Re: ASCII Pfiles (was: Color)

From: Daniel A. Koepke (
Date: 01/16/01

On Tue, 16 Jan 2001, Peter Ajamian wrote:

> Well, if you're gonna go to the Linux numbering scheme it might be
> better off to simply skip 3.1 and move to 3.2 as the first stable 3.x
> release, that way you can get on track quicker and keep the stable
> releases even...

I was going to avoid skipping any version numbers so we don't get
questions like, "Where is 3.1?" if we just jumped straight to 3.2.  It
seemed like the best compromise between eliminating 3.0 confusion and
avoiding version jumping.  There's some justification in treating 3.1 as a
hybrid stable-development release.  My vision of things:

    Version 3.1 -
        Major feature is any real bugs in 3.0 gone.  Some code that does
        not work exactly as it should will also be reworked for 3.1 (these
        fall under the category of "feature bugs" -- i.e., things that
        should be there, but aren't, such as fixing up the issues with mud

    Version 3.2 -
        First 3.0 release that can be considered nearly feature-complete.
        3.2 MUST, at the minimum, propose and implement some solution to
        OLC, probably by direct incorporation of GenOLC.  The included
        front-end should be minimalist.  This might disqualify both Oasis
        and obuild.  The idea being that it's a functional, but not
        flashy, interface to the OLC capabilities of CircleMUD.  Better
        interfaces (Obuild and Oasis) are available from the FTP site,
        after all.

    Version 3.4 -
        Address ASCII pfiles and rent files.  Remove the reference to the
        then-defunct e-text system, unless we want to provide the ability
        for binary player files.  This should make CircleMUD 3.x truly
        feature complete.  Minor features might be added to the 3.x
        branch in successive releases, but these should mainly fall under
        3.4.x.  At this point, the 3.9 (i.e., pre-4.0) branch should

    Version 3.9[1] -
        The more we talk about it, the more it sounds like one of the more
        ambitous server projects around.  Not because it proposes anything
        exceedingly difficult or time-consuming (although it does), but
        because it proposes to do something that few, if any, other muds
        have done: really use the technology we have to great advantage.

In short, in my view, 3.1 is, itself, an interim release, between the beta
3.0 and the real deal (3.2) which bridges at least one gap between feature
completeness and CircleMUD 3.x.  It's safe, I think, to classify 3.1 as a
development release that is, by coincidence, more stable than 3.0.  Just
so long as we don't get into the habit of such hybrid releases, all should
be well.


[1] George suggested Circle^2 or Circle^3 as project codenames.  Here's
mine: Hyperbolical.  A reference to both hyperbola and hyperbole.

   | FAQ: |
   | Archives: |

This archive was generated by hypermail 2b30 : 12/03/01 PST