> What's next?
> 
>   (in next mail)

We're about a month shy now of 2 years since the 3.8.2 release.  So,
I'd like to start gliding down toward the next one.  We may not have
massive _forward_ movement since then, but we've moved mountains
laterally at least.  There are a few minor bits I'd like to get added
before finalizing it, but I think we're hitting the run-up.  Any bugs
we can find and fix would be icing on the cake as well.


3.8.2 was released shortly after I brainwashed Richard into passing me
the reins, along with all the power fame and wealth that goes along
with that.  It was consciously made as an "OK, here's our current
incremental step since the last release, before massively uprooting
everything" release.

And since then, we _did_ uproot everything.  We replaced the entire
build system, reformatted the entire codebase, dropped our [pretense
of] support for a ton of older systems (and in fact almost everything
but recent systems active users care about trying to keep running),
removed various historical features wholesale, redid the command-line
long args to double-dashed, and have an ongoing stream (well, in
occasional spurts as time allows) of code reorganizations and cleanups
and [hopefully] improvements to future maintainability.  And a bunch
of other stuff I'm forgetting or considering too small to mention.

With all those changes, this is our first chance in a long time (and
hopefully last chance for another long time) for clean breaks with
legacy.  We've done pretty much all of them that I had floating around
my head; I'm still looking around for anything else, but I don't think
there's anything major or pressing left.  I'm open to ideas though.


As part of the reworked everything and retired legacy, I'm proposing
to bump the major version for the first time since Claude released 3.0
more than 20 years ago (I don't know exactly when; 3.2 was no later
than 1994 though).  Of course, you could argue that unless your name
is part of the acronym for the program, you don't get to do that sort
of thing.  But I'm too power-mad to pay attention to such limits.


And so, the upshot is that I'm currently hoping for us to make a 4.0.0
release, probably sometime in the next couple months.  If anybody has
suggestions for things they think should be done before (or instead
of) that, or other notions, let's have 'em.  Otherwise, get your
preparations made for what's sure to be one of the top 15 parties of
the half hour, when we unleash our first composite major version on
the world.


-- 
Matthew Fuller     (MF4839)   |  [email protected]
Systems/Network Administrator |  http://www.over-yonder.net/~fullermd/
           On the Internet, nobody can hear you scream.

Reply via email to