Re: [Tagging] Long Tail ( was Removal of amenity from OSM tagging)
On Mon, May 18, 2015 at 1:24 PM, Clifford Snow cliff...@snowandsnow.us wrote: On Mon, May 18, 2015 at 9:13 AM, Janko Mihelić jan...@gmail.com wrote: This talk by Richard Fairhurst suggests that 5% of mappers do 95% of work. So it's more important to find those few dedicated mappers and make their life easier, than to cater to the 95%. I find the proposition that we find ways to engage the 95% more compelling. The 5% have already figured out the system. Figuring out the system is not the entirety of the issue. This was the crux of a talk I proposed for a SOTM US a few years ago that was rejected- how we can more use our limited resources to get a better result in the map by focusing on (or growing) the top mappers, rather than on one off casual mappers. Sadly the politics of this view was just too radical for the talk to be accepted (as I was told later). - Serge ___ Tagging mailing list Tagging@openstreetmap.org https://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] Blatant tagging for the renderer: bridges abandoned railways
On Mon, Mar 9, 2015 at 4:03 PM, Bryce Nesbitt bry...@obviously.com wrote: The core problem is: railway=abandoned Refers to railway service, and does not describe what's on the ground. What's on the ground could range from a bit of residual lead arsenate herbicide, up through a highly visible gravel trackbed with bridges and culverts and bits of railway artifact scattered about. Is that the case?/ If so then I suggest that such objects be moved to a more suitable home where they're less susceptible to deletion, such as a database of historical objects. If there's nothing visible then such objects are likely going to be deleted at some point by someone who is walking/driving by and doesn't see the railroad. I've had this issue myself in NYC where there's no tracks, because the objects were mis-tagged as being on layer 0 rather than underground. I deleted railroad tracks that I could not observe. Apparently the tunnels still exist, so the issue was resolved, but I can imagine this becoming a source of conflict. I know that the railway community in OSM is very passionate. I'm wondering if there's not a better way to get their mapping needs met. - Serge ___ Tagging mailing list Tagging@openstreetmap.org https://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] Deprecation of associatedStreet-relations
It'll be interesting to see how the German community handles this as an excercise for other communities. I think that handling this in a local way is the right move. - Serge On Sun, Jan 25, 2015 at 8:57 AM, Jo winfi...@gmail.com wrote: Sometimes I remove stuff by accident realize it later and instead of going back just replace it, because I did more stuff in that time and don't want to lose my work. If you care about preserving the history of that object, you could do the following in JOSM: Create new layer Download data around what you deleted Merge downloaded layer into working layer You'll get a conflict and can retrieve the object by resolving the conflict. Polyglot ___ Tagging mailing list Tagging@openstreetmap.org https://lists.openstreetmap.org/listinfo/tagging ___ Tagging mailing list Tagging@openstreetmap.org https://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] Tagging Voting system- time for reform?
There seems to be conflation of this list as having some kind of administrative function. It doesn't. This isn't an OSMF working group, it's a discussion list, and as such there is no administrative function for this list beyond the boundraries of the voting process on the wiki. In OSM, official tags have no greater status than unofficial ones. If you'd want to change that, you'd need to change things in OSM at a far more fundamental level, and (frankly), I'd be very hesistant to see this happen. I do think there'd be value in some practical tagging cleanup- moving from 2-3 tags meaning the same thing to a single tag, through some agreed-on process, but I don't think this is the right forum for it. - Serge ___ Tagging mailing list Tagging@openstreetmap.org https://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] Tagging Voting system- time for reform?
