According to Richard Rathe: > ACGI is better in that the server is freed up to handle other > requests. This does NOT mean the ACGI will process more than one request at <snip> Actually, I have seen (and tried) an ACGI which worked quite well. What it did was to declare a pointer. Then, as each request entered the program, a new link was established to perform the operation. Once the operation had expired, a flag was set stating as much. This had some really nice features. It allocated new locations only if there were no free locations (ie: One pass through the list to look for open areas and if one was available - use it. If not, then create a new one.). The requests were handled in the order in which they came in (more or less). And nothing was ever lost since each request was given it's own small memory space. Without going into a lot of detail, there was a main program and a couple of subroutines. The main program alternated between checking for incoming requests, calling each of the individual subroutines, and returning information. :-) If I remember correctly, it was one of the examples which Chris (the maker of Quid Pro Quo) gave out. ***** Want to unsubscribe from this list? ***** Send mail with body "unsubscribe" to mac-perl-request@iis.ee.ethz.ch