RE: New Technologies -- MIDRANGE-L

> -----Original Message-----
> From: Joe Pluta [mailto:joepluta@PlutaBrothers.com]
>
> > From: Metz, Zak
> >
> > All in all, it's a
> > learning experience, and I'm nerd enough to think it's fun. I'm
> > lucky enough to have time to play--at nobody's expense.
>
> More people should do this.  People who actually do this
> stuff on their own
> time are more committed and so have a better chance of making good
> decisions.  Of course, you don't get paid for it, so you have
> to WANT to
> learn it.

As someone who skipped the college experience, I would have to agree. You have 
to want to learn. Wanting to learn is what led me to starting the website and 
myriad other projects. I have very few delusions about ever making money with 
them, I got the day job for that. Wanting to learn is how I made the best use 
of my time at IBM, earning seven internal certifications along the way. It's 
that child-like quest for knowledge that keeps me going in life. I actually 
have very poor mental health, so I simply have to have projects like this to 
keep my busy.


> > Knowing my database layout intimately, creating the SQL statement
> > was a very quick process, no more than five minutes to create a
> > working version, albeit a slow one. Optimizing the query as
> > documented in my post took around a half hour, including writing
> > the post. Total time of about 35 minutes or so.
>
> I didn't get that feel, Zak, I thought it was hours or even
> days of work.
> Now, don't take this badly, but the fact that a
> self-acknowledged SQL expert
> still had to dig through joblogs to do this is a little
> problematic in my
> mind, but I guess in a way it's not so different from program
> dumps or trace
> logs.

The blow-by-blow in my initial post was indeed performed in about a half-hour, 
entirely while I was writing the post. Designing a database that made it that 
simple took many days and iterations.

I don't know that saying I had to "dig through joblogs" is any more accurate 
than saying you have to "dig through compile listings" to debug RPG. One 
joblog, one compile listing. It ain't that bad, just a different place to look 
with different kinds of messages. Also, for the record, I could have optimized 
this query without the debug info. I went through the exercise for the benefit 
of the original poster.


> > The SQL performance issues I've seen
> > are generally related to lack of skill, not system deficiencies.
>
> Of course, I'm still reluctant to agree that it's that easy.
> In fact, the
> only reason some junior programmers can learn SQL debugging
> at all is that
> they learn it in college.  But then again, that cold harsh fact is a
> reality, and one that needs to be considered in any busniess decision.

It's easy for me. I'm not saying it would be easy for everyone. Dealing with 
datasets is really a different paradigm and requires some adjustments in 
thinking, as anything new does. The IBM books have all the info you need. No 
need for college. But I agree that if your coworkers don't have the skill set, 
it's a bad decision to jump into SQL. I don't use it at the day job.


> > I know I can
> > write any and all of that faster in SQL than RPG. I know that
> > from experience. YMMV.

> And this is indeed where SQL shines.  Queries.  SQL is very,
> very flexible
> at querying data.

Agreed. I don't try to use it where RPG makes more sense. The right tool for 
the job...


> As you point out, there are still trade-offs.  I just wanted
> to point out
> that you need to make those decisions with your business
> goals in mind.
> Just doing it "because it's there" is a pretty weak excuse,
> unless you can
> honestly justify the learning curve argument.  Which, in your
> case, Zak, you
> obviously can.
>
> Joe

Agreed, and at the day job, I certainly don't have the time to kill to go off 
on a tangent exploring something that might not work in the end, nor would I 
recommend that everyone run out and start writing in SQL. As another poster 
stated, it's another tool in the toolbox and isn't right for everything, and 
depending on your business, may not be useful at all. But, it might just be the 
right tool, so it's worth knowing a little about. And who knows when I'll have 
a job where these skills pay off in a big way. It's happened to me before.

Z