[asterisk-dev] About asterisk development plans

[David Backeberg]

On Wed, Mar 18, 2009 at 5:02 AM, Jon Bonilla <manwe at aholab.ehu.es> wrote:
> El Wed, 18 Mar 2009 07:58:19 +0100
> "Olle E. Johansson" <oej at edvina.net> escribió:
>> in my svn directory). The cost of evaluating a new version is high and
>> 1.6 release policy is not acceptable, since the core keep changing too fast.
>
> That's a point. It is not expected Asterisk 1.6 to be suitable for production
> environments.

Let's rephrase those as: 'evaluating a new version costs too high for
that person' and 'asterisk-1.6 is not suitable for that person's
production environment'. That may have true for more people perhaps a
year ago but as development and releases continue it will mostly be
only true for those unwilling to set up a lab, test it out, run a
pilot, and burn it in.

The features that are going in are making asterisk run faster, scale
better, and in my experience, things are 'just working' in
applications I need in ways that they didn't used to. With any version
of any piece of software you should do testing for your particular
usage, set it up the way you want, and then when you've sufficiently
tested it set it loose in production.

Did you honestly grab asterisk 1.2 when it was new and put it straight
into production? If so, send me your resume so I can remember to not
hire you. I think sweeping generalizations like "asterisk 1.6 is not
ready for production" are ridiculous.

No upgrade is painless. If you want to compare 1.4 to 1.6 conversion
against say Windows XP to Windows Vista conversion, I don't think
they're even in the same league.
Most of the changes are nitpicky ways that you change the arguments in
your dialplan and flags to applications. You fix them once, and then
your application works. AGI development is way more awesome than it
used to be, and I want to kiss whoever made

cli> agi set debug on