Re: Emacs development

[Thien-Thi Nguyen <ttn@gnu.org>]

[-- Attachment #1: Type: text/plain, Size: 3041 bytes --]

() Eli Zaretskii <eliz@gnu.org>
() Sat, 21 Feb 2015 20:43:50 +0200

   You could help understanding how these are related, by
   posting questions that might bother newcomers, for which
   there are no ready answers in the available documentation.

You need not look further than these (contemporary) threads:

- Move verilog-mode and vhdl-mode to elpa.gnu.org
- Proposal for a closed-buffer tracker

both of which...

   > [...] in one repo or another [...]

   We only have one repository these days.

...pose questions on which administrative domain (or "repo" for
short) is best aligned w/ the respective code.  So, maybe there
is only one repository in the strict technical sense (i haven't
checked), but certainly from the contributor pov, there is some
choosing among perceived alternatives required.

IIUC, in these examples, the contributors are both old and new,
but the nature of the questions are the same: "To decide where
to land, i want to know what is the landing zone like.  What is
the landing zone like?  If i land in one place, what travel (or
travails :-D) must i undertake to visit the other place?  What
are the border-crossing fees?  What kind of visa will i hold?"

And most importantly: "How volatile are the answers to the above
questions?  Is the overall relationship worth the commitment?"

Traditionally, and for the longest time (before ELPA and before
lexical-binding), the degenerate answer to both "relationship?"
and "expectation?" was "coherence".  This concept was easy to
grasp, judge, implement and compatibly uphold because of Emacs'
"fully singular" nature: one repo / dynamic-binding.

The overall cost of commitment was mostly fixed, w/ flexible
payment schedule as a nice bonus -- you could start w/ "‘setq’
groves in ~/.emacs" and work your way towards the mysteries of
‘(require 'cl)’ at whatever pace fit your fancy.

But those days are behind us now -- no complaints from me, i
hasten to add! -- and the context of the shared fruit plays a
larger part in its recipients' experience than ever.  Emacs is
no longer a simple punchbowl; it (and GNU ELPA and * ELPA and
gnu-emacs-sources and the Ferreted Wisdom of the Internet) is a
N-dish feast.  If i place on the table TTN-HACK, will it curdle
the lips of those who enjoy CORE-FEATURE-24?  CORE-FEATURE-25?
Or how about CORE-FEATURE-25 mixed w/ ELPA-ONLY-HACK-2012?  And
so on.  Of course the most important pondering almost goes w/o
saying: How can i protect my master-chef self-image?  :-D

"But ttn, you fool, you egoistic throwback!  Coherence is always
the answer -- that hasn't changed any!  You just have to look at
it in the right way (otherwise You're Doing It Wrong)."

Well it's true: the moiré i see the moiré i get confused...  :-D

-- 
Thien-Thi Nguyen
   GPG key: 4C807502
   (if you're human and you know it)
      read my lisp: (responsep (questions 'technical)
                               (not (via 'mailing-list)))
                     => nil

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 197 bytes --]