Oh, Andy, my goodness; that old trick. My sentences are complete and grammatical. I recently had another volunteer read this thread and he had no trouble following it.

Inside it, the owners can and do fell timber. When those trees are down, they plant more trees. This is called landuse=forest.

Outside it, the owners can and do fell timber after they obtain a timber permit, then they often (and this is the reason the DID cut the trees) develop the farmland from felled trees into orchards, vineyards, greenhouse_horiticulture, apiaries and container farming (berries, tomatoes...). Some keep trees up to harvest acorns, dissolving out the tannin and making a kind of "oak meal" (I'll see if I can find the folks at the farmer's market who are awaiting for organic certification for that). Some live on the land while farming it, some simply farm it, living elsewhere. This is called landuse=farmland.

I tag what I observe. OSM doesn't have a good landuse tag of something like "live_on_farmland" which is what these largely are (though aren't exclusively, as some farm these lands without living here, I'm not perfectly sure which).

Please answer: Have I done something wrong with my tagging here? Or are you simply suggesting that I might have?

Please answer: Are OSM's landuse tag values too simplistic to well accommodate the complex kind of landuse this is? (Hint: I think so, but I haven't proposed a new tag, so I use the ones we have and try to supplement them as best I can).

Please answer: Can you see bees in visual imagery?

Please tag: the polygon(s) with better values if you actually know what it (they) should be and assert them.

How would you tag this if you don't have the time (or access) to micromap land uses within the observed area? Is there a way for me to map my observations without needing to conduct further research?

Continuing a trend around here, I've changed the shopping center's landuse tag from commercial to retail, as that is a more precise tag. (Commercial might be "office parks" or similar, but shopping centers like this one fit exactly the definition of "retail" in our wiki).

"Done for now" around here!

No, that is not what I am saying; it isn't local government, it is the property owners. Would you re-tag this area to something else, which would in effect say "I'm asserting away these property rights and agricultural activities by tagging based on what I can see, rather than what the owners of this property do or can do?" If so, I invite you to do so, but I can discover no such appropriate tags, so I have tagged as best I can — and continue to do so as I discover more distinct and specific agricultural activity in the area: another local example of the "first draft, then improve" ethos found throughout OSM. For example, there are many apiaries in the area producing honey, too, that I have yet to map: another agricultural activity on this farmland. Can you see bees in visual imagery?

It is not that something (like farming) "does not exist" (as you say). These property owners either can or do farm here. Either timber (though some may not have the "last permit"), orchards, vineyards, greenhouse_horticulture or other. Some owners also live on this property, making it "part residential, part agricultural," yet OSM doesn't have a landuse tag for that sort of "combined landuse." Perhaps better developing that is a strategy to "fix" what you perceive as a problem here (I don't see one, though I do see room for improvement to add more specific farming, so I do as I add it when I discover it).

It may be that you are being polite or conciliatory by extending me the courtesy of not knowing, but I look at your original message: you say it "looks" a certain way, you ask "is it really...?," you say the "imagery suggests," you say what YOU "would expect round here" (why is that?), you "guess," you say "probably," you say "the edges don't correspond to any change on the imagery at all," (yet the boundaries, like many in OSM, are both real and invisible) you "suspect" (more than once). Andy, I don't "suspect," I actually know, I live here, I visit with people here (they are my neighbors and friends) and I have been to these farms, vineyards, orchards.

I believe I have answered how this land is actually used, and why I used the landuse key value of farmland to tag how the land is used. You assert that what you see in visual imagery is the only thing that defines how the land is used (which would be much better captured with a landcover or natural tag, and I use natural tags when/as appropriate). I respectfully disagree with the assertion that visual imagery is always a or the definitive criterion for what defines landuse. I certainly am aware of the "map what is" and "on the ground" tenets of our project but "what is" here is farmland, so I have tagged it as such. Correctly. Thank you for your concerns, I hope I have addressed them.

Hi Paul: node/70170520 is in Georgia, so I don't know about that. And 6468638563 is indeed the Tesla Supercharger and it looks like you tagged it properly (as I suggested, but I'm not the ultimate authority on tagging, usually our wiki is).

