Mailing List Archive

Hoi,
I forgot to indicate that the relative costs of mentoring a GSoC person are
not high. I would even argue that it would benefit the WMF and MediaWiki
when we pay this from within the regular budget.

Growing our community should in my opinion be a key goal.
Thanks,
GerardM

2008/12/10 Gerard Meijssen <gerard.meijssen@gmail.com>

> Hoi,
> When a Google Summer of Code project is about aspects of usability, then we
> have on the one hand the cost of mentoring and on the other hand the benefit
> of three months of work done by typically really dedicated people. We have
> seen that people who started in a SoC program continued to be part of our
> community after their project.
>
> In my understanding the Stanton grant intends to improve our usability.
> Some people say that we "wiki people" make unusual decisions, have non
> traditionals approaches. I would applaud it when the Stanton grant allows us
> to do more then is traditional and consequently achieve more then is
> traditional.
> Thanks,
> GerardM
>
> 2008/12/10 Ting Chen <wing.philopp@gmx.de>
>
> THURNER rupert wrote:
>> > hi,
>> >
>> > on http://www.mediawiki.org/wiki/Summer_of_Code_2008 there was a
>> > statement that such efforts are restricted by "mentoring-manpower".
>> >
>> > now that there are real people and a budget dedicated to improve
>> > usability, could it make sense to leverage that effort by bounties
>> > given in a way comparable to google sumer of code?
>> >
>> > imo this might also used to attract additional donations from
>> > individuals, as it is simple paraphrasable. and usability is a real
>> > pain to a lot of people.
>> >
>> > kr, rupert.
>> > -----------------
>> > http://wikimedia.ch/donate - exempt your donation to wikipedia from
>> swiss tax!
>> >
>> > _______________________________________________
>> > foundation-l mailing list
>> > foundation-l@lists.wikimedia.org
>> > Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
>> >
>>
>> If you are inferring the Stanton grant, as far as I know not. That grant
>> is very closely defined and we cannot use it for other issues.
>>
>> Ting
>>
>> _______________________________________________
>> foundation-l mailing list
>> foundation-l@lists.wikimedia.org
>> Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
>>
>
>
_______________________________________________
foundation-l mailing list
foundation-l@lists.wikimedia.org
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l

THURNER rupert wrote:
> hi,
>
> on http://www.mediawiki.org/wiki/Summer_of_Code_2008 there was a
> statement that such efforts are restricted by "mentoring-manpower".
>
> now that there are real people and a budget dedicated to improve
> usability, could it make sense to leverage that effort by bounties
> given in a way comparable to google sumer of code?
>
> imo this might also used to attract additional donations from
> individuals, as it is simple paraphrasable. and usability is a real
> pain to a lot of people.

GSoC has never produced anything useful for us, so I don't know why you
think it would be a good model. Our record with contract development has
also been pretty poor.

Our most active volunteers are adept at collaborating online in public
forums. That's because they're self-selected for that trait and they're
adequately motivated that they can perform useful projects without
supervision. Remote workers brought in via GSoC and contracts seem to have
difficulty integrating with this culture, and either perform their work in
isolation or collaborate only with their designated mentor.

The idea of a collection of remote workers paid by the line of code might
sound nice to an online community member, but I'm beginning to think that
it's risky at best. A software development team working in an office
together might be old-school, but at least the management practices are
established, with good results commonly produced.

-- Tim Starling


_______________________________________________
foundation-l mailing list
foundation-l@lists.wikimedia.org
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l

On Thu, Dec 11, 2008 at 11:02 AM, Tim Starling <tstarling@wikimedia.org> wrote:
> GSoC has never produced anything useful for us, so I don't know why you
> think it would be a good model.

But it has produced useful results for a number of other open source
software communities, so I don't think it's prudent to kick the idea
out the window just yet. What's needed to make a system like GSoC work
is to have a large team of prospective mentors, effort to integrate
the volunteers into the community, etc. I've seen other groups who
have been very successful from this program, adding a large portion of
students as active committers after the program is over. If MediaWiki
hasn't been as successful, it might be worthwhile figuring out what is
being done differently from those projects which are more successful.

> The idea of a collection of remote workers paid by the line of code might
> sound nice to an online community member, but I'm beginning to think that
> it's risky at best. A software development team working in an office
> together might be old-school, but at least the management practices are
> established, with good results commonly produced.

I've heard the idea of code bounties discussed before, maybe that's a
model that's worth reconsidering. Take your existing development staff
and ancillary contributors and offer firm cash rewards for specific
features. Ideally new contributors will join the effort in pursuit of
these rewards. Again, something to consider.

