Stephen,

>Yea I use pessimistic buffering and set reprocess to 1
>I put the data in the same table so I don't have to use to many tables over the network
>

With the idea I propose, each data file is smaller (because of normalization). IOW, space does not have to be reserved on every record in every table to store lock information. This will lower network traffic. Only a small subset of records will normally be locked at any given moment.

>I try and keep things simple or (encapsulation) the way I decipher it is, keep everything internal to itself where applicable

Yes, simple is good. Speaking of internal to itself, one advantage of a rlocks table is this. Let's suppose you want to know what locks currently exist on *all* tables. You only have to query one file to get this information.

>
>and as always you never ever have to listen to what I say

I'm not sure why you think I don't want to listen to what you say. I had no intent to offend, but merely wanted to kick in some ideas. My apologies...