Re: Survey for all GNU Midnight Commander users

> Actually, it seems to me he asked if it was used.  What purpose he's
> collecting the data for I am not sure.  Perhaps it's good to take a step
> back, and consider what's best for the whole userbase.  I agree there is
> room for different opinions.  Just wanted you to understand mine.  :-)

The reason is to find out what is used.  When I get bugreports about bugs
that are many years old, I start wondering if anybody has used that
feature before.

I don't want to spend my time fixing features that nobody needs.  I don't
want other developers to spend time on that.  I don't want the manual for
mcserv to be translated into Italian, if not a single user uses it, not
just in Italy, but in the whole world.

I don't want potential contributors to be scared by the overly complex
sources.  But I do want them to test their changes.  Less unused features
means more testing for the code that is actually used.

If somebody contributes code that e.g. break the directory tree, I 
want to be able to say "No, you cannot do that.  There are at least 3 
users who use directory tree".

The Windows issue is a bit trickier, because I cannot ask somebody to 
install Windows or cross-compiling environment to test his or her changes.

Other projects that have Windows ports (Mozilla, Apache) usually rely on
contributors from the Windows development community.  I don't see such
community here, so the last question may seem a bit "rhetoric".  If there
is nobody to maintain the port, the users won't get it.

However, if somebody actually uses the native Windows version, then there
is working code outside our tree, and there are potential contributors.
Then maybe it makes sense to take this code, as long as it doesn't break
anything else.

So far, mcserv and the directory tree got no votes.  The Windows port and
the background operations got 2 votes each.  Its interesting because the
background code crashes for me (try overwriting a file in the background),
but I didn't get a single bugreport about it.

Pavel Roskin

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