This description here is severally lacking context in the mapping activities, and a person whom didn’t even map is attempting to just highlight how the ways were tagged, as opposed to the context of the mapping activity.
It is clear from the changeset that most of the changes are not directed specifically at the tag, but to bring significant improvements to the basemap in terms of data.
The level of detail was also improved to include various other parking aisles, driveways (as in service=driveway2), footways and various other items.
Fact: (Context) The way serves a closed neighbourhood and does not allow traffic to connect to other neighbourhoods. In other words, this is a destination type of access.
Tagging Opinion: This is not tagged as residential because of the above condition - no through access that allows access to other estates/neighbourhood.
Tagging Opinion: This is as driveway2 because it is a minor access road, mainly meant for access to those houses (as a destination).
Fact: A significant portion of those roads are actually newly-added, and not re-tagged as opposed to what’s stated. This is one example of how the perception of normal activities is casted as wilful damage in terms of data contribution.
Fact: (Context) This way is a service road in an industrial estate, which services as a link way to parking lots and other service roads.
Fact: service=driveway2 does not restrict its’ use to residential or business property, nor restrict it from being used to define service roads linking to other service roads. The service road meets the definition in service=driveway2 in two aspects - providing access from the highway to an area used for (1) parking and (2) accomodating motor vehicles. There is no problem using it for a nonresidential service roads as claimed.
Fact: (Context) Way is a service road that happens to go around the property.
Fact: There is no restriction on service=driveway2 for mapping property perimeter roads. What is the issue here that is being complained on is un-clear and cannot be responded to.
Fact: This is a minor service road that provides access from the highway that accomodates motor vehicles.
Author’s Opinion: Is this user trying to refer to the specific exclusion clause in service=driveway regarding “Paths in or around a parking lot … perimeter of the lot…”? Having such exclusion for a definition is indeed difficult to map for, as it challenges even advanced users.
Fact (Context): Both were minor service ways driving into other parking areas.
Fact: Both service ways meets the definition of service=driveway2 by being a minor service road proving access from the highway to an offstreet area for parking.
Author’s Opinion: One service way was originally not tagged service= (which can either be un-classified service way, or that specific exclusion clause in service=driveway). In any case, ambiguity is not good.
Author’s Opinion: The other service way is mistagged as service=parking_aisle, which only applies w.e.f. from the parking area.
In conclusion, still good improvements to tag that accurately describes the purpose of the service way.
Opinion: (Context) In Japan, most of the alleys are driveways serviceable by Kei cars or small <bicycle/motorised> vehicles (for deliveries). Most of the service ways are also located between properties, and does not really serve the back of a building. The choice to use service=driveway2 and not service=alley is because it is the primary access to the building, and not the rear entrance.
Fact: The way still meets the definition of service=driveway2 given that it is leading to an offstreet area able to accomodate motor vehicles (Kei cars).
Author’s Opinion: Technically nothing wrong with mapping it as service=driveway2 instead of service=parking_aisle, because the road itself does not function with the goal of parking but more as a result of allowing students to temporarily practice their “driving skills”. However, on second thoughts this can also be tagged using the other tags. However, do bear in mind that this is just one or two ways in the entire driving school’s circuit, and the mapping was done quickly as the main focus is not on the driving school but the bus routes.
Author’s Opinion: I think the expectations of users here are very high, given that this is a community mapping project using un-paid resouces and goodwill of the mappers. This seems like trying to nitpick other users for each and every node/way/relation added, and expecting users to know every bit about the Wiki’s complex definitions and tagging.
Fact: (Context) The service way serves some residential further down the way, and classifying as track is thus not accurate.
The point they seemed to be making on the wiki is that all highway=service should ideally have a service=*, which I don’t necessarily disagree with. However, as far as I can tell, they actually used service=driveway2 as a catch-all for any kind of service road that lacks an established service=* value – essentially, service=yes – and eventually took to using it on just about every service road they touched. (You can query Overpass for service=driveway2 ways before and since the DWG intervened.)
This is not accurate.
Fact: Service ways are tagged when they matched the definition of service=driveway2.
Opinion: The part where it is stated “…actually used service=driveway2 as a catch-all for any kind of service road that lacks an established service=* value” is an opinion. Nowhere in the Wiki entry defines it what you stated.
