RE: alternative to WebFacing -- MIDRANGE-L

Joe,

How do you get hold of your eDeployment book?  I don't read many books any
more, but I'd be interested in yours.  (Didn't see anything on quick scan of
your site.)

I added a couple other comments inline (>>) and a 2 1/2 page long discussion
at the end.  I actually wrote most of this last night, but have been
thinking off and on throughout the day about how to phrase some of the
things properly.

jt

-----Original Message-----
From: owner-midrange-l@midrange.com
[mailto:owner-midrange-l@midrange.com]On Behalf Of Joe Pluta
Sent: Friday, June 22, 2001 1:55 PM
To: MIDRANGE-L@midrange.com
Subject: RE: alternative to WebFacing


jt wrote:

I think this could be a direct application and one of the biggest benefits
of the iNation.  What would it take to eliminate the interactive penalty?  I
think it would require:

that people quit wasting time complaining about it
some brains volunteered to work out the details
someone to lead the charge
some computer time
some way of providing long-term support
---------

jt, this part is already done, at least for relatively simple display files.
I've written a tool that will analyze your RPG and DDS and replace the
screen I/O with a call to an API.  Boom.  In one step, I have unbound the
business logic and the API.

>> Sounds pretty good.


I suppose you could also use this to modify the
code to communicate with a SPECIAL file, but there are obstacles there - how
do you return all the indicators, including the command key indicators?

>> I haven't done it, but I would just pass the *IN array as a field in the
PLIST specified on the F-spec.  It would behave just the same as the INDARA
keyword.


Anyway, the issue is not whether it can be done - I've already done it.  And
I "wrote the book on it", with my eDeployment book.  My programs run quite
happily on a 0CPW model 270, and they're quite fast.  I'm working on
automatically generating JSPs and thick clients and XML wrappers, but that's
really fluff.  The hard part is done.

So now what?

=========================================




Good question.  I have a few questions, an observation, and a correction to
a prior post:


I'm sure you noticed that I was bringing up the topic of OSS development in
the 400 Community.  Many have cited that as a weakness of the platform and
of this Community.  I have somewhat followed the development of your
software in these posts, and I'd always assumed that you were looking to
develop this into commercial software.  Are you suggesting you might donate
the code you've developed, if the circumstances were right?  I'm writing the
rest of this on the assumption that's what you were suggesting.

If so, that would be exceedingly generous; perhaps more than you realize.
I'd want you to go into something like this with your eyes open, for you own
sake of course, and because I'm not going to take _any_ responsibility for
suggesting you do this.  I'll be glad to suggest OSS might be a way to
improve things in the iSeries Community.  But I'm going to make every effort
to avoid suggesting, to any particular individual, that they contribute
time, code and/or money to it.

Also, I had an unstated goal in attempting this kind of project, which was
to try to galvanize some of the Citizens of the iNation into some action.
The reason for getting the backing of the iNation was two-fold.  Part was to
have some kind of semi-stable organization to support any software that was
developed.  But the other reason was that there are potentials to the
iNation that will never be realized, IMHO, unless the Citizens jump in and
do their part.  I've tried various things to jump-start some kind of action
(going back to first of this year, on IGNITe).  Met with little success, so
far, so thought a specific project like this might show a concrete example
of what collaboration can be good for.  Do you see this as a desirable goal,
too??


I want to take a brief intermission to tell a little story, and do a little
bragging on myself in the process.  I don't brag often, but if I can claim
I'm making a point by it, I can use that as an excuse...;-)  Maybe a couple
people outside my family have heard this story, so when an opportunity like
this presents itself, I gotta go fer it!

I probably won't make a big impression on you, Joe, because I saw your post
a while back on your "resume" of vast knowledge and work experience.  I
recall your writing Operating Systems (several, I think) and, I'm pretty
sure, also developing a lot of AS/SET for SSA.  I've never done anything
like that, so you might want to skip to the next section...:-)

In my first job, about 20 years ago, I was a one-man shop on a Univac mini;
I was given free use of the computer in my off-hours; and I was sort-of a
Cro-Magnon Dweeb.  (I was a dweeb before there was such a thing as a
dweeb...;-)  So in my spare time I developed an SDA-like utility.  (I'm
proud to say mine had the ability to center text where SDA didn't get that
until about CPF 7.0 or 8.0!)  I also deciphered the hex code to be able to
generate Univac's equivalent of a DFU, from the SDA-type screens.  As I
mentioned in a prior post, I developed a pre-compiler to take a DDS-like
data dictionary and generate RPGII code.  It was a bit more than that
though.  Since my pre-compiler generated C-specs as well as the I/O specs
from the database and screen definitions, it was something like a real, real
crude version of a proof of concept for something like Synon.  Being small,
modular RPGII programs it ran on a 64K machine with no virtual memory.  (For
any young folks out there:  a K used to be a unit of measure that stood for
1000...ROFL!)

