Yet another blog for JBKempf - Tag - GSoC

State of VideoLAN

Jean-Baptiste Kempf — Wed, 11 May 2011 12:42:00 +0200

It's been a long time

It's been a long time since the last time I wrote something long enough around here.
But the cool thing is that it is due to being working quite a lot on VLC and different projects...

So, how well is VideoLAN? How well is VLC?

In this post, I will come back on a few points that happened in our last year.

1. VLC

VLC is doing quite well actually, but some things could be better...

1.1.0

VLC 1.1.0 was out last year, in June.

This release was a tremendous success, and probably the one where we've had the best success for launch.

The focus on GPU decoding and some HD codecs was very well welcomed by users, as was the improved Qt interface.

Of course, 1.1.0 had a lot of bugs, since people still refuse to test our betas and RC versions, and as usual, 1.1.1 was ready and out in less than a month.

1.1.x

The 1.1.x-bugfix branch is quite stable and feature-full.

We still have notorious issues with PulseAudio, DVD Unicode path and Font caching on Windows. But all of them have been fixed in the git master.

1.1.0 to 1.1.9 updates

We've done 9 minor additional releases in 11 months from 1.1.0 to 1.1.9. And 1.1.10 is imminent...

While, this is a bit annoying for the users, seeing the poor update mechanism in VLC, this is mostly due to numerous security issues found and fixed in VLC.

This shows that a very small team, like VLC's can care enough to support and have a strong security focus... Even when people are volunteers.

Numbers

Since the switch to SourceForge, for mirroring the downloads, VLC has been downloaded 237 millions of times, in 11 months.

The repartition per country, is the usual one as seen on my numbers post. And 57% of those downloads were done through the upgrade systems of VLC; the rest came from various websites, including the VideoLAN website.

VLC team and development

Unfortunately, the VLC core team is still very small, and I might say smaller than before. The Bus Factor of VLC is still too low.

However, the number of next-to-core VLC developers and the number of VLC contributors has increased quite a bit.

With this increase, the process for reviewing patches and the quality of code entering VLC has improved a lot. This is good for VLC's maturity.

On the same topic, a lot of code has been cleaned up and outdated modules have been removed.

VLC 1.2.x

I will speak of VLC 1.2.x in a later post. But, you should know that VLC 1.2.0 is in a good shape and development is still happening at a very fast pace.

2. VideoLAN

But VLC isn't the only thing happening in VideoLAN.

The VideoLAN association is in a good shape, having fixed all the assets issue we might have had in the past and allowing developers to work together.

Websites and machines

The main VideoLAN website was redesigned and simplified a lot to stop confusing our users. The users feedback was quite good, on this part.

The other services have been cleaned, removed and improved a lot (major software upgrades, spam fighting, uptime improvements).

The main external websites are now:

Software upgrades: libdvbpsi, libdvbcsa

libdvbpsi has seen one major upgrade and a change of license: it is now LGPLv2.1.

libdvbcsa has seen one major upgrade to increase speed.

Phonon-VLC has seen several upgrades in the last year and is now perfectly working on Linux/KDE, Mac and Windows. Most of the rough edges have been fixed now and it will gain maturity this summer, once again. The amazing work from KDE folks, like apachelogger, is to be noted.

libbluray and libaacs

We have welcomed 2 new libraries into VideoLAN: libaacs and libbluray.

Those 2 libraries are focused on Blu-Ray integration for video players, and are still in early development.

Events

On the communication and community work, VideoLAN schedule has been quite full too.

We've been to various events like FOSDEM, CeBIT and LinuxTag.

We've been part of GSoC 2010 and GSoC 2011, like every year since a few years, and we've been also to participating to the first Google Code-In.

And finally, we've celebrated our 10 years of open source in February.

Conclusion

VideoLAN and VLC are now quite mature projects and the last year has re-stated this matter of fact.

VLC 1.1 was quite an important success for the users, and VLC 1.2 is on the way.

However, the fact that the core team is still mainly composed by a handful of volunteers can be worrisome for the future.

We need your help and we are quite confident for the future!

Evaluation of Google Summer of Code 2008 for VideoLAN

Jean-Baptiste Kempf — Thu, 27 Nov 2008 05:17:00 -0800

This is a post to summarize and evaluate the work that we did this year, compared to last year too.

