This is needed because Fare’s backward-incompatible changes to ASDF keep breaking projects.https://twitter.com/ngnghm/status/942269688242954240 …
-
-
-
Replying to @tapoueh
I suspect any urge for that is cured by reading the sources.
2 replies 0 retweets 2 likes -
Replying to @xach
Reimplement from scratch the 80% of it (maybe less?) that are actually useful?
1 reply 0 retweets 0 likes -
All parts of ASDF are useful. Every bit of complexity is there for a good reason. And every "extra" feature is used by at least one person that's not me. (Except for the finalizers, now that QRes is dead.) The ASDF page links to two known failed "forks", mudballs and asdf-lite.
3 replies 0 retweets 2 likes -
Maybe it’s like autoconf and other build tools, the problem to solve is so complex that no simple solution is possible to design. I have a hard time believing it, and also don’t want to spend time on it... Good luck. Please fix it someway.
1 reply 0 retweets 0 likes -
Indeed ASDF plays for CL a bit of the roles that autoconf, pkgconfig, make, ld.so and libc play for C. With significantly less complexity than the equivalent parts of C tools, despite supporting 16 implementations on a variety of OSes. Blame me for any mission creep.
1 reply 0 retweets 2 likes -
I don’t know the goals, can’t comment on mission creep. Thanks for ASDF, I’m happy I didn’t have to solve those problems. Breakages are frustrating though.
1 reply 0 retweets 4 likes
In case anyone cares, I explained the mission creep of ASDF from @telent_net's 2002 creation to my 2014 version in my ASDF 3 article: https://github.com/fare/asdf3-2013 (beware: long version is 26 pages).
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.
Read my blog!