Re: Large MOOs

• Date: Mon, 18 Nov 1996 22:05:24 PST
• From:
  Will Orton <worton@romt.com>
• Content-Type: text/plain; charset="us-ascii"

At 05:13 PM 11/18/96 PST, John Towell Ph.D. wrote:

>But why would you store these objects containing information in the MOO
itself?
>Just keep the information in external files and use FUP.
>Or better yet, keep the information in a regular database and query the 
>database from inside the MOO. At NIU I have a MOO running that has an 
>object (called the Query Machine) that will take an SQL query, send it
>to a Sybase Database running on the same machine and bring the result of
>the query back into the MOO.  No need to bloat the MOO database with
>information that could be kept outside.

Right, this is starting to look like it may be the way to go. It may be
beneficial to code some built-ins and link to an external database through
an API rather than using FUP to avoid some system overhead.

As someone else mentioned, this also allows the possibility of keeping the
MOO and the actual data on separate machines, so that the load may be
better distributed between db lookups and WWW interface generation. The
ability to access the raw data (through SQL or whatever) without having to
hack up MOO may be an advantage in my specific case as well.


-Will

• Prev: Re: Large MOOs
• Next: Re: Large MOOs
• Index: MOO-cows
• Thread: MOO-cows