[antlr-interest] are ides a good thing? (was: Seriousdoubtsonusage of incrementalparsinginides)

Scott Stanchfield scott at javadude.com
Wed Apr 27 03:58:22 PDT 2005


> I work for an end user. There's only about 4 coders all told, 
> and we tend to work somewhat individually.
> 
> And I find I spend most of my time talking to the users, 
> including quite often asking the question "are you sure what 
> you're asking for is sensible" (a lot of our work is statistics).

And what do your coders do? Sounds like you're not one of the main coders,
but more of a requirements person. NOTE: THIS IS A GOOD THING, but you're in
no position to make statements about "all coders"

> John Mitchell said I was unusual, in that I don't tend to 
> "see" things visually (he didn't put it that way).

I think John's point was that different people use different parts of their
brains in different ways. Some are more visual than others, and that says
nothing about the quality of their work.

> where the author said he noticed that much of Unix history is 
> very verbal. And it wouldn't surprise me in the slightest if 
> most of the top Linux hackers were verbal people...

"Hackers"... Enough said.

> (Oh - and yes I am a (very) abstract person. Didn't I say 
> that educational academics were surprised "recently" to 
> discover that maybe 50% of people never make the cognitive 
> leap from concrete to abstract thought? :-)

Do you have any sources for any studies you keep quoting?

I would tend to agree on that one, but you really need to back statements
like that up.

I could claim that a recent study showed that "90% of programmers are more
productive with IDEs". Of course that "study" would only be my personal
experience at the past couple of jobs I've been at, converting people to
using eclipse from vi/emacs, and with nothing but their remarks as defining
"more productive". Without knowing the source of such statements, they're
pretty useless.

Later,
-- Scott




More information about the antlr-interest mailing list