Somehow it looks like you deleted the shopping center polygon (landuse=commercial) but that's OK, I added it back in.

So, it all looks OK now, thanks for your addition to the map!

Steve

I appreciate the time volunteers take to answer questions in OSM, DWG included.

Ian, again, these are not tweet-length topics, they require words. Either folks have the time to engage with me in dialog, or not. I do not wish to abuse that privilege, so I continue to better endeavor toward the concision you request.

"Concision" is a technique linguist Noam Chomsky talks about which is used by communication networks, propaganda arms of governments and others which often limit the speech that is able be uttered so that only a "party line" gets through. As I am more frequently repeatedly asked to engage in concision, that's how this feels: that fully answering the question is replied to with "TL;DR." I have taken ten years to make a better map. I believe somebody can take ten minutes to have a conversation with me to continue to do so into the future, especially when I am asked a question and expected to reply to it.

I did offer another venue (besides changeset comments) to have this discussion, as it obviously needs to be had. When and where that takes place, I will use words to do so and offer my patience to others who offer their words in return — it's the least I can do.

I am respectfully patient as I wait for Andy's reply to answer my questions of "Do you agree or disagree?" and "Does that answer your questions?" He asked me questions to address his concerns and I have answered them, I believe.

Thank you.

You are quite welcome, it is my pleasure to help new users.

If you are interested in "pictures," I encourage you to learn about Mapillary and OpenStreetCam. (You probably already know that using Google Maps and its "street cam" images are not to be used in OSM, as that would violate OSM's ODbL / open database license).

I'm glad you are adding local (to me, anyway) charging_stations to OSM, especially as 1) electric cars are the FUTURE! and 2) these are in my area, and I have an electric bicycle (I charge it with solar) and one of my cars is part-electric. (Another car of mine got 53 MPG when I bought it new in 1988!) Plus, Big Sur is an smart and awesome place to put a charging station: it's quite rural, gas is expensive (at Gorda, further south, gas is often over $7 a gallon) and it's a great place to have a burger, stock up on supplies, shop the nice gift shops or take a nice hike while you wait for your batteries to charge.

If you need any help "fixing the issues," simply ask and I'll be glad to offer help.

Enjoy your future mapping!

Steve

Ah, yes, thank you for pointing out that you DID enter a node where you say. (It is node/6468638563, which you can use a browser to see with node/6468638563).

OK, so you're almost done. 1) Please remove the name=Tesla Supercharger tag from the "surrounding polygon" (the one tagged landuse=commercial), as that isn't correct, it isn't the name of the whole shopping center. (Again, I don't know if it has one, but it's OK omit a name=* tag from OSM if we don't know the name of something).

2) The node you entered isn't really correct (yet), even though it has decent address tags (and a start_date). But in order for OSM to really "know" this node as a charging_station, you'll need to add those tags I suggested above to the node. The amenity=charging_station is the most critical one to add, the others are more minor "window dressing."

As I read our charging_station wiki (and you would benefit by doing so, too, especially as you might add more of these charging stations), I don't see any subtags that logically apply to the other "status" things you want to enter, like Permit Found. However, the "state=proposed" tag does a good job of describing that it is "on the way soon." Also, I suggest you look up "construction" at wiki.osm.org to see how we show that things are under construction, should that be the case. (It really is important to NOT use construction tags unless construction REALLY has begun). For things like "permits, etc." those aren't really relevant to OSM, but if you like (as you discovered before, and it was helpful to me), you can use either a note=* tag or a description=* tag to tag things like that.

Good luck, have fun!

Steve

Andy, these are not tweet-length discussions. They require more than a few words, so I use them. As I address your questions, but risk (or engage in) loquacity, I recognize I may try your patience, hence my early offer of an apology, perhaps better characterized as a warning of "lengthy reply ahead." Either way, it shouldn't take anybody "all day" to read it, that's hyperbole, really, and I'm trying to stick to facts and truth and fully answer your question, not engage in hyperbole.