Fact: Wiki entry attempts to compare the definition to aid understanding, but ultimately the definition of service=driveway2 is meant to be Keep It Stupidly Simple (KISS) with a single line definition.
It avoids the mess that…
You must satisfy the definition of A; and
Where A has a definition that consists of 20+ lines; and
You must check that it is not B which results in C; and
Where C can still be A or D.
… which results in determination loop A and !B → C → A/D → C… (repeat)
For readers, A (service=driveway), B (perimeter ways), C (service=), D (service=unclassified)
@JaLooooNz snuck these tags into OSM incrementally over the course of three years, but that doesn’t mean we have to spend three years catching up to these edits.
This is again an biased opinion by using the verb “snuck”. Tags were not snuck in, but were actually discussed, documented and used with the Any Tags You Like principal. All tags were added when the entire area’s mapping is improved.
Just by having a different view from yours, some people are trying to define/mis-representing themselves as the global community (if it is truly global, these issues should never even pop-up).
Last of all, these users didn’t even contribute to any mapping unlike the evidence above, but yet is trying to subject their personal preferences on other’s mapping activities.
<This reply took 4 hours to prepare. Any further replies will need to wait till next week as unlike the people commenting on their opinion here, my replies are based on facts and evidence and I have already spent significant weekend hours responding to these effortless allegations.>
Since this user wants to highlight the tag use only picked-up after the dispute, I would want to highlight that this is not true. The tagging scheme has already been in use as early as 2014, just that it was tagged with service=driveway and not with the service=driveway2 tag (which is required to be used due to the definition change in 2019/2020). (=v1 and >= v1) Thus, do take care reading other’s view of the chronology regarding tagging prefences.
Your participation in this thread is not required. Re-hashing old arguments, accusing established, respected community members of bad faith, waving around ATYL like it’s a legal loophole, and leaving walls of text only serve to highlight further how absurd it is that we’re still talking about a single mapper’s quest to establish a tag that they can’t even define.
The only contribution we’re looking for from you is a one-sentence summary of what service=driveway2 means so that we can decide whether to remove it entirely or replace it with something. Since you keep side-stepping this very simple question, I’m pretty confident that the result will be a removal without replacement.
Claims you made are highly dubious, questionable and provided in large volume but often unrelated to reality. For example lets take a clear case of such problem:
To quote myself:
You still failed to either withdraw this claim or at least provide evidence that would make some sort of sense. I would advice to spend time on either of that rather than writing walls of text.
Or even better, answer
question.
As tag that is not definable and explainable is not useful, no matter how many procedural mistakes were made by people disliking it.
Although there are driveways in Japan judging from the descriptions in OSM wiki, Japanese speakers are not familiar with the “driveway” concept (I guess, for most Japanese, “driveway” sounds like a road very suitable for driving for fun, e.g. a scenic parkway). So I don’t understand the necessity to differentiate service=driveway & service=driveway2, and I’m rather doubtful about the necessity of tagging driveways in Japan (I mean tagging highway=service alone is enough).
This is just my personal opinion, but I am confident that Japanese mappers & data users care little about service=driveway2 in Japan, because it is not understandable and thus useless for them. And I will support the bulk removal in Japan because the tag brings hard-to-understand differentiation that is not necessary to the local community.
Anyway, I’d like to say thank you to @Minh_Nguyen for informing the Japanese community of this topic.
Seeing the OSM objects with that tag, those are “shared” service roads from residential/unclassified roads to houses or properties. And those roads do not have a specific purpose (undefined purpose). So I feel those are not even “driveway”, so I prefer the “highway=service alone is enough” approach.
May I try to answer? I went back to the original wiki page of driveway2, which gives some sort of definition (some sort only, because the page also talks about driveway, which is where the whole thing has a flaw, in my view): driveway2 is a catch-all value for all situations that feel like driveway but are excluded by the formal definition of driveway.
The whole point being that this may/might point at ambiguities in current definitions, but does not define a very compelling category.
And I’d like to tell the designer of this value that nobody questions his/her seriousness or dedication to OSM. Just the significance of this attempted new value. Not you; just the value. Maybe you can propose an improved definition for existing values, or a new attempt at a new value, avoiding the various pitfalls this one encountered.
It seems to be a catch-all, but the name appears to be deliberately confusing. Perhaps not a classic trolltag, but a troll-value.
I think we have spent enough time feeding this particular troll.