--Andrew Whitwoth

_______________________________________________
foundation-l mailing list
foundation-l@lists.wikimedia.org
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l

Hoi,
Last year Nikerabbit was enrolled in a Finnish Summer of Code project. He
did a ton of great work for Betawiki as part of this project. The Liquid
Threads project was a GSOC project. It is used by the WikiEducator project
and as such I would rate it successful.

When you look at our own bigger projects, SUL took a crazy amount of time to
materialise, we are still not able to produce predictable data dumps. When
you look at commercial projects, at least 50% of such projects fail to meet
expectations. The notion that classical "in the office" projects do better
is not one I share.

When we are to do a proper job for summer of code projects, obviously all
our existing developers are most likely to do the better job. Nikerabbit's
project is a case in point. If there are observations why such a project
does not work out as well as we hope, we should address those issues. The
most important thing achieved with a summer of code project is not only the
software but also the experience given to what we hope will be a developer
who stays with our project after his project.
Thanks,
GerardM

2008/12/11 Tim Starling <tstarling@wikimedia.org>

> THURNER rupert wrote:
> > hi,
> >
> > on http://www.mediawiki.org/wiki/Summer_of_Code_2008 there was a
> > statement that such efforts are restricted by "mentoring-manpower".
> >
> > now that there are real people and a budget dedicated to improve
> > usability, could it make sense to leverage that effort by bounties
> > given in a way comparable to google sumer of code?
> >
> > imo this might also used to attract additional donations from
> > individuals, as it is simple paraphrasable. and usability is a real
> > pain to a lot of people.
>
> GSoC has never produced anything useful for us, so I don't know why you
> think it would be a good model. Our record with contract development has
> also been pretty poor.
>
> Our most active volunteers are adept at collaborating online in public
> forums. That's because they're self-selected for that trait and they're
> adequately motivated that they can perform useful projects without
> supervision. Remote workers brought in via GSoC and contracts seem to have
> difficulty integrating with this culture, and either perform their work in
> isolation or collaborate only with their designated mentor.
>
> The idea of a collection of remote workers paid by the line of code might
> sound nice to an online community member, but I'm beginning to think that
> it's risky at best. A software development team working in an office
> together might be old-school, but at least the management practices are
> established, with good results commonly produced.
>
> -- Tim Starling
>
>
> _______________________________________________
> foundation-l mailing list
> foundation-l@lists.wikimedia.org
> Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
>
_______________________________________________
foundation-l mailing list
foundation-l@lists.wikimedia.org
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l

On Thu, Dec 11, 2008 at 11:37 AM, Gerard Meijssen <gerard.meijssen@gmail.com
> wrote:

> Hoi,
> Last year Nikerabbit was enrolled in a Finnish Summer of Code project. He
> did a ton of great work for Betawiki as part of this project. The Liquid
> Threads project was a GSOC project. It is used by the WikiEducator project
> and as such I would rate it successful.
>

That's great, but neither Betawiki nor WikiEducator are the WMF, nor
are those functionalities being used (beyond the localizations provided
from BW) within the WMF. That's precisely what this is about, making
use of the GSOC-style of development for MediaWiki itself.

When you look at our own bigger projects, SUL took a crazy amount of time to
> materialise, we are still not able to produce predictable data dumps. When
> you look at commercial projects, at least 50% of such projects fail to meet
> expectations. The notion that classical "in the office" projects do better
> is not one I share.


SUL required a massive amount of work to coordinate, not to mention the
task of coming up with a model that not only works, but is scalable to the
WMF's needs and also actually make sense. Good data models are essential.
Dumps are another thing that requires careful work and coordination. Poor
execution leads to poor results. It's bad enough having someone e-mail the
list(s) once every month or two saying "new dumps please," but imagine if
we provided dumps that were just inherently bad.

I fail to see where you get this 50% of commercial projects fail. Without a
reliable source, I have to assume you've just made up this number and have
never worked in software development.

When we are to do a proper job for summer of code projects, obviously all
> our existing developers are most likely to do the better job. Nikerabbit's
> project is a case in point. If there are observations why such a project
> does not work out as well as we hope, we should address those issues. The
> most important thing achieved with a summer of code project is not only the
> software but also the experience given to what we hope will be a developer
> who stays with our project after his project.
> Thanks,
> GerardM
>

Tim never said that GSOC is a bad model. Nor did he say that it produces bad
results. He simply said that based on previous experiences, it hasn't worked
_for us_. And it's true. Looking at last summer's projects, I don't see a
lot of
results:

1) HTMLDiff - ended up as a highly experimental, still somewhat buggy,
disabled-
by-default feature that was an i18n mess (and still might not be 100% fixed)
2) Category Redirects - what ever happened to this? Was this ever merged
from
branch to core? The branch hasn't been touched since August, at the least.

