November 2008


I thought I had a novel idea when I told a friend this weekend that Sun & Apple could make an interesting pair.  However, it seems that this isn’t a novel idea after all.  In 2006 it was reported that Sun had tried to acquire Apple once and considered a merger with Apple on two other occasions.

How the times have changed.  Apple, now with a 25x higher market capitalization than Sun, would be in a slightly different bargaining position.

Sun is trading at ~$3B, which is slightly above their cash and near cash balance of $2.6B. The likelihood of Sun making it through this global recession as a standalone company is, frankly, not a sure bet.  So who would make for a good suitor?  Apple, HP, IBM, Oracle, Microsoft and Red Hat (merger?) come to mind.

Of these, I think Apple is most interesting (maybe not the best fit, but most interesting).

Both vendors have deep experience in hardware at different ends of the product spectrum.  As such, there would be minimal real overlap across product families.  Sun’s Systems business (servers & storage) would be a great way for Apple to get deeper into the enterprise.

While both vendors offer premium priced products, it’s arguable how much longer Sun brand can uphold the premium.  However, take one of those Sun boxes, slap an Apple logo on it, along with Apple’s styling, and that premium looks to be in tact.

Sun has a strong developer following, while (the cool crowd of) developers love using Apple products during development.  To me, Sun’s developer following is Sun’s single most valuable asset.  Sun gets developer love like no other company I’ve seen (well, except Apple, but the love is more general than simply developers).  Any company considering acquiring Sun needs to have street credibility with these developers or risk losing the most valuable part of a Sun acquisition.

Oh and there may be the minor issue of Apple’s proprietary software strategy vs. Sun’s open source software strategy. ;-)

What do you think?

Paul Krill has a balanced view of the Microsoft-Novell Linux cross-licensing deal from two years ago.  It’s worth a read.  So too are the comments:

Posted by Linux Fan:

“Say what you will about the agreement’s effect on the community, and there are legit issues there, but get the facts straight. Novell’s market share is up significantly since the deal. Schestowitz is just plain wrong on that point. Check IDC’s numbers. Their share of bookings and of server counts is increasing every quarter and it is coming primarily at RedHat’s expense.

Large organizations in particular seem to like the deal. They don’t care about religious FOSS arguments. They want software to work together with as few headaches as possible. It is that simple.”

Posted by Alan:

“I agree with Corrin. It’s unfortunate that all criticism of Novell-Microsoft seems to come back to Schestowitz and BN. There are plenty of reasonable arguments for criticizing MS-Nv, Roy just happens to be making them the loudest and most obnoxiously (and stretching them for all they’re worth).

So you get people like some here who say that “only paranoid crackpots think this is a bad deal, everyone knows it’s good for Linux!”

The way I see it, it’s pretty obvious why the community was upset about the patent part of the deal, especially when you have Novell salesmen telling customers that they’re the only ones who can legally sell Linux (I’ve heard that firsthand from our Novell rep).

Think about it: You’re a FOSS developer. You write software and release it GPL so everyone can use it.

Then Microsoft claims they own it because of patents. Then Novell comes along and licenses your software from Microsoft, in a sense legitimizing their claims of ownership. Now the software that you wrote and released for free is making Microsoft money. Wouldn’t you be pissed?”

Head on over to Paul’s story and have your say.

I’ve been a big fan of RedMonk’s unconference at CommunityOne over the past two years.  The conference is largely driven by the attendee participation.  So here’s your chance to submit a session, panel topic or lightning talk that falls into one of these categories:

  1. Next-Generation Platforms – Development and deployment in the cloud; virtualization.
  2. Social and Collaborative Platforms – Social networks and Web 2.0 trends.
  3. RIAs and Scripting – Rich Internet Applications, scripting and tools.
  4. Web Platforms – Dynamic languages, databases, and web and application servers.
  5. Mobile Development – Mobile platforms, devices, tools and application development.
  6. Native Development – HPC, chip multithreading, code base development, customization and contribution.
  7. Open – Any and all subjects are fair game.

Get your entry in before December 11, 2008.  Oh, and for those of us facing tighter travel budgets in 2009, CommunityOne will also be taking place in NYC (March 18th).  This is of course in addition to the usual event held in SF (June 1, 2009).

Roy Schestowitz’s Boycott Novell blog reported that organizers of the National Conference on Free Software 2008 at India’s Cochin University attempted to suppress the peaceful anti-Novell protest.  Apparently the state police even confiscated mobile phones (for a short period of time) so that pictures of the incident could be deleted.

It does appear that conference organizers, and not Novell, called the police. Some folks commenting have questioned why Novell would be allowed to sponsor a Free Software conference.  Others place the blame on Free Software activists seeking attention through a publicity stunt.  The truth likely lies somewhere in between.

However, what I find more interesting are the calls to boycott Novell, SuSe and Microsoft products in India.  I’m not sure if this is truly possible yet.  Considering that India’s IT economy is driven by the outsourcing needs of Western clients, many of whom still rely on Microsoft technologies, can India truly turn its back on Microsoft?  This is clearly a point in time statement.  When India’s IT economy is driven more by the needs of domestic clients, rather than the needs of clients outsourcing to India, the importance of Free Software will surely rise.

My hypothesis is that domestic Indian clients will be more accepting of, or even insist on, Free Software than Western clients.  Part of this will be driven by the skills present in the (Indian) market.

