2 ms·
Oh for sure. To a hacker ( or a progrMmer, or the guy who's going to have to maintain the code, possibly when it's scaled to 40,000 servers), code beauty is cr
by bigiain 16y ago
Oh for sure.
To a hacker ( or a progrMmer, or the guy who's going to have to maintain the code, possibly when it's scaled to 40,000 servers), code beauty is critical.
To "an ideas guy", a prototype is _way__ more important.
But I hope the ideas guy partners with or hires someone like you, or like Tom Christiansen - the guy who got the "write it at least twice. always throw the first version away" meme firmly stuck in my head back in comp.Lang.perl.misc in the mid 90s.
Modern startup techniques, particularly the "fail fast" mantra, strongly suggest that quick prototypes, even if they are crappily coded prototypes, are often a great idea - and are often within reach of inexperienced or non technical founders.
(and yeah, I know what you mean by throwaway code ending up as critical production systems, just this morning I got mail from a cron job trying to run a crappy perl script I wrote over 10 years ago - it's monitoring critical infrastructure for a project I did for a company I haven't been in touch with since 08, and from the grapevine I hear they got sold in 09 or 2010 and I've never even met the new owners...)
- singular 16y agoCool, I'm glad you see my point :-) - it's just that there seems to be almost a meme going round, not just on hn, that 'hey this stuff just doesn't matter because the customer doesn't care' which seems fallacious to me (see other posts for why :-) I think the term 'code beauty' is a poor choice though; that seems to me to implies something subjective, when in fact I fundamentally believe code quality dictacts product quality.