeping it around is generally not a good
idea.  Developers are as attracted to deprecated API as car drivers are
attracted to old clunkers.

>>> 6. Drop all geolocation support

+1.  Same argument as with the plugins.

>>> 7. Fix all crashes and security related bugs.

I can't comment on this one.

>>> 8. Make package maintainers _happy_

+1, very important.  When I was active within Hugin, I migrated almost
all downstream patches from Debian, FreeBSD, Fedora.  Result: the
maintainers paid more attention to our release cycle and kept packages
up to date in their distros.  Some of them ended up contributing to
Hugin's build chain, documentation, and even code base.

>>> 9. Release version 2.1 as announced in 2011

+1.  Release soon, release often.  Even if the improvements are minor.
But beware of regression.  Encourage feature branches, and merge back
and release when the features are ready.

>>> 10. Celebrate with champagne.

Nobody can oppose this one, I guess :)

>>> (11. Find a new maintainer)

have you considered how more decentralization may reshape the role of
the maintainer?  Is a maintainer really necessary?  Github empowers
anybody to maintain their own forks.  make it a "federation" of
maintainers having a dialogue via pull requests on github.  Why
overburden yourself?

>>> Any thoughts?

Make it as easy as possible for others to take ownership of Rawstudio,
and they will come.

(1) build tools:  I have built Rawstudio on my machine, but it still
looks like mysterious magic to me.  CMake would help a lot.  I had
started something a while back, but it is an effort i would have to
start from scratch, and not having used CMake for the past four years
does not help.

(2) documentation:  When I asked the mailing list where to dig for my
colour bug, you gave me a beautiful walk through where to look.  You got
me there in very little time.  This kind of information should not get
lost in the archives of a mailing list (do not worry, I keep it handy so
that when I have time I have a map to remind me where to jump in).  Get
some sort of a wiki started, document the build system, the
dependencies, the architecture, etc, all in human written text directed
at noobs like myself.  Assume that you are writing for occasional,
sporadic contributors that need to be brought up to speed.

(3) focus:  To the extreme, focus on your needs.  Only your O/S, only
your camera; only the use you make of Rawstudio.  Working for others
will overload you and will not help the others in the long term either.
 Give a man fish and he will survive for a day.  Teach him how to fish
and he will survive for life.  Use your precious time to empower others.
 Try making your solutions generally applicable, so that others can add
support for other O/S, other cameras, and to a limited extent other use
cases (this may conflict with the narrow focus on RAW processing only,
so you will need to be selective about it).  Empowering others,
including your children, is the best way to make sure the software
survives your retirement.  Your children may become future maintainers
:)  Mine is well on the way to develop an interest
<http://vimeo.com/112571974> but has competing interests
<http://vimeo.com/111993345> and needs <http://vimeo.com/88614631>.  I
just make sure he does not fall for the dark side, where technology is
defined as consumer technology only.

Sorry for the longish dump, it is easier to write long than concise.

HTH
Yuv

Attachment: signature.asc
Description: OpenPGP digital signature

_______________________________________________
Rawstudio-dev mailing list
[email protected]
http://rawstudio.org/cgi-bin/mailman/listinfo/rawstudio-dev

Reply via email to