That and way/217799896 should be primary because that is the route for the EO83 traffic to access the roundabout. The opposite side stays as secondary because it has to cross another major junction before the traffic can join the EO83.

As for the second junction, and any grade separated junction, I was thinking more of osm.org/#map=18/51.70614/-0.38238 - I could not classify the crossing roads as motorway_link because it made no sense when it is sandwiched between two unrestricted roads. so there is a case for using secondary_link in this grade separated junction. I am also sure that there was a traffic flow relation at some point: in any case they exist because some routers … some routers sometimes recommend illegal turns without them.

Hello there,

For example: way/437409814 is not a motorway link because the tertiary side road merges into a motorway slip road in the process of joining a secondary road. It should realistically be secondary.

At way/561182489, the link roads that cross Tatoi should be a secondary link, because they are not part of a motorway. You also must be sure not to merge ways without checking that there is no turn restriction or traffic flow directive, which exist to reinforce the norms of Greek driving law.

Unfortunately, there is a user who thinks it is okay to map motorway junctions so crudely (see osm.org/#map=18/37.80333/22.73178), despite the fact that the remainder of the slip roads are motorway_link since there is "no escape".

I am very sorry for sounding harsh, this concern could have not come at a worse time when there is a serious community dispute over road names and numbers — it is so tempting to suggest organised editors to sit out while the crisis continues. My position is that "Odos" and "Leoforos" go to "name:prefix"=*, and that National Road (EO) numbers go to ref=* because they are in the law. I think that is a reasonable compromise.

^ Aww man, my grammar is awful. This could have not come at a worse time when there is a serious dispute over road numbers and road naming.

It seems that 69617882, 69617940 69618528 have unintentionally deleted valid data.

In this changeset, you deleted the relation for National Road 3.

You deleted a provincial road relation, but why? I think JOSM is behaving badly.

In this changeset, you removed a perfectly valid tunnel name that was signed.

I am not happy with how you assign some of the side roads to a motorway when it is really not, such as the slip roads on the Athinon.

Hi there, I kindly request that you be careful to ensure that the road numbers and formats are correct: you can discard the nat_ref value if it is wrong (the main motorway is numbered the A1 instead of the EO1).

That's odd. I thought that if SNCF Mobilités were merely running trains on the network, then they could not be listed as an operator.

There is no ULEZ charge if a diesel vehicle meets the Euro 6 standard, and that is why the charge applies to vehicles below Euro 6.

I will change that, but I had to improvise the values because the key doesn't document conditional charges properly, and I don't think we ever had access rules based on emission standards.

Hi, I think I have done my best to portray the road condition (surface and smoothness) despite the current policy being that villages be linked by at least one tertiary road. If you like, I can change the smoothness to "very_bad" instead of "bad".

Hi there, I have done the north of the city, and added the necessary editing notes there to clear up any confusion for others. I happened to have enough time today.

I am sorry about that. I will get onto it in the next changeset. I will say it is Bing, based on the OSBL.

Hi there, I think the name:UN=* tag may no longer be needed. The now-obsolete FYROM name could go to old_name:UN=* and local language variants.

Dang! It was supposed to be DigitalGlobe Premium Imagery as the source.

Hi, it's a named junction.

Why you downgrade roads between villages and hamlets to a dirt road? The current consensus is to connect all villages and hamlets with a tertiary road, and using surface=* and smoothness=* to describe the condition.

Source for the above message: way/32502470/history.