Re: Tcl/Tk - Observations from a newbie.

From: Victor Wagner (vitus_at_45.free.net)
Date: 08/22/04 

Date: Sun, 22 Aug 2004 18:57:05 +0000 (UTC)

USCode <uscode@dontspam.me> wrote:
>> Why would you assume such a silly idea? Tcl/Tk is mature enough both as
>> language and as community. I think that most valued goal is not to scare
>> away old users, rather than attract new ones.
>>
> I'm not clear on what part you consider "silly" ? Trying to attract new
> users to Tcl/Tk?

Trying to set up development priorities looking to new users.
Development priorities should be set to achieve technological
superiority. If there would be techological superiority, new users would
came anyway and learn any language we, old users, are used to.

>> Fortunately, Tcl community is not commercial software vendor which would
>> go bancrupt without stream of revenues from new users and doesn't care
>> much about old ones, which are no better than new ones from the point of
>> view of marketing department.
>>
> Why can't we attract new users AND keep old users at the same time?

Because ways to attract new users, you are suggesting are marketing
crap. Old users would be scared away with it.

>> As far as I can understand, new features appear in Tcl/Tk when someone
>> already experienced in Tcl programming need them to solve new problem or
>> deploy Tcl in new area.
>>
> I think the TCT is open to suggestions and new features can also appear in
> Tcl/Tk by suggestions to them from the community.

You just think. I have some experience. If you want your suggestion to
be accepted, you have to write a patch. Moreover, you have to follow
coding style, and make sure that no other people's code breaks.

Otherwise you can seat there forever and wait for someone to implement
your suggestion.

Of course it can happen that some of TCT would kick himself at forehead
reading your suggestion, and cry "How I couldn't think of this before".
This is only way how just suggestion can make into next release - at
least one of active project members would find your suggestion
excellent, and make it into runnable code.

>>
>> I believe that Windows is not a most important platform for Tcl/Tk.
>> It should be there, because people want windows ports of apps, and it is
>> pain to develop them in such developer-hostile environment as Windows.
>>
> Never said it was the most important platform. But IMHO there is a large

You suggested to look into some silly DLLs for implementation of some
silly controls. Since source code of these DLLs is not published,
only way to use them is use system they are part of.

BTW, I'm programming GUI for fifteen years and I sincerely don't
understand what is so cool in toolbars.
First of all, what is so special in the toolbar? Why have it be a
separate widget? What the difference
between just frame with row of iconic buttons and thing which is called
toolbar?

And every time I see a toolbar in somebody else's application, it looks
very cryptic. It is much harder to associate these 16x16 icons with
their function than to remember appropriate number of keyboard
shortcuts.

> pool of Windows developers that would be interested in Tcl/Tk with a few

Sincerely, I'd rather sentence all those developers to not to touch
anything containing microprocessor for the rest of life.

They do not understand basic principle of programmable equipment - USER
SHOULD NEVER FORCED TO DO ANYTHING BY HAND. If program allows to do
something using keyboard or mouse clicks, it should always allow to do
same thing programmatically, so if user finds out that he have to repeat
same sequence of action more than twice, he can turn it into macro.

Second basic principle which is often violated by Windows application
(and things like Gnome/KDE too) - never try to attract user attention
more than problem deserves. User has other important things to do than
repeatedly click only button in your system-modal dialogs. 

-- 
Actually, typing random strings in the Finder does the equivalent of
filename completion.
	-- Discussion on file completion vs. the Mac Finder
Quantcast