Re: Any PB & ADO gurus?
don cole wrote:
>
> ken mortenson wrote:
> > Another thing (again not the fault of Euphoria) is I wrote a windows program
> > right after my skipper submission using win32lib. When I updated to a newer
> > version of win32lib my program didn't work anymore. So I won't use win32lib
> > because I don't trust it to not break my program in the future.
>
> This could easly be fixed with
>
> include win32lib_ver3.2.exw
>
> include win32lib_ver4.exw
>
> etc...
Your making me shudder Don! That is a solution, but is it the right one?
Remember, this is Don Quixote your talking to. Making something work is
not enough. As a programmer I feel great responsibilities. Programmers
seem to know, to a greater or lesser extent, that they have a responsibility
to produce a product that satisfies the customer. Yes, they all know they
have a responsibility to make things work. But they have another
responsibility that usually doesn't get the attention it deserves. They
have a responsibility to be humble.
Your solution, in my opinion for what it's worth (IMOFWIW), deals with the
first two but fails to address the third. Let me explain why if I may.
I strongly believe in the concept of the humble programmer. I often talk
about ugly code and the two are related. Ugly code isn't just funky syntax.
It isn't just avoiding spaghetti code. Ugly code is fragile. Everything
breaks, but solid code breaks in ways that are predictable and managable.
Your solution certainly might be managable but increases the workload of
any (you can't always assume this to be yourself) that follow you. Your
solution is a sledgehammer which is fine, but for me it's not the right one.
I hope I haven't offended and I know I haven't explained myself well.
BTW, my attraction to Euphoria is partly because I see it embodies much
of the wisdom I feel my experience has given me. I also see a lost
opportunity cost. There is a reason VB is third in popularity. There is
a reason it's not number one. There is a reason Euphoria is 47th. Euphoria,
could go in either direction. With the right changes, I see that it could
go to number one and stay there with no challengers. The wrong changes will
leave it languishing below the predominate languages. I see a lot of people
advocating the wrong changes.
Nothing gives you more raw power than machine language (best written
with a macro assember of some kind.) But it's not popular. Why? No, it's
not because ML is hard. It's really not any harder than Euphoria. It's
because it's difficult to be expressive in ML. Euphoria's strength is
that it is expressive. Algorithms are easier to write than in other
languages. But it has some of the shortcomings of ML too. This is best
expresses by the INCLUDE keyword. Include makes your program one long
linear list (ignoring loops and branches for the moment) which is really
all it ever is to your machine. People are not machines. Machines do
some things better than people and people do some things better than
machines. We need to play to eaches strength.
People naturally conceptualize in a modular way rather than linear. We
divide and conquer. This is our strength.
...and that's a lot of hot air. Don, you do push my buttons! That's
probably a good thing. My best response will be by example. So I need
to get busy on my project which I hope will show you some of what I've
talked about here. I'm not doing the fun stuff yet because of some of
the show stoppers I working on, but once I'm past that I should have
code to upload within a short time after.
Again, thanks for your help. Don't let my bloviating fool you into
thinking I don't greatly appreciate your responding because I certainly do.
|
Not Categorized, Please Help
|
|