A model failing in one use case doesn't indicate a failed model overall; it
simply
means it doesn't work in this situation. :)

-Chad
_______________________________________________
foundation-l mailing list
foundation-l@lists.wikimedia.org
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l

Hoi,
When you consider that the WMF and its projects have not benefited from
Betawiki and its functionality, then you prove again for how little
internationalisation and localisation count within MediaWiki development.
Consistently you will find Siebrand or Nikerabbit as top developers for
creating MediaWiki patches. When you exclude Betawiki from what "we" are
about.... really sad.

MediaWiki development is not only development within the Wikimedia
Foundation. There have been several projects that should have benefited the
WMF projects but did not. There was a presentation on usability at Wikimania
Boston and in Alexandria and none of the lessons learned materialised in
"our" code.

Let me be clear, I am happy that SUL finally materialised and with 430 users
active under my global account, I benefit from it a lot. However, the
functionality it provides is minimal. It does not allow me to share many of
my preferences. It works, it is great but it could be so much more. There
are always "good" reasons why thing are like they are. Given that we under
invested last year relative to the budget, I am amazed by what has been
done, however I am not satisfied with the results.

As to experience as a developer, I have some 20 years of experience as a
consultant. I have seen some great projects and some really bad ones. There
have been over time several publications that inform how many projects have
gone bad. With 50% I am being nice.
Thanks,
GerardM

2008/12/11 Chad <innocentkiller@gmail.com>

> On Thu, Dec 11, 2008 at 11:37 AM, Gerard Meijssen <
> gerard.meijssen@gmail.com
> > wrote:
>
> > Hoi,
> > Last year Nikerabbit was enrolled in a Finnish Summer of Code project. He
> > did a ton of great work for Betawiki as part of this project. The Liquid
> > Threads project was a GSOC project. It is used by the WikiEducator
> project
> > and as such I would rate it successful.
> >
>
> That's great, but neither Betawiki nor WikiEducator are the WMF, nor
> are those functionalities being used (beyond the localizations provided
> from BW) within the WMF. That's precisely what this is about, making
> use of the GSOC-style of development for MediaWiki itself.
>
> When you look at our own bigger projects, SUL took a crazy amount of time
> to
> > materialise, we are still not able to produce predictable data dumps.
> When
> > you look at commercial projects, at least 50% of such projects fail to
> meet
> > expectations. The notion that classical "in the office" projects do
> better
> > is not one I share.
>
>
> SUL required a massive amount of work to coordinate, not to mention the
> task of coming up with a model that not only works, but is scalable to the
> WMF's needs and also actually make sense. Good data models are essential.
> Dumps are another thing that requires careful work and coordination. Poor
> execution leads to poor results. It's bad enough having someone e-mail the
> list(s) once every month or two saying "new dumps please," but imagine if
> we provided dumps that were just inherently bad.
>
> I fail to see where you get this 50% of commercial projects fail. Without a
> reliable source, I have to assume you've just made up this number and have
> never worked in software development.
>
> When we are to do a proper job for summer of code projects, obviously all
> > our existing developers are most likely to do the better job.
> Nikerabbit's
> > project is a case in point. If there are observations why such a project
> > does not work out as well as we hope, we should address those issues. The
> > most important thing achieved with a summer of code project is not only
> the
> > software but also the experience given to what we hope will be a
> developer
> > who stays with our project after his project.
> > Thanks,
> > GerardM
> >
>
> Tim never said that GSOC is a bad model. Nor did he say that it produces
> bad
> results. He simply said that based on previous experiences, it hasn't
> worked
> _for us_. And it's true. Looking at last summer's projects, I don't see a
> lot of
> results:
>
> 1) HTMLDiff - ended up as a highly experimental, still somewhat buggy,
> disabled-
> by-default feature that was an i18n mess (and still might not be 100%
> fixed)
> 2) Category Redirects - what ever happened to this? Was this ever merged
> from
> branch to core? The branch hasn't been touched since August, at the least.
>
> A model failing in one use case doesn't indicate a failed model overall; it
> simply
> means it doesn't work in this situation. :)
>
> -Chad
> _______________________________________________
> foundation-l mailing list
> foundation-l@lists.wikimedia.org
> Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
>
_______________________________________________
foundation-l mailing list
foundation-l@lists.wikimedia.org
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l

Gerard Meijssen wrote:
> Hoi,
> When you consider that the WMF and its projects have not benefited from
> Betawiki and its functionality, then you prove again for how little
> internationalisation and localisation count within MediaWiki development.
> Consistently you will find Siebrand or Nikerabbit as top developers for
> creating MediaWiki patches. When you exclude Betawiki from what "we" are
> about.... really sad.