This was our second year of the Google Summer of Code project.

Google Summer of Code 2008

For the second time, VideoLAN was an association mentor for the Google Summer of Code project.

The 2008 edition of GSoC had a few improvements and change in the way Google and we handled it. Let's detail ours:

VideoLAN, VLC && x264

This year, we decided that VideoLAN would be a mentoring association for BOTH VLC and x264, instead of doing the same confusion as usual between VLC and VideoLAN.

This meant that we had to work with the x264 team to find ideas, mentors and define common rules. This was an extra amount of time, but for a good reason!

VideoLAN rules

After the last year Google Summer of Code summit, we learned quite a few things, and I decided to be a bit more strict during the summer of code. This is Google money, what about using it correctly? Here we go for the major changes:

More publicity,
Qualification tasks. I cannot emphasize more on the goods this brought to us,
Better reviewing from developers, with common rules to follow on how to give points,
More involvements of students in the community (presentation mail on mailing list, mandatory public wiki page),
Mandatory commit and status mail every week to mentor+admin,
A more active special mailing list!

Results

Those changes were great...

Selection

We received many candidacies, with higher quality and we had the possibility to chose 14 students!

One of them stopped before even beginning, because he had found a new job. So we started with 13 students:

9 for VLC media player
4 for x264

Detailed results on VLC:

Just going to review every students and whether their code was merged or not.

Please keep in mind that coding on VLC is not as easy as it might seem. And the code is evolving quite fast, so sometimes, it is difficult to keep stuffs up-to-date.

Lukas Durfina

Lukas was my student and was working on the Fullscreen Controller and Phonon work. The Qt fullscreen controller was in VLC release 0.9.0 and the Phonon work is almost merged in 1.0.

Geoffroy Couprie

Geoffroy had the difficult task to rework on the WinCE/Windows Mobile port of VLC. Most of the work has been merged in the upcoming VLC 1.0.

Antoine Lejeune

Antoine was working on the Maemo port of VLC. The code has been merged in VLC 1.0.

Kaarlo Henrik Räihä

Kaarlo is a very very active member of our community and was working on very difficult aspects of Vista integration, namely DXVA. The project was very difficult and successful for the most part, but it isn't merged yet in VLC.

Fabio Ritrovato

Fabio was working on MTP device integration. The code is good and the project is successful and the merge is pending (hopefully for 1.0)

Mirsal Ennaime

Mirsal is a long time coder in the community and was rewritting the UPnp client and hopefully the server for DLNA. The project was also successful, but the merge is pending because of some services discovery changes in our code. It should happen for 1.0

Dylan Yudaken

Dylan was working on PVR improvements and on timeshifting. The project was also very successful and Dylan was a delight working with. Some things have been merged, some are going to.

Eric Dudiak

Eric reworked our very bad MacOS X interface This was very difficult but the outcome was quite good. The mergins is still pending too.

Vladimir Belousov

This project was a failure.

Google Summer of Code Summit

This summer of code summit was great, because it was 2 days and we had way more time to discuss with other people.

I, however, have found it not enough focused on the GSoC, but maybe this was because I didn't choose the right discussions.

Conclusion

This Google Summer of Code was a HUGE success for VideoLAN and we had some very very good output, some of which you already have on your Computer!

Thank you Leslie!

Reflexions about the first part of this SoC

Jean-Baptiste Kempf — Tue, 08 Apr 2008 23:16:00 +0200

Maybe no one cares, but still here are some random thoughts about what happened so far for us. I write them now, because I might forget those when all will be over.

Introduction

This is the second year for VideoLAN as a mentoring organisation for the GSoC.

In our first year we didn't outperform for many reasons (see here for details

VideoLAN is not a very stuctured organisation.

Candidacies

But this year, we received 85 candidacies (+8 ineligible) against something between 40 and 50 last year... (I fear that I don't have the exact number )

I know the GSoC is getting bigger every year, but still this is a big increase for us. Moreover, some of the students did some really good patches (one almost solved the main subject of one of our idea ) and the application level is better...

Why ?

