Project Community Enhancements for 2010
I mentioned in a comment last week that — pending Board approval of the 2010 budget — there are a number of items which we’re looking at for 2010 that we at the Eclipse Foundations are pretty excited about. These are some major pieces of work that we are going to be focusing our time and energy on.
There will also be a number of significant programs for the commercial members, but those will be the topic of a separate post.
- Build and Test: The team here is aggressively begging companies for the hardware contributions required to take on more of the build and test requirements for Eclipse projects. If we are successful, we will happily provide additional infrastructure and support for build and test activities for projects. But don’t forget that the hardware is actually only a relatively small part of the total solution here. The resource management that the webmaster team will take on will be a significant amount of additional work for Denis and team.
- Git: Denis and the webmaster team have been working hard on getting read-only git mirrors up and running on eclipse.org. But we really look at this as just the starting point. We view git as the SCM solution of the future at Eclipse and hope to have it up and running for as the main SCM repository for as many projects as possible by the end of 2010. Our reason for doing so is pretty simple. From what we can tell, the use of git at Eclipse is going to make it easier for contributors to make and track changes to the Eclipse codebase. And anything which gets us significantly more contributors is a very good thing.
That said, there are a couple of items that are out of the Foundation’s control that the community will have to help with.
- The CVS Team Provider that has been in the platform for years is really good. In fact, it is so good and so stable that it makes CVS very usable from within Eclipse. For git to gain traction at Eclipse, the EGit Team Provider is going to have to get good, and do so quickly. If we want to see git adopted to a significant degree, my personal belief is that we are going to have to ship EGit with Helios. My guess is that you, the community, are going to have to help make that happen through use, feedback and contribution. (Note that I do not know what the plans of the EGit project are. This is my personal opinion.)
- The Eclipse Foundation doesn’t tell projects what to do. So if git is going to be adopted at Eclipse, the projects are going to have to vote with their feet. In other words, we are going to have to see a number of the large and mature projects bite the bullet and make the transition to git. Because if there is little or no evidence of this within the projects, then the investment by the EMO will be for naught.
- In some ways, the most difficult part of this will be deciding which of the existing SCM systems we’re running to shut off. I personally have a major problem with the idea of running CVS, SVN and git for anything other than a short and constrained period of time. The main reason for this that having multiple SCM systems is a barrier to contribution for the community. If someone needs to learn three different SCM systems to interact with and contribute to three different Eclipse projects, then we have failed our community.
- “Eclipse Labs”: We’re looking at creating an Eclipse Foundation affiliated forge where projects which are based on the Eclipse platform, but are not interested in hosting at eclipse.org can run their projects. There will obviously be some constraints (e.g. only Eclipse Foundation projects can use the org.eclipse namespace, and be part of the release train) but Eclipse Labs will hopefully over time form a powerful complement to the projects hosted at Eclipse.
- “Get Involved”: We are going to be adding some variant of a “get involved” menu item to the left nav for project home pages. We want to make it easier to contributors to understand how to get involved with every Eclipse project, and we want to help projects to learn and follow best practices on how to facilitate more community involvement. Obviously, this may require some additional work for some projects, but the increases in community involvement, contributions and project diversity will be worth it to all of us.
- Artifact Repositories: We want to make it easier for adopters of Eclipse technology to find and consume the great work that’s happening in the Eclipse Foundation projects. For this reason we are going to be looking into what technologies would make it easier to consume Eclipse software. To be blunt, we don’t know what this might be at the moment. It could be Nexus or Buckminster or Tyco or something new from the B3 project. We just don’t know and we won’t be making any final decision for six or seven months. If people are interested in getting together for a BOF at EclipseCon, I think that would be a great way to move the conversation along.
As you can see, this is a pretty significant list of projects for 2010. Denis, Wayne, Ian and their respective teams are going to be carrying the ball on these. So comment at will, but please also consider buying them a beer at EclipseCon 🙂
Dear Bjorn: Go away
Bjorn,
Your latest post deserves a clear response.
Your former colleagues at the Eclipse Foundation have tolerated your public abuse quietly because we are professionals, and we honestly thought that you would tire of it. Apparently we were wrong. But the time has come to say it: You are a jerk. Please go away. You quit the Foundation, you have zero commits since April, and we tire of your sniping from afar.
