Re: DMoz Relational Implementations category: chaff from wheat

In an attempt to throw the authorities off his trail, Leandro Guimaraens Faria Corsetti Dutra <leandro_at_dutra.fastmail.fm> transmitted:
> I maintain the
> http://dmoz.org./Computers/Software/Databases/Relational category,
> including the Implementation subcategory.
>
> It has been a long time since I have wanted to check the
> 'relationalness' of several listed systems, but haven't neither the
> time nor the expertise.
>
> So I would like to ask someone more knowledgeable to take a
> look and tell me which systems should be considered 'quasi-relational'
> -- meaning not relational at all, but at least more interesting than
> SQL -- and which should be dumped altogether.
>
> Any hints accepted, but please justify.

If you're going to have a pointer to /rdb, then you should also point to

• Rand/Hobbs /rdb <ftp://ftp.rand.org/pub/RDB-hobbs/>
• NoSQL <http://www.linux.it/~carlos/nosql/>
• Unity <http://www.bell-labs.com/project/wwexptools/unity/>
• Jinx <http://wwwtios.cs.utwente.nl/doc/jinx-doc/>

Which are all implementations of much the same sort of thing.

Unfortunately, they are pretty weak in some ways:

• Nulls aren't treated uniformly
• Views can exist, but only as programs, not as 'virtual tables.'
• They are quite weak on data integrity and "data independence" in general.

[Quoted] They represent an interesting approach to a data query language, and I find it disappointing that this approach has been pretty much ignored/lost, but they are only marginally "more relational" than xBase :-(.

-- 
(format nil "~S_at_~S" "cbbrowne" "ntlug.org")
http://www3.sympatico.ca/cbbrowne/rdbmsmisclinux.html
Rules of the Evil Overlord #117. "No matter how much I want revenge, I
will never order an underling: ``Leave him.  He's mine!''"
<http://www.eviloverlord.com/>