I think the best way to consider the local roads website, is as yet another tool in our armoury. My thoughts are that if a Rural road appears on there, where no information previously existed, then its probably correct. It may be, and probably applies to the changeset above, is that while the road is now listed as a rural road, any physical improvements may come at a later date. Whether we tag on physical appearance is another issue.
In this changeset, I noted some new RR signs & paved roads while riding down the 1184. Naturally I did not have the time to ride down every one so I referred to lmp.drr.go.th before plotting, as it helps with the length of RR from the main Hwy. In this case it drew my attention to the fact that there were a number of new Rural Roads in this area, and from our outdated aerials, I could see that many previously plotted tracks, were now 2 lane asphalted roads, carrying a RR ref.
While I accept that after checking Google streetview, the one way mentioned does look like a residential road, I chose to use the info on the database, and per our Wiki, gave it a Tertiary status.
Seeing a new sign physically on the road, supersedes the database in my mind, but there will always be exceptions as Alaska Dave and I remember a road in Udon that carried different numbers on the blue signs, as were on the mileposts … for at least 2 years !
However, by way of example, yesterday I added the South Western section of the new Chiang Rai bypass. https://www.openstreetmap.org/changeset/58234217#map=16/19.8483/99.7940 .
This a major 4 lane dual carriageway and is now open for traffic, but you won’t find any evidence of it on the rural roads database, or roadnet ! It’s mostly farmland on all the aerials we are allowed to use. Only Google satellite carries evidence of its construction, (and has it plotted as a road too)… but we can’t use their images & data, right ?
So I think I can confidently say that the two “official” road databases are not current and continually updated, which may go some way to answering Stephan.