The first wave of Indian IT professionals were taught on predominately non Free Software (i.e. Microsoft) technologies.  These professionals didn’t have access to computing resources as readily as we are used to.  The second wave of professionals are in a completely different boat. They have easier access to a broad range of computing resources and information about Free Software.

What do you think?

Some of you will remember that I wrote about an interesting startup, Tasktop, earlier this year.

Tasktop is a commercial product from the creators of the Eclipse Mylyn tool. Some of you will know/remember that Mylyn is a task-focused UI for developers using Eclipse. Tasktop applies this idea to general purpose knowledge work (i.e. not just for developers).  As I wrote previously:

“So, for instance if I was working on a customer issue for work, researching for a blog post and writing a paper for school, I could switch between these three tasks and Tasktop would open/close the appropriate files, emails, webpages etc. based on the task I’m working on. Searching for files becomes simpler because they are associated with tasks. Also, Tasktop tracks how much time I’ve spent on a given task.”

When Wesley Coelho from Tasktop emailed to give me a heads up on the Tasktop Pro Autumn 2008 release, I realized that Tasktop is a great example of an “open core” business model.  The core of Tasktop is Eclipse Mylyn, an open source product.  However, Tasktop Starter and Tasktop Pro add additional features and capabilities through commercial, closed source products.  Very smart choice if you ask me ;-)

Tasktop offers a 30 day free trial, so give it a shot and leave a comment with your thoughts.  A new version goes live on Wednesday.  Amongst other new features, you can now run TaskTop on Linux and with Firefox.

SpringSource announced that it’s acquiring G2One, the company behind Groovy and Grails.  Groovy is an open source dynamic scripting language that runs inside a Java Virtual Machine.  Groovy uses Java-like syntax, so it’s often positioned as the scripting language of choice for developers who already know Java (and/or don’t want to learn a scripting language like PHP).  Some have even gone as far as suggesting that Groovy will become the dominant language for the JVM (ahead of the Java language).  Grails is a Groovy-based framework.

SpringSource is joining the list of Java application server vendors adding support for Groovy.  IBM has been supporting Groovy with WebSphere sMash, and Project Zero since July 2007.  Sun’s GlassFish recently added support for Groovy in GlassFish v3 Prelude.  JBoss has integrated Groovy with JBoss Seam.  I couldn’t find any product specific info from Oracle or BEA about Groovy.  Oracle/BEA offers developer documentation that describes how to use Groovy, but nothing about Groovy support within an Oracle/BEA product.  (Oracle/BEA readers, please correct me if I’m wrong).

What does this acquisition mean for customers (considering or using Groovy)?  According to the SpringSource acquisition FAQ:

“SpringSource has built a global support and governance operation for the Spring Framework. This infrastructure, coupled with G2One’s experts and some investment, can deliver a 24×7, worldwide support network for enterprises investing heavily in Groovy and Grails. Additionally, there will be some immediate product enhancements and we will be investigating the creation of enterprise grade Eclipse tools for Groovy and Grails development.”

Selling support for Groovy (a language) or Grails (a framework) doesn’t sound like a great business to me.  Look at Zend, the company behind the PHP language.  They’re making money from ancillary products around the PHP language, not simply from supporting the PHP language.  Next, there isn’t a whole lot of revenue to be had from support of a framework.  SpringSource knows this better than most vendors.  The Spring Framework is widely used, but the majority of customers are in production without any support from SpringSource.  This is why SpringSource has introduced Spring products, because as I’ve been saying, products are valuable, support, not so much.  But maybe SpringSource will introduce enterprise grade products with Groovy & Grails?

It’ll be interesting to see how the acquisition plays out. Best of luck to G2One and the SpringSource team.

I must confess that I haven’t paid attention to the Alfresco open source barometer study in the past.  However, the title of the press release “Alfresco’s Open Source Barometer Reveals that Enterprises Are Using a Mixed Microsoft, Java and AJAX Environment” caught my attention.

Reading the results, and more importantly, understanding the methodology, I’m disappointed that this study is billed as an “open source barometer”.  In fact, it is a study of respondents interested in an open source ECM solution who have decided to evaluate/use Alfresco.  To claim that the results from this very narrow set of respondents is representative of the broader open source user community is a stretch at best.

For instance, just because a respondent is going to deploy Alfresco on Tomcat on RHEL does not mean that Tomcat or RHEL are the key production environments at the company.  Different departments, divisions and sites can, and do, make different technology decisions.  In fact, even within the same IT team, at the same location, different infrastructure is often used depending on the application being deployed.

In another example on chart 20 of 30, the report claims that 33% evaluate on Windows, while only 21% deploy on Windows.  The figures are 53% & 64% for Linux.  While this may be true for Alfresco evaluations/deployments, this is not the case for all open source.  For instance, Windows represents much more than 21% of JBoss production deployments (as Microsoft & JBoss would tell you).  And heck, Windows represents much less than 21% (i.e. 0%) of production RHEL deployments!

Or the sub-heading on the press release: “Over 50% of Windows respondents use Java Architecture for Content Management.”  This statement, while based on results from the study, isn’t completely accurate.  It’s equivalent to Microsoft surveying SharePoint users and claiming: “Over 99% of Windows respondents use .NET Architecture for Content Management”.

Morale of the story, understand the survey methodology before looking at the results.

This report is much more useful for Alfresco users and vendors considering partnering with Alfresco than it is for the broader open source or software community.

Next Page »