So I don't really know exactly why, but here are a few guesses:

Last year, GSoC was a bit of a surprise for us. This year, we know way better what we can expect from students/mentors.
Last year, VideoLAN was mentoring only for VLC media player, this year, we have VLC media player and x264 (we have roughly 70/15 applications for VLC/x264)
Ideas were done mostly the same way, but we put more of them on our wiki this year. Ideas are more sized for a summer (see first reason)
We added ideas cross projects (x264-VLC, VLC-other codecs libraries, VLC-FreePV)
We added qualification tasks (VLC: compile VLC, provide a patch on the trunk, x264: more complex qualification tasks). This highered the quality.
We have tried to work with other projects (some FFmpeg mentors can read our applications (and vice-versa), we accepted FFmpeg and x264 qualification tasks as an "ok" for us, ...)
We announced everything earlier and were more organised between mentors (ml, IRC, meetings...)
We have a former student who is a mentor now.
I have more time than last year during that phase... :D Must be because the sun is better in California than in France...

Many ideas were recommendations that came from the Mentor Summit, so I must say that this summit did help us a *LOT*.

Hope this can help some people...

Google Summer of Code: VideoLAN

Jean-Baptiste Kempf — Mon, 17 Mar 2008 18:10:00 +0100

GSoC

Hello to everyone,

I am glad to announce you that VideoLAN is a mentoring organisation for Google Summer of code 2008.

Students, ideas

If you are an interested student, please check the ideas for VLC and x264 on our wiki.

Original ideas are more than welcome.

Timeline

Please check the SoC timeline. http://code.google.com/opensource/gsoc/

Rules

Unfortunately, we have to unforce a few rules to avoid failures and not-serious enough students: Rules.

Those rules are flexible, especially for x264.

Contact Us

If you are interested in anything around SoC, please contact xtophe (vlc), Dark_Shikari (x264) and me (vlc and other weird requests)

Everything is quite flexible, but I would like to focus on quality more than quantity.

May the code be with you.

Google Summer of Code: Mentor Summit [fr]

Jean-Baptiste Kempf — Thu, 08 Nov 2007 00:59:00 +0100

Bon, j'ai pas mal posté sur le Summer of Code en anglais, voilà un petit résumé pour les gens intéressés, mais qui sont allergiques à la langue de Shakespeare.

Google Inc. (GOOG) organise, depuis l'année 2006, un événement durant l'été, le Google Summer of Code, pendant lequel Google paye des étudiants à travers le monde pour travailler sur des projets Open Source. Le but de cet événement est double: améliorer la qualité globale des logiciels Open Source que Google utilise régulièrement et recruter des éléments de qualité.

Google ouvre son programme aux organisations Open Source qui le souhaitent, en sélectionne une centaine et les dote d'un nombre de projets possible, en fonction de l'intérêt du projet, de la taille de la communauté, et du nombre d'étudiants intéressés. Cette année, 130 projets étaient représentés, avec plus de 900 étudiants et plus de 500 mentors.

Le taux de succès de travaux d'étudiants s'est élevé à près de 80% et à produit plusieurs millions de lignes de codes.

Dans les projets sélectionnés cette année, beaucoup sont d'origine européenne, et la France était très bien représentée.

Cependant, la partie la plus intéressante de cet événement consiste en la rencontre entre les administrateurs et mentors des diverses organisations. En effet, à la fin de l'été, le 6 octobre, environs 150 mentors et administrateurs parmi les meilleures organisations ont été invités au siège social de Google, à Mountain View, en Californie, afin de discuter de l'avenir de l'événement pour l'améliorer, mais aussi, de travailler à la coopération entre les projets Open Source, dans une rencontre de type non-conférence.

La communauté française était bien représentée, notamment à travers les projets FreeNet, FFmpeg, VideoLAN, Sip-communicator, mais aussi des membres de KDE, GNOME ou Eclipse. Votre serviteur était présent, lui aussi.

Des conseils pour l'organisation du Summer of Code 2008 ont été échangés, et l'événement sera reconduit, et sera de taille plus importante. En outre, il a été suggéré d'allonger la rencontre entre les organisations, puisque les échanges, les partenariats et les projets annexes et communs ont fleuris en marge de cette rencontre. En effet, malgré des manifestations comme OSCon, le FOSDEM ou LinuxTag, une telle concentration de personnes des projets Open Source reste exceptionnelle, et le style de la rencontre est propice à ce genre d’échanges.

