Editing bits of roads with loads of routes on them is extremely tedious. I had to remove two pieces of road from ~10 different relations and replace them with another piece of road.

There must be a better way to define routes than relations.

I've been adding some addresses.

I've been using nodes for the entrances tagged with the addr:*=* tags and building=entrance.

The reason for this is because some buildings have multiple "bits" in them. I've seen others dividing the building into smaller sections instead.

I'd say the node way is better, but I just wonder if it's good practice.

An idea that I saw on the mailing list (thank goodness I don't subscribe to that) was that of different layers of data. The idea would be choosing what "layers" of data you want.

Certainly it would make editing a lot easier as some areas are unwieldy to deal with.

Thinking about it, XAPI does (did?) this.

Does every road need to have a maxspeed tag? What methods do people use for collecting speed limits?

I've been looking through bugs on MapDust, there are rather a lot of fairly useless bugs with no description, but I've also fixed at least 3 genuine bugs.

Seems like a nice system altogether.

I've just requested mkgmap's packaging in Fedora. See https://bugzilla.redhat.com/show_bug.cgi?id=592645. Any advice is welcome

I think the United Kingdom label should be a wee bit higher, somewhere near Newcastle would be more central I think.

I have changed Great Western Road in Glasgow to be highway=trunk. Feel free to change the lower regions nearer to the centre of Glasgow to be primary again.

This was done on the evidence of green signs along the road instead of white ones.

I very much like Crschmidt's work on the data browsing thingy. That is all.