Anyway, my goal was 100% computer-generated code, and I got as far as
generating maybe 75% of simple maintenance programs.  I was in the process
of adding field validation code from the data dictionary, and implementing
subfiles, when three things happened:  I decided I wanted to change jobs, I
heard about the S/38 that did all these same things.  Also, as I saw that
this program-generator could actually be written, I started realizing what
it would take to turn my hobby into a business.

I wasn't real swift about a lot of things (one constant challenge was
getting along with _all the people_ in the department... all 3 of them...:-)
But I did recognize that, while I had the technical side, I didn't have any
business skills, and didn't know how to market anything.  Didn't even begin
to know how to support a product.  The guy I was partnering with had less
skills, in all these areas, than what I did.  (Luckily all this wasn't a
complete waste of time.  By a series of flukes of nature, this project got
me a recommendation to the Manager of a S/38 shop, where I got my start.)


Besides the bragging, the two points to all this, Joe, is to say I've got a
pretty good idea what you have invested in your revitalization methodology,
and to call into question your statement that the hard part is done.  I'm
sure you mean that the technical problems are beatable.  With your
experience, I'm sure you know how hard it will be to deploy and support the
code.  That's why I asked in an earlier post, what's the best way to get
something like this in the hands of customers and then support them.

I don't know what you'd had in mind, Joe, or if you're looking for a
monetary return on your tremendous investment in time and experience.  So
it's hard for me to even guess, in your particular case, what you'll do
next.


But I'd say that there might be an opportunity to start an OSS Community
with a project like this, and there are a few additional things I think
would need to fall into place for that to happen.  As I said before, I think
these are some of the pieces that are needed:

some brains volunteered to work out the details
someone to lead the charge
some computer time
some way of providing long-term support

*** CORRECTION ***  Instead of "someone to lead the charge" I should have
said "three people to lead the charge".  Of course, that requires that more
than three people volunteer their brain-power to "the cause".  But other
than that, having three people in charge makes a lot more sense.  This would
prevent one person from getting a wrong idea that takes the whole venture
down the drain.  It also adds continuity in case one of the leaders can't,
or chooses not to, continue on with the project.

I think I'll probably elaborate on this some more in a separate post.


Joe, if you see the sense of this suggestion, you probably see why I could
never suggest you take up one of the leadership roles, and especially never
suggest you donate software you've already developed.  I know for a fact how
much you've invested in developing that code; and the time and your years of
experience are just a tiny fraction of that investment.  You don't work on
something that hard without having a tremendous emotional investment in it
too.

But since I'm suggesting that collaboration will produce superior results,
by definition, that won't necessarily include all the code that you've
developed, Joe.  Maybe none of it.  It could be that your contribution would
be your experience, and a whole new code base would need to be developed.
Depends, of course, on the design goals that are initially developed.


So to answer the question "Now what?"

I can only answer that for myself, and even that's not definitive because it
depends a lot on what other people do.  But I'll probably wait and see if a
team of leaders emerges, and see if some semblance of organization is
developed.  At that point, I would see if a good mix of, and the right
number of, people volunteer their time and brains.  At the same time, it
should start to become apparent if the iNation will choose to get involved
or not.  If it looks like most of these things have a chance, then I'll be
one of the volunteers.  I won't be one of the leaders because I don't have
practical experience in OSS development; plus, I'm the one suggesting this
approach.

I'm not going to make any suggestions to you, Joe, but I think I can spell
out at least a few of the various alternatives.  You can continue on with
your project, very successfully, as you have been (I assume as commercial
software).  You could start your own OSS project based on your code.  You
could try to partner up with some of the other brains of this Community and
just see what comes out of it...

It may seem like I favor the option for you to lead, or at least join, in a
Community effort to develop some new OSS.  Nothing is further from the
truth.  For example, if I signed up for this myself, would I fix up my old
code-generator and then donate it?  I don't know that at all.  It's easy to
get swept up in the enthusiasm of this kind of thing, and end up being
overly generous.  I wouldn't do that and sure wouldn't be fool enough to
recommend someone else do that.  For a lot of reasons...  But one is that
it's a short-term strategy.  If anyone ends up feeling they've been
over-generous, and that they've ripped themselves off...  Well their
involvement won't last.  And one of the main tricks to accomplishing
something like this is sticking it out until some benefit actually gets
delivered.  So I won't jump into a decision like this lightly, myself.


I know I didn't really answer the question "Now what?"  But hopefully I made
enough points to make this worth your while...:-)

+---
| This is the Midrange System Mailing List!
| To submit a new message, send your mail to MIDRANGE-L@midrange.com.
| To subscribe to this list send email to MIDRANGE-L-SUB@midrange.com.
| To unsubscribe from this list send email to MIDRANGE-L-UNSUB@midrange.com.
| Questions should be directed to the list owner/operator: david@midrange.com
+---