Martin, Let me elaborate on what I mean by this not being the right forum. I agree with you that it should be. The problem is that in my time on this list- I've seen some pretty wacky ideas that go against what I think most OSMers would consider good tagging. I'd be worried about the results. - Serge On Sat, Jan 24, 2015 at 8:07 AM, Martin Vonwald imagic@gmail.com wrote: Hi! 2015-01-24 13:21 GMT+01:00 Serge Wroclawski emac...@gmail.com: There seems to be conflation of this list as having some kind of administrative function. It doesn't. This isn't an OSMF working group, it's a discussion list, and as such there is no administrative function for this list beyond the boundraries of the voting process on the wiki. In OSM, official tags have no greater status than unofficial ones. Fully agree. I want to quote one of our core values: OSM is not a hierarchical organisation; almost everything can be done without need for central sanction or even post-hoc approval. If you'd want to change that, you'd need to change things in OSM at a far more fundamental level, and (frankly), I'd be very hesistant to see this happen. If someone wants to change this and actually succeeds, I'll not be around here any more. And I guess I won't be the only one. I do think there'd be value in some practical tagging cleanup- moving from 2-3 tags meaning the same thing to a single tag, through some agreed-on process, Fully agree. but I don't think this is the right forum for it. Isn't it? Well, then at least it should be. Best regards, Martin ___ Tagging mailing list Tagging@openstreetmap.org https://lists.openstreetmap.org/listinfo/tagging ___ Tagging mailing list Tagging@openstreetmap.org https://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] Feature Proposal - RFC - addrN:*
On Thu, Jan 15, 2015 at 8:29 PM, Friedrich Volkmann b...@volki.at wrote: In that scenario, I'd much prefer to see two nodes, each with their address, and each tagged as an entrance. What you prefer certainly depends on your needs. Adresses on entrances are fine for routing, maybe for visual representation too, but if you want to run a script generating a list of building sizes and addresses, you need addresses on buildings. I'll explain how we dealt with this issue in NYC later in the mail. What benefit does this proposal have over simply using a list separator? A list separator is fine as long as there is only one key. Unfortunately, there is no simple addr=* key. The addr key can be used on its own to denote a complete address, without using subkeys. The subkeys are preferable in most cases since they can be used to construct the keys themselves, but are not strictly necessary. There is an addr:city=* key for the city, Is there a building in your dataset that lives in two cities? an addr:postcode=* key for the postcode Postcode's an interesting one- but again, in actuality, do you have a building that has two postcodes? an addr:street=* key, and addr:housenumber=* key, and others. These are the two we care most about in this discussion. And here's where we simply say: addr=val1;val2;val3 If you're in North America or a European country, it would be something like housnumber street name Let's say 123 Foo and 567 Bar addr:123 Foo;567 Bar We can omit the city because the city is the same (if you have a real life counter example, please show me) and we can omit the postalcode for the same reason. Here you do not need to count semicolons, neither do applications. You can check address for address. Which solution do you like better? Maybe I'm mistaken, but it's my understanding that this solution is to address the exceptions in the data. I live in New York City, and we do have some buildings with multiple addresses, so this isn't a theoretical for me. We already dealth with this. There will be some exceptions, but not many. Even in a city like New York City, with over a million buildings (litterally), the number of multi-addressed buildings which we In most cases, going back to your first question, the solution was to use naked address nodes placed inside the building polygon. You asked how one would retrieve addresses from a data processing perspective. The answer is that in these cases, you *might* want to use some kind of building relation, and then you'd have the building as a relation and the nodes inside it, but what we did in NYC is to simply add naked address nodes inside the building polygon. This adds an extra step in data retrival, but it's not that bad. As an aside, it may be useful for someone to create some kind of an API or SDK on top of OSM to make it easy to retrieve these broad categories of data which may be represented in different ways. - Serge ___ Tagging mailing list Tagging@openstreetmap.org https://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] Feature Proposal - RFC - addrN:*
