Re: [Gnu-arch-users] What are version numbers?

[Pau Aliagas]

On Wed, 10 Sep 2003, Tom Lord wrote:

> But there's way too much flexability in the system to require that
> isomorphism.   You can use config names to map from release names to
> versions.   You can use N-component version numbers.  You can throw
> tags into the mix...
> 
> It's a cataloging system, as in what libraries use (but sans the
> dictionary of pre-defined terms which, currently, would at most
> contain the keyword "devo" when used as a branch label).  As a
> cataloging system -- it just goes half-way from "unstructured" to
> "cataloged", relying on a cultural context to evolve and meet it
> halfway.

I still think as I did one year ago:
-you don't need categories--branches--versions

Why? Because the projects do not have any implicit relation among them; 
the only relation is arbitrary, the relation you "feel" when they have 
similar names. And this can be accomplished equally with free names.

Let me explain it in an example:
-I create projectA--prod--1.0
-later on I create projectA--prod-1.1

The only possible formal relation is that projectA--prod-1.1 is a 
descendant of projectA--prod--1.0. But I could name it as I liked. In fact 
I can name it as I want and have this ancestor relation -i.e. 
projA--prod--1.0.

For me things would be much easier to understand if you only had a project 
name, free syntax. You could abrowse the tree using regular expressions if 
you kept a disciplne; if you don't, it's up to you.

Forcing this schema is forcing policy, but for what's worse, for no gain. 

The only use I've found of version numbers is in config projects, where, 
if you do not specify a version, it looks for the newest one. And this can 
be accomplished in the same way with free names!

So my opinion is: "Let the user decide his naming schemes".

Pau