You again assert that "OSM doesn't care about local authority zoning." In my opinion, this is patently false, given OSM's definition of landuse and how (in many, perhaps most cases) zoning logically maps essentially directly onto OSM's definitions of landuse. Not only is that a simple semantic truth, this has been true the entire decade I've been in the project. Even as our wiki definitions of landuse have "richened up" over the years, I have seen nothing that changes that, or even a discussion close to changing that (except this one, which I characterize as a not-entirely-surprising attempt/move towards such a rule change, as I've seen these before). However, it doesn't seem to be applied map-wide, but (in this case) appears to be singled-out toward me. Worldwide, OSM would be vastly more blank than it is now if contributors hadn't entered landuse polygons which accurately describe landuse. The mere fact these may (or may not) have come from zoning data is wholly irrelevant to anything; a total red herring. Why? Because they accurately describe landuse as OSM defines it.

In your original question (first Comment) of "Is it really all farmland in OSM terms?" My answer was " Yes, it is as OSM defines it." Do you agree or disagree?

Here (succinctly) are the differences between the two areas:

On 41172401, the owners of this property have rights to "timber production" (and as I mention, appear to have done so on about 15% of it, this area of felled trees is correctly denoted natural=grassland). They have obtained a state timber permit, which tips landuse into one best described with the OSM tag "landuse=forest."

The owners of the property described by the large farmland multipolygon (adjacent to 411772401) have rights to farm their property, and some do, especially after they fell trees which would interfere with what most consider farmland (harvesting trees IS agriculture, though OSM better tags this landuse=forest as noted above). However, some have NOT obtained state forestry permits to log the trees. 99 times out of 100 they WILL be granted these permits, should they apply, especially given the zoning of their land. Some do, and orchards are planted, or vineyards are started. Some don't, leave the trees, harvest the "downed wood" (without cutting, trees do "shed branches" in wood-productive ways) while in some, not all cases, also live in residences on the property in a residential manner. This is agricultural landuse of a certain kind, similar to forestry, but not accurately described by landuse=forest (no chainsaws felling trees), so landuse=forest is distinctly not applied. AND, the use of this land as "could grow orchard, could plant vineyards, could start greenhouse_horticulture..." is also extant here, whereas on other lands (which are zoned residential or industrial, for example), this is absolutely not the case, so this agricultural distinction is captured with OSM's tag of landuse=farmland.

(The maybe-1%-of-time a timber permit is not granted is if there is an endangered species in a creek below and the logging activity would create too much silt, for example. BTW, I know this as steelhead salmon spawn in my backyard ocean-flowing creek and State Fish & Game simply draws the line at timber production on these after a fish census is taken).

Does that answer your questions? Might I ask you to answer my single question above, please? Thank you.

NeverFollow: Welcome to OSM and your first edit!

Rather than mapping the entire landuse=commercial polygon (the shopping center area) as both the name of the shopping center (that's not correct) and the name of a yet-to-exist Tesla Supercharger station, I have some suggestions. You were helpful with your description=* text, so thanks for that.

First, take the "Tesla Supercharger" name tag off the shopping center polygon. I've been to this place many times (Target, Del Taco...) but I don't know if the shopping center has a name (the entrance sign simply lists the names of the stores). It might, but if we don't know the name of something in OSM, we simply omit the name=* tag. That's OK.

Second, use a node (sometimes called a "point") to map the charging station, rather than the whole polygon of the shopping center. For the node's location, the rather precise lat-long coordinates you gave seem to indicate it will be in the far eastern lobe of parking, in the area just to the south of the building south of Del Taco. What to put on its tags? I suggest:

amenity=charging_station
name=Tesla Supercharger
state=proposed
start_date=2019
capacity=14
charging_station:output=72 kW

That's a good start. I also recommend reading our wiki at amenity=charging_station to see if there are other subtags there that might be applicable.

Map well, have fun in OSM!

Steve in Santa Cruz

OK, I'm resigning this issue here and now.

