As for the freerider problem. My intention is to get as many developers as possible to work out of OpenACS CVS and work collaboratively on the core and, if, for whatever reason they do not want this, that they feel the need to provide their enhancements in structured way to the community as a whole. For myself it is too much hassle to work on the OpenACS core which I'm going to use for all my clients and then take the code from some other source where I'm not (directly) able to collaborate and especially not familiar with. This was the case when I looked at dynfields where I thought, wow, nice piece of code, but do I really have the time and nerve to integrate it right away with contacts and the OpenACS core, especially as the original developer would know considerably more about a proper integration than I do.
So maybe the comment regarding the freerider was too hard, but as community we are too small to let even bits of code which are excellent not make it into the toolkit because of our inabilities to properly work together.
Now, what could be possible actions:
- Frank already offered his help with Marketing and his project open site is a great show of the fact that he knows what he is talking about. Maybe you could encourage some other people in joining you to market OpenACS out of the niche market and spearhead this effort? I'd be more than happy to provide help as good as I can.
- With 5.3 we should have a "report our installation back to OpenACS". Maybe this could be modified to allow a reporting of P/O installations to Frank as well. This way we would at least begin to know how many people are actually using OpenACS.
- We should provide a place where people could work together around projects they are interested in. See my previous postings about a Bazaar or the effort to get a Wiki going.
More later :).