(Sorry to pile on in English, but this has been a fascinating conversation.)
Americans would agree with this statement too. Even in many rural areas, a controlled-access divided highway (motorway) would not be financially justifiable, but a limited-access divided highway (expressway) is still such a major upgrade that a map needs to show it more prominently than the alternative conventional roads.
The local community was only willing to adopt a functional definition of highway=trunk with the understanding that renderers would visualize expressway=yes at the same time. In North America, there’s a long tradition in print cartography of depicting the physical characteristics of an expressway or motorway as a double line, reserving the most prominent color for motorway.
By coincidence, the mappers who care most about this distinction also care deeply about realistic-looking route shields, so they are satisfied by OSM Americana’s implementation of both improvements, and OSM Carto isn’t a major impediment. However, in Italy, there may potentially be more reluctance to adopt a tag that OSM Carto doesn’t support because of that project’s different priorities.
Yes, highway=trunk is being applied to some highways, such as the strada di grande comunicazione Firenze-Pisa-Livorno, that the U.S. community would probably classify as highway=motorway without question. However, if we had to avoid highway=motorway because they don’t meet an even higher performance standard, but the highways of this type don’t form a coherent functional network, then we’d apply highway=primaryexpressway=yes. In relative terms, this is the approach that we’ve partially implemented for some local networks such as Santa Clara County expressways, though it’s still a subject of debate.
I agree, the American situation is not comparable to Europe and I don’t think we necessarily should align with what was chosen there. Distinguishing motorway like roads from others is quite significant for our context.These gaps in low zoom maps are not necessarily a problem rather they reflect the situation on the ground
Not just American, from what I see pretty much every country outside Continental Europe is using Trunk as a functional description. But in EU it works quite well as it is.
To be honest, those visual gaps are actually useful as they are, to know which areas in Europe are less easily accessible. Visually closing them wouldn’t give us more information: everyone already knows the network is dense and there’s at least one basic road reaching pretty much any place (excluding maybe some peaks in the Alps). Gaps don’t mean a place is not connected to the main network, they mean it’s going to take long to move around there…
Detto questo, tornerei alla proposta che avevo lanciato nel post #157, vorrei sentire le opinioni della comunità sul tema della continuità delle Strade a Scorrimento Veloce.
I’ve noticed Google Maps does something similar, albeit their criteria are different. They probably rank roads by traffic volume. This is more evident with the Traffic layer.
In general stubs are less noticeable because even at lower zoom levels Google tends to show what in OSM would be considered the Primary network, so “Trunks” are never really alone on the map.
Bing does something similar, but their low level render is quite poorly detailed so it’s of no use anyway.
Other providers as ViaMichelin and TuttoCittà show no stubs, their network appears consistent but it’s in fact wrong, they often draw corridors which make no sense.
I think this stubs issue could easily be solved in OSM by rendering the Primary network as well.
Ciao a tutti. Collegandomi al thread di Wilhem, vista la difficoltà nel definire regole comuni adatte a tutte le caratteristiche di strade presenti su OSM, mi atterrei a un approccio di pesi minimi, che tenga in considerazione le diverse caratteristiche che una strada assuma nel suo percorso, delineato da una indicizzazione comune o ricollegabile (es. su una strada statale, le diverse nomenclature SS11, SPexSS11, SR11 …).
Un tratto di strada qualificabile come truck, a parere mio, dovrebbe descrivere su un percorso continuativo dell’ordine minino dei chilometri, le caratteristiche comuni di una SSV: un limite di velocità medio/alto e costante, incroci sfalsati, zero ostacoli permanenti al traffico (incroci a raso con rallentamento forzato del traffico, rotatorie, semafori, attraversamenti pedonali, ecc.).
Già la presenza di più incroci a raso su una tratta di pochi chilometri, la presenza di zone residenziali o percorsi pedonali contigui, dovrebbe essere un indicatore di attenzione, che descriva un determinato tratto di strada come primario, al netto del fatto che la strada nella sua interezza, descriva per la maggiore una SSV, quindi con più tratti identificabili come truck.
La presenza di pochi incroci sfalsati, quindi, su uno stesso tratto che ne descriva altri anche a raso, non è un parametro sufficiente per identificare una SSV, dal momento che se ne possono rilevare anche in strade residenziali. Gli incroci con strade di servizio potrebbero essere esclusi dal calcolo, purché non creino un impatto evidente sulla circolazione dei mezzi.
Suggerirei di tenere il tag trunk solo se si tratti di incroci isolati, che non comportino un importante rallentamento del traffico. In caso contrario, se la strada nella tratta adiacente subisce un cambio evidente di caratteristiche, come una riduzione delle corsie per senso di marcia, o un abbassamento importante del limite di velocità, suggerirei il cambio al tag primario, in modo che dia evidenza già a colpo d’occhio del cambio di caratteristiche della tratta.