[Seaside] Database

[Stephen Pair]

Adrian,

In a real deployment of BerkeleyDB, I wouldn't load the data into the
client...instead I would bring it into the server and manipulate it
there.  Indeed, the objects that form the indices have to be brought
into memory at some point, so why not bring them into Squeak where you
can manipulate them with the full expressive capabilities of Smalltalk?

Also, I am getting close (probably 2 more weeks) to releasing my
ChangoPVM and database which will make these things much easier (that's
why I've been frantically publishing things that it will need out to
SqueakMap).  I hope that you might consider using it (if your time frame
allows)...I will be using it in a commercial context.

- Stephen

> -----Original Message-----
> From: seaside-admin@lists.squeakfoundation.org 
> [mailto:seaside-admin@lists.squeakfoundation.org] On Behalf 
> Of Adrian Lienhard
> Sent: Friday, December 06, 2002 4:12 PM
> To: seaside@lists.squeakfoundation.org
> Subject: [Seaside] Database
- Berkeley DB (http://www.sleepycat.com/). Plugin for Squeak:
http://spair.swiki.net/27 from Stephen Pair The DB may be very good in
some areas (according to the webpage..). But eg. for doing a fulltext
search it's quite useless because all data would have to be loaded into
the client application... And second, it can't be used over the network.

So, I really don't know... I think to tend to Postgres and try to use
Tantalus. On the other hand it would be probably more comfortable to use
a object database.

What are your experiences?