I like the gumball analogy.
I'm more awake today. I wasn't complaining last night, I was just
stating my criticism in an emphatic way :-)
As you say, Mark & Co just started writing a little soft PBX system to
"scratch their own itches" at the time.
Mark was so kind as to share it with the rest. The fact that it grew so
wildly fast (and made them very rich) shows that there was a GREAT NEED
for something like this. The other side of the coin is that there's a
risk of becoming a Frankenstein system with lots of ugly patches and
configuration syntaxes with small variations and basically lines upon
lines of code where everybody adds something to scratch their own itch
in their own particular manner.
And 90% of Asterisk users like it. Probably they just want a quick and
dirty PBX to make cheap long distance calls over VoIP and a few
extensions within their company.
I'm not sure what improvements version 1.4 will bring (except for
having graduated AEL2 into the "real AEL") but being a 1.x version it
can't bring architectural changes. I haven't looked at the code but I'm
sure it will be much better.
I checked out OpenPBX again today. They seem to have started with a lot
of enthusiasm and good ideas but they are going very slowly and I'm not
sure they are differing much from Asterisk's architecture after all.
I also took a look at Freeswitch. Now, here's a guy who knows software
engineering!!! I'm afraid it's still too new and featureless to be of
any competence to Asterisk... but who knows in a few months...
I'd love to help with the coding but at the moment I'm too busy... I
wish I were like that X-Men guy who replicates himself, then my third
clone would be the Asterisk clone!
BarZ
Michael Collins wrote:
Asterisk is what you make of it. If you don't want certain
applications to run on a certain instance/machine then you should
"noload" them in modules.conf.
Barzilai still has a point. Noloading various applications doesn't
address the underlying architectural issues. The fact of the matter is
that Asterisk has grown quickly and therefore many features have been
attached like the proverbial gumball getting bigger and bigger.
The improvements from 1.0.x to 1.2.x were significant and hopefully the
jump to 1.4.x will be equally significant. However, the needs that
Asterisk fills today are broader and more complex than they were just a
few years ago. It was impossible for Mark & Co. to predict everything
that Asterisk could possibly do, so they couldn't necessarily engineer
the system with all of those things in mind. Now that we have several
years' experience and thousands of Asterisk systems in production there
is a clearer picture of the needs to be fulfilled - the "itches" that
Asterisk scratches, as the TFOT book puts it.
I for one hope that Asterisk 2.0 isn't too far off in the future. I'd
love to see it redesigned and re-engineered from the ground up, taking
into consideration the many years of experience garnered by the Asterisk
community. (Disclaimer: this is NOT a complaint! I love Asterisk and I
just want it to be even better than it already is.)
Having new development team members is definitely a good thing. I'm
looking forward to 1.4 and more.
-MC
_______________________________________________
--Bandwidth and Colocation provided by Easynews.com --
asterisk-users mailing list
To UNSUBSCRIBE or update options visit:
http://lists.digium.com/mailman/listinfo/asterisk-users
