For the longest of time I have been an avid supporter of open source projects. I still am. However, there is one inherent problem with not being paid for your development time: There is a distinct lack of commitment and dedication in comarison to closed source and commercial applications.
Take everyone’s favorite IDE OnBoardC, for example. The application is good enough, but there is definitely room for improvement and optimization, particularly in the area of code generation.
But we’ll save compiler optimization for another day… There is one major problem with OnBoardC that really bugs me (along with many other OnBoardC users): The lack of Bird integration. When you look at the OnBoardC source code, it looks like it is almost child’s play to do this… “Then why hasn’t it been done, Ryan?” you may ask. Simple, really; It is a lack of dedication on the projects administration’s part.
According to the OnBoardC Developer’s Charter the following must happen in order for a change to the source code to be made:
1. A bug or feature request is posted appropriately on the Sourceforge project page.
2. A member of the OnBoardC project must propose a vote for the upgrade to happen.
3. A second developer seconds this.
4. Via the Sourceforge Developer’s mailing list (which, by the way, has been inactive as of 21st December 2005) a vote is made on whether the proposed change should be enacted.
5. If it is approved by more than half of the project members the Point Of Contact is chosen and they are in charge of handling the task. At this point they are in charge of handling the code, obtaining the Roger and committing the code.
As for Bird integration, we are still on step 4, and it seems it will stay that way indefinitely. I originally purposed a vote per the charter, and Steve Little himself has seconded a vote, yet where is it?
Now, granted, Steve and Dave Beers have other things going on, I fully understand this. I just feel that administrators almost have an obligation to be the driving force in their project and actually DO SOMETHING. If an administration is too busy, then appoint another administrator who is able to spend more time on the code.
For example (And I’m sure he’ll be mad at me for even mentioning his name for another project) John Wilund, the now primary developer of SrcEdit, would make a wonderul OnBC administrator. The SrcEdit bundled with OnBoardC version 2.51 is an ok source editor, but a bit outdated. When you download the latest SrcEdit from the Trac page, the improvements will make you very happy. What’s more is that the SrcEdit development is far more active than OnBoardC development has been in a long time. If you were to E-Mail a bug report to John, you can expect a bug fix very soon, and this fix will work as stated. John is in the rare situation where his area of work affords a lot of free time to programming.
Now I’m pretty sure John isn’t up to working on OnBoardC, on top of a number of other projects he ahs, but I still believe that OnBoardC development could use some fresh blood.
To put this somewhat ranty post into focus, I have a moral (and there is the collective groan from all of the TamsPalm readers): For any open source project to survive, not just OnBoardC, it as to be active and most importantly needs a strong administration that can fuel this activity.
I first wrote this in a ©2007 Ryan Rix “Moment of Brilliance” and, originally, I had no real plans of posting this. It is a bit harsh to David and Steve, but hopefully it will spur some sort of action (as long as that action isn’t a defamation of character lawsuit ) But maybe it will just end up pissing Dave and Steve off, which, believe it or not was not the intent of this post.