gnu-arch-users
[Top][All Lists]
Advanced

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

[Gnu-arch-users] post mortems


From: Thomas Lord
Subject: [Gnu-arch-users] post mortems
Date: Tue, 06 Sep 2005 11:08:49 -0700

Anonymous wrote off-list:


> I'm a daily user of tla since about 1.5 year. Thanks a lot for it.

I'm glad.  I like it a lot too.   It has flaws:

1. It is indeed too slow on big trees.
2. It is a pain to have to set up a revision library.
3. The namespace is indeed too inflexible -- this one doesn't
   bother me personally (I think it's a feature for most purposes) 
   but there are reasonable use cases for which it is a needless
   obstacle.
4. Intra-file delta-compressed storage is becoming an anachronism.
   In other words, the fact that `mkpatch' is invoked for every commit
   is increasingly inappropriate, economically speaking.
5. The in-tree patch-log mechanism is a bad fit for some kinds
   of project.

The `revc' codebase (arch 2.0) lays down a foundation to fix all of 
those problems and in a very simple way while simultaneous fixing
other minor annoyances (such as the need for better "librification").
It's incomplete in the sense that the merging features we know and
love from Arch haven't been made available as a layer over its existing
capabilities.  I doubt it will ever be completed although I'll 
excerpt and forward some of this to the list just to stir things up
a bit.

Unfortunately, I won't be able to take revc any further.   There's
no money in it.  The baz fork ignored an opportunity to cooperate
on review and long-term planning (understandable in retrospect -- 
they regarded the fork as a short-term thing only, to be replaced
by baz-ng or whatever).   Having observed how the baz-vs-tla crap
played out, and how this resembles so many other intrusions on the 
volunteer community by corporate projects, and how I am unable to
*pay* volunteers for their best contributions, and how -- to even
come close to competing against baz -- I have no time to work with 
volunteers whose skills are still developing.... given considerations
like that I have come to have severe ethical concerns about engaging
with the volunteer community at all: it sets up vulnerable,
well-intentioned young engineers for exploitation and dis-education.

(It was disagreement about those ethical concerns around labor issues
and the proper handling of volunteers over which I separated myself from
the GNU project and the Free Software Foundation.)


> [hey, why don't you put up a page with paypal links?]


I'm disinclined to set up a web page and solicit donations in general.
That may seem odd and perhaps it is.   I'm trying to steer a moral
course through troubled waters and to focus on finding a lasting, 
sustainable, just outcome.   Back when I was officially busking (I used
to have a page with paypal links for arch) many individuals gave.  Often
enough it was students and others of limited means.   I think I
represented to them a symbol of how a free software hacker could
get by in the world.   I represented how projects can succeed.
But it was not a realistic image that I projected.   The biggest
structural problems that lead to my situation need to be solved
by people of greater means and greater influence in the industry.


-t








reply via email to

[Prev in Thread] Current Thread [Next in Thread]