The idea, if I understand it, is to allow for some arbitrary number of values for an address. That's an important goal as we increase the number of addresses in OSM. I do have some questions/concerns about this specific proposal. As I examine it, it serves one very specific purpose, which is a building with two addresses. In my experience, this setup is often (not always) associated with a building with two entrances, each associated with an address. In that scenario, I'd much prefer to see two nodes, each with their address, and each tagged as an entrance. The other way I see these entrances used in real life is that one business or residence within a building uses one address, while another uses a different one. Here again, a POI would be more accurate and easier to parse. This leaves us with the scenario with a building which has both addresses associated with any entrance. So here we essentially a list of values. To that end, I don't see why we can't use the existing OSM list value separator, the semicolon, so the address is: addr=val1;val2;val3 This is advantageous to data users because without this, they would have to look for N arbitrary tags, as in addr, add2, add3, etc. What benefit does this proposal have over simply using a list separator? - Serge ___ Tagging mailing list Tagging@openstreetmap.org https://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] Distinction between amenity=restaurant and fast_food
Folks, Reading this threat, I think I agree with problems brought up around amenity=fast_food It's very subjective, and if you actually look at the DB, you see that many times, people are mis-labeling fast food establishments as restaurants. I see this with McDonalds, and Burger King, specifically. The NYC government classifies restaurants differently. They have the cuisine of the restaurant, but then they also classify it as casual/not-casual. What do folks think of this as an alternative classification? - Serge On Fri, Dec 12, 2014 at 3:51 AM, Philip Barnes p...@trigpoint.me.uk wrote: On Fri Dec 12 2014 04:49:56 GMT+ (GMT), John F. Eldredge wrote: I would not be surprised to find some establishments where the emphasis is on food up through mid-evening, but the kitchen closes earlier than the bar does, leaving the final hours of operation to offer only beverages and perhaps some precooked snack food. -- Many foodie pubs in the UK operate in this way. Phil (trigpoint ) -- Sent from my Jolla ___ Tagging mailing list Tagging@openstreetmap.org https://lists.openstreetmap.org/listinfo/tagging ___ Tagging mailing list Tagging@openstreetmap.org https://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] Bitcoin: Distinction of purchase through website and cash register/Point of sale
Anita, Frankly a large majority of the bitcoin edits are unauthorized (and probably copyright violating) imports. Discussions of website allowing bitcoin seem to fall in that same category. There have been few complaints but I think it's inevitable that if the imports continue, someone will complain and the DWG would be asked to step in. So I'd say unless you visit a store and know for sure they take bitcoin there, leave it out. - Serge On Thu, Aug 14, 2014 at 6:53 AM, Anita Andersson cc0c...@gmx.com wrote: Since payment:bitcoin=yes is a de facto and used tag and since payment:website:bitcoin=yes is not, I would suggest a combined usage of payment:bitcoin=yes and payment:website:bitcoin=yes until the new tag is chosen by more mappers for their use cases. I'm considering using the combination for the moment so that backwards compatibility is maintained. ___ Tagging mailing list Tagging@openstreetmap.org https://lists.openstreetmap.org/listinfo/tagging ___ Tagging mailing list Tagging@openstreetmap.org https://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] Kindergarten, Childcare and Preschool
This is one of those areas where we are all just splitting the definitions in complex ways. What is a day care, and what is a real school? That depends on your definition of the terms. You can skip down to my proposed solutions if you already agree with me. If you don't yet, here are my reasons: 1. We can't use age, because age isn't universal across countries. 2. We can't use the length of the class/instruction/program because that will differ too 3. We can't use whether or not it's government run/offered, because in some European countries, they offer day care for infants 4. We can't even necessarily use the idea of instruction, because this is also extremely cultural, and even within a single country, you often find different ideas So we need a solution. I'm going to suggest two: 1. We keep the tags uniform but we localize the meanings Day care, nursery school, kindergarten. We know what these terms mean, so let's use them in whatever way is natural for the mapper. Benefits: This will be good in that it's simple and natural. Downsides: It's going to make data processing very tricky though, especially across areas. 