It is no secret that the Eclipse Foundation is a 501(c)6 and is supported financially by members. But to say that the Foundation does not care deeply about the open source community is a pure fabrication. It attacks the personal and professional reputations of all of us who work hard at the Foundation for the entire community.
Since you left, the team has made great strides in pulling together a number of wonderful, committer-oriented prgrams for 2010. Included amongst the items already committed to the Board is a branded Eclipse forge hosted elsewhere (e.g. IP Policy free), Git support for the whole community and major improvements in hosted build and test at eclipse.org. All initiatives that you failed to make much progress on in four years on the staff here.
If you want to go create your own forge somewhere, go ahead. Your steady acid drip of negativity will serve your new community well.
Some (Ottawa) Holiday Cheer
Calling the Eclipse Ottawa community!
The Eclipse Foundation would like to invite everyone in the local Eclipse community to join us at Marshy’s on the afternoon of December 18th. If you can make it, please add your name to the meetup page.
We hope to see you there!
This Week’s Announcements
Hearty congratulations are due to Microsoft, Soyatec and Tasktop for their announcements yesterday regarding Eclipse and Silverlight, Azure and Windows 7. It is great to see Microsoft providing support to the ecosystem to ensure Eclipse remains a great development environment for the Windows platform (which still represents almost 80% of our users on any given day).
Much of this progress results from work led by Vijay Rajagopalan, a Principal Architect in the Interoperability Technical Strategy team at Microsoft. Vijay has been working closely with the Eclipse Foundation since just before EclipseCon 2009 and we’re very happy with the progress we’ve made.
Collaboration Is Just Good Business
Last week at the Open World Forum, I had the pleasure of attending Michael Tiemann’s talk on the “Road to the Digital Recovery”. Michael kindly summarized his key points on his blog.
The most interesting sound bite from Michael’s talk was his estimate that $1 trillion is wasted every year in ICT spending. His logic is:
…18% of all applications are abandoned before ever reaching production, 55% are “challenged”, meaning they are either late, missing key functionality, buggy, or some combination of the three that results in measurably degraded performance. Back when this study was done, the scope of the analysis concluded that $78B/year was being wasted by US CIOs on “bad software”, but that is the tip of the tip of the iceberg: with global ICT spending over $3.4T USD in 2008, money wasted on “bad software” now exceeds $1T USD per year.
I actually think that the reality is worse. Much worse.
Michael’s conclusion is based on the number of projects which fail to accomplish or only partially accomplish their stated objectives. Everyone in the software industry knows that these types of failures are a simple fact of doing business-as-usual. The proposed solution is to lower the percentage of complete or partial failures by improving the quality of the software being built through using open source processes and techniques.
While I have no disagreement with that conclusion, I think that it is missing a huge part of the problem, which is that we all collectively waste massive amounts of human and commercial capital by building too much software. The sheer amount of wastefully duplicated effort endemic to the ICT industry is staggering.
Note that I am not referring to the software which provides companies any source of competitive advantage. I am referring to the software infrastructure which every company needs to merely operate in their particular industry. In every single industry you will always find some amount of software required by 100% of the players, for which 0% get any sustainable or even measurable competitive differentiation.
For one example, imagine the scenario where a new government regulation in the (say) insurance industry requires all of the companies in that industry within that jurisdiction to implement a new set of procedures. Pretend that there are 30 companies impacted. Even if the implementation project within each of those 30 companies was executed flawlessly, the wastage is 30x, because none of those companies achieved any customer differentiating value from their efforts. Multiply this scenario across all of the companies in all of the industries and the wastage of human effort, skill and imagination is depressing.
In my view, the future impact of open source on the ICT industry is not simply to make software better quality. It is to reduce the amount of wasteful effort squandered on implementing and re-implementing and re-implementing yet again the same bags of stuff that our current corporately-silo’d development structures require.
Open source communities such as Eclipse, Apache, Linux, et al offer enormous potential cost savings to industry. By establishing the licensing, IP management, governance and development processes to enable cross-company collaboration, these communities open the possibility that much of the “operating systems” of various industries could be built and shared. This will require some cultural shifts, but I predict that the business and economic rationale will inevitably drive companies in this direction.
Life at Eclipse 