Developer Forum

Why do we have standards? And who sets them? And who protects them?

When I first encountered standards bodies in the computer industry I
thought they were a good idea.

Then I encountered X.25. A standard for data exchange. The only
problem was there were 4 groups who really defined the X.25 standard,
and they all had a vested interest. OK there were other representatives
on the committee but the people who did all the work and made all the
decisions were suppliers who were selling X.25. And guess what, even
though they all agreed on a standard, each vested interest implemented
the standard in such a way that use of it became almost proprietary.

As for SQL - another great idea. But who really defined the standard?
Not the user base. Not the hardware companies. But the vested
interest relational database companies. And what does that really mean
to the standard - it is the lowest common denominator of the facilities
available from the vested interests. And again they have all extended
beyond the standard to make their products proprietary.

So when anybody says "standard" to me I always think of who set up the
standards bodies and why was it set up. Just think about it. What does
ASCII really mean? And why doesn't it work worldwide? What does ANSI
stand for - and is it really the best idea, or the idea that gives the
best profit to the capitalists.

I'm sorry, but until the users are allowed to define the standards and
the capitalists are kept out of the loop I will treat any "we adhere to
xyz standard" with the contempt it deserves.

Sorry Serge!

Malcolm

-----Original Message-----
From: XXXX@XXXXX.COM [mailto: XXXX@XXXXX.COM ]
On Behalf Of Andrew Hamm
Sent: 15 June 2004 04:47
To: XXXX@XXXXX.COM
Subject: Re: On the use of ANSi




Many years ago I have fond (no, not) memories of reading the ANSI
database details in the Informix manual, and repeatedly shaking my head
at the pointless OCD assumptions.

Have examples? Not any more. Am I going to re-read the manual just to
make a list of silly assumptions? no.


sending to informix-list

alcolm weallans wrote:
There is nothing restricting the user base from participating.
But who has enough time to do it?
> Not the hardware companies.
Why would they? Let them deal with SCSI and ISA and PCI..
There are also ISVs who particpiate in the SQL Standard committee
> And what does that really mean to the standard - it is the lowest
> common denominator of the facilities
Unfortunately true. There are two problems here:
1. Like any more or less democratic process, reaching a decision takes a
long time in a standard - too slow for the needs of customers in thsi
fast paced industry.
2. Bringing proposals to the standard in beforehand reveals a company's
intention.
Example:
In 2000 I canvassed reactions for SELECT FROM INSERT in the DB2
newsgroup (I try to at least ask the user base). But I would have rather
bitten my tongue than tipping of Oracle and MS how serious this was.
Nonetheless we have started the process of standardization - and I'm
confident it will go through.
By contrast MERGE first went through the standard, then into Oracle and
DB2 (and then was extended upon by both companies which will go back
into the standard)
Standards are relevant as long as there is a need to communicate.
Who decided on the width of railway tracks? The shapes of powerplugs,
batteries, etc... They likely didn't start out the way they are today.

There is always a fight between proprietary and open.
Philips screwdriver vs. special tool only your mechanic has.

> What does
ASCII works great for what it's worth. UNICODE gets the job done.
Note that SQL also is an ISO standard

Users are allowed. There is nothing preventing you from founding a User
Interest Group for SQL, go through whatever formalities are needed to
register and start submitting papers.

Hey that's OK. Believe me I'm bitter when it comes to not being able to
use my cell phone when I fly "home" in two weeks, and I need an adaptor
for my razor.

I have one consolation though:
As technology matures standards stabilize.
There will always be hold outs (like Brits driving on the left side) and
the odd traffic sign, but by and large with commoditization comes
standardization. DBMS are on the way towards standardization.
If a DBMS wants to be part of the party it has to comply.
That's why SQL Server and Oracle added ANSI JOIN syntax and the WITH clause.
Successful quasi standard features get absorbed (modified to fit) like
Sequences and Identity.
I see loose casting rules happening in teh future. The user base (!)
seems to like them. What can we do.

If you want to see it happen be on the receiving end of an SAP request.

There also was a really interesting article by Jim Melton somewhere on
OTN on the future of the SQL Standard.
I suppose it wasn't biased enough. Seems to be purged.

Cheers
Serge
--
Serge Rielau
DB2 SQL Compiler Development
IBM Toronto Lab

malcolm weallans said:

Right on, brother! Death to the capitalist pigs! Mind you, looking at how
hard it is to round up a few people willing to commit to the relatively
minor amount of work needed to run, say, a user group, I can only imagine
that organising an actual vendor independent standards committee would be
... challenging?

--
Bye now,
Obnoxio

"C'est pas parce qu'on n'a rien ?dire qu'il faut fermer sa gueule"
- Coluche

"I'm trying to see things your way, but I can't get my head up my ass"
- JCH

"Ogni uomo mi guarda come se fossi una testa di cazzo"
- Marco

http://www.yqcomputer.com/ ~esr/faqs/smart-questions.html
sending to informix-list