View Single Post
  #20   Report Post  
Old 15-10-2005, 01:48 AM
Ted Byers
 
Posts: n/a
Default orchid database?


"?" wrote in message
rg...
On Fri, 14 Oct 2005 15:22:20 -0400 in
Ted Byers wrote:

[snip]
I'm familiar with mysql (In fact it has become a somewhat important
part of my current job). I tend to keep away from Java. With the
appropriate libraries seems to be a decent, if slow, tool for providing
user interfaces. It seems to be wholly inadequate for data processing.
I'm old school C/sed/awk/perl/DB2 SQL including implementing state machine
parsers in perl and awk.

I understand completely. My first programs were written two and a half
decades ago in fortran on punch cards. Remember those?

My first preference for programming is C++, but I find java today provides
decent performance. Sun has come a long way with it, even though their
support for generic programming is severely crippled. It used to be as slow
as molasses in January, but now it s reasonable. It isn't nearly as fast as
a binary produced by a good C++ compiler, but it is adequate for the problem
domains I use it for.

The first attempt, which was successful, was a problem tracking
application done in REXX with one of the GUI toolkits for REXX
in the early 90s (DB2 was the backend for the data).
The second attempt was a connect four game in Visual Age Small Talk.
I'm afraid that my viewpoint on object oriented programming was
indelibly tainted by someone more focused on the method than
the result.
It doesn't help that I'm one of those folks that doesn't see an issue
with inflicting an obscure positional grammar onto end users :-).


In this case, if we're talking professional software development, we'd be at
odds. I have never had a software project fail and a large reason for this
is that I take the user's needs and desires into account, producing an
interface that is intuitive and very easy to use, even if the user's skill
level is that he can only turn the computer on and use a word processor as a
glorified typewriter.


The fact is that spreadsheets are modelling tools, wholly inappropriate
for
trying to maintain a database. But it is tempting to abuse them in this
way
because it is so easy to use them to manage data. Using a spreadsheet to
manage data, though, is rather like using a hammer to drive a screw. You
can do it, but doing so is usually harder, and always less efficient,
than
using the right tool, and it will eventually lead to significant
problems.


In this case a spread sheet is a substitute for graph paper that is
less likely to be water damaged :-).
(Except it still gets water damaged because I work from printouts.)

I have seen whole computers fried because somebody spilled water on a
spreadsheet. ;-) Computers and water don't mix with good results.

My M.Sc. supervisor never used a text editor. Instead, he would write out
programs on paper, and I would have to enter them, and then debug them, for
him. And this was not that long ago.

I'd wager that Open Office doesn't have something like MS Access largely
because there are several open source DB products including, but not
limited
to, MySQL and postgres.


I have been corrected. Open Office does have a database called base. So
that means there is even less reason to use a spreadsheet. ;-)


When I say "Access" I really mean something like QMF and that
horrid GUI forms tool for DB2/2 from a decade past.
From what I've seen of Access is that it provides a more up to date
implementation of such functionality. However I'm not entirely
certain as it's almost always interfacing to the most braindead
database schema I've ever seen :-).

MS Access is a very good product, and quite appropriate in many contexts.
In some cases, I wowuld, and in fact have, used it. However, for most of
what I find myself doing lately, I have to use something more pwerful, such
as MySQL.

Cheers,

Ted


--
R.E. (Ted) Byers, Ph.D., Ed.D.
R & D Decision Support Solutions
http://www.randddecisionsupportsolutions.com/
Healthy Living Through Informed Decision Making