You need to select in the left menu “Roads” in the “TEN-T Regulation 2024” panel
open The legend to understand which is core and which is comprehensive
see that some of it is under construction/planning or projected and in that case:
existing old roads are highlighted as part of the network
the projected road has approximated arbitrary position on the map
If we all can possibly decide to use it all across Europe as it’s a reliable officially decided plan by the EU, that would likely be great news and then we could see how to do it in more detail and decide what is trunk and what is primary and if motorway should be on top of that or simply a mark of quality and not importance only
Because the Core network is basically a motorway, but many roads from comprehensive network are motorway as well. So we’d have to make one thing certain along this OSM approved definition:
highway trunk The most important roads in a country’s system that aren’t motorways. (Need not necessarily be a divided highway.)
highway primary The next most important roads in a country’s system. (Often link larger towns.)
Are motorway considered the MOST IMPORTANT, or the trunk is and simply motorway is only when its quality is high enough, but it is considered of the same highest importance as trunk
Because of this definition i bring up the fact, that in TEN-T network Core network is recognized as the MOST IMPORTANT and comprehensive network is less important.
Do accept this or we consider both Core network and comprehensive network being parth of the most important network and hence be tagged trunk?
Otherwise comprehensive network should be tagged primary
Additionally because both Core network and comprehensive network are not always finished we can then decide of those old roads that are not yet rebuilt or the most important right now or will be only once finished
There are more little details other than GRB BLR and RUS do not participate in TEN-T
Turkey has only comprehensive network so there it is the most important network by default
Sometimes the map is little bit vague
The classification is not set in stone, EU and signatory countries periodically decide changes
Most important for me is to see if there can be any european international consensus on this idea and if that can be more important than indyvidually national for every separate country
the poll is very similar to Sonda nastrojów trunk w Polsce: and similarly misses winning option: “Other” (as I complained two weeks ago in post there)
I expect that you cannot create “TEN-T = highway=trunk” or a similar rule and apply it to OSM and get great results.
Still, such rule will be probably better than “highway=trunk is for expressways forming bunch of disjoint highway=trunk separated from each other by lower grade roads”.
I expect to overlap to be be present here, especially in areas where motorway network is really unfinished.
trunk are for the most important roads in national or wider network (and should form consistent network rather being bunch of scattered segments*)
*yes, I know that say Germany is using for expressways/motorroads and Poland was doing it recently (maybe is doing it again, I cannot keep up with highway=trunk discussions and edit wars and edits)
highway=motorway are for roads of a specially high quality
In theory you could have badly planned or not yet fully constructed highway=motorway used only for a minor local traffic. Or ultra-important highway=trunk trouncing highway=motorway, or highway=trunk of a very similar importance to motorways. In such places motorways tend to be at least planned (or on some alternative connection)
So overall, there is overlap - but typical trunk will be less important than motorway. With this effect being especially strong in areas that completed motorway network. But not all =trunk are less important than all =motorway
As far as I can see the Core Network roughly means “the most important roads in the European system”. That is a very different thing to “most important roads in a country’s system”. In at least some countries, if we said that only the Core Network qualifies as highway=trunk,that would make the network much sparser at that level. That in turn would compress the existing network between unclassified and primary and likely lead to a loss of information, as roads that are currently distinguished from each other would end up in the same category.
For example:
Ireland would in be left with only 3 trunk roads. Even the road between the 2nd and 3rd largest cities in the country is not in the Core network.
There are no Core network roads in southern Portugal - even the very busy Algarve motorway is not Core, and neither are the motorways connecting to Lisbon and Spain. So that would be maybe half a million residents with no trunk connection to anywhere, not to mention the importance for tourism.
in Spain, the cities of Granada and Jaen are not connected to the Core network. That’s at least another 400k residents cut off from the trunk network. And the road from Malaga to Cordoba, which is also part of the logical route from Madrid to Malaga, is not classified as Core. That would leave the population of 1.8m in Malaga province without a realistic trunk connection to the capital (as nobody would divert as far west as Seville to stick to the core network).
Maybe the fact that a road is part of the core TEN-T network could be expressed via a new tag if people are interested in that, but I don’t see any benefit to using it as part of the existing OSM tags.
As for the 2nd option, I only had a quick look at Ireland. It looks like the comprehensive network is very similar to the existing mapping of trunk in OSM (apart from unbuilt sections) - I guess both essentially reflect official classifications. But I don’t know enough about other countries to comment on that option.
By the way, is that an unattributed use of OSM I see?
… Unattributed and outdated, severely. E.g. Vittorio Emanuele II has long been renamed per Italian convention to Vittorio Emanuele Secondo. The oneway road (upper red arrow) has long been closed off… the western 3rd remains. The geometry of Bovio curve (lower red arrow) has substantially changed, is connected now to the trolley busway, no access but bus and emergency. The main ways in the orange elipse are secondary, yet show as black instead of brown (used to be primary). Why, when identically tagged, since years?
But, it makes me strongly think of the int_ref=Enn network for Europe where the E55 and E80 cross ways on the primaries, trunks and motorways around town.
only if the authorities in charge have proceeded with the legal procedures to rename it, otherwise it is just wishful thinking by the statistics office, who do not have the right to rename streets.
