The TrustNet server was originally made to service LSL scripts. This has worked nicely for now, but has a number of inconveniences. For example, the format used to talk to clients is rather ugly and hard to extend. Things had to be sacrificed to make it easier to deal with from LSL. Now that not only the in-world scanners talk to the server, some changes are needed.
I’m reworking the server code to be able to produce output in multiple formats. In fact, a good part of the work is already complete. The new code will be able to produce output both in LSL-friendly and XML formats. The later has vast advantages over the former, such as being able to return data with newlines in it, such as comments. It also allows to get better performance — the server will be able to accept multiple commands in one batch and process them as one transaction. This should result in a better performance on the website.
The reason this is being done is because the website is actually a TrustNet client like any other, and doesn’t have a direct access to the database. The current format puts limits on the data that can be transmitted, and it’s starting to become inconvenient to code website support in the server according to limits that only apply to LSL clients anyway.
These changes shouldn’t affect the server’s functionality. If something breaks, please IM me about it.