[Haskell-cafe] Investing in languages (Was: What is yourfavouriteHaskell "aha" moment?)

[Brandon Allbery]

On Sat, Jul 14, 2018 at 11:28 AM Paul <aquagnu at gmail.com> wrote:

> Better IMHO to have less infrastructure code. Better is to hide all
> “machinery” in compiler.
>

Hiding all "machinery" in the compiler leads to perl 5, PL/1, and similar
monoliths. Which, if they do manage to catch on, eventually get discarded
because the "compiler" can't keep up with the rest of the world without
becoming a completely different language… which will move everything into
the ecosystem so it can keep up.

Monoliths have one advantage: people can ignore all the stuff going inside
the monolith, and therefore think they're easier to work with. Until they
no longer do what those people need, and they get tossed on the trash heap,
never to be seen again. The languages that stick around, that are still
used, are the ones that are extensible instead of being monoliths.

-- 
brandon s allbery kf8nh                               sine nomine associates
allbery.b at gmail.com                                  ballbery at sinenomine.net
unix, openafs, kerberos, infrastructure, xmonad        http://sinenomine.net
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.haskell.org/pipermail/haskell-cafe/attachments/20180714/d639729a/attachment.html>