2. We keep the tags uniform, but we localize the definitions in our editors An editor shows Kindgerten but the tags specify that it's a school, for age 5 (this would be the US definition). Benefits: It's easy to parse, easy and standard to define, easy to process. Downsides: It's going to be work to define this for each language, and it's not something we could automate. I've added a bunch of day care and school objects over the years and have no strong feeling about how they're tagged, so long as the tagging is relatively uniform. - Serge ___ Tagging mailing list Tagging@openstreetmap.org https://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] Feature Proposal - RFC - Tag:building=kindergarten, accept and document usage of this tag
Maybe this is a US-centric view, but in the US, a kindergarten is a grade of school. In other parts of the world, does kindergarten mean day care? - Serge On Mon, Jul 21, 2014 at 12:13 AM, Никита acr...@gmail.com wrote: Hello! This is second time I send email. First was rejected for some reason. Tag:building=kindergarten - accept and document de-facto usage of this tag. Proposal page: http://wiki.openstreetmap.org/wiki/Proposed_features/building%3Dkindergarten Tag description page: http://wiki.openstreetmap.org/wiki/Tag:building%3Dkindergarten ___ Tagging mailing list Tagging@openstreetmap.org https://lists.openstreetmap.org/listinfo/tagging ___ Tagging mailing list Tagging@openstreetmap.org https://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] Synonymous values in the shop key
Perfect, so now we have a reason to keep both tags! - Serge On Wed, Jul 16, 2014 at 3:18 PM, Steve Doerr doerr.step...@gmail.com wrote: On 16/07/2014 20:11, Serge Wroclawski wrote: there are no delicatessens in the UK. http://www.yell.com/ucs/UcsSearchAction.do?keywords=delicatessenlocation=united+kingdom -- Steve --- This email is free from viruses and malware because avast! Antivirus protection is active. http://www.avast.com ___ Tagging mailing list Tagging@openstreetmap.org https://lists.openstreetmap.org/listinfo/tagging ___ Tagging mailing list Tagging@openstreetmap.org https://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] Synonymous values in the shop key
On Thu, Jul 10, 2014 at 5:00 AM, Mateusz Konieczny matkoni...@gmail.com wrote: There are some values of shop key that seem to be synonymous with other tags and I want to confirm whatever it is true. Unless mentioned otherwise all mentioned tags are values of key shop, numeric values is occurrence count according to taginfo. Before - are values that IMHO should be replaced by value after -. fish (368 and documented as replaced by seafood), fishmonger (2106 and documented as replaced by seafood) - seafood (2110) delicatessen (108) - deli (4101, documented) A deli and a delicatessen are not the same thing. I'd say a delicatessen is a cuisine of restaurant, and a deli is a type of shop. - Serge ___ Tagging mailing list Tagging@openstreetmap.org https://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] Reviewing the use of addr:housename
On Thu, Jun 19, 2014 at 12:31 AM, Paul Johnson ba...@ursamundi.org wrote: On Wed, Jun 18, 2014 at 2:15 PM, Serge Wroclawski emac...@gmail.com wrote: Perhaps there's a call for a building name tag which may or may not be the addr:housename tag? I think that's called name=* Paul, if you'd read the actual instances of addr:housename that I provided earlier on this thread, then you'd have seen that the name field is already being used for the POI itself (as it should be). What we have in the United States are examples where people will put in the name of the structure, such as the mall that a store is in as the addr:housename. - Serge ___ Tagging mailing list Tagging@openstreetmap.org https://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] No abbreviations in names edge case
But I like the idea of having a key for the signposted spelling available. Of course I could bring up the fact that FDR Drive in NYC is spelled FDR Drive and F.D.R. Drive, and F D R Drive, depending on which sign you look at. :) - Serge ___ Tagging mailing list Tagging@openstreetmap.org https://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] Reviewing the use of addr:housename
