[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: custom vs. derivative (Re: packages.gz corrupt, missing packages and other issues)



Hi,

On Tuesday 01 April 2008 15:31, Andreas Tille wrote:
> > 2. What is the meaning you once gave to CDD, where is it documented?
> Easy: Try the first Google hit "Custom Debian Distributions".  If in
> doubt follow the link I gave in one of these mails in this thread I have
> given.  There were several active discussion on this mailing list in
> the past - feel free to search the archive.

Well, I hoped for an answer from you, not a pointer to search engines.

The first hit on google for me was http://wiki.debian.org/CustomDebian which 
says "Custom Debian Distribution (CDD): a subset of Debian that is configured 
to support a particular target group out-of-the-box." 

AFAIK subsets of something can be inclusive and exclusive of other stuff, so 
that first hit on google doesnt help.

Also, the same page states lists Skolelinux / Debian under "CDDs released with 
Etch:" which is wrong.

> > And, isnt it possible that this meaning changes over time?
> Yes.  It is perfectly possible.  But this should not happen silently
> to avoid confusion. 

But it did :-)

> It makes no sense if people use terms with different 
> meanings.  That's why I talked about "tolerance" to give room for
> a change. 

IMO it makes even less sense to bring this topic up, if someone merely asks 
how to use simple-cdd here.

It's (alsmost) completly pointless IMO. Or at least, it shouldnt be the main 
part of the answer.

> > Also I dont think there is agreement, where a CDD is or should be.
> Well, there is a definition.

No. There are many definitions. 

> > Also, I personally see this list as a list for discussions about CDD
> > development and about the tools useful for this. And as that I dont
> > tolerate discussions about (for example) using simple-cdd to build
> > whatever non-CDD someone has in mind, but I think this _exactly_ one of
> > the things this list is for.
> OK.  What would be your conclusion to do?

Move on. 

> > Maybe it would be wise to introduce a new term, for Debian based
> > distributions, which are not CDDs according to our definition. That would
> > prevent us from using "(official) CDDs" and unoffical ones. So I suggest
> > DBD (Debian Based Distributions) for those :-) What do you think?
> I think we should find a new name for what is currently called CDD and
> it desperately needs the term "internal" in the name.  Anything else is
> missinterpreted because I learned in the past that people refuse to
> read definitions and the term CDD is simply missleading. 

So you plan to create a new term and new definitions and hope that those will 
be read? 

> When I had 
> my talk about Debian-Science at Chemnitzer LinuxTage two people asked
> myself after the talk why I try to build just another Debian derivative.
> I've talked about the fact that Debian-Science is internal several times,
> I even argued to other people who wanted to know the difference to
> other scientific distributions that we have a major advantage: We are
> plugged into the huge Debian distribution and thus are on the sholders
> of a giant.  

I tell people that Debian Edu is (or rather: soon will be) Debian.

The concept of a distribution within the distribution is not yet widely 
understood.

I believe one way to make it understood better, is to give people the 
definition in short sentences, instead of pointing to google or cdd-dev.


regards,
	Holger

Attachment: pgpHEyz7WIbyQ.pgp
Description: PGP signature


Reply to: