dce/rpc named pipes port domination

David Brodbeck DavidB at mail.interclean.com
Wed Jan 9 06:02:02 GMT 2002


>From: Lukas Smith [mailto:smith at dybnet.de]
>Sent: Wednesday, January 09, 2002 7:39 AM
>To: samba-technical at samba.org
>Subject: dce/rpc named pipes port domination

>samba is not a project to fork from easily, if at all
>samba is huge and samba is way too established

--> And has already forked once.  That suggests it's not impossible.

>samba is one of the key linux (open source) projects (imho)
>so if samba prevents other projects from working when installed (as
>default) then those projects will probably not suceed you can't compete
>with samba

--> I disagree.  Yes, it's harder to compete with a popular package, but it
can be done.  Used to be *every* distribution shipped with inetd -- it was
about as popular as a Linux program can get.  But now most of them ship with
xinetd instead.  (And you can't realistically install both of them and have
them both work.)  I wouldn't be surprised to see, for example, syslog-ng
replace syslog in the same way, in spite of having a completely incompatible
configuration file format.  It's that much better.

The way I see it, this is partly how the open source community works.  If
you disagree strongly enough with the approach of one group that's doing
something, you can always work on a parallel project.  If other people agree
that it's better, eventually you'll reach a critical mass and it will become
widely accepted, perhaps even replacing rival products.

I agree that it would be *better* for Samba if this could be worked out
without a code fork, but if not I think the people making the complaints
need to implement their ideas and *prove* they are better instead of
complaining about conspiracies against them.  Eventually you have to put up
or shut up.




More information about the samba-technical mailing list