[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Geoff Kuenning <address@hidden>] Re: undeniable bug in /usr/bin/spell

From: Dan Jacobson
Subject: [Geoff Kuenning <address@hidden>] Re: undeniable bug in /usr/bin/spell
Date: 22 Nov 2001 09:59:55 +0800
User-agent: Gnus/5.0808 (Gnus v5.8.8) Emacs/20.7

Geoff sent me a message,

--- Begin Message --- Subject: Re: undeniable bug in /usr/bin/spell
The problem is that you have been misled by false statements made by
FSF and whomever made your RPM.

/usr/bin/spell is not part of the ispell package.  As such, I can take
no responsibility for incorrectness in the script you have, nor for
the lack of documentation.

The bug you found is indeed undeniable.  Your two-liner with 'sort -u"
is the correct fix, although technically it's still wrong because it
doesn't accept the "-b" switch of spell, but it at least does
approximately the right thing.

> $ which spell|xargs rpm -qif
> Group       : Text tools                    Source RPM: 
> ispell-3.1.20-13mdk.src.rpm
> Ispell is the GNU interactive spelling checker.

The statement in the last line is false.  Given the history of FSF
with respect to ispell, I have trouble calling it anything other than
a lie.  Ispell is not "the GNU interactive spelling checker".  I
permit FSF to distribute it, despite their insistence on being
difficult and uncooperative.  But it is completely false for them to
imply, as does the above statement, that the program has any
connection to GNU or to FSF.
    Geoff Kuenning   address@hidden   http://www.cs.hmc.edu/~geoff/

--- End Message ---

Bonus conclusion from Dan: "best to have correct frontend and man page
for spell distributed along with the ispell package, that way we
wouldn't worry about each distributor doing a slob job, with bugs
reports still going to GNU."

Yes, indeed, a one liner /usr/bin/spell frontend and a tiny spell
manpage would mean that for 5 minutes of effort, GNU/Linux would have
"spell" included in its vast armada of utilities.

I could CC: this to mandrakesoft, but if Geoff adds the script "at the
source", then there would be no need to fix it in every GNU/Linux
distributon... plus repoting a bug to mandrakesoft takes more than a
CC, so never mind that.
http://www.geocities.com/jidanni/ Tel+886-4-25854780

reply via email to

[Prev in Thread] Current Thread [Next in Thread]