Daniel Stenberg wrote:
> And then there are some that are more advanced HTTP/HTML mirroring tools
> ...

Heh no, not that sort of thing. Here's a screenshot of the present
development version:

http://telcontar.net/store/hosted/lists/HW13.png

It looks a lot nicer than it functions; even right-click in the address
bar tends to randomly freeze it for about ten seconds. Authentication is
working but flawed (e.g. it will quite happily get stuck in a futile
loop that you have to fight to get out of).

> I guess nobody has wanted it badly enough to do it themselves, or
> perhaps the existing tools are good enough.

I still develop HW on Mac OS 9, and I realise I'd have to abandon that
to use any recent version of wxWidgets and most likely to use cURL too,
although cURL could run on Mac OS 9 if you had at it with a bunch of
power tools, sledgehammers and maybe bits stolen off WebTen's UNIX
environment :P

> If you start it and announce your work and progress, I'm sure you'll get
> users and fellow contributors if this sort of thing is what people want!

It's a circular argument if you put it that way. To get any progress to
announce I'd have to learn C again, learn C++ on top of that (for
wxWidgets), learn wxWidgets enough to build a program that does
something, and learn how to call on libraries and drive cURL.

Why would I do all of this if the idea is dead and worthless to
everyone, since I could drag myself to patch the last maddening bug in
HW (the drop-down menu keeps clobbering history items) and then retire it.

I'd rather start under the premise that trying to write working C++ is
worth it. And not just working C++, but code that people won't vomit
over. We all know that it takes time and practice to write decent code
and who'd want to work on or maintain my first ever C++ code?

> Personally I'm a command line person and I prefer using curl as it is or
> at times when the complexity requires it - from within shell or perl
> scripts.

The screenshot should indicate why I think a GUI is better for the tasks
I have in mind. I don't mean just grabbing a bunch of files at once, for
which the command line is going to be faster.
Received on 2007-01-08