How good are people at making decisions? http://danluu.com/bad-decisions/
-
Show this thread
-
A side effect of the lack of data on practices is that charismatic people with compelling stories have outsized influence. How many folks learned Lisp because of this story? The two most valuable startups of that era (GOOG&AMZN) were Blub-based, but this story still has currencypic.twitter.com/mxZJd5kiu5
6 replies 21 retweets 86 likesShow this thread -
Replying to @danluu
I disagree on GOOG. Yes, the languages were blub, b/c their killer idiom was at another level: dist-sys. Protobufs,stubby,mapreduce,gfs
2 replies 1 retweet 6 likes -
Replying to @dbentley
That's the point! PG focused on languages and claimed Lisp would give companies a technical moat, but the companies that actually built technical moats did so on blub. The next startup to build great infra was arguably FB, despite using the blubbiest blub that ever blubbed.
6 replies 1 retweet 30 likes -
I think PGs argument was for "hackers" as founders (i disagree, but w/e). Goog/FB both founded by hackers; PG misjudged what langs hackers would hack in
1 reply 0 retweets 0 likes -
Replying to @paulbiggar @dbentley
One could've written a signaling argument about hackers and Lisp, but that's not what PG did. He went on at length about how Lisp lets you ship features so quickly your competitors can't keep up, about how it lets you do things you can't easily do in other languages, etc.pic.twitter.com/iRvvUDn8Pv
1 reply 0 retweets 3 likes -
Should have said "the lesson I took from PGs article". Ironically, I don't think Lisp has the advantages he claims anymore. The rise of SPA and horizontal scaling and larger teams has not been kind to it
1 reply 0 retweets 1 like -
Replying to @paulbiggar @dbentley
Fair enough. I think we've also seen that, contra multiple PG essays, popularity is a kind of strength since it leads to a rich ecosystem, which lets people concentrate on solving business problems instead of re-inventing technology.
2 replies 0 retweets 3 likes
The underdog technology that's a secret weapon makes for a better story than the boring technology with nice libraries and good examples on stackoverflow, but the latter appears to be a significant advantage even if it's harder to tell a viral story about it.
-
-
Underdog secret weapon is real, imo. Just not always valid, and hard to tell when it is and isn't. Our current gains from OCaml / Elm are pretty insane
1 reply 1 retweet 5 likes -
Replying to @paulbiggar @dbentley
I think you get large gains that come with a large cost, esp. for Elm but also for Ocaml. One of the core concepts in Elm from when I last used it (signals) was removed from the language and it's still pre 1.0 and at risk of significant conceptual changes.
1 reply 0 retweets 0 likes - 7 more replies
New conversation -
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.