IIRC service=feeder has been suggested in the past for more “major” service roads, if this or another drop-in value is not considered by most to be a suitable replacement then I’d vote delete it as service=driveway2 clearly doesn’t have enough meaning to be worth keeping.
Given the apparent vagueness of the usage (which is so complex that it must be hidden behind enough obfuscating text that you could paper a wall with it) it is probably trying to cover too much and should really be two or three different concisely defined values. If that’s the case retaining the confusing tagging in the interim isn’t really an advantage as other highway=service without service=* would need to be checked for the new usage too.
I agree that it appears the best definition of “driveway2” appears to be “highway=service where no service=* values are adequate” which leads to the natural conclusion that, in accordance with the rest of the OSM schema, all instances of service=driveway2 should more properly be either service=yes or service=service (or removed, because we don’t really need to just write the word “service” many times to confirm that a road is indeed a generic or ill-defined service road and not just laziness).
I understand the urge to completely and unambiguously classify everything in the world with detailed tags, but at a certain point it’s just busywork. In fact I’d sooner err on the side of removing all cases of service=driveway entirely than continue splitting hairs about different minor paved surfaces.
Unfortunately for anyone in favor of “driveway2,” the OSM database isn’t a reflection of local or personal ways of defining English words, and when a number like 2 is used in a tag key it’s usually assumed to be a way of defining secondary values before the use of semicolons in values was commonplace (like address vs address2 or name vs name2). Even if there was a big need to highly classify driveways, which there isn’t, and even if there was consensus about what the word driveway should mean as distinct from the current meaning in the wiki, which there isn’t, adding a number to the end of an existing tag or value to modify its definition is the worst possible way to accomplish the stated goals.
I vote to either mass remove driveway2 since its utility seems very limited, or mass replace with a different generic value that seems to align with the user’s stated goals without undoing work, like service=yes or service=service.
If you meant addrN scheme, then it was bad example because it has clear use case and can’t be adequately described with semicolons as address tags are not expected to be contain lists of values.
I vote to either mass remove driveway2 since its utility seems very limited, or mass replace with a different generic value
Agree in general, but suggest to replace with service=parking. From what I see [1], [2], driveway2 is for service roads that go around parking lots. For me, that is either service=parking or service=yes (meaning tag deletion).
@JaLooooNzdriveway2 is quite misleading and non-explanatory name and should be replaced with something else. Do you have good use case that can’t be achieved using other service=* tags, primarily with =parking and =driveway? After that we can try figure out better name for this feature.
You’re correct @batyrmastyr , addr2 would be hard to merge into addr with just semicolons, I just mean that its meaning is “second address field compatible with and extending addr on the same element” rather than “another separate type of address with a similar but distinct definition and only one should be used at a time on this element” – we don’t use motorway vs freeway tags for example to accommodate regional naming, and we don’t invent cafe, cafe2, and cafe3 tags to accommodate different variations on the definition of “cafe” (which could be anything from an outdoor coffee shop to an indoor deli to a place where you can make a reservation and pay to pet owls and no food is served, thanks Japan!)
Confusingly, per wiki consensusservice=parking_aisle is intended only for navigable ways inside a parking lot only used to access the parking spaces contained by that way, and not the access ways in/among the parking lot which should be left as plain unspecified service roads. There’s basically a hierarchy inside of large parking lots where service roads let you get in/out/around the lot, and parking aisles are the lowest tier of road only used when trying to actually park. Even a single road with parking spaces used for no other reason than parking wouldn’t be a parking aisle since it would be its own main way in and out of the parking area (itself.)
We agree that the advocate for driveway2 seems to be suggesting that these parking lot access ways should be specified as service=driveway2 instead, as a sort of third tier in between a proper service road and a parking aisle, among other possible uses, but many others including myself are saying that such an arrangement would only add to the confusion and not fix anything, for many reasons already discussed.