I'll politely, respectfully and correctly answer the way/41172401 issue addressed. As that polygon (which abuts the large farmland polygon noted above) is properly tagged landuse=forest, OSM data consumers can correctly conclude that active timber production either is going on or could go on exactly in the area described by that polygon.

That such timber production activity isn't displayed in imagery seems largely to hinge upon assumptions of:

1) The imagery is recent and/or accurate, (which many times it isn't) and more importantly,

2) Imagery as evidence for how land is used isn't a sole criterion for landuse, as landuse describes land use, not land cover (which is what imagery displays) and land use includes how land can be used as well as how it is used.

A simple question should help make my point: if a swath of land in the real world is "active timberland" and yet visual imagery only shows some fraction of the land as "felled trees," should OSM not tag the full swath as landuse=forest? I believe it should: the land USE is forestry, even if "those trees I can still remain standing in imagery" haven't yet been felled. Should OSM reasonably be expected to "chase the chainsaws" with updated imagery, displaying the patchwork of fallen trees as they move through the forest? Or would it be better to tag the swath landuse=forest and accurately describe this as "active timberland?" I believe the latter, and our wikis seem to concur.

Quoting from our Landcover wiki: "One of the most difficult cases for landcover mapping is woodland/forests. See Forest for more details. There is not currently a good tag to describe a landcover of trees as opposed to a landuse of timber production for which landuse=forest is appropriate or natural=wood for primary unmanaged woodland. The tag landcover=trees has been proposed for this purpose. In addition, landuse and landcover are often confused. For example landuse=grass actually describes a landcover, not a use. This causes problems as one can not describe an area of railway land as being primarily covered with grass as the landuse tag is used for both purposes." So, please, let us candidly conclude that OSM continues to have problems here in harmonizing its syntax and semantics. (We do).

And, land cover is often expressed with natural=* tags, much less so with landcover tags (which I believe are still, technically a proposal). In fact, our Landcover wiki says that land cover is often expressed with the landuse=* tag, as "many types of Landuse imply a certain type of landcover." (And, "some do not").

It is apparent in visual imagery that the significant central portion of this polygon (perhaps 15% of the overall area?) has had some timber production, as it has no trees (unlike the surrounding forest), was created by me about five years ago as landuse=meadow and has been touched by user:bdiscoe. As noted in our county wiki (and database histories) user:bdiscoe made a large-scale "hit and run" style edit in this area, with a similar smearing of semantics between landuse and landcover (apparently in his quest to remain on some leaderboard of large-contribution OSM editors). He did not change tagging here, but it appears he added some nodes to improve accuracy, which is fine. However, as landuse=meadow implies some sort of agricultural sense (cattle grazing, hay cultivation...) and neither is apparent, I just changed the tag on this area from landuse=meadow to what I believe is a more accurate tag: natural=grassland (until such time as OSM has a sense of more accuracy). This continues a trend I began last year to more accurately change meadow to grassland where no agricultural use of the land is known.

Am I really to understand that my assertion in this changeset comment of "the data in this changeset have now become nonsensical and incorrect according to what OSM defines as a valid multipolygon" is responded to with: that's OK, these wholly incorrect data will "be dealt with eventually?"

I find that to be a most unsatisfactory response. In my opinion, this isn't a waste of time, it is about presently-incorrect data in OSM being called to the attention of their author so that they may be corrected, if not immediately, then posthaste. The author's response seems cavalier and unhurried, even as he embarks on an apparent campaign to delete similar data (his May 12, 2019 deletions).

These data were edited and left in a faulty state. That is all I assert, and I even offer/suggest a methodology to correct them. So, let's correct them. Now or soon as this edit was recent and its errors were newly discovered. Not at some hazily-defined future time when the author decides to to so "if I remember" or "eventually." I find that disingenuous and a failure at the fair and reasonable scrutiny of other OSM contributors.

I disagree with "nothing more needs to be said about it." This feels like a stifling of open dialog, the kind of dialog which OSM welcomes and frequently engages in to solve errors in the map. An error in the map is all we're talking about here. So, again, let's solve it. Sooner, not later.