I certainly wouldn't exclude Nikerabbit and Betawiki from the list of
projects which benefit the Foundation, but I think he probably would have
done it anyway. I said that new developers are poorly integrated, he
wasn't a new developer.

I do count David McCabe's project LiquidThreads among those that have not
been useful for Wikimedia.

-- Tim Starling


_______________________________________________
foundation-l mailing list
foundation-l@lists.wikimedia.org
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l

I see Summer of Code is kind like gambling or a long term investment...

The metavid.org project was partially inspired by my SOC participation
as a student for mediaWiki in SOC 06. While metavid efforts will likely
not directly benefit Wikipedia until 2009. Metavid pioneered a lot of
the technologies that are making their way into the software. For
example archive.org shipping oggz_chop:
http://metavid.org/blog/2008/12/08/archiveorg-ogg-support/ and or the
sequencer efforts currently under way with kaltura.

I mentored two students this summer, one for metavid under wikimedia's
SOC. While its true as students coming fresh to the code they do need a
lot of hand holding and it would have been more productive and easier to
just "do it myself". I see it as more an effort for community building
and a future investment in getting people to think about participating
in open source.
The other project mentored for xiph was mostly a failure.. but the
project concept has been picked up by a community member and is now
ready for integration testing with mediawiki upload system :)
http://www.firefogg.org/

I think its fine to have SOC projects "fail" in getting code onto
production servers while succeed in testing the waters for high risk
areas and plant seeds for improving the free software ecosystem.

I think wikimedia should continue to participate and maybe avoid high
risk by not committing core developers to be mentors. This should be
more possible now that technical staff is expanding beyond a few
stretched extremely thin people.

peace,
--michael

Chad wrote:
> On Thu, Dec 11, 2008 at 11:37 AM, Gerard Meijssen <gerard.meijssen@gmail.com
>
>> wrote:
>>
>
>
>> Hoi,
>> Last year Nikerabbit was enrolled in a Finnish Summer of Code project. He
>> did a ton of great work for Betawiki as part of this project. The Liquid
>> Threads project was a GSOC project. It is used by the WikiEducator project
>> and as such I would rate it successful.
>>
>>
>
> That's great, but neither Betawiki nor WikiEducator are the WMF, nor
> are those functionalities being used (beyond the localizations provided
> from BW) within the WMF. That's precisely what this is about, making
> use of the GSOC-style of development for MediaWiki itself.
>
> When you look at our own bigger projects, SUL took a crazy amount of time to
>
>> materialise, we are still not able to produce predictable data dumps. When
>> you look at commercial projects, at least 50% of such projects fail to meet
>> expectations. The notion that classical "in the office" projects do better
>> is not one I share.
>>
>
>
> SUL required a massive amount of work to coordinate, not to mention the
> task of coming up with a model that not only works, but is scalable to the
> WMF's needs and also actually make sense. Good data models are essential.
> Dumps are another thing that requires careful work and coordination. Poor
> execution leads to poor results. It's bad enough having someone e-mail the
> list(s) once every month or two saying "new dumps please," but imagine if
> we provided dumps that were just inherently bad.
>
> I fail to see where you get this 50% of commercial projects fail. Without a
> reliable source, I have to assume you've just made up this number and have
> never worked in software development.
>
> When we are to do a proper job for summer of code projects, obviously all
>
>> our existing developers are most likely to do the better job. Nikerabbit's
>> project is a case in point. If there are observations why such a project
>> does not work out as well as we hope, we should address those issues. The
>> most important thing achieved with a summer of code project is not only the
>> software but also the experience given to what we hope will be a developer
>> who stays with our project after his project.
>> Thanks,
>> GerardM
>>
>>
>
> Tim never said that GSOC is a bad model. Nor did he say that it produces bad
> results. He simply said that based on previous experiences, it hasn't worked
> _for us_. And it's true. Looking at last summer's projects, I don't see a
> lot of
> results:
>
> 1) HTMLDiff - ended up as a highly experimental, still somewhat buggy,
> disabled-
> by-default feature that was an i18n mess (and still might not be 100% fixed)
> 2) Category Redirects - what ever happened to this? Was this ever merged
> from
> branch to core? The branch hasn't been touched since August, at the least.
>
> A model failing in one use case doesn't indicate a failed model overall; it
> simply
> means it doesn't work in this situation. :)
>
> -Chad
> _______________________________________________
> foundation-l mailing list
> foundation-l@lists.wikimedia.org
> Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
>


_______________________________________________
foundation-l mailing list
foundation-l@lists.wikimedia.org
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l