I think there's an excellent point which is being danced around, which is that there's a conflation between a building's name and the addr:housename. Perhaps there's a call for a building name tag which may or may not be the addr:housename tag? - Serge On Wed, Jun 18, 2014 at 2:17 PM, Fernando Trebien fernando.treb...@gmail.com wrote: Reading from the wiki: This is sometimes used in some countries like England instead of [1] (or in addition to [2]) a house number. From that, I originally understood that one would use housename: [1] when a house number does not apply (when houses are identified by names/non-numeric codes, not by pure numbers) [2] as a generic field for additional addressing information (mostly as a substitute to addr:door and addr:unit, which never really took off despite being approved for 3 years now; this interpretation may be incorrect though) However, if you have a building (even building=house) with a name, you would have building=*+name=[building name], right? In Brazil some people have suggested that one would use addr:housename for a building's name when the element which is a building is also an amenity (e.g. building=yes+amenity=police), in which case the name tag would refer to the amenity, not to the building. I don't think this is the intended usage of name and addr:housename, but let me know if I'm wrong. On Sat, Jun 14, 2014 at 2:24 PM, Andrew Hain andrewhain...@hotmail.co.uk wrote: The uses of the tag addr:housename in the database[1] does not match documentation[2] well. A high prroportion of uses are accidental; there were some bug reports[3][4] against iD, which used to have a housename box at the beginning of every point for entering addresses, pointing out doubtful use by new mappers. The top 100 values of the tag tell their own story. -- Andrew [1] http://taginfo.openstreetmap.org/keys/addr%3Ahousename#values [2] http://wiki.openstreetmap.org/wiki/Key:addr [3] https://github.com/openstreetmap/iD/issues/1525 [4] https://github.com/openstreetmap/iD/issues/2124 Bloc 1 265 (empty string) 531 1 472 bloc 385 2 379 3 373 4 335 5 293 6 279 7 269 9 263 8 258 10 240 12 224 11 211 edificio 209 Mairie 206 16 202 15 192 14 189 18 189 s/n184 Edificio 179 Heidehaus 168 San Antonio II 164 17 164 Taman Cantek 159 Rathaus159 20 155 19 152 13 146 casă 146 22 143 21 139 Berg Studentby 138 25 136 26 135 23 132 Rose Cottage 130 24 129 A 124 28 122 Taman Ridgeview Phase 12 121 The Cottage121 27 118 The Lodge 116 B 111 31 103 29 100 32 99 30 98 C 98 Pfarrhaus 96 Garages95 Arcaden94 Haus 1 94 33 91 Nöhren Hof 89 The Bungalow 88 36 88 Edificio␣ 88 Haus 2 88 Vestergård 87 Casă 86 Lidl 84 34 84 35 84 D 84 Østergård 83 40 80 38 77 Ældrecentret Kærgården 77 Магазин76 C.C. Condado Shopping 75 37 74 CSI Warehouse 73 Bahnhof73 42 71 44 71 Березники 71 39 70 45 70 The Coach House70 41 69 Stevnshøj 69 Gemeindehaus 68 E 68 McDonald's 68 Golden Haven Memorial Park 66 Gemeindeamt65 Unit 1 65 43 64 Højgård
Re: [Tagging] [OSM-talk-be] generalized survey : proposed wiki update
On Sat, Jun 14, 2014 at 7:03 PM, André Pirard a.pirard.pa...@gmail.com wrote: Hi, Following this discussion here is a proposed clarification to Key:source http://wiki.openstreetmap.org/wiki/Key:source. The goal is to define the word, make date mandatory, use ISO format, define per source tag meaning. Is there any objection or suggestion for changes? OSM doesn't have mandatory tags. We work by consensus and agreement. You can't impose a tagging system on changesets. - Serge ___ Tagging mailing list Tagging@openstreetmap.org https://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] Bitcoin ATM (amenity=atm | currency:XBT=yes)
On Tue, Jun 10, 2014 at 5:36 AM, Janko Mihelić jan...@gmail.com wrote: What if it gave you paper money for Bitcoins? Would it be an ATM then? What if there's a shop that only works with bitcoins? Is it a shop? An ATM isn't a machine that gives you paper money, the term is broader than that. These semantic arguments are going to underly my broader point, later on in this email. I don't get how tagging something correctly is advertising. I'm not that stubborn, if people are against tagging Bitcoin ATMs as amenity=atm I'm fine with that. An alternative could be amenity=bitcoin_atm. Maybe there should be a vote. The core issue is two part: 1. The community process for tagging is one based on the understanding of regular person. Using dictionary definitions or quoting wikipedia is exactly why this is it's getting so much pushback. If you want to make some definition for a new type of machine that is a bitcoin machine- go ahead. Here's a suggestion for one: amenity=cryptocurrency_kiosk currencies=bitcoin;litecoin And your tagging problem is solved. 2. The bitcoin community has generally been skirting the rules Bitcoin mappers have been doing everything from copying other maps outright (violating copyright), to geocoding against Google and then placing that in OSM (violating copyright) to geocoding against nominatim and then using that (really bad quality data). A while back, when I would see suspicious Bitcoin data, I would try to contact the mapper and if it it was confirmed that it was bad- delete the data. Sometimes the users told me they didn't know anything about OSM, or OSM rules about what should or should not be on the map. Other were outright rude to me about it- saying that I was part of the banking conspiracy, etc. It's my experience, and the experience of many others, that the Bitcoin community overall (not everyone, but as a group) has been really uninterested in OSM as a whole, and