Re: [Chicken-users] other leading Schemes

[Brandon J. Van Every]

Shawn Rutledge wrote:
> The end-users didn't know or care what language it was written in - it
> was just a Windows .exe file that happened to put up a funky-looking
> UI.  :-)
>
> If there were a nice portable (Windows, X/GTK or QT, MacOS) widget
> library that isn't a hassle to use, I'd use Scheme in my present job
> too, for a couple of simple internal-use-only applications that need
> to be written one of these days.  I would want the same result - be
> able to make a standalone .exe on Windows.  Using native widgets on
> each OS would be a plus, too.

T'would be nice.  There's this basic problem of infrastructure though.  
Not enough people have gotten themselves into a position where they can 
choose any language they want.  Even fewer are going to take the time to 
create cross-platform GUI toolkits, once having gotten to that place.  
It's easy to see why languages such as Scheme get stuck in the "binding 
to other languages' libraries" problem.  Too much work to do everything 
all over again in an unpopular language.

On the bright side, binding to some extant C GUI is within the realm of 
the doable, for those interested.

On the negative side, it almost never ends up being of interest to me 
personally, because the GUIs always have bad OpenGL support.  Games also 
don't need the elaborate GUIs that other apps do.  Periodically I find 
myself looking around at widget libraries.  Invariably I find bad or 
nonexistent OpenGL support.  I am also reminded how boring widgets 
actually are. 

If an inventory of community skills caused 3..5 "GUI guys" to get 
together and do something, then it would be a win.  I'm realizing a 
further problem with the inventory idea though.  It's not enough to get 
people with similar skills and interests in contact with each other.  
There also has to be leadership.  No leadership, no project actually 
happens.

I don't know what can be said about the problem of leadership.  I can 
lead a build engineering effort.  I can lead a 3D engine effort.  I can 
lead a community skills inventorying effort.  But the number of things I 
can personally lead is finite.  This is true of anyone.  Leadership is a 
scarce resource.

It's fair to say that open source is driven by leadership.  Someone 
decides there's something they want done.  The only successful projects, 
are the ones where 1 key person toughs it out initially to make it 
happen.  Someone who is bound and determined.  Then it's possible for 
others to amplify their efforts.  But first it takes that 1 person.

Also, some people are great at producing code, but they're bad leaders.  
Like, they don't treat people well.  You'd think it would be hard for 
that sort of leader to survive in open source land, but I have seen it 
happen.  I don't know why people put up with certain people.  I figure 
either their code output is extremely valuable to others, or the project 
has "selected for" exceedingly easygoing followers.  Myself, there are 
certain things I just won't put up with, that I don't feel anyone should 
be putting up with.  It's no accident that I've stuck around in both the 
Chicken and CMake communities.  They both have easygoing leadership.

I'm going to start a new thread on this subject of leadership.  I think 
this discussion is drifting from the Subject line, but I suppose it does 
have the word "leading" in the title.  :-)  Er, maybe we see where this 
leads?  Ok I'll stop now.


Cheers,
Brandon Van Every