Deleting erroneous data is one thing. Taking the necessary time to re-introduce correct data is another thing. Both are good things. However, editing, then uploading data so they are in error is not a good thing, I believe everybody in OSM agrees. (Yes, we all make mistakes, I do as well, this appears to be a mistake, let's correct the mistake).

These data are not in error because they are in dispute with the real world (that is one thing and can be harder to solve). They are in error because they are an erroneous data construct in OSM. IMO, these should be corrected as soon as possible.

Hi Andy: I apologize in advance for the length of this, it appears to me to be necessary. If there is a superior venue to continue this conversation (a talk page, a mailing list...), I invite you to point the conversation there.

"Looks quite odd," how, exactly? (That's rhetorical, you need not answer it. I ask it that way to make the points below).

OSM's "terms" (definition) for farmland say "An area of farmland used for tillage." That is an extremely broad definition, and technically makes even MORE area on Earth which is not so tagged in OSM likely quite deserving of this tag. Moreover, the definition of "tillage" in my dictionary is "the preparation of land for growing crops," making OSM's definition of farmland to be:

An area of farmland used for the preparation of land for growing crops.

That is rather precisely what this is: these lands are owned by people who prepare this land for the growing of crops, and in some cases already do (and in fact, there are landuse=orchard and landuse=vineyard areas superimposed where these are found: I know because I have either visited them, entered them, or both). These orchards and vineyards both have and do exist here (they come and go over the centuries, but they are both here and are allowed to be here). In fact, any of these lands are allowed to "use the land" like this, and hence we are speaking of "land use" which in OSM is (properly, in my opinion) captured by the landuse=* tag.

(BTW for many years I have owned land here — this is literally my backyard — and both lived on it and cultivated food/fruit, herbs and flowers here. True, my particular property is zoned residential, but as the source of many of these data are areas which are zoned "RA," meaning "Residential-Agricultural," they have both landuses. As OSM doesn't really allow "both" (a limitation in OSM's tagging schemes which we might eventually improve to capture this semantic?) we are left to "choose one," and in fact, some of these lands do vacillate between landuse=agricultural to landuse=residential and back again. If you truly know what's what — although, I do live here — and you wish to set this Zoning=A multipolygon to residential, of course you are welcome to do so, but it isn't RA, and I would assert that change as "less correct").

If you are "looking" (at imagery) to suggest how the land is used, this might be better dubbed "land cover." Land cover is "what you see," contrasted with land use, which is "how the land is or can be used." This is a frequently blurred semantic in OSM, and may be at the root of what appears to be our disagreement here (if it actually is a disagreement and not a misunderstanding). It may be that yet another re-definition of terms in OSM is what is not-so-subtly going on, as these polygons have existed in OSM for most of its existence with no (or very little, witness your concerns here) contention, misunderstanding (as they meet our wiki definitions, both landuse=farmland and as well-describe in our local wiki) or request of their multiple authors to examine (as you do), "is this really...?" Yes, it is as OSM defines it. And yes, we've even gone so far as to further describe these data as "capturing zoning with landuse (is a good first step)...then replacing these with (more precise data) correctly supplement them." Again, if you have more precise data, I invite you to supplement them. I have, by adding the orchards and vineyards I have and that process has been underway for many years. Finally, while I don't take credit for this (I stand shoulder-to-shoulder with many, in true OSM fashion), years ago (after my large-scale corrections of nmixter's import around v3), bestofosm.org awarded Santa Cruz a Gold Star Award, saying it had "nearly perfect landuse." That may not be saying much, but what I heard is that we're at least on the right track. Can improvements be made? Of course, they have, and they continue. It may be presumptuous to say this to you especially (I mean no disrespect), but as our landuse wiki now says today (and has for most of OSM's existence), "See Landcover for general information on tagging the physical appearance of land and Proposed features/landcover for a proposal to more clearly separate the concepts."

I don't understand why you would say that government zoning information is not useful to OSM, as zoning logically maps quite closely (if not exactly) to landuse — as OSM defines landuse. As described in our local county wiki, "capturing zoning with landuse (is a good first step)...then replacing these with (more precise data) correctly supplement them." I have always believed that this is exactly how OSM is intended to be "developed," and this is exactly how and why I have contributed tens of thousands of edits to do so. Have I done something wrong over the last decade as I have?