has been just dumping things into the database in a way that is not only bad data, but is potentially dangerous for OSM (if there are copyright violations). I have a side project (which is currently on the back burner) which is able to show which Bitcoin data is highly suspicious of this kind of either copying or geocoding. I haven't deployed it yet, but I think if/when I do, it would show a very large percentage of Bitcoin data is either of low data quality or is copied or geocoded from another source. My conclusion (without having run the data) is based on limited data I've looked at, and Bitcoin mappers I've spoken with. I would like to see Bitcoin mappers to start collecting data like the rest of us- by hand and direct observation, then I think Bitcoin proposals would be less contentious. - Serge ___ Tagging mailing list Tagging@openstreetmap.org https://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] [osm_sk] Re: Aktualizace: Tags for Czech/Slovak address system
On Wed, Mar 19, 2014 at 6:30 AM, Petr Morávek [Xificurk] p...@pada.cz wrote: Oh, here we go again... You are wrong. It's nothing like addr:country, it's not duplicating any information, and the polygon approach is not applicable. I would really appreciate, if people would read on the local terms used in this import before adding their wisdom that completely ignores local conditions (aka truth on the ground). I did my best explaining at least the basics (in English) in my previously referenced email to imports list. If someone is too lazy to read that, then his/her claims and suggestions regarding this specific import will most probably be worthless (as is the above example). The only explanation I've seen is that It's too hard to write software to use admin boundaries. That is not a good reason in my opinion. Are there other reasons? You see a lot of people who are generally agreeing that the import as proposed is not good. Imports should in line with the community at large, and the community has concerns which are being shunted into strange discussion about addr:place vs addr:borough, addr:district, etc. An administrative boundary would solve this problem, and allow the import to be agreed on. Is it more work for a parser? Yes. If you don't want to do an import, you don't have to. There are projects like https://github.com/openaddresses/openaddresses which appear happy to take addresses outside of OSM, but I think that adding them to OSM has lots of benefits. - Serge ___ Tagging mailing list Tagging@openstreetmap.org https://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] [osm_sk] Re: Aktualizace: Tags for Czech/Slovak address system
My opinion is, as I've stated before, that having the boundaries of the municipal area in OSM makes sense, but having it in each and every object absolutely does not. This would eliminate the tagging debate, make OSM cleaner, make the objects more flexible and easier to manage in the future. It's a win all around. - Serge 2014-03-10 2:38 GMT-04:00 Andrew Shadura and...@shadura.me: Hello, I think it's important to add both tagging@ and talk-cz@ to the loop, as this question needs more serious consensus, in my opinion. I've left the original message below, just in case anyone wants to translate it from Czech directly. In short, Dalibor proposes to use addr:place and addr:borough as a more featureful replacement of addr:suburb, which I think is not very justified and is against common practices elsewhere. His proposed scheme is: addr:conscriptionnumber=220 addr:housenumber=220 addr:street=K úvozu ref:ruian:addr=28413113 addr:place=Lochkov část obce addr:borough=Praha-Lochkov městská část addr:city=Praha addr:postcode=15400 addr:country=CZ source:addr=cuzk:ruian The Czech term 'část obce' here, in my opinion, exactly matches what is 'a distinct section of an urban settlement (city, town, etc.) with its own name and identity, e.g. annexed towns or villages which were formerly independent', which is a definition of a suburb. However, addr:place is commonly used to define a part of address which has usage similar to the street part, but isn't related to the street. Speaking of addr:borough, the only difference I see between the proposed usage of it and what would be otherwise addr:suburb is that the official name of a municipal district may be not the same as the name of a locality. Dalibor, please correct me if I'm wrong. I wonder, do we really need to introduce new tags now and redefine the meaning of old tags, or maybe we can fit this into the existing model somehow? I think that maybe it's enough to have the districts and boroughs as properly tagged boundaries, and to have addr:suburb set the the official name of a municipal district, what do you think? Anyway, I'd like to also hear the opinion of non-Czech or non-Slovak members. On Sun, 9 Mar 2014 20:28:40 -0700 (PDT) Dalibor Jelínek chrab...