Hi everyone,
I am grateful to the past and present developers of the
SimpleMachines forum software (which this forum runs on) for all the time and energy they spent on it for
many years. You did a great job, there.
Presently, however, I am not convinced that SMF is being well maintained. There are critical security bugs in the current stable release which have been made public, addressed to a wide audience of both white and blackhat hackers early this month (two months after they were privately anonunced to the developers, as the person announcing them now states). To date, there is no public reaction to those by SMF, and attempts to discuss these vulnerabilities and possible counter measures are actively prevented on their support forum. While it's perfectly fine for a software which is based on volunteer work to develop slowly, this is not the right way to handle security vulnerabilities, be it a professional or volunteer run project.
Moreover, SMF development seems to have slowed down a lot during the past two years, while there are known issues which really should be addressed (an embedded anti-spam mechanism which works, support for newer versions of dependencies such as PHP and MySQL, an extension / mod interface which is not patching, a more complete API, and more). While there was (and is?) a major version upgrade in the works for two or three years, it has not yet made it to a production ready release. At the same time, the source code repositories for the current stable release are not open, which is somewhat unusual for an open source software project.
Some of these issues
may be better addressed by
ElkArte, which forked off SMF a year (?) ago and has since undergone massive code changes. Whether those are, for our needs, to the better or worse I need yet to test. Which is why I'm currently in the process of setting up a test install on the staging site. This should happen once
initial bugs (note that I'm testing a beta pre-release) have been sorted out.
If we should decide to go with ElkArte (it is definitely too early to tell at this point), there is an SMF importer available (which I am going to test, too) which has apparently undergone some testing on this migration path before.
So I'm mostly posting this to get some initial feedback on whether this sems like a good or bad idea from your POV, and to see whether I missed some major road blocks.
(Lack of response will not discourage me.)
Thanks,
Tom