GSoC Summit

Jean-Baptiste Kempf — Thu, 04 Oct 2007 01:45:00 +0200

meeting

Meet the VideoLAN team in San Francisco during the Google Summer of Code summit, friday, saturday 5th or sunday 6th, october!

I'll have a few t-shirts, if you want one :D

GSoC Results: Open Letter to Leslie and Google Team

Jean-Baptiste Kempf — Tue, 11 Sep 2007 00:00:00 +0200

Introduction

VideoLAN was selected as one of the mentoring organization, on one of my proposition, and we have had seven slots and seven students.

Our experience is kind of interesting, because we have faced not-showing students, disappearing student, disappearing mentor, successful students and very successful students.

Our experience might also be interesting, because VideoLAN and VLC media player are very near to the bazaar in the http://catb.org/~esr/writings/cathedral-bazaar/ Spectrum. In fact, unlike KDE, MoFo or Debian, we have no leader, no formal meetings, no scheduled regular IRC meetings, no foundation, and not a lot of money. Most of the development is done on the free time, and there is no actual guru acknowledging all the patches.

However, the source code of VLC is not that small ( around 300,000 lines, for all the modules, and 200,000 lines for the core, libvlc, and the various bindings and a lot more of external libraries) and VLC is quite popular with millions of downloads. Moreover, the ambiance in the Team is quite nice and the IRC channel has a lot of developers laying around all days (since this is the main synchronization way for the project ). Moreover, entering the VLC codebase is not that easy compared to other projects.

Google Summer of Code

So this was the first year of our participation to the Google of Code project, and we worked a bit to welcome the students, we have had a lot more candidature than we thought at the beginning, and the choice was difficult, but we finally choose a few ones, that seemed good folks and good workers. It took me a lot of time on that, but well, we thought that this would be ok.

In my opinion, some of the biggest problems of VLC are the audio part ( not enough filtering and no gapless sound ), the absence of fullscreen controller, and subtitles support. We wanted to address that in this SoC.

Work and results.

So we have had a few surprises, some good, some a lot less good.

Richard Guo

This student had a great resume and proposed us a great technical solution to our fullscreen controller project. The code wasn't the most difficult one, so that would be an easy project. He had a very good mentor, in a technical point of view. So we thought that all would be good.

However, this student never coded anything. Because of his San-Francisco-Area based University, he thought he could just ignore us, ignore our mentor and lie during the first two weeks by mail. A disappearing student! We have mailed him, welcomed him as all the other ones, but he did not do anything...

I am really clueless about what we could have done with this one. So he did not fail, he did not start at all. Very sad :'(

Pierre d'Herbemont

Only his work and collaboration in the team made the whole VideoLAN GSoC a great success. He worked on a hard subject (a MacOS-X Framework so other applications in OS-X can now use VLC), that he succeeded very brilliantly. But he also worked at the same time on the Symbian port of VLC, he rewrote some part of libvlc and has began a rework of the playlist. His code is great, he is nice ( he came to one of our dinner ) and did not need a mentor at all... He is now full part of the project and will go on with us.

All the code has been merged with the trunk of VLC.

Philip-David Lamparter

He did work on the implementation of ASS/SSA subtitles in VLC using an external library that he had to develop and finish.

He was late because of several problems, and some timing not very synchronized with the SoC, but he did some great work and his work was very good. His library supports a lot more formats than it used to when we first met him and the integration to VLC, not perfect is quite good.

The work has not yet been merged to the trunk, but should be soon integrated.

Biodun Osunkunle

Biodun is a student in audio and sound effects. His skilled in coding were not that good at the starting of the project, but he had great ideas and a lot of will, and was very nice. Therefore, we trusted him and gave him a great mentor. Anyway, his project was not easy, because we have not good skills in audio filtering. A lot of work was to compare the different algorithms and choose them. Before implementing them.

However, the mentor had to go and disappeared... I took the mentoring of the student ! Me ! I have just finished my studies ! I could have worked as a student in the SoC ! Waouw.

So, he was a bit late, but he managed to make the first half ( a spatializer filter, with strange effects ) quite nice, and we did merge it already in the trunk.

He also made a good part of the virtual surround and upmixing ( dolby prologic 2 like ) filters, but the implementations are not ready to be merged in the trunk yet. All the matlabs files for the algorithm have been committed. However, he had some fun, and I am quite sure he will finish this part of the work, and will stay on the team to code new funny things!

