RE: Tabs in SEU ... I am SICK of it... PLEASE STOP! -- MIDRANGE-L

Sorry, but I guess that's my cue to join back in the discussion...;-D


No NEED for ANYONE not interested to read this looooooooong post, of course.


| [mailto:midrange-l-admin@midrange.com]On Behalf Of rob@dekko.com
| Sent: Thursday, October 10, 2002 3:22 PM
| Subject: RE: Tabs in SEU

| I had a gent here who wasted 45 minutes trying to debug a CL program.  He
| used SEU.  Soon as I pulled in up in Code/400 I could see the problem
| immediately.  The line he thought was doing something was actually a
| comment.  Simple things like color actually do make a huge difference.

I had a similar experience.  That was back at my first 38 job, at Liebert in
'82...  I'm not AGAINST colorizing code, but there's an old saw that says
"it's a poor workman who blames 'his' tools".

At the same time, there's a LOT to the saying that "a good craftsman is ONLY
as GOOD as 'his' tools", too...  Therefore the heated debates.


Some of the comments that struck me from the past few days' debate:

"It does take some getting used to because old habits are hard to break - we
know how to do things instinctively in SEU"

"just have to give it a bit of time"

"Frankly, if I could have gotten away with it I would have removed (SEU) ...
after we paid for Jon Paris to come down here and train."

"The code-l list is very busy and strongly supported by IBM."


These comments, coming from the STRONG PROPONENTS FOR Code..

..well, I guess I'm not seeing this as all that strong an endorsement.




I'm glad some have found sufficient spare time to get good at using Code.
My experience, primarily back when it was labelled V3R1 and V3R2..  Well, it
was quite a bit different...  The MAIN reason I was looking to migrate was
to use the verifier.  (Not that I'm in the jetset crowd that needs to be
able to do code offline, but for the speed and efficacy.)  Well I saw the
same MCHxxxx (pointer offset error) that several reported to the Code
Newsgroup...  Iirc (and I believe I do, but icbw), that messages about that
error never got a very serious response, let alone a fix.


But the documentation at that time was a joke.  I did have a couple minutes
to look again at V4R5, and the documentation had improved substantially...
But that misses the point.

I was productive on SEU in one day, and fairly proficient in two.


I know.. I know....  There are SO MANY MORE FEATURES in Code, that it's
simply not possible to make the design intuitive.  Well, if documentation is
an afterthought, as it usually is in the *nix world, well then the design is
very likely to be counter-intuitive.


Maybe I could stand to revisit Code, of course.  (Wonder if it's still
slower than SEU on program loads, for example.)  But, burned once, I'm not
so eager as I was back then (when I had more time and inclination).



But my understanding is We're supposed to be migrating to WDS..
(..ooops, did You FEEL the BREEEEEZE.. as THAT acronym flew by...!)
....I mean Eclipse.





My other point is that one (Wo)Man's features is another's overhead, for one
thing...  And there's an assumption that anyone who is not dedicated enough
to learn all this crap is not really a professional, for another...  Well I
recall a very informal survey by iSeries Network about a year ago...  SEU
got 80%, Code 5%, and various others made up the difference...  Some were
people afraid to learn new things, no doubt...  But some, like me, gave Code
a look back when it was (to be charitable) less-than-prime-time...  Compared
to *nix and Windoze standards, I guess it made the grade.. but compared to
400 standards it seemed a lot more like a Beta version, to my eye.

(My personal preference, btw, was for Flex/Edit...  But glad I didn't spend
too awful many man-days/weeks? messing with THAT product...!)

I'll have to admit that part of my inclination for using SEU is not wanting
to waste time on bells and whistles that don't ACTUALLY get the job done
BETTER...  There's a cost/benefit to learning, and I'm paying my own way...
Another part is that I have plenty of other things I'd just-as-soon learn
rather than how-to's and workarounds of using a particular editor,
especially if it's going to be around less than a few years.

And it still seems pretty shambolic, to me anyway, that M$ Word (although
lacking a syntax checker) is in many respects a better editor than what I've
seen of Code.

Maybe Eclipse will do better, but sure hope that 90+% of the $40M that went
into it is for marketing...!



Jmho (perhaps not expressed humbly enough?).

jt