Red Hat's Bluecurve (was: Red Hat 8.0 is 'official')

bscott at ntisys.com bscott at ntisys.com
Tue Oct 8 00:41:05 EDT 2002


On 7 Oct 2002, at 3:50pm, pri.nhlug at iadonisi.to wrote:
>>> Minor nit: I know the inside story about why there was a 7.3 and can
>>> only say that it had zero to do with the problems or lack of problems
>> >with 7.2.
> 
> The basic issue is that Red Hat only bumps major release numbers when
> there are backward (or is it forward?  Or both maybe?  I forgot) binary
> compatibility issues.

  Forward.  A binary built on Red Hat Linux release "x.y" must work on any
Red Hat Linux release with the same value for "x" (assuming dependencies are
solved).  For example, a binary built on Red Hat 7.0 will not run on 6.2,
because the compiler changed (and GCC, as a rule, does not maintain
compatibility with anything).

  Backwards compatibility means that a binary built for an older release of
Red Hat Linux should work on a newer release.  This is accomplished by
including various "compatibility libraries", i.e., newer releases of Red Hat
include libraries compatible with older releases.

> I think the fact that they stuck with the .0, .1, .2 release numbers is
> purely coincidental.

  I do not think it was conincidence.  It appears that Red Hat could give us
three releases before so many "neat new things"  made the pain of breaking
binary compatibility "worth it" in someone's eyes.  If the number of "neat
new things" introduced per unit of time is roughly constant, we would see
this behavior.

> Actual, most distros have been pretty good about not doing that.

  *cough* Slackware *cough*    ;-)

-- 
Ben Scott <bscott at ntisys.com>
| The opinions expressed in this message are those of the author and do not |
| necessarily represent the views or policy of any other person, entity or  |
| organization.  All information is provided without warranty of any kind.  |




More information about the gnhlug-discuss mailing list