@gmail.com wrote: Ahoj, dovolte mi přispět do diskuse a objasnit naše stanoviska. Předně ono je to o hodně složitější. Fakt. Zejména Praha. Viz http://cs.wikipedia.org/wiki/%C4%8C%C3%A1sti_Prahy Ale od začátku: addr:place jsme začali používat a úspěšně používáme pro malé obce a části obcí (tady je třeba říct, že část obce je termín z registru RÚIAN a znamená v lidských termínech prostě malá vesnice, co nemá místní úřad a patří pod jinou obec), kde nejsou pojmenované všechny ulice. Před nějakou dobou totiž začal Nominatim úspěšně hledat podle addr:place, když nenašel nic podle addr:street. Takže běžné adresy do malých vesnic jako je Libív 5 najednou šly najít, pokud měly addr:place. Teď se snažíme doplnit všechny adresy podle RÚIAN a tam používáme addr:place i ve městech, kde jsou používané ulice. A nevidíme žádný problém, protože addr:street máme pořád a navíc jméno části obce, což je ve městě rovno (podle nás i podle RÚIAN městké čtvrti). Jako bonus je, že se dá najít i dům podle čtvrti a čísla popisného, což je informace, která je uvedena v katastru. Jenže nad tím je ten zmatek s většími městskými částmi. Praha je sice extrém, ale ostatní statutární města jsou taky nic moc. Naštěstí v RÚIAN je to o maličko jednodušší: Tady je jedno pražské adresní místo: http://vdp.cuzk.cz/vdp/ruian/adresnimista/28413113 Do OSM ho přepíšeme: addr:conscriptionnumber=220 addr:housenumber=220 addr:street=K úvozu ref:ruian:addr=28413113 addr:place=Lochkov část obce addr:borough=Praha-Lochkov městská část addr:city=Praha addr:postcode=15400 addr:country=CZ source:addr=cuzk:ruian Rovnou upozorňuju, že neplatí, že by takhle podobné byly addr:place a addr:borough všude. Tahle adresa ale má ještě další vyšší celky (dle RÚIAN) správní obvod - Praha 16 městský obvod - Praha 5 A tady asi nastupuje addr:suburb, který by mohl mýt jedním z těch obvodů, ale spíše bychom potřebovali dva. Mohli bychom použít district, ale ten máme už využit jako okres a ani jeden z těch obvodů není okres. Borough je definován tady http://wiki.openstreetmap.org/wiki/Key:place a vůbec nám nevadí, že zatím není addr:borough, to můžeme později dopsat a adresní tagy většinou vznikají jako dvojčata addr:něco a place=něco. Navíc použití suburb v OSM je prostě blbě. Suburb je periferie, předměstí. Že ho OSM poutíbá ve významu městského obvodu, části je sice možné, ale my se nechceme přidávat k špatnému používání anglických slov. Mohli bychom použít quarter místo place. To by bylo asi logické, jenže ne z pohledu RÍUAN, kde place je část obce, což znamená malou ves na venkově a čtvrť ve městě. To bychom si v tom pak udělali pěkný hokej. neighbourhood už je
Re: [Tagging] Feature Proposal - Voting - opening hours holiday select
Indeed- I have no idea what summer holidays are. I know what federal holidays are, I know what some religious holidays are, but summer holidays isn't something I'm familiar with. - Serge On Fri, Mar 14, 2014 at 7:03 AM, Philip Barnes p...@trigpoint.me.uk wrote: How do you define summer holidays? surely on its own it is not helpful. It will require a database of when holidays are, based on location. A business which serves multiple local authorities will straddle all those areas holidays. Phil (trigpoint) -- Sent from my Nokia N9 On 14/03/2014 8:48 Martin Koppenhoefer wrote: Am 13/mar/2014 um 19:06 schrieb Pieren pier...@gmail.com: It's unclear if your proposal is opening_hours=SH(summer holiday) or opening_hours=SH (then you should correct the wiki because the tag template is using the first version) IMHO summer_holiday would be preferable because we should avoid abbreviations cheers, Martin ___ Tagging mailing list pier...@gmail.com https://lists.openstreetmap.org/listinfo/tagging ___ Tagging mailing list Tagging@openstreetmap.org https://lists.openstreetmap.org/listinfo/tagging ___ Tagging mailing list Tagging@openstreetmap.org https://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] Feature Proposal - RFC - opening hours open until
I've never seen a business have a closing time specified without an opening time. Can you provide some examples? - Serge On Mon, Feb 24, 2014 at 4:09 PM, Robin `ypid` Schneider ypi...@aol.dewrote: Good evening I would like to discuss one little detail in the opening_hours syntax. It is about facilities which do not have a open time specified (only a closing time). So I created a short proposal: http://wiki.openstreetmap.org/wiki/Proposed_features/opening_hours_open_until Feel free to add your thoughts to the discussion page. -- Live long and prosper Robin Schneider ___ Tagging mailing list Tagging@openstreetmap.org https://lists.openstreetmap.org/listinfo/tagging ___ Tagging mailing list Tagging@openstreetmap.org https://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] Feature Proposal - RFC - All You Can Eat
On Sat, Feb 15, 2014 at 7:57 AM, John Packer john.pack...@gmail.com wrote:
Actually, all_you_can_eat:type=* describes the *way* the food is served.
