Been reading some of the comments to the recent GNOME Foundation press release about OOXML involvement. Reading things like the comments on osnews and linux.com leaves me in a state of disbelief.

When you have people like Michael Meeks and Jody Goldberg, who have been working on free software office software for a longer time than most others in the community has even been using free software, it is sad to see people trying to claim they know more about the respective qualities of the ODF and OOXML standards than they do. The claims of being corporate shills sounds even more hollow considering these people have been working on free software office software for much longer than they have been paid to do so. Having known especially Michael for many years and had many a beer with him I it saddens me when people in the community doubt his word. Michael care about free software, but whats maybe more important in this context is that he cares about technical excellence.
And thus he is not willing to claim that the ODF spreadsheet specification is better than the OOXML one when he, based on his eight years worth of hacking experience doing spreadheet software in the context of Gnumeric and OpenOffice, knows it not to be the case.

The question worth asking here is if the community has come to a point where we no longer care about quality and technical perfection in what we do, instead we are going to join the proprietary software vendors in a game of FUD and mud slinging. When I started out with free software it felt as the ethos of the community was that we are not a community of people blinded by pride or arrogance instead we happily adopt and evolve the best of what is out there. If that means adapting to standards and formats originally developed by Microsoft then so be it.

Michael has been tirelessly been promoting OpenOffice for quite a few years now, even working on it and promoting it at a time when most GNOME developers felt it was a bloated and unmaintainable piece of software and most the code comments where in German. Through his and his team’s effort OpenOffice has gone from being something screaming Windows 3.11 when looking at it to something that actually looks like it belongs on my GNOME desktop. And there is no longer a half hour wait for it to start up.

I use OpenOffice.org every day doing my job today. Back in the day when I was at Oracle I couldn’t do that. Not because Oracle refused me to do so, but simply because most of Oracle’s documents where in MS Office format and at that time the OpenOffice filters where not good enough to do the job, so I had to dual boot into Windows in order to write reports for customers, fill in expense reports and so on.

Even in my current job, doing linux consultancy work around free software technologies like GStreamer, Telepathy and Webkit I would not be able to use OpenOffice if it wasn’t for the excellent MS Office compatibility. The vast majority of documents and proposals I receive from our customers are in those formats. I haven’t gotten any OOXML documents yet, but it is only a matter of time and when it start to happen I would gradually be forced back to my old dual booting ways unless OpenOffice supports OOXML in a good way.

So Jody if you are reading this, please don’t listen to the detractors and instead keep up doing such great work. I for one would be sad if I have to install windows on my laptop again because you stopped pushing OOXML in the right direction.

Had a great time at GUADEC this year, better than expected even as I was feeling a bit worn out and tired before it got rolling. Meeting the other members of the community again and hearing about all the plans and efforts underway is incredible energizing and gives renewed energy for ones own part of the greater effort.

Seems tighter integration with internet services and technologies was the topic of the conference.Havoc’s Online desktop of course catching the most attention, but there where other efforts presented too, like Spyro and of course the work the Songbird guys are doing. I also think Conduit will have an important role in our new direction. Of course the question now is what ideas we from the GStreamer community can bring to the table to move this even further forward.

Currently visiting Tim and Heather at their place in Bristol, but flying down to Barcelona this evening. Think the coming weeks will be hectic as in addition to getting the new company rolling I will be packing for my end of August move to the UK. On top of that I have Zeeshan and his girlfriend Ansku coming down for a visit mid-August.

Managed to catch a slight cold here in sunny brittain, hopefully it will clear up once I get back to the sauna which is Barcelona.

After a lot of back and forth I finally got Linux installed on my PS3. Ended up installing Yellowdog Linux. Not sure how much I will be using it though due to the lack of OpenGL support and access to the video card. Would have been nice to run Elisa on it though, but I guess once we get around to doing a DirectFB backend for pigment.

Linux on the PS3 feels a little ‘raw’ due to the graphics issue, and while the Sony people I talked to about it put the blame on NVidia for refusing them to make the GPU available in the hypervisor I think the value for Sony in having people being able to learn how to program for the PS3 under Linux should be big enough to do the needed legwork to get the GPU available with OpenGL support. I could see people even using this box as their primary desktop machine in such a case.

Suprised me also to see that YDL ships GStreamer 0.10.4 on the box, which is rather old at this point. A brand new distro like YDL 5.0 should have been able to update this. My next target is to try and compile the Fluendo plugins on the box to see how they perform. The combination of 64-bit and BigEndian might of course reveal some more bugs even.

DVB and Elisa

Got myself two WinTV NOVA-T-500 cards today. I gave one of them to Philippe so he could come through on his promise to get DVB support into Elisa if I got him such a card. The card is a nice dual tuner one, so its perfect for also making sure we can support watching one program and recording another.

On the topic of donating hardware to people. I have already promised to pitch in on the Delta 44 card that Jono is buying for Tim which I think should also help us resolve the issues that Davyd Madeley reported in his latest blog entry regarding his Alesis io|2 card with GStreamer and Jokosher.

