From: "John D'Errico" <>
Newsgroups: comp.soft-sys.matlab
Subject: Re: About posts
Date: Mon, 15 Feb 2010 14:22:03 +0000 (UTC)
Organization: John D'Errico (1-3LEW5R)
Lines: 28
Message-ID: <hlblab$fhe$>
References: <hlbafe$epl$> <hlbjpj$869$>
Reply-To: "John D'Errico" <>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 8bit
X-Trace: 1266243723 15918 (15 Feb 2010 14:22:03 GMT)
NNTP-Posting-Date: Mon, 15 Feb 2010 14:22:03 +0000 (UTC)
X-Newsreader: MATLAB Central Newsreader 869215
Xref: comp.soft-sys.matlab:608128

"Jan Simon" <> wrote in message <hlbjpj$869$>...
> Dear Sadik!
> Even a "Solved" flag would not satisfy all needs. Sometimes the OP thought the problem is solved, but some experts mention important limitations of the presented solution, e.g. if inputs are empty, NaN, Inf or for repeated elements.
> Kind regards, Jan

Exactly. A response to a post does not mean that it
is done with. A response might be completely wrong,
something that often happens on the internet. I saw
a basic rootfinding question answered by multiple
people the other day on StackOverFlow, all of whom
failed to point out that fzero was the answer. Some
suggested writing a code in Matlab to do bisection,
a couple of suggestions for Newton's method, two
people suggested Fsolve. Nobody bothered to
mention that fzero is the tool in matlab for this

The OP might respond to their own post and say
they thought they had the answer. Many times this
is still wrong.

Even sometimes when a response is perfectly correct,
it is appropriate to expand on a response to add some
valuable fact.