Typically the street signs have not been replaced, so there are good arguments for “name=Viale Vittorio Emanuele II” even if the street has been officially renamed (e.g. it could be tagged as official_name=Viale Vittorio Emanuele secondo).
To quote myself. Why this nonsense had to be, don’t know. The most (expletive) a series of allies numbered from 1 through 46, the last now named ‘Vico quarantaseiesimo Garibaldi’ and the one doing that all over the place unencumbered by preserving the signed names so Vico 46 Garibaldi remains searchable in OSM.
Thank you for all the feedback. One can’t know unless asked ;D
I see that highway categerisation is a lively matter in every country and everyone make their own choices along their own rules i gues.
@Mateusz_Konieczny the “other” option is quite obvious yo gues. I was thinking to add even more options and allow multie choice, but better to keep things simple and work from that when things actually get complicated.
But in general i don’t think people are really interested to discuss the matter or looking for unification.
@alan_gr yeah, good points. I’d classify connecting cities and and to the core network as primary while the core network as notionally important doesnt have to go through the cities itself. It’s tough to imagine when one lives in a city where motorways goes everywhere, but yeah, If a city is not connected via a motorway or EU officials didnt decide to put core network near it, then I don’t think that road needs to be the most important and hence be tagged trunk
It’s actually a very good idea to design trunk outside of population centers and in western Europe i can see exactly that and in Spain there are even more motorways than the TEN-T netweork require,
While in the eastern EU motorways are greatly under construction and using the old roads network (where in near proximity motorways are planned, coz not even under construction) as trunk don’t be suprised to see driveways of casual residential homes going right to the trunk without even a sidewalk present.
But yeah, every community in their own country likely wants to do it their own way.
I happen to be in one where you can’t make heads or tails right now.
@SekeRob yeah, Unattributed. But if they dont use OSM, we’d all have more difficulties to compare anything. Even if they got some updating issues, is that really important? Especially that nor core or comprehensive network goes to this kinda outdated OSM map tiles.
Why does it have to be all or nothing? Why can’t one of these official schemes be informative rather than dictating each individual classification? If the community can’t depart from an objective literal reading of some standard because it would bring neverending acrimony, look what the alternative has gotten you.
Well, if you dont have administrative entity that decides road importance, then who does? A few voting OSM mappers?
SO the current problem is exactly that 50% people voted one thing and the other 50% a different one.
A few people decided to go through the option they liked and had 50% + 1 majority, but then suprise suprise, turned out those people had their own ideas and started reverting one another mapping…
So yeah, sticter rules dependent on what administrative entities decide and not the mappers might be a solution to that.
A decision to leave some mapping logic independently to every country community do help, but then example: at border crossings you wonder why highway classification changes at it.
This is not so hard. At least one other country with no applicable government classification system has figured this out. Yes, there are occasional disagreements on specific roads. We spent a lot of time coming up with the exact language that described what we meant by importance. Yes, these written standards for importance use adjectives that are subjective.
If there is a government classification that is a good fit for OSM highway classifications, great! If they aren’t a good fit, we should not use them. But, having mappers decide how to map isn’t some odd idea. In fact, it’s the foundation of exactly what OSM is.
yeah, i can see that since 2021 most of the states are still in the draft phase ;d
But yea, you dont have TEN-T. Maybe you have got something else, but still chosen to make your own rules.
I’m also happy for you, that it’s decided with a consensus :)
It’s a plan, not a finished system. Core roads should be finished until 2030, but comprehensive until 2050. It’s a long time so many of comprehensive roads may simply never become one.
It’s not known. It’s not signed on the ground so many people probably have no idea which roads are even part of it. In fact it’s so obscure, that we don’t have any tagging for it yet.
Many are still in draft, or not started, because there has been very little disagreement. If there is disagreement, mappers in that state have the starting point to finish off. Often the reality is that the first person to write things down ends up getting to set the rules because other mappers shrug and say “seems reasonable to me”. It’s a fair bit of work and thinking to come up with these things and there’s little penalty for getting it “wrong” on a road here or there. The bulk of the debate centered around the exact (subjective) wording of what we considered a trunk. Getting the philosophy down (with great effort) was how we ended up nearly eliminating messy debates.
I do want to point out that we do have several “something else” candidates. You have administrative classifications, i.e. Interstates vs US Highways vs state highways vs county highways. There’s the National Highway System. We have Federal Highway Administration functional classifications. And, you can go by physical characteristics. There are probably one or two others that we looked at and rejected because they weren’t a good fit.
The laziest thing we could have done is to just pick one of these systems and map them to OSM. Interstate = motorway, US Highway = trunk, state highway = primary, county highway = secondary. Voila! No fuss mapping with no debates. Except, the result would have produced a (from our perspective) clearly wrong outcome that was incoherent between states and nonsensical to any data consumer trying to show a common-sense hierarchy of the road network.
I would encourage mappers in other places to consider these government systems carefully and use them only if they are a good fit for OSM.
And what would’ve happened is debates about which of the overlapping systems to align to, endless discussion, edit wars in the meantime. That’s kind of what happened after the TIGER import hewed to one of the official systems. No system is perfect.
If you seek a rational basis for road classification, you have to start from practical end user needs. Otherwise, what’s the point of road classification? The officials already have their own official maps of official classifications to work from.
The most important roads in a country’s system that aren’t motorways. (Need not necessarily be a divided highway.)
