I have been searching for that post for quite some time and I have been
unable to dredge it up. I don't remember when the post was made, or
who the poster was (though I think it was Gorbag), nor do I remember
anything about the actually article.
I wonder if you guys could help me out? Does any of this ring a bell?
Any sort of link to this paper would be greatly appreciated.
(Please don't email me -- would you mind posting instead? I have some
difficulty retrieving mail with this email address. Thank you).
rsobr...@eudoramail.com wrote:
> I remember reading a post in comp.lang.lisp that referenced a software
> engineering paper. The gist of the paper was that managing a software
> project was an NP-complete problem: the cost of completely specifying a
> project is the same as completely developing it.
>
> I have been searching for that post for quite some time and I have been
> unable to dredge it up.
Possibly because...
> I don't remember when the post was made, or
> who the poster was (though I think it was Gorbag), nor do I remember
> anything about the actually article.
Now that is what I call a complete specification!
OK, possibly your condition arises from a rather natural process in
which perception actually involves quite a bit of interpretation, the
best example being people paraphrasing text as they attempt to read it
aloud verbatim.
I offer this because your first paragraph reminded me (via reverse
pharaphrasing) of:
http://www.developerdotstar.com/mag/articles/reeves_design_main.html
The cost of coding and design being the same is merely implicit in them
being the same, so I could see you taking away that synopsis. And I did
mention the Reeves papers here in the past year.
hth, kt
--
Cells: http://common-lisp.net/project/cells/
"I'll say I'm losing my grip, and it feels terrific."
-- Smiling husband to scowling wife, New Yorker cartoon
> I remember reading a post in comp.lang.lisp that referenced a software
> engineering paper. The gist of the paper was that managing a software
> project was an NP-complete problem: the cost of completely specifying a
> project is the same as completely developing it.
[...]
> I wonder if you guys could help me out? Does any of this ring a bell?
It does, but I don't know whether it's the right bell:
Large Limits to Software Estimation
http://www.idiom.com/~zilla/Work/kcsest.pdf
Mathematical Limits to Software Estimation - Supplementary Material
http://www.idiom.com/~zilla/Work/Softestim/softestim.html
Paolo
--
Why Lisp? http://wiki.alu.org/RtL%20Highlight%20Film
The Common Lisp Directory: http://www.cl-user.net
Thank you so very much! These are the ones!
(Thanks to Madhu for replying and KT for the Reeves paper).