Chris Date has recently discovered a notification I sent him
in the late 80’s of an interview Bill Gates gave at that time to the now
defunct DBMS magazine. It was not much later after Microsoft joined with Sybase
to sell SQL Server. I thought it was perhaps apropos Codd’s recent passing to
remind readers of it.
Given what transpired from the Microsoft trial and Gates’
part in it, we now know that it is not knowledge and intelligence that spells
success in the industry, but rather luck, combined with – as Dustin Hoffman
once said of the entertainment industry – bullying and ruthlessness. Not
exactly the attributes that society should be rewarding, but there you have it.
Anyway, if the technical people don’t know any better (see the many
examples on this web site), what should we expect of businessmen masquerading
as technical?
If Gates were not rich, would anybody pay any attention to
such drivel, such poor articulation? See if you can identify all the ignorance,
fallacies and misleading squeezed in the following interview section.
“DBMS: At some point with SQL Server go further and
let the database engine enforce integrity, or at least referential integrity?
Gates: The type of integrity we support in SQL Server
is far, far more general. Referential integrity is a small part of it. SQL
Server allows arbitrary integrity checks, okay? Referential integrity. I mean …
I don’t know if you really want to talk about that. I’m pretty sure that
there’s nothing shipping that has it. I’m not sure that people reading this
know or want to know what those words mean. I mean, unless you’d have a long
article about it.
DBMS: We’ve talked about it in pieces in several
articles, and we actually have a piece scheduled later in the year about it.
Gates: The word referential integrity?
DBMS: Yes.
Gates: The word referential integrity? I mean, you
realize there are no relational databases. Did you write an article about that?
DBMS: It’s come up in several articles.
Gates: There never have been, and there never will be
in the next decade, any relational databases. [Ed.
Comment: Not if Gates has anything to do with it.]
DBMS: Do you think it’s important to fully conform to
Codd’s rules and be a full relational database?
Gates: No!
DBMS: Why not?
Gates: What do you want me to do, go through it rule
by rule?
DBMS: We actually went to the length of publishing
Date’s rules of what constitutes a distributed databases in our article on
distributed databases in the DBMS magazine right there…
Gates: Right, but that doesn’t mean people understand
them when you write that stuff down. You think I know what it means? I’ll sit
here and read this article.
DBMS: You don’t have to read it now.
Gates: Do you think I have any sense about why I’d
care?
DBMS: Well, I think that one role this magazine has is
to talk about what the state of the art is, what the relational model lets you
do, and why you should care about these things.
Gates: The key issues have to do with, well, not
having redundant data, not building into the data a big set way of querying the
data. I mean, those are the big benefits, and in addition most databases
labeled “relational” give you this arbitrary query language that operates with
some efficiency.
Actually, you can have and arbitrary query language against
any database, but they end up enumerating every record in the database to
process the queries. Do some SELECT against a hierarchical database that’s not
going down the hierarchy? Yeah, they can do it. The nice thing about relational
databases is they let you do queries on a very ad hoc basis in an efficient
fashion. Now, this notion of not having data redundancy requires somebody who
sets up that database to understand the entities and what’s independent of
what, and what’s driven by what – it takes a little bit of expertise, and
developers have gotten relatively good at that stuff. But I don’t think you’ll
see the relational rules being the key competitive factor in the microcomputer
world.”
Posted
06/20/03
[ABOUT]
[QUOTES]
[LINKS]