You say "if people want the zoning info they can go to the local authority." What OSM "wants" (apparently) are data which conform to the tags and conventions it documents in its wiki. So that is what the (northern California) community has provided to Santa Cruz County (and other counties, like Monterey, where I was involved in a farmland data import before OSM had import guidelines, this, too, is documented in our wiki if you care to look). They were (and are to some degree) a "first draft," intended to improve, and they have and do improve. Again: have I (and others) done something wrong by following our wiki documentation about landuse? (I ask this full-well-knowing that there are many in OSM who "wish" that OSM better contain/display something resembling "more accurate land cover, given the vast visual imagery resources OSM can access," yet that is a distinctly different topic than landuse polygons which have already entered OSM. Importantly, it doesn't make these polygons wrong.

If there actually is a trend in OSM to "convert" areas entered as landuse to something more resembling land cover, because of what can be seen in the relatively-recently-available wider use of imagery data, I welcome that conversation. But I do not believe it fair or correct to question the entry of perfectly valid and correct data as "is this really what you mean to enter?" when it is clear (from the persistence of the data and great pains to explain how they entered and why they remain and are supplanted as they are) that they are correct.

In 2014, after I spoke at SOTM-US on how I spent between 2011 and 2014 correcting the mess of national bicycle routes as then-entered (largely by banned member NE2) into OSM as WikiProject USBRS, I experienced the DWG contacting me as Paul Norman and Serge Wroclawski had lunch with me after that 30-minute presentation, discussing "proposed" (bicycle routes, in this case) in OSM. I was keen to listen and because I endeavor to create more light than heat, subsequently developed WikiProject methodology for "proposed" (e.g. "very high bar standard") which live on, years later. I say this not to boast of my accomplishments, but rather to show my earnest diligence to listen to changing trends in OSM. If there is a trend in OSM to "convert" what are perfectly valid landuse (multi)polygons now entered into what more closely hews to a land cover semantic, that's fine. Though I ask you be forthright about that and concur with me that exactly that sort of "we're changing the rules" (again) is what is going on. I largely find these larger-scale rule re-definitions to be disruptive to OSM, though I do "go along to get along," realizing that a certain amount of this change (or pain, depending on perspective) helps the data in the project grow towards more of what we meant to do, rather than what we actually did do. They put erasers on the ends of pencils, as we are human and do perform mid-course corrections.

But please do not let us pretend that doing so (changing the semantics of landuse to more closely represent land cover) isn't such a mid-course correction: it is (or would be) exactly that. It is NOT a misunderstanding on my part, nor is it or was it mis-tagging in a way that is "probably actually not very useful to OSM" (as you say). It is the way that OSM defined its tags at the time, as it does now and continues to do so into the foreseeable future until such time arrives that they are re-defined (as landcover or to more closely hew to that semantic) with consensus of the community. I don't see that conversation resolving itself anytime soon, though I do continue to listen (and witness this lengthy reply) participate in it.

After being edited in this changeset, relation/1015348 is made up entirely of inner members, a nonsensical OSM relation construct. In short, this national forest is now wrong not because of its data but because it was edited in OSM to become a relation that makes no sense.

Before deleting an outer relation member (especially using the edits-relations-clunkily iD editor), try:

1) Installing JOSM, a superior relation editor,

2) If the assertion is that the "forest area that was mapped wrong...and can be mapped more accurately" please find RIGHT data to replace it with. The source (at usda.gov) was properly tagged, remains an active link, and the shapefile.zip at https://data.fs.usda.gov/geodata/edw/datasets.php?dsetCategory=boundaries was updated as recently as last week. You will need to install into your JOSM environment the opendata plug-in to open the shapefile, but this both documented and not difficult.

Rather than deleting data (which you assert are wrong), please follow their source tag and improve them with more current, more correct data (asserted by our employees, the USDA, to be right).

