Re: Was the release of Debian 2.0 put on Linux Announce?
Hi,
>>"George" == George Bonser <grep@shorelink.com> writes:
George> I agree. If I am speaking nonsense, just tell me what part of
George> my idea is nonsense and why rather than a blanket statement
George> along the lines of "The thoughts of the user community are of
George> no importance to us" which is the message I heard.
I was not going to respond to this thread any more (I really
fail to find your arguments convincing), but this gross
mischaracterization is something I have to correct.
I assume you were referring to my posts, though you were too
polite to say so.
I never said (nor did anyone on thisd mailing list) that "The
thoughts of the user community are of no importance to us".
I do refute that "Since the users are the ones you all are
doing this for, you are working for the users, by gar, and what they
say is of over riding and paramont importance, and do what they say
or else"
Yes, I exagerate. But no more than the user-unfreindly spin
that peole have been putting on my statements.
George> That hurts as a user and leaves one feeling that they can not
George> contribute the results of their life experiance which may be
George> quite a bit more varied than a 20yo developer.
Whatever gave you the idea that anyone is saying that
non-developers can not or do not contribute? Bug reports, and request
for enhancements, and feedback on user interface issues are
important. The community which I write things for is larger than
those who write code.
However, remember that the people who work on free software
projects may not share your views, and indeed, may have different
agendas from your own.
s far as the version numbering scheme goes: Major versions
imply a significant (major) change in the distribution; in the past,
a.out-->elf and libc5-->glibc have been considered major changes that
warrant a major number increase. Whetner a change is "Major" or not
is subjective; and indeed, so far, the changes that have qualified
have been one time changes and unlikely to create a rule set from.
There is an unspoken implication that there may be release
boundary incompatibilites at a major number change; you may need to
upgrade a significant number of packages on the system; and, in the
past, that may mean that you can't mix and match packages between
different releases; in the future, we shall do our darndest that such
upheavals are prevented.
Minor changes just cause the minor number of the release to be
bumped up; and the changes in the system are minor; in the past, that
has meant you can have a hyubrid system; in the future, we shall try
extend this around major version changes too.
The version numbers are determined based on an subjective and
empirical measure of the changes involved, as determined by the
developers; and not on pilitical motives like "that makes our version
number higher than anyones".
I still reject the idea that we should change major versions
faster to increase our market share. I do not want to achieve market
share that way. The same reason applies to the number assigned to
slink: I would assign the number based on how different slink was,
(and that depends on how much of the transition to the FHS, and the
new way of doing releases, and the restructuring of the Archive gets
done), rather than how well CD manufacturers can market the upcoming
Slink.
If forced, I shall ask the version number to be bumped up to a
billion or so to fix this marketing problem once and for all.
So pardon me for putting technical reasons ahead of marketing
ones, and maybe this shall be the death of Debian if the project does
it this way. But making decisions based on marketting rather than on
technical merit is as likely to kill Debian as anything else.
As Debian has always averred, anyone can use Debian as a base
for their own distribution, of this attitude offends or displeases
anyone. And the version number of that distribution can be set to
whatever the distributors like. (Is this being too offensive and
confrontational? -- In the free software community that I matured in,
if one did not like the current way, one created the method that
worked better. But that is a dying philosophy, and doubtelless I
shall be villified for bein harsh and selfish and unreasonable. So be
it)
manoj
--
Let them obey that know not how to rule. -- Shakespeare
Manoj Srivastava <srivasta@acm.org> <http://www.datasync.com/%7Esrivasta/>
Key C7261095 fingerprint = CB D9 F4 12 68 07 E4 05 CC 2D 27 12 1D F5 E8 6E
--
Unsubscribe? mail -s unsubscribe debian-user-request@lists.debian.org < /dev/null
Reply to: