Re: Suggestion for file type detection approach

[Donald Henson <wepin wepin com>]

On Fri, 2003-12-26 at 09:38, Dalibor Petricevic wrote:

> O.K. a little OT:
> WHO do you mean when you say "users targetted by GNOME"?!
> What about users already using GNOME?
> Less and less I like the whole idea of content based file recognizing; 
> don't have to say anything that hasent been proposed already, but sure 
> as hell I'm thinking on this one. Some interesting things have been 
> heard in this thread(s).
> 
> Cheers,
> -- 
> Dalibor Petricevic
> Iskon Internet d.d.

I'm a user. I want a system that will allow me to get my work done
without getting in the way. Personally, I couldn't care less if file
types are determined by file extension, content, or the phase of the
moon as long as it works reliably. I could even accept an occasional
mis-identification if there is an easy means to correct it. However, a
mis-identification that prevents me from continuing with my work is
unacceptable.

Earlier, someone mentioned a file that was mis-identified as an HTML
file. Someone else suggested that you should just add emacs or vi to the
file association list and press on. Well, I'm here to tell you that most
ordinary users, people like me who just want to get some work done, have
never even heard of emacs or vi and, even if they've heard of it don't
have a clue as to how to use it. They *might* know how to set up a file
association but probably not. What might seem like the ultimate simple
answer to someone who has worked with Unix for 25 years may seem like
the ulimate black magic to a newbie user.

I don't know if there are any developers on this list but, if you're
here, please keep the newbie user in mind when making design decisions.
Just my $0.02.

Don Henson