Capabilities: RTF editing, Graphs, flat-file database

[jbv]

Sarah Reichelt :

> Just to add to Ken's answer about data storage:
> The one-card-per-record method will get rather unwieldy and slow if you
> have 60,000 records. I wouldn't recommend that method for more than
> about 1000 records.
> Another very popular way is to use a single data field with one line
> per record, and each field in the record separated by some delimiter,
> usually tab. The whole field can be loaded into memory allowing very
> fast searching, sorting etc.
> Also, you might want to consider XML which saves your data externally
> in a file that can be read by many applications, but uses Rev's fast
> XML library for searching, editing etc.
>

BTW, when do you guys decide to use an SQL database instead of a single
file loaded in memory of a flat-file database ?
Is it related to the number of records ? the number of items per record ?
The complexity of the treatments to be done on the records ?

In several apps I developped (either offline and online) it was hard for
me
to decide (except when the SQL DB already existed of course), as I always
found the Rev search & sort functions quite fast & powerful (slightly less

powerful than the same functions in OMO that I still miss).
AFAIR, about 2 years ago, someone mentioned on the MC list using a well
indexed flat-file DB of about 700,000 records, with no disturbing delays
in
searching & sorting...

Any opinion ? Thanks,
JB