Re: TRM - Morbidity has set in, or not?

"David Cressey" <dcressey_at_verizon.net> wrote in message news:tdo9g.1728$_B5.1298_at_trnddc01...

> Marshall,

> I'm going to start a new discussion, rather than divert this one. There
are
> several ideas that come together here,
> including some that you have brought up in the past.

> First, the idea that the index IS the database, from the article the OP
> referenced.

I also said something similar at one time, intrigued by the statement in the patent that there are not indices.
I said the data is the index. :-)

There is also that "saying" that the log is the database and the database is an optimized access path to it. :-)
http://weblogs.asp.net/aaguiar/archive/2004/05/04/125857.aspx http://flipdb.com/

> Second, the idea of "content based addressing" from some of your old
posts.

We are talking about relations, aren't we ? Codd said that maybe ? Maybe many others before Codd ?
What is the difference between a function and a relation ? I've read about relations in a book published around 1920. The ideeas in that book were older than that. Why do we name columns and not rows ? :-)

> Third, the idea that the index and the table are redundant, but not
> "harmful redundancy" in the sense that term is usually used to mean.

Redundancy at the storage level is not "harmful redundancy". Redundancy at the logical level is not "harmful redundancy". Unmanaged redundancy is "harmful redundancy".

> Fourth, hardware associative memories.

Cellular automata ?

> Fifth the question of whether a "column store" is any more fundamental
than
> a "row store".

Why do we name columns and not rows ? :-) What is fundamental about an implementation issue ? :-)

> Sixth the question of whether Starkey didn't use the "index IS the
database"
> concept in Interbase.

I don't know about that.
I've heard that some people claim they can read a book by reading its index table. :-)
Maybe we should say "THE index is THE database" ? I heard Interbase/Firebird is based on multiversion, compressed btrees and bitmaps.
I heard Netfrastructure is based on some kind of in memory "column storage" - a kind of segmented storage.

I heard that the i386 can organize memory as we like - array, paged, segmented, or a mix. :-)
I heard that some file systems allow sparce files and writing/deleting any part of a file. :-)
I heard that one could put some kind of translator on top of that. :-) An XML one ? :-)

> All of the above has little to do with the TRM, but may merit discussion
in
> its own right.

Very little, indeed. :-) Received on Mon May 15 2006 - 11:03:25 CEST