Re: [9fans] off topic: troff book

["Boyd Roberts" <boyd@planete.net>]

From: <saroj@bear.com>
> 
> No, that is not true. Knuth felt that troff was burdened with
> couple of generations' legacy at Bell Labs, and he wanted to
> start with a clean slate without fear of hurting any Bell Labs
> colleague's feeling (about troff).

nonsense, research does not work like that.  if you're afraid
of 'hurting' people's feelings you'll get no research done
at all.  do you understand the scientific method?

do you think einstein would have held back not to 'hurt'
newton's feelings?  are you mad?  admittedly, newton
was dead, but it _could_ hurt his standing, retrospectively.

> And finally, he demonstrated that even large programs can be
> explained in full detail (unlike most programs written by early
> Bell Labs guys), though Pascal is not a great language on which
> WEB (the implementation language of TeX) is based.

LP is nonsense.  you write the code.  the code is the 'truth'.
you add comments to explain the global view and stuff that's
not 'obvious'.  not 'obvious' is a judgement call.  you get
'judgement' after you've spent a few years in the caper.

unfortunately, that last sentence is not always true.

then again, no documentation is better than bad/incorrect
documentation.

i fought tooth and nail with my old man about what was
the point of 'pure' research.  i thought it was pointless.
finally, i realised he was _right_.

he did surface physics.  kinda useful now, given we're at
the limit of how many transistors you can fit on a given
chunk of silicon.

do you think he thought about 'hurting' the feelings of the
guys who built the electron gun (at york) that 'no-one could
fix' when the problem landed on his desk and he fixed it?

'doing a boyd' maybe, but somebody's got to do it.