>Almost a year since I last asked this question, and there have been zero
>updates to the Cricket tree since then.
Yeah, I kinda went down the bog with trying to adapt the
no-inline-html patch for grapher.cgi — which ends up as major
brain surgery on the grapher every time :-( — and eventually
put it on the backburner.
I don't think Cricket is dead, it's just been in hibernation. If
we can drum up some interest again — and it won't take all
that much; there's quite a bit of low-hanging fruit — we
should be able to get up to speed again.
Just off the top of my head, I might suggest we stop being so
damn paranoid about stability; simply because it tends to be
such a drain on the poor saps trying to maintain it (burnout has
been a problem for this project).
Lets try to put in the smaller, more low-impact, patches in the
tracker — and any other small fixes — and push a 1.0.6-pre1
release just to let the world know the project is still alive;
and then put some more radical stuff into 1.1. If the policy is
“Anything that makes sense goes” and “makes sense” is
defined as anything that gets no objections on this list, people
might be able to work on stuff like the multi-threaded collector
(or, my pet research project, the daemonized collector), support
for the latest RRDtool, or a grapher with all the HTML ripped
out, without the obsession with making it perfect before checkin
that saps resources and interest.
If we go crazy and the result is significant lessening of
stability, we can tack on a stabilization period afterwards to
fix what we broke.
And just to kick things off I'll try to dig up my old devel
machine and polish up the grapher.cgi changes to check in on a branch.
Francois, you up for polishing genDevConfig a bit and see if we
can have it ready to include by default in a Cricket 1.1 release?
A vendor abandoning a platform does not mean we have to follow suit
(if anything, we might continue the support it just out of spite...)
— Jarkko Hietaniemi