Indeed. Seems to me at least this is "log for nerds who want to see code line by line", not so much something the average user would need to see. Presumably once stat pages are constructed any autodetected concurrencies would be reflected there, in a more graphical and user friendly way than this.
I'd call it the "log for the programmer who wanted to see exactly what the algorithm was doing." I'd like to either have something easier to understand in the either in per-user logs or a separate log like this one, and/or a map where one could browse the detected concurrencies, likely with a "u=" query string parameter to restrict to autodetected concurrencies for a given user. Again, that's for down the road.
After efficiency improvements, I think the priorities should be:
- Nail down data formats and convert (if needed) to them. I worry about this step until we get a route editor up and running with the new format. I don't think implementing a new editor is especially difficult, but I'm hoping someone else will take it on. I can continue on data processing work with the old format, but I do think we should make the switch before starting highway updates and adding new systems. There's a huge backlog of fixes to existing systems, several in-development systems that are probably close if not ready to activate, and a good number of people itching to contribute new things. It should be trivial for me to load in data in a new format if and when the format change happens - just a couple lines of the Python code will change.
- More maps. Both Google Maps and the image files like the old CHM project produced. I think everything needed to support these is in the DB now, but I'm sure I or others who implement these parts might want additional information or information in more convenient formats.
- Start generating some stats. I think this is mostly best done on the Python side when a site update is running, when so much information is in convenient to access data structures. The stats could be displayed in log form to start, then added to the database to be put onto pretty web pages. I'll probably start with things similar to what we see in the old CHM project, but I think there's all kinds of room to do interesting things here.
- Lots more of the web-facing infrastructure. I'm generating some pages that have the PHP code that queries the DB and puts up the basics, but I'm hoping others with better web design skills than me will take the lead on this.