A driveway is a minor service road providing access from the highway to an offstreet area used for driving, servicing, parking, or otherwise accommodating motor vehicles. << Mandatory Clause
A driveway is a minor service road leading to a specific property. ← Mandatory Clause
Driveway will typically lead to residence or business but may lead also to research institute, court, military installation, construction site or an abandoned property. ← Mandatory Exclusion Clause (other properties do not apply)
Use this tag on ways in conjunction with highway=service on a driveway, especially in any of the following scenarios: ← Mandatory Clause
Primarily provides direct access to a house, garage, or carport; a small number of parking spots may be available for occupants ← Mandatory Clause that contradicts with first Clause
Primarily provides space for picking up and dropping off passengers – also use covered=yes if it passes under a porte-cochère ← Mandatory Clause
When not to use service=driveway tag: ← Mandatory Exclusion Clause
Paths in or around a parking lot (amenity=parking) are tagged with highway=service without service=* on the entrance and exit ways, as well as any way that forms the “trunk” or perimeter of the lot, connecting multiple parking aisles (service=parking_aisle). ← Mandatory Exclusion Clause with no rational nor basis behind clause, and is not prescriptive
The delibrately confusing part only comes in because there are objections to generalising the service=driveway entry by removing specific problematic clauses. However, both definition (A) and definition (B) are driveways.
@JaLooooNzdriveway2 is quite misleading and non-explanatory name and should be replaced with something else. Do you have good use case that can’t be achieved using other service=* tags, primarily with =parking and =driveway? After that we can try figure out better name for this feature.
What you and a few others are trying to state is that there is objection to two tags having a similar namespace (service=driveway and service=driveway2), but different definitions. I think the proper term for this is naming confusion but disagree that it is misleading / confusing / non-explanatory.
I would just like to highlight the following use cases that there are various gaps that cannot be covered in the existing proposals, which is covered under “Service ways linking to offstreet area”…
Specific examples includes…
Service ways that forms perimeter of parking lot (as per today, it must be left un-tagged as required by service=driveway) ← Definition problem in service=driveway
Service ways leading to parking areas
service=parking can address this case to a certain extent
Offstreet areas that are not covered by service=driveway ← Definition problem in service=driveway that is valid only for selected types of property
i.e. service ways linking to park / recreation area / industrial areas, and any other offstreet areas not in that list.
Service way linking to other service ways or a batch of properties ← Definition problem in service=driveway that is valid for access to ONE property
This is a common occurrence in rural areas whereby the same service road services multiple properties
Service way only for the purpose of driving
i.e. driving school that was also highlighted previously
i.e. tourism / theme park area for driving only
Avoiding highway=service without service=* tag - allow the preceding condition to indicate that service way is not yet classified
Given the apparent vagueness of the usage (which is so complex that it must be hidden behind enough obfuscating text that you could paper a wall with it) it is probably trying to cover too much and should really be two or three different concisely defined values.
The tag is precise if you consider just the sentence definition, ignoring the comparisons which others demanded to know the “difference”. The “obfuscating text” is meant as comparison and contrast to other existing tag definitions, which is not really required if the definition is analysed standalone.
Shared driveways are documented in the wiki as pipestems and have three competing tagging schemes, all more descriptive than driveway2.
What you are attempting to do here is override the generally understood meaning of driveway with one from a niche legal (re)definition in one corner of the United States. How the Californian legislature twists the meaning of a word is not generally useful to a global project and is doubly useless to a project that prefers British English anyway. You have been told this before and a repeated refusal to respect OSM’s longest standing tagging principles comes across as bad faith. Yes, you may say it’s an alternate definition, but in that case we may as well call it service=motorway under a new definition of motorway as “anything a motorised vehicle can fit down”.
driveway
noun[C]
a private area in front of a house or other building onto which you can drive and park your car
…
driveway | American Dictionary
noun [ C ]
a short private road that leads from a street to a person’s house or garage (= building where a car is kept)
driveway
noun
: a private road giving access from a public way to a building on abutting grounds
… First Known Use 1845, in the meaning defined above
Which is a slightly broader definition, but doesn’t really cover service roads whose purpose is to lead to other surface roads.
If we need a point of distinction I’d probably say that per the Cambridge definitions parking on driveways is usually expected, but the service ways to these larger facilities do not (usually) permit parking on them, they lead to parking or other service ways types.
IMO the argument for driveway2 is like passionately arguing that something should be tagged as a duck because of the feathers, webbed feat and general beak shape, while completely ignoring the fact that the animal in question is actually a goose. Arguing about mandatory and optional clauses in a project that doesn’t document things that way is somewhat missing the point.
For specific claim that we should be following specific definition of driveway from specific chapter of a legal document from California, defined there solely for purpose of that specific chapter, and not intended to match general definition of driveway… see one specific section on this topic on OSM Wiki talk page
Also why California? Rather than another definition of driveway from another chapter of another legal document from say Alaska or Poland?
