jump to navigation

EE6 Public Draft APPROVED February 26, 2009

Posted by Sacha in JBoss.

Good news, the EE6 Public Review Draft has just been approved by the JCP Executive Committee.

Looking at the results gives some interesting information:

  • The Apache Foundation voted NO, but didn’t do so based on the merits of the specs, but because of their long-standing issue with SUN wrt the SE 6 license. While I support Apache’s decision to vote NO to any SUN-led JSR until SUN gets their act together, we didn’t take such a strong stance and will only vote NO to any future SE JSR proposal (unless the ASF gets a satisfactory proposal from Sun obviously).
  • SAP voted YES to the specification but commented that “[they] would like the Spec Lead to consider putting more emphasis on architectural rigor regarding a single consolidated and extensible component model to be used across the platform – right now there are three (EJB, JSF and JSR 299).” While I don’t think there are any problems with the current approach, I can see why SAP might want to dispatch things differently. The good thing is that SAP is part of all of those specs (EE, EJB, JSF, 299, etc.) so we are waiting for their specific advices.
  • As always, I like to keep the fun for the end. SpringSource, the Switzerland of middleware, courageously voted NO… hum, no, they voted YES, hum, no… actually they voted ABSTAIN! SpringSource always had a hard time positioning itself wrt J2EE/EE. Consequently, this is no surprise that they opted for a non-risky position where i) they don’t vote NO to EE (not good for their karma), but ii) don’t back it either. It gives them the freedom to criticize the spec when they see fit. Opportunism at its best. Last but not least, they add this comment: “We would have preferred to see a dependency injection model for SE, as we proposed in 2007.” ?!? SpringSource never contacted us for a JSR DI specification, so I am not sure what “spec” they are referring to. Maybe some back-doors discussions with other vendors. In any case, if these were back-doors discussions, they should remain so and I find it strange to use it as a public comment to justify an ABSTAIN vote. In the land of privacy, you don’t become Switzerland overnight – and I know what I am speaking about…

Also, I’d like to officially thank our colleagues at Oracle, Google and IBM for their deep involvement between December 2008 and February 2009 to make sure JSR-299 fitted to their requirements. In just a few weeks, they’ve worked with no agenda other than solving problems. Thank you.




DVD collection: be my Google! February 16, 2009

Posted by Sacha in /dev/null, IT.

I am spending some time these days (actually week-ends…) trying to understand what’s the best way to move from a horrible collection of DVD to a sequence of 1 and 0, stored on a hard drive, which I can then play from various location (read: what’s the best way to watch moviesย  in the 21st century).

(Just for the record, this is all about PAID DVDs, paid content (I’ve those ugly DVD boxes filling my place) – so all of this IS legal.)

I am trying to find a solution which:

  • keeps content in an “open” format (i.e. not a proprietary or encrypted format)
  • keeps maximum quality (i.e. I want the initial copy to be a “master” which I can later format-convert if needed)
  • reduce the IQ required to process new DVDs
  • provides a nice GUI once you actually want to consume those bits

Easier said than done. I’ve actually done my homework quite well, but I still have some doubts or things I haven’t been able to figure out.

So I figure out I would share those with you in case you have experiences you can share as well. Once my solution will be ready, I’ll obviously report it here in great details – I’ve already had several requests from people wondering about the same thing.


  • There are plenty of well organized CD databases that exist so that you can get the list of the songs, picture of the CD, etc. when you rip those. Such as what iTunes is doing. Question: does the same thing exist for DVDs? BTW, I am asking for a TRUE solution based on a UID, not some ugly tool using web-scrapping algorithm to match the DVD name to some name-related movieon Amazon.com.
  • Are you aware of any decent ripping software (Windows or Linux) which will automatically open-the-dvd-tray/get the real movie name (see above)/rip the DVD in a folder named according to the movie name/eject the DVD; and this for multiple drives in the same system? If not, I’ll have to code it, fine, but I hate NYH syndromes.
  • Have you tried the TVIX HD M-6500A device? If yes, what are your thoughts on it?
  • What ripping format is best for a “master”: ISO or direct copy of the DVD structure in a folder (content of TS_VIDEO)? I know there are no quality differences, I am just wondering which one might be preferred for other reasons (size, easier to consumme by devices, etc.)
  • Thoughts on the NetGear ReadyNAS NV+? or on the ReadyNAS Pro?


Oh, and remember one thing, I live in Europe (Switzerland), which means that unlike in the USA, there are no decent ways to buy your movie content directly in electronic format from a 3rd party. While some ISP or telco do provide some kind of offering – it never reaches anything close to a decent specialized shop.

It seems the movie industry still doesn’t really understand what “ease-of-consumption” means. They missed the market-flip from Audio-CD to online-mp3-shops and instead prefered to complain during years that their content was being hacked on peer-to-peer network (also known as “listening to the radio” IMHO) instead of keeping the leadership by doing it themselves. Will the same drama happen for Video content? What is sure is that they are clearly not answering the needs of a growing portion of their market. How come I have to spend so much engineering time to design such a simple solution? Instead, what I am trying to achieve should be the de facto way of obtaining content.


Thanks for your help, comments are open!



Mobicents 1.2 GA hits the tarmac February 16, 2009

Posted by Sacha in JBoss.
add a comment

Since the last Beta, the Mobicents team has done an impressive job nailing the GA down.ย  From the Release Notes:

This GA release includes the ‘all’ JBoss Application Server configuration, which hosts a cluster-enabled Mobicents Sip Servlets container. The Mobicents SIP load balancer is also included. Note that there are some limitations when clustering is enabled and clustering is only for the Sip Servlets container. For more information see http://www.mobicents.org/clustering.html. To demonstrate clustering and mid-call failover we have included a predeployed UAS example in the ‘all’ configuration called ‘simple-

So go and get it now!



Is it still possible for FOSS companies to raise VC capital? February 16, 2009

Posted by Sacha in Finance, JBoss.

Good question, especially in those difficult times.

And it seems the answer is a victorious YES. Here is a short list of some of the 2009 rounds:



Russia moving to Fedora February 11, 2009

Posted by Sacha in IT.
add a comment

A recent news didn’t get as much visibility as it should have gotten IMO, maybe because the original PR was in Russian.

The Russian Federation (aka Russia), which was developping its own Linux variant, decided to redirect its efforts towards Fedora – and hence RHEL for its production system.

That’s a significant move.



Autogoal (as in “don’t speak too fast Jonathan”) February 9, 2009

Posted by Sacha in IT, JBoss.

Just one year ago, on the 26th of February 2008, Jonathan Schwartz said:

Q:Turning to MySQL, how long do you think it will take before MySQL is fully integrated into Sun? We saw with Red Hat and its JBoss acquisition that this can be a slow, laborious process….

A: First, the personalities involved. Let’s just say that integrating Marten Mickos into a company might be easier than assimilating a few of the JBoss personalities. Marten is a joy to work with and will make this integration work.

No comment.

Onward Marten,