On Wed, Dec 01, 2004 at 10:43:22AM -0500, Joe Schaefer wrote:
Stas Bekman <[EMAIL PROTECTED]> writes:
One needs to go through a deprecation cycle before any backwards compatibility in the same generation of the project can be dropped.
Heh, a search for "deprecation cycle" on marc's [EMAIL PROTECTED] archives comes up empty.
I'm not sure really what you expect. That no API changes can be made
during 2.1 development unless they were predicted ahead of time by N
years and marked with a red dot?
I expect the API changes in the 2 adjucent major version number (2.0 - 2.2) not to affect backward compatibility. But obviously my expectations are wrong on the httpd land, like with many other things.
2.2 will not be backwards compatible with 2.0, that is documented in
VERSIONING. Exactly *how* it is not compatible depends on exactly what
gets changed. Exactly *what* gets changed depends on round tuits and
itches getting scratched by individual developers.
How can we predict that Mladen Turk will come along and do a bunch of proxy work ahead of time? Should we say instead to Mladen Turk "Hey, nice code. But we'll have to deprecate <this> and <that> - come back in three years when 2.3 opens and you can do your work then. Thanks for calling!". Formal process has its place, but I don't see it here.
You can't predict that someone will come and do the innovation. But you can easily prevent those innovations breaking other people's perfectly running code and build obstacles to upgrades.
I'm not talking about this particular proxy change, but in general. At the moment I have no personal problem with it, and I don't know if anybody else have. So I rest the case.
-- __________________________________________________________________ Stas Bekman JAm_pH ------> Just Another mod_perl Hacker http://stason.org/ mod_perl Guide ---> http://perl.apache.org mailto:[EMAIL PROTECTED] http://use.perl.org http://apacheweek.com http://modperlbook.org http://apache.org http://ticketmaster.com
--------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