Thanks @InsertUser for this analysis! I ran an overpass query of ways tagged service=driveway2 which are connected to service=driveway to see how often service=driveway2 might be tagged as a pipestem in this way. However, it only pulled up 46 ways that matched this criterion, so it does not seem that this is the predominant usage of this tag.
I would too (as someone who was not following this forever, and thus has not yet entered this-stranger-is-attacking-my-pack-I-must-defend-it mode , so am still open to hear actual reasons for the tag) like to see @JaLooooNz simple explanation (instead of other people guesses) for what is the purpose of service=driveway2?
As far as I can understand from graphs on that wiki, it is supposed to mean “It is highway=service which was surveyed and found to NOT match ANY of other documented service=* tags”.
But that is just my guess, and could be wrong as any other. I’d thus like to hear confirmation, or (short and concise!) definition from @JaLooooNzthemselves (instead of current defensive-mode-grabbing-at-straws mode they seem to currently be operating from; which is actually quite understandable flight-or-fight mode of operation, which I’ll for the moment try not to prejudice against them).
That being said, I actually do see some use for tag having such semantics (as in my guess), as it would help distinguish between:
highway=serviceintentionally not havingservice=* tag, as none of specific ones match – e.g. service road leading to a gas station, or a service road leading to parking aisles (but not being service=parking_aisle itself) etc.
and
highway=service which does not have any service=* tag because it is not (yet) mapped in enough detail (e.g. someone might later micromap better and add service=parking_aisle or service=drive-through etc.)
For example, such distinguishing may allow apps like StreetComplete to ask for service=* details on all highway=service ways which miss service=* tag.
In any case, I agree that service=driveway2 is horribly chosen name and that this at least should be changed. I’d however suggest completely removing it only as a last resort if the actual tag use is not verifiable. Needless to say, @JaLooooNz could help greatly here by confirming if that (bolded above) simple definition of mine is correct interpretation (or if my interpretation is not correct, stating in unambiguous terms what is the actual simple (one sentence) definition of it), thus defining whether renaming or removing that tag is correct way to proceed (keeping it as-is seems quite unlikely, barring appearance of some extraordinary as-of-yet-unshown information).
If that what I’ve described above is indeed its function, it should be service=yes (as others have suggested too); and I’d also suggest adding several more-precise values for popular subgroups (e.g. service=parking_link etc.) to minimize the use of such too-generic service=yes (remember shop=yes?)
Your focus on avoiding dataloss is commendable, but you aren’t the first to propose this approach. Unfortunately, we’ve been trying to get an answer for over two years and don’t seem to be any closer to one, hence what amounts to a pile-on this time. We can continue to run out the clock in hopes of a perfect solution, but all that means in reality is that service=driveway2 will have become a de facto established tag, to everyone else’s dismay.
I would summarize the plea for service=driveway2 as “This is a driveway, too”.
In Nederland, a driveway (“oprit”, “inrit”) has variants.
Most are short private driveways to a single house or garage (“oprit”)
Some are slightly longer driveways eg to the back of a property (“inrit”
Some are much longer, but still lead from the road to a property (house or complex), usually starting with a gate (“oprijlaan”)
Some are shared driveways, servicing multiple properties (“gedeelde inrit”, “gezamenlijke oprijlaan”)
Some shared driveways turn into a private service road giving access to multiple regular driveways. That is when we no longer call it a driveway, but still a (private) service road, even if some access is usually granted.
This service road may be public access: still a service road, but definitely not a driveway.
Some have another exit: then it’s still a service road, because the main function is to give access to the driveways to the properties.
If such a service road gives access to a common parking, we will definitely not call it a driveway. It’s a service road (“dienstweg”) or access road (“toegangsweg”).
Long story short, I think it is fine for the Dutch situation to come up with more specific service=* values for the non-driveway variants of service roads. I don’t think they will gain much usage, but you never know.
I don’t think the value driveway2 will be seen as adequate for these variants of service roads.
BTW Nederland has the concept of “A driveway construction”, which means that a regular street, where it links to another road, has the construction of a driveway entry (sloped, table across sidewalk, different paving) and this has legal value, because traffic on driveways or roads constructed as driveways has to give way to the larger road.
, so am still open to hear actual reasons for the tag) like to see 