From: John Joyce <dangerwillrobinsondanger@...>
Date: 2007-03-21T23:56:12+09:00
Subject: Re: Object/Relational Mapping is the Vietnam of Computer Science

Reality was uglier than that. More comparable to the browser wars and  
implementations of html of the 90's.
But worse.
All the database makers had their own ways of doing things. SQL was a  
pseudo-standard middle between systems that always offered their own  
proprietary things as well. Like html and C/C++ and JavaScript and  
everything else out there, vendors slowly merged toward more fully  
supporting a standard by customer demand. You don't think Microsoft  
was the first to think up the idea of using proprietary things to  
lock in customers, do you?
Databases were the original home of neckties running IT.
Everyone was calling their SQL the real one, but none were the same.  
(still aren't really) because they don't want you to jump ship so  
easily! Long term licensing revenues with low support costs.
On Mar 21, 2007, at 11:44 PM, Eivind Eklund wrote:

> On 3/20/07, Gary Wright <gwtmp01@mac.com> wrote:
>> I'm curious as to why query language development got hung up on SQL.
>> I've read a little bit about Tutorial D.  Is SQL simply
>> another example of pre-mature standardization?
>
> From memory, and as far I've understood it: Sometime during the early
> 80s there was a rush to get relational databases out there, with
> several vendors producing databases and wanting an interoperable query
> language.  As a result of some sort of differences, two different
> groups were formed, both creating their own attempt at a standard
> query language.  Over time, a bunch of rivalry formed between the
> groups, and each of the languages they developed had advantages in
> some areas and disadvantages in others, so it wasn't possible to pick
> a single "winner", and everything kept bickering back and forth.
> Prestige entered, and none of the groups were politically able to give
> in to the other.
>
> Along came SQL, which was a toy language made by a single developer to
> just show have SOMETHING.  It was worse than either of the proposals
> from the other groups, from a usability point of view (though with a
> couple of extra features I don't remember what was that were useful),
> and it was a possible political compromise.  So, by shooting down BOTH
> groups for something that was worse, a solution was possible.  Thus,
> we standardized on SQL.
>
> Take with a grain of salt - this is based on renderings of history
> I've read on the net, and reproduced from my spotty memory.  Yet it's
> a nice story that fits very well with my feeling about SQL ("Nobody
> could really have meant that language SERIOUSLY, could they?")
>
> Eivind.
>