** Reply to message from "Todd Walton" <[EMAIL PROTECTED]> on Fri, 11 Apr 2008 10:12:29 -0500
Java's reflection mechanism is one way to find out what is and isn't available and I think CORBA's naming services was/is a way to find objects you may need but don't have locally. So the tools are there but not often used. I've not delved into reflection much but my brother has and he loves it and uses it often. There used to be alot of stuff going on with object orientation and distributed objects in the mid 90's. But, there's a company who found that OOP was a threat to their product since OOP hides the underlying platform and makes it much easier to make cross-platform applications. So they came up with their own non-standard C++ language and APIs which were unOOP( object-like ) and tied closely to their platform APIs and they dumped( really cheap ) these on the market. The result was that only in the high end is where these things still exist and are used. Doug > On Thu, Apr 10, 2008 at 3:23 PM, James G. Sack (jim) <[EMAIL PROTECTED]> > wrote: > > The statement Humans are way easier to 'program' than computers. > > is interesting to interpret. > > I had an "Aha!" moment re this. A significant part of my work is > technical documentation. In a sense, I program humans. When I > document a procedure for doing something I have to consider what are > the essential steps, condense them for efficiency if possible, and > present them in a straightforward and unambiguous way that someone I > don't know can follow. > > There are things I think I can take for granted, and things that I > explain just for completeness' sake or "just in case". In other > words, I have to manage a set of primitives. I can say "change the > password in PWM" and just assume that the target audience will know > how to do that. If they won't know how I can either explain it in the > procedure or I can put that in another document or section, sort of > like calling a procedure in a library. > > Thinking about it, if a person doesn't know how to do a thing that > I've assumed they do, what are they going to do? They're going to go > looking for the procedure. They're going to Google it, pull up a man > page, ask somebody, something like that. Maybe that's something > missing from programming today: the ability to look up unknowns. What > if you didn't have to explicitly import libraries? Most languages now > let you call a function defined elsewhere in the code and you don't > have to tell the program what line it's on. Why not the same with > libraries? And what if you could say "do this" and if the computer > doesn't have a library for it it goes out and asks other computers > around it? Package management systems kind of do this now. In my > Kubuntu if I say 'kate', for example, it'll say "You don't have kate, > you have to install it using apt-get install kate". > > I think if we're going to talk about programming humans we should > concentrate on technical documentation, and forget about the > simplistic "go pick up that ball" type statements. > > Another issue is the level of statements you can give a computer > versus a human. In this regard, I think everyone understands the > issue, and I think progress is being made in this regard. From > machine language to assembly to C style "high level" to languages like > Java that do garbage collection without you having to tell it to and > languages like Perl that mimic human language syntaxes to a high > degree without me having to tell it what "until" means. This is a > problem under control, if not licked. > > -todd > > > -- > [email protected] > http://www.kernel-panic.org/cgi-bin/mailman/listinfo/kplug-list Doug -- [email protected] http://www.kernel-panic.org/cgi-bin/mailman/listinfo/kplug-list
