Joe Pluta wrote:
And I expressed my opinion! Isn't the Internet great?
Sort of. You basically approached this with the "Nope, IBM should keep
it VARYING and automatically adjust the size of the index." Despite
that I had already given my reasons why I didn't like that as much, you
persisted in making me prove that my way was better -- without giving
ANY reason why your way was better. In other words, put the burden of
proof on Scott, and then every time he wastes his time explaining it,
come back and give some alternate solution to the problem so that Scott
has to come up with another example or another reason why his way is
better. On and on and on, debate about it for days...
On the other hand, some people actually appreciate my opinions! What I find
really interesting is that it's usually the "industry experts" that get mad
at me. Funny world, isn't it?
It's not that I don't value your opinion, Joe. It's that I don't want
to sit here and debate about it.
It seems that every time I get into the same conversation that you're
in, it turns into a long protracted debate. I just wanted to say my
opinion, not fight about it.
I like to have the length of the prefix explcitly stated. That's my
opinion. I've already given the reason that it would let me use the
length with shorter fields. I've also said (at the start of this
thread) that it'd enable me to use it with the SQL Scratchpad option.
(Right now I have to use a data structure with 4 bytes then a series of
data.) That'd save me a little bit of coding.
It'd also make it easy to use the *VARY 4 option of a *CMD object.
It'd also work nicely with a lot of the API data structures, where they
often give a binary(4) length followed by a bunch of data. For example,
the dirent structure used in IFS programming -- it has a 4-byte integer
called d_namelen followed by a variable-length d_name field that
contains the filename. I could just stick a VARYING4 or VARYING(4) or
whatever in there, and I'd be good to go. It'd save me work.
Lets say you're a C programmer (or COBOL, or CL or Java or whatever) and
you're calling an RPG program that takes a VARYING field. You're not an
RPG documentation trying to understand the parameters you have to pass.
(Sort of like the way RPG programmers are often struggling through C
documentation -- but in reverse). You see that the field takes a
VARYING parameter. Should it be a 2-byte prefix? a 4-byte prefix?
Sure, you can figure it out, or you can ask someone, but wouldn't it be
much easier if it were explicitly stated to be 4 bytes? or 2 bytes?
Wouldn't that be a lot cleaner and easier??
I can go on and on about this. But I don't want to. I don't want to
fight about it. I don't feel that it's my job to waste hours of my time
trying to convince *you*. It's not your decision, Joe.
Barbara understood what I was saying, and responded right away on
Friday. She also gave a reasonable argument as to why she wanted
VARYING(2|4|8) instead of my suggestion of VARYING4, VARYING8. And
that's fine... It's not a big deal to me, it was just a thought.
What I don't want to do is sit here and spend hours gainsaying
everything you say, while you spend your time (probably less since your
messages aren't as in-depth as mine) gainsaying everything I say. It
accomplishes nothing. But it's the way you have always been in these
forums.
I really could give a rats ass whether you're an industry expert, or
whether you think I am. Go back and look at the same sorts of arguments
that we had in 2001 or 2002 when nobody considered me an industry
expert... didn't matter, we still fought about it. The only difference
is that at some point, I got tired of these pointless ongoing debates.
midrange.com