New queue element for GStreamer

One of the things which reduce the user experience of GStreamer today is that the current element used for buffering is not very nice. Meaning that if you have bandwith problems with streaming audio or video it will be very noticeable and give you a less than nice experience. Wim just made sure we took a big step forward in that
regard today with the new ‘queue2’ element commited to CVS. While it doesn’t take us all the way there it do improve upon the current situation a lot. This work combined with Wim’s recent code magic to improve RTSP support should take us a big step forward towards being a very good library for doing network streaming clients.

That the RTP work has been highly anticipated seems obvious based on the huge number of questions about RTP support we get onto gstreamer-devel mailing list. The same goes for the windows support which seems to have a lot of people testing it out. If only we could get more people onto testing our MacOSX support also to make sure it gets rock solid. Hopefully when we get a new release of gst-plugins-good out with both the MacOSX and Windows support included it will get us more users also on the Mac platform.

Jokosher

Getting a new piece of software stable and widely usable is hard and one quickly realize that working on the developers machines is not the same as working for everyone else. Which is was nice to see this blog entry by Niels Kjøller Hansen where he talks about latest version of Jokosher working for him. Still a lot of work remaining for Michael, Laszla, Jono and the rest of the crew before Jokosher has reached its full potential. I guess there are two major items on the todo going forward as I see it which need being taken care of, one being support for more advanced multichannel soundcards for recording, the second being making sure all the needed GStreamer plugins are in gst-plugins-good so distributions can package them easily. Today a few of the needed Jokosher plugins are in CVS of plugins-bad which of course makes it harder for people to get everything they want/need.

Tribal Trouble

I have blogged before about the game Tribal Trouble which is available for Linux. I was happy to see that they decided to
release their sales statistics recently. The most interesting point is probably that Linux sales accounted for as much as 11% of their total sales. I think that is a high enough number to make doing a linux port interestinng, especially for smaller game publishers.
The nice guys at Oddlabs also released the terrain generation engine they made for Tribal Trouble under the GPL.

I also came accross this recent Ryan “icculus” Gordon interview on linuxgames.com. If Linux ever becomes a mainstream games platform I think Ryan personally probably deserves something like 75% of the credit.

Just discovered the photo tagging feature of facebook. It lets you click on various people/objects in your photo and facebook then tries to give you a square box centered around the object to clicked on, in other words they have some basic shape recognition stuff built in. What you usually do is mark the people who are in the picture, and if they are your friends on facebook they will be told that a new photo with themselves are online. Also the image will be added to the link with photos of the person in question on their facebook profile page. You can also tag other people’s photos, but they will have to approve your tagging before it becomes public.

While simple I found it pretty nifty actually. We will hopefully be enganging in a huge EU funded project at Fluendo starting in January. If that project goes through it would be nice to see if we can add a feature like that for video’s as part of that project as collaborative editing and metadata handling is a big part of the project. Of course the technical challenges are much bigger in the context of video so it might not make the cut as it wasn’t concretely planned for in the current proposal. But no matter what the project will be a huge boost for the Pitivi project and the features we offer in Pitivi.

Read David Trowbridge‘s blog entry about the recent blogging campaign posting one of the AACS keys with interest. As someone who has to deal with software patents and DRM systems on an almost daily basis as part of my job I would have to say I disagree on his conclusion that such efforts have no value and should just be considered worthless online ‘whining’.

There was a similar thing happening with CSS back in the day, where people posted the DeCSS source code everywhere and put it on t-shirts etc. While it might not have caused the whole DRM eco-system to fail it did play a part in the content industry starting to question the value of DRM systems. If we get a similar story now with AACS it could be the decisive blow that ‘educates’ the content industry on the futility of pursuing DRM systems.

The fact that we have had DeCSS out there now for many years has also been a useful tool for countering many of the arguments made by for instance the DVD CCA, regarding why DRM systems are needed. In the DVD CCA FAQ there is a question about why CSS is necesarry. They are saying that without CSS the content industry would be hesistant to release their content. Well DeCSS has been out for a long while now and there is a long list of DVD rippers out there using it, available for all major operating systems, yet that fact doesn’t seem to have put a dampener on the willingness of the movie studios to release more DVD’s…..

So David might feel that the ‘Civil disobedience’ in regards to AACS is worthless, yet I say that discrediting the major DRM systems out there actually do serve a purpose in terms of educating people about the futility of such systems, and also demonstrates that the assumption that without effective DRM systems sales will collapse is a false one.

And being someone who has to deal with people doing hardware, software and content in this area I can tell you that the ‘educational effect’ of more and more DRM systems getting broken is slowly changing the mindset and level of understanding of the industry.

So while I as part of my work have to deal with the laws and regulations as they are today and have to work within the constraints they set that doesn’t mean I don’t respect the efforts of those being able to operate in less restrained ways than we can as a company. The world is a complex place and a working democracy is much more than just voting and mailing your local politician.

Well then you are in luck. Edward just pointed me to the budget airline bmibaby (pronounced ‘be my baby’) which offer budget fligths directly to Birmingham which will save you quite a bit of both time and money compared to flying into a London airport (cost me a total of 70€ in train tickets the last time going back and forth to Wolverhampton/Birmingham). If you are flying to London I do recommend Stanstead over Luton though, unless you are comfy with the UK public transport system as the train from Stanstead goes directly to Birmingham, while you need to switch twice when going from Luton by train.