I'm not sure where I got this (I sketched my understanding with a pen, leving my notes "source-less.") It was from maybe six months ago about how "orange" was being introduced into the VTA light_rail lines when BART SV arrives in the 2020s and it didn't call Orange MV->Winchester, it called Orange MV->Alum Rock. Baypointe becomes a "transfer station" as it serves all three (Orange, Blue, Green). This "truncates" Blue and Green at North 1st and Baypointe. (Orange is East-West, a bit SE towards Alum Rock), Green is (N-S, with a SW jog to Campbell) Baypointe->Winchester and Blue is (N-S with a SE jog to Santa Theresa) Baypointe->Santa Theresa.

It seems like you (or the signs? photo appears to be Baypointe...) say Orange will be a direct replacement for Green, while Blue will remain as it is today. Are we clear? (I honestly say no: I don't know which segments will make up Orange, though if VTA has chosen a single place to put Orange on a sign in prepping the future for BART SV, Baypointe station is a good choice).

We can look for more VTA Orange line signs or we can ask VTA to more clearly articulate its (BART SV route change plans), maybe both; that would be a two-pronged approach to get to the bottom of this. Sorry to be confused, it's like having two watches and not knowing what time it really is. Basically, "any help appreciated" and I'm looking around for answers myself (VTA's web site, BART's website, maybe...). I haven't tagged anything Orange...yet. Kinda wrapping my head around the future, really, as I'm surprised to see VTA putting Orange on light_rail signage this far before BART SV.

I add my voice to Mateusz that we use name=* as we document it our wiki. And to and to Martijn's that if there is a wiki that keeps track of (or does its best to do so, as there is in this case) something like the status of a transit network, a state's railroads, a collection/tables of motorways/highways at a political level, etc., that before sweeping changes are made, these wiki should be consulted, followed and updated when required. Thank you.

OK, I sorta get it. Thanks, Steve.

Much has happened since 2008-9 when SCC's Countywide Bike Plan (CBP) was adopted, including a lot of work in 2016-7 to revise the CBP. I've just taken a look at that (and probably should have a year or three ago) and also find a dearth of bicycle route numbers on many of the related documents. I'm not sure what's up with that, though if you are seeing "11" on signs in Mapillary, these network=lcn routes do "still" seem to be around. I'm going to spend some time poking around "what's up" with these, perhaps I'll have more to say later. Though this discussion is getting long; perhaps I'll private missive you with what I learn. Thanks again for "waking up" some apparently needed attention and good communication about updating various sorts of routes around here.

I've been "watching" as VTA prepares for BART and "going to orange," but yours (above) is the first confirmation that I've seen (either on a map, not a planning map, an actual route map) that there really is an "orange" right now. So, thanks for that, I did not know VTA had started doing either "orange" or putting colors on signs like they're doing. Thanks for your great communication!

(BTW, I agree with all you say about "planning documents" and unsigned_ref. True, dat.)

Thanks for your answer: ground-truth bolstered by Mapillary and OSC images. Wow, not only bike route signs, but BEGIN and END signs for them, too! (I'm impressed with VTA for being so complete with its signage).

I don't want to sound publicly contentious, but when you say you're considering removing routes which aren't signposted, that seems a very (overly?) strict application of our on-the-ground rule. There are plenty of routes (and boundaries) in OSM which are not well signed (or even signed at all). Does OSM really want to remove them all? Do we want a map devoid of routes (and boundaries) which aren't clearly marked on-the-ground?

Have we settled whether or not if VTA (or any government) "publishes" a route (in written form, for example, declaring its existence) but hasn't signposted it, that route doesn't belong in OSM?

What about VTA's light_rail routes? Those aren't signed "green" or "blue" along the train routes, yet we agree those routes with color designations should be in OSM. How do we know what color and composition are the routes? VTA publishes those, they don't signpost them on-the-ground. I wonder where the determination is made about routes (or boundaries) when they can't be seen on-the-ground, but "everybody knows" (due to the publication of geo data or public-domain maps) where those routes (or boundaries) are. Thanks in advance for sharing your thoughts.