Upgrade Georgian transnational highways to trunk

Wikiproject Georgia Abkhazia South-Ossetia proposes scheme for Georgian roads proposed by moszkva ter:

This scheme leads to the fact that now there is no trunks in Georgia. All trunks going to Georgia from neighbouring countries breaks off at Georgian border.
Also this scheme proposed tagging roads as trunks based on its physical characteristics. This contradicts assumption at wiki:

I propose reformulate rules as follows:

In following this definition upgrade to trunk highways included in routes:
https://www.openstreetmap.org/relation/5941631 (all)
https://www.openstreetmap.org/relation/28929 (all, and also road going throuth Kutaisi that not included into route but complement it in direction from East to West)
https://www.openstreetmap.org/relation/3346319 (part from border with Turkey to Poti)

See scheme: http://openstreetmap.ru/?mapid=1502017408

Are there any objections?

——

На странице википроекта Georgia Abkhazia South-Ossetia представлена схема тэгирования дорог в Грузии предложенная moszkva ter:

Эта схема приводит к тому, что сейчас в Грузии вообще нет дорогу уровня trunk. И все trunk, идущие в Грузию из соседних стран, обрываются на грузинской границе.
Также эта схема подразумевает тегирование дорог на основе их физических свойств, что противоречит принятому международному соглашению и, в частности, такому предположению в вики:

Я предлагаю переформулировать правила следующим образом:

И вслед за этим поднять до уровня trunk дороги, являющийся частью маршрутов:
https://www.openstreetmap.org/relation/5941631 (все)
https://www.openstreetmap.org/relation/28929 (все, а также дорогу, идущую через Кутаиси, которая не включена в маршрут, но дополняет его для движения с востока на запад)
https://www.openstreetmap.org/relation/3346319 (только часть от турецкой границы до Поти)

Смотри схему: http://openstreetmap.ru/?mapid=1502017408

Есть какие-то возражения?

You should also post your proposal in the communication channels used by the mappers of Georgia.

Both groups in Facebook unavailable. Mailing list have just two message from 2009.
Therefore I send personally invites to forum topics by osm to mappers who draw Georgia.

I somehow find the old rule more correct, as most of these roads are very far from anything you could call a “highway”. On the other hand, when I see how it is handled in the neighbouring countries, your rule would be more persistent.

If I could decide, I would correct it in the other countries. The “European Road” does not mean much, as you can clearly see on the example E97 Kherson - Aşkale. Its route follows several disputed areas and a normal person probably could not cross most of these borders.

So the importance of the road should only be one factor, another one clearly should be the actual quality of the infrastructure.

This voting fix that general defenition of tag ‘highway’ should be “importance for the road grid (hierarchical position in the interconnecting network) instead of physical attributes”.

Wide used approach when ‘highway’ may be increase or decrease by one level when its quality is very good or very bad. The most common case is upgrading ‘trunk’ to ‘motorway’ when ‘trunk’ have some extra attributes.

But I agree with remark about roads formally classificated as European or Asian routes but which are not in fact because they cross disputed regions. Therefore we could not link to official classification and use more common rules from Key:highway page.

In additional to my previous proposal I’d like also upgrade to ‘trunk’ the most popular road go from Russia to Erevan (Armenia) via Tbilisi.

Yes of course, if there is already a decision, then you are absolutely right. You just should make sure, that you also upgrade those parts, which are actually a motormay right now, that they can be distinguished somehow from the rest of the roads.

I am not a big fan of this rule though, as I was a “victim” of it one day myself. I followed a “main road” with my car and it was barely usable. I was happy that the car and me survived :wink: What I expected to be a 20-min ride, turned out to take 3 hours. So I may be biased.