This morning I was in the Sam Ramji keynote at EclipseCon, crammed in the front row in an equally crammed ball room. As the keynote was starting, I noticed that Sam’s had already put up a post describing the announcements, which was great. I always love seeing blogs used in addition to (or instead of in the best case) press releases.
Helping out with Identity Management in Higgins and SWT on Windows
In summary, the announcements are that Microsoft will continue to help the Eclipse Higgins project out more when it comes to CardSpace interoperability and also help out with WPF performance enhancements for SWT (this means making it easier and/or better to use Windows native look and feel in Eclipse-built applications).
What’s clear from the question I asked Sam during the keynote is that there’s not going to be any Microsoft employees committing code to Eclipse. Instead, Microsoft will simply help others do this. This matches the usual model of interacting with open source – through third parties instead of directly – that I’ve come to expect from Microsoft, and it’s just fine, really. This is the kind of thing we should probably expect ongoing from all parts of Microsoft when working with open source.
The motivation here from Microsoft is the one we’ve heard over the past two years with partnerships with JBoss, SugarCRM, and other open source stacks that find themselves running on Windows. The point is, as long as you’re paying for the Microsoft platform – Windows, ActiveDirectory, or even the System Center stuff on the optimistic end – Microsoft is OK with you running open source on top of it. The promise of announcements like this and the ones in the past is that Microsoft will even spend time and money to make sure they work well.
From the Eclipse angle, it seems all gravy: getting help to optimize Eclipse UI’s on Windows will make the Eclipse GUI and GUIs built off it better and getting Microsoft “on board” with the bus of identity management ambitions of Higgins is great. The Eclipse folks I talked with seemed happy.
Lingering Mistrust Warranted?
As I tell all people, the biggest challenge for Microsoft with things like this is the long trek to getting to the benefit of the doubt from the open source crowd. Your analyst here tends to fall into this camp having fallen victim to quick of skimming of the recent interop memo from Microsoft and missing the key “not for commercial use” clause, having to rely on Dalibor to point it out (thanks!).
Indeed, as I IM’ed Paul Krill during the keynote:
An analyst was skeptical of Microsoft’s efforts. “We need some more fine print to make sure there’s not any weird clauses like that ‘not for commercial use’ stuff in the Microsoft interop memo,” from a few weeks ago, said analyst Michael Cote of RedMonk.
That is, if there’s some fine-print to be found, it deserves some scrutiny. But, to be fair here, there doesn’t seem to be much room or reason for fine-print wizardry here. Still, as mentioned I’ve missed the boat there in the past: the RedMonk community is usually good at hunting down the bad apples in the fine print barrel.
Assuming the work is as straight forward as it looks on the tin, it seems all good to me. While there’s not code committing here, the point is that Microsoft would be committing to help out Eclipse by way of answering questions and running tests. They’ll answer the phone when Eclipse calls up, not only answering the questions, but doing even more with lab resources like testing and assuring that interop works. If things are that simple, then sure, good job all around.
Disclaimer: Microsoft is a client, as is the Eclipse Foundation.