Posted by Ed Murphy on 11/19/07 05:10

steve wrote:

> My point is MS is attempting to make application development easier at
> the expense of database technology. There is nothing in LINQ that
> advances db technology one inch. It is pure utility. There is nothing
> I've read concerning LINQ that indicates that anyone remotely
> connected with it has any idea of relational ideas/technology. And why
> should they, it was not a requirement for the job. Had they the brains
> to understand that relational technology is the key to overcoming the
> impedance mismatch and leads to a simplified programming model for
> application development, they may have
> taken a completely different approach. Their holy grail is making sql
> server invisible and what message does that send to the database
> community? The day that the LINQ group recognizes the idea of a true
> table type will be the day a new crew comes aboard for database
> development:) I hope it's soon because net is a marvalous platform, to
> good to waste on medicore thinkers.
>
> www.beyondsql.blogspot.com

Straw man. I did not ask about what LINQ explicitly does, but
rather what it suggests:

>> Look at the comments, in particular. If the type /could/ be named at
>> design time, at both the database and application layer, then would
>> your Holy Grail have finally been achieved?

My objection is not so much to your general idea of variables of
type table-with-given-columns (I've recently worked with some systems
that could be cleaner if such a thing were available; currently they
work around it using temp tables); more to your specific use of D4 in
all your examples, as opposed to a pseudo-code extension of SQL.

[Back to original message]