Thursday, May 8, 2008

When will Sun acquire Liferay ?

I have been told last night about that announcement: In a few words, Sun and Liferay are collaborating on a common set of components that will be reused by both platforms.

To me it's a perfect admission of two facts: Liferay failed at implement new technologies such as JSR286 and Sun failed at creating an ubiquitous and visible version of their portal.

Liferay benefits from Sun technologies that they were not able to implement due to the huge legacy of their code base. Brian admitted it publicly in a post ("we are not smart enough, we don't have enough man power, and we don't have enough energy to build an innovative product").

Sun on its side distributes a rebranded version of Liferay and kills its existing portal. Why ? Sun open portal is too tied to the operational environment and their identity management product (read here big fat momma). Sun realized that they wouldn't be able to increasing their product visibility while making it more adaptable to more diverse environments.

The real question on everyone's mind though is: why is Sun moving in this direction now? in my opinion most of the benefit goes to Liferay, so the hypothesis that Sun could acquire Liferay seems very valid to me even if the announcement claims it won't happen ("No, there are no plans for an acquisition.") and Sun does not yet have vested interests in Liferay. We can see this as a last attempt from Sun to revitalize their portal offering by trying to quickly leverage Liferay's community. Will this work for either side of this partnership? Only time will tell...

At least I am not wondering anymore about why our proposals submitted to JavaOne have been refused :-)


Luc Boudreau said...

Julien, the whole quote was : "The truth is, we're not smart enough, we don't have enough man power, and we don't have enough energy to build an innovative product **without you guys**", talking about the community.

This suggests that they ARE smart enough to do it, since the community backs them up. They don't admit they are not smart enough alltogether.

I just think it's important not to take such a quote out of context. You guys have a strong support coming from your community too.

It is said, An emperor without a kingdom is a lonely fool.

And you guys are no fools.

That said, I'll keep a close eye on the future of Liferay. The next years will bring forward a big boom in the Enterprise Portal world. Business Intelligence is blooming and EPs are the favorite frontend for this. (I'm using Pentaho along with JBP)

Thanks for pointing this new though.

Bryan said...

Luc beat me to it.

When I saw the quote you provided ("we are not smart enough, we don't have enough man power, and we don't have enough energy to build an innovative product") I thought that was a fairly stark and odd admission.

So naturally, I followed the link you provided to see if that was actually said... and come to find out, no it was not.

If you're going to purposefully take quotes like that out of context and attempt to use them in a dishonest and malicious way then you shouldn't be foolish enough to provide a link to the resource that exposes what you're doing.

Hani said...

Actually, your talks got rejected because they were vendor pitches, and offered no new content over previous submissions on the same topic. Many/all vendor portal pitches got rejected because believe it or not, people get very annoyed when they go to a talk and it ends up being 'buy my stuff'.

I know it's easier and more comfortable to try and blame external forces and to assume that you're great and the world is against you. In this case it isnt so, your submissions were of inferior quality and of limited interest.

Contrast this for example with the large number of JBoss (and JBoss related) talks which did get accepted (mostly from the hibernate crew, who seem much smarter than you lot), where they knew the target audience and ensured the content is compelling.


Thomas Heute said...

Julien was missing an obvious ";-)" also known as Smiley.
Out of context it's just funny, the link to the blog post make it back in the context, that's the role of a link no ?
My bad, i originally sent that part to Julien with a smiley (purposely out of context).

For Hani, thanks for the kind words, as always. There was the smiley after "refused".
Now your explanations are worse than i thought... Product pitch yeah... My abstract didn't even mention JBoss (Portal). The title was: "Java EE Portals demystification" and was explaining all the integration technology and its challenges.
Now that said, i know you are receiving tons of abstracts and have to select few, with such an announcement it's also obvious that there is a natural filter for competition. I don't mean to say that my abstract had to be accepted, but let us think that it couldn't go through anyway.

Back to the *real* story, what is Sun interest in the merge ?

Julien Viet said...


I did not personally submitted anything to J1 this year because I did not have any interesting idea to submit (you can check).

I doubt that out other submissions were product oriented as we avoid it (because we know if will be rejected), but I will check.

Hani said...

Just to clarify, I don't specifically remember reviewing any JBoss Portal specific talks. My point was a general one about talks being rejected, and people always assuming its because Sun prefers certain vendors over others.

For the last 2 years there have been external reviewers involved in the talk selection process (I'm one of those external people). Portlet/Portal talks are usually tricky because they fall in a weird hole between EE and Web tracks.

There's a very simple criteria for selecting portlet talks. It's basically:

1) No vendor/product pitches
2) No old content

For example, there was the usual 5-6 talks about JSF/portlet integration. Boooring! Similarly, there were a couple of 'intro to portals/why you need a portal/how to write a portlet' type talks. Again, this is content from 3 years ago. If there's this sort of content at J1, people will get back to bitching about how J1 never talks about any new or cutting edge stuff.

The talks that had a good chance were JSR-286 oriented talks. Ones that discussed new features, use cases, examples, etc. Either that or anything that felt 'new' or at least somewhat interesting in the portlet space.

Yes I did see the smiley, but just like the link you showed, it's like saying 'no offense but...-insert offensive comment-'. You cant disavow responsibility and disown your words that way. Believe me, of all people, I know that!

Julien Viet said...

fair enough, Hani

Personally I think it is hard to submit interesting talks about Portlet because of the restricted number of related specification (JSR168 / JSR286 / JSR303). Personally I would like to see more specification about portal / portlets.

On the other hand when I look at the number of J1 slots about JavaFX, I think that the selection process for J1 is very biased.

Raju said...

Julien... it appears you are a bit shaken by the announcement. Your post very much looks like a knee jerk reaction than an unbiased opinion.

Under estimating your opponent might not be a good thing for your community. Well only time will tell who makes the most out of the newly announced collaboration between Liferay & Sun.

Paul Hinz said...

When will Sun acquire Liferay? The better question is, what are the motivators in the industry which will benefit both companies. They are:

1.) The ability to engage in this first of it's kind open source business partnership.

2.) The need for the next generation in web presentation technologies.

These two aspects will change our industry very rapidly. It is better for companies who firmly believe in the advantages of open source to work together for the common good - and to each build upon those common components their own value for customers.

Sun recognizes the substantial work and capabilities of the Liferay community. Their expertise and true commitment to open source is one we value and their community is one we are excited to be humbly working alongside.