Reply – Re: Forum improvement
Your Name
Subject
Message
or Cancel
In Reply To
Re: Forum improvement
— by io7m io7m
While I'm currently indifferent as to whether or not the project needs a new forum... An attempt was actually made last year due to Nabble falling over under heavy load.

Basically, the requirements of anything we host ourselves are:

* Security - The project's source code lives on that server. We have mirrors in lots of places, but any compromise would likely still be bad news.
* Support for posting over mail. Sven uses the forum as a mailing list, as he said.
* Importing data into the new system in a manner that preserves URIs. Put anything into a search engine about JOGL and you'll get hundreds of results linking to specific forum posts. That's something we don't want to mess with!

phpbb has an attrocious security history (as do the majority of PHP projects). They claim to have improved things, but I'm skeptical that systemic problems like that can ever be fixed. They claim to be able to import data from Nabble, but I don't see any claim that URIs will be preserved.

We tried punbb.org (http://punbb.org), because although it was still PHP, it was a lot smaller and had a better security history than any of the others. It was a bit too minimal though: It didn't have any support for mail or importing data from Nabble.

We then tried SMF (http://simplemachines.org). Their security history at the time was "better than phpbb", but I don't know any more than that. They did have a mailing list plugin to allow posting via mail, and it gave every appearance of working. When it came to Nabble, however...

http://jogamp.org/log/irc/jogamp_20130701050539.html#l80

So it kind of stalled there.

If there was an option that satisfied the first two requirements, I'd not personally be opposed to putting the Nabble forum in a read-only state and starting a new forum on a different subdomain (if it's even possible to do this with Nabble). This would preserve the old links but direct all new posts to the new forum.

Like I said, though, I'm indifferent. If people are picking software projects based on the software used to drive the project forum, then we probably don't need them!