Sourav Pal

This is a deception. He had to work on a very difficult part of the code (RTSP/RTCP, etc...) but he was a PhD student, and he knew a lot of things on the different RFC. We were impressed and he was granted the part.

The thing is that he interacted very little with us, and he started in a wrong way, but never sent us code, nor use git, nor SVN and did not answer correctly our mails that asked code, so we couldn't correct him.

After the mistakes been seen, we review the objectives and re-put him back on tracks, with new schedule. But he did not interact more, and finished just before the deadlines saying that he had been sick for some time, which may be right.

However, the code was not enough functional, he hadn't warned us when he fell sick, and this was a problem since the beginning so we decided to fail the student. On this case, I think that we could have asked some weekly updates, but I am unsure that this would have changed anything.

Søren Bøg

This student was chosen because he came with his own funny idea of overlay video filter. The project was cool, he was nice and skilled. He worked with dionoea, and almost all went very good, except the fact that some of the extra tasks weren't done. The code is not committed yet in the trunk for performance reasons, but this might change very soon.

Hugo de Jesús Garza Gómez

This student had to work on a Matroska muxer, which is not a very easy task, but far from being the most difficult, when we compare to Pierre, or even Biodun or Søren. He was working with one of the best guy from matroska...

He was late at the beginning, but this did not worry us too much, since the start was the difficult part, then it should be quite straightforward.

However, he almost disappeared after the midterm payment, and left the work at a 40% completion... We mailed him a lot and offered him some help, he worked a bit more on it but the school started again, and he was unseen. We failed the student.

Conclusion

We are a bit so-so in the results of this project.

At one hand, we have had some very successful students that made some great things and that did join the team and will work with us.

At the other hand, two students did not finished the code and their code won't be integrated and left in the middle of the work. One said that he was very sick, I can't check that. However this is a bit of a shame.

Some more code need to be integrated in trunk, and will be inside VLC before next release. This is very good news.

We are waiting for our T-shirts. I haven't had any confirmation about the payment, so I hope the tax form was correctly set, but I am not sure, and this worries me a bit, since I wasn't the one doing it.

We did not get invited to the mentor summit, which is quite normal according to our size. However, I may be in SF at that time, maybe I can grab an invitation, right? I'll ask the team. If I can, I'll blog that. If I can't, I'll cry alone in my room :D

KTHXBYE

VideoLAN: What I have been doing lately?

Jean-Baptiste Kempf — Mon, 23 Apr 2007 14:06:00 +0200

Hello to everyone,

This blog has been silent for a couple of weeks. I am so sorry. Let me explain you why.

Google Summer of Code

I had to finish all the stuff about the new students, a lot of discussion, reviews and choices for GSoC. All the team trusted me on the choice, so, that was difficult. But I am very happy with everything we have come up with.

Thanks Google!

Holidays and elections

Two weeks ago, it was easter's week end,
Last week, I was in Bretagne to help for one of my friends' wedding,
Yesterday, I was in Paris, for the French presidential elections.

Therefore, I wasn't home enough to work and document for VLC.

0.8.6b release

As some have noticed we had a small release on VLC branch. This is going to be the latest on this branch, I think. But I had to do win32 binaries and source tarballs... We still have issues with Vista, but well, until drivers are correctly written, we cannot do what we want...

0.9.0 - Qt4

I was chasing bugfix and not really adding new features. This will change in the next few days.

VideoLAN and Google Summer of Code 2007

Jean-Baptiste Kempf — Thu, 15 Mar 2007 11:23:00 +0100

Google summer of code

This year, Google proposes one more time the summer of code program, and I thought that VideoLAN should participate on it.

Preparing it

So, this year, I wrote the wiki ideas page, bother the other devs (sorry guys :D) to get, select the best ideas (with the team) and discuss with everyone. We found mentors and backup mentors and build a few very interesting projects.

Then, last week, a night, when I was in Paris for the Journées FedeRez , I applied and answered to the questions on the application form.

And I waited.

We are in!

This morning, I received a Google mail, and we appeared on Google Summer of Code. So let's Rock!

Ideas can be found on our wiki. If you are eligible, apply!