Re:SQL on AS400 was(Modernization and multi-member files)

[Frank.Kolmann@xxxxxxxxxx]

Jon
It was I who used the incompetant word, not Nathan.

I have no idea what SQLCLI is and have never used it.
Originally SQL (or should I call it embedded SQL) was a separately 
purchased product.  Without this product I found the SQL-like
interfaces (OPNQRYF, QRYMGMT) etc. very cumbersome.

When the company purchased BPCS they got 57xx-ST1 LPP.
It is this product that I see as SQL. 
Perhaps the SQL precompiler is a legacy of this separation of
the imbedded-SQL into a separately purchased product.
There is the 'native' for want of another word, access to the
AS400 database and there is SQL.  They coexist.  If the
S/38 had only SQL as a database access method I reckon no one
would have adopted it.

I have always had small issues with the SQL precompiler,
using different compile commands, different parameters on the commands.
It seems that to make the 2 data base access methods available
on a common compiler is next to impossible.

But surely WDSCi can be fixed to have SQL variables in the Outline View.
Is this not a level of competency.

Frank Kolmann


Message from Jon Paris <Jon.Paris@xxxxxxxxxxxxxx> on Fri, 18 Jul 2008 
11:30:44 -0400
> On 17-Jul-08, at 8:49 PM, Nathan Andelin wrote:

> > THE ENTIRE IMPLEMENTATION of SQL on the AS400 smacks of incompetence
> > and I believe it was not ROCHESTER that was responsible, but rather 
> > the
> > influence of the 'big-iron' in TORONTO.
> > (rant rant rant) apologies
>
> First of all I don't agree with Nathan's "incompetence" remark - I 
> have had many issues - particularly with the pre-compiler but 
> "incompetent"?  Anyway there are two things fundamentally wrong with 
> this particular argument ...
>
> First, you can't hang this one on Toronto.  DB/2 for the i is a home 
> grown Rochester product and that in part was why it lagged behind the 
> other DB/2 implementations for so many years.  They could not take 
> advantage of the DB/2 Universal code base for enhancements to SQL.  I 
> believe that these days they may be able to pick up some of the DB/2 
> code from Toronto but that is a relatively recent thing.
>
> Not that Toronto Lab (under the Software Group and lately Rational) 
> haven't done some monumentally dumb things related to "i" - but let's 
> at least criticize them for what they _have_ done!
>
> Jon Paris
>
> www.Partner400.com
> www.SystemiDeveloper.com