lizard wrote:Maybe the best form of contribution are examples. We all know the situation, we want to quickly program a .pdf output with cairo for example. "Wait, there was a demo in the example files that came with fbc..." Then you can load this example, change few things and your pogram part is ready, without reading tons of sites and long explanations. Thats programmer style.
I agree 100% with this. In fact we don't know by advance when we'll need this or this information on what matter.
We have examples coming with the fb distribution of course but they can't be exhaustive unless someone decided to grab everything valuable everywhere (coded in fb), then sorted it, and maintained the information on the prerequisite (a missing dll for any reason with time passing would break the code ..etc.).
This is not how I would do then. For me there is a need for some automated referencement of the published stuff. I started this for testing the concept and it should work. May someone end it, or me if I find the precious time! and we would have something. Of course referencing the past content is not enough, it supplements the distributed examples, but new stuff should always feed the base.
In case someone wants to know how I project in an unknown future to reference the things, here the proof of concept (yet quite in disguise at the time):viewtopic.php?f=8&t=24884&p=222463#p222463
This things sort the text by frequency, so if well used you may sort any topic by content (if cairo is cited 12 times as well as pdf in the same topic, chances are that you found your grail). This is more the kind of job for the admins, but we can certainly make a well illustrated proposal first.
Ok this is just an idea among others. I know that it's not obvious how it differs from the search function of the forum, but in my mind it does by a real ability to know all the context.