Patrice,
That's a hell of a write up! I knew the background to the SQLServer product,
but not in so much detail!
I also have a write up on the Ashton Tate/Microsoft/Sybase relationship, to
give a little more info on the "early days":
Back then the PC database Market was dominated by Ashton Tate's dBase
product. dBase IV had just been released and had been having a difficult
time overcoming industry criticism that its memory management did not meet
Ashton Tate's claims. They wanted to start to build on what they saw as
their solid foundation in the PC market and start moving onto bigger
systems. They moved to bring out a version of dBase IV on UNIX and in 1988,
together with Microsoft, jointly announced a deal with Sybase to port
SQLServer down to the intel platform using the new IBM multi-tasking
operating system OS/2.
Sybase's interest was obviously to license more copies of SQLServer and
penetrate a new, lower end market. Ashton Tate's goal was equally clear - to
promote dBase as the database application for everyone. It's ease of use was
designed to make it open to fairly inexperienced users while its programming
language was to make an ideal tool for developers. By buying the Server
Edition, developers would be able to connect to more serious data sources,
like SQLServer, to store their data. (A vision remarkably similar to another
PC database that was to come to market later).
Microsoft's interest in the deal was less clear, although everyone seemed to
assume they were only out to sell copies of LAN Manager, their network
software built on OS/2 - the operating systems they had just coded for IBM.
After all - as the contemporary gossip went, Microsoft knew nothing about
databases!
The deal gave Asthon Tate full marketing rights to the product, which was
badged with the Ashton Tate logo. Microsoft's cut came from sales of LAN
Manager and Sybase's came from commission of every copy, who retained all
intellectual rights on the SQLServer software. Ashton Tate were licensed
only to add to the product.
Ashton Tate had apparently, signed a clause in the deal, that committed them
to deliver dBase IV Server Edition by a specific date. Failure would result
in the company loosing any rights in the deal. This indeed occurred and full
marketing rights switched to Microsoft almost immediately.
So that is was happened in a more detailed summary than was on Page 8.
Microsoft did pretty well out of the deal don't you think?
Regards
Mark
-Original Message-
Patrice J
Sent: Thursday, May 03, 2001 09:49
To: Multiple recipients of list ORACLE-L
More info since this is s-u-c-h a fascinating topic. In case you're
wondering, it's past the end of the work day here for me, so I'm not wasting
time... ironically I'm staying late to study for a SQL Server 7
Administration exam!
Here goes, mind you this is from the lead developer for MS SQL Server, so
keep that in mind.
Inside SQL Server 6.5, p. 5
"Ashton-Tate, Microsoft and Sybase would work together to debut SQL Server
on OS/2. (This was the first use of the name "SQL Server". Sybase later
renamed its DataServer product for UNIX and VMS "Sybase SQL Server", the
name by which it is known today.)"
Continuing with the quotes on the history:
p.8 loosely summarized, MS and Sybase dropped Ashton-Tate because that
company was focusing on their dBase IV Server Edition product, and SQL
Server 1.0 was not much of a priority for them. At the time though Microsoft
still saw SQL Server primarily as a way to push LAN Manager. So Sybase
appears to have been the key developer for this project, the ones with the
most stake in it.
p.9. in 1991 SQL Server 1.1. was released to work with Windows 3.0..
Here the shift begins.
p.10 - exact quote: "It was a great day in the SQL Server group at Microsoft
when in early 1991 Microsoft's agreement with Sybase was amended to give
Microsoft "read-only" access to the source code, for the purpose of customer
support." [...] "As a small group of developers at Microsoft became adept
with the SQL Server source code and internal workings, Microsoft began to do
"virtual" bug fixes. Although we were still not permitted to alter the
source code, we could identify line-by-line the specific modules that needed
to be changed to fix a bug. Obviously, when we handed the fix directly to
Sybase, high-priority bugs identified by Microsoft got resolved much
quicker."
Continuing, still p.10:
"After a few months of working in this way, the extra step was eliminated.
By mid-1991, Microsoft could finally fix bugs directly." They still had to
let Sybase review the fixes before applying them, because Sybase at that
time was still nominally the owner of the base code for the database engine.
p.11 on the purpose of SQL Server at that time: "No hard limit was
established, but in general, SQL Server for OS/2 was used for workgroups of
50 users or less. For larger groups, customers would buy a version of
Sybase SQL Server for higher performance UNIX