If the value is buffet, then people go to the food to get it; if the
value is rodízio, then waiters go around the restaurant offering samples
of food to each table;
I think type is the wrong word, and I hate subtags, so why not simply
serving={buffet|whatever...}
So now, what's the difference in /serving style/ between rodizio and dim
sum?
if the value is conveyor_belt, then people sit around a rotating table
which carries the food(probably always used for sushi); and so on.
Fine so far, but I cannot see a need for a separate
all_you_can_eat:opening_hours subtag when the normal opening_hours tag would
serve the purpose.
It should only be used for special cases. For example, if a cafe has an
all-you-can-eat happy hour every friday afternoon, then you might include
all_you_can_eat:opening_hours=Fr
14:00-18:00.
This all seems like too much microtagging..
- Serge
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] [Talk-us] Feature Proposal - RFC - Marijuana
On Thu, Jan 2, 2014 at 10:08 PM, Martijn van Exel mart...@openstreetmap.us wrote: I would look at how these places are already tagged in, say, Amsterdam. I know, I should know, having lived there for 20 years, but I don't :p Colorado will be full of coffee shops. - Serge ___ Tagging mailing list Tagging@openstreetmap.org https://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] Bitcoin and Online shops
Actually the problem is that most of these businesses adding themselves are doing more harm than good. 1. Many users copy data from Google - There was even a video on coinmap encouraging users to do so 2. There are many users who find the location via nominatim, which doesn't add accuracy to our geocoding, it just makes it worse 3. Many of the businesses are entered incorrectly. Overall, I wish coinmap had its own database. - Serge On Tue, Nov 26, 2013 at 4:04 PM, Janko Mihelić jan...@gmail.com wrote: Hi, Openstreetmap has been contributing to the Bitcoin revolution with this map: http://coinmap.org/ the problem is that lots of online businesses want to get on the map, and I don't know what tags to suggest. Should we invent something like office=online? Then it could be further specified with online:shop=clothes, online=pizza_delivery, or something like that. Janko ___ Tagging mailing list Tagging@openstreetmap.org https://lists.openstreetmap.org/listinfo/tagging ___ Tagging mailing list Tagging@openstreetmap.org https://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] Bitcoin and Online shops
On Wed, Nov 27, 2013 at 3:29 PM, Janko Mihelić jan...@gmail.com wrote: 2013/11/27 Jean-Marc Liotier j...@liotier.org Commerce, e-commerce... What is the difference nowadays ?Are there any activities left that do not have an online side ? I have a feeling you've got something there. office=commerce seems enough. The difference is that for OSM, we map feature we can see. We don't map PO Box address locations, for example. If a business has an office, then sure, map it as office=foo but it's not a shop. I think we need solutions to high density office mapping, though- and OSM as it is today is not a great fit. - Serge ___ Tagging mailing list Tagging@openstreetmap.org https://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] Proposal for new tag: landuse=plot
On Wed, Sep 18, 2013 at 6:22 AM, SomeoneElse li...@mail.atownsend.org.uk wrote: Lukas Hornby wrote: In particular defnition seems to be key and I can confirm my definition is from a British perspective. Which is fine, because OSM uses British English names for things except in rare cases. The rare cases include when a word means something different in British and American English. Part of the problem was this proposal didn't explain the proposal other than by using the same word as the tag, which left the reader to use the terms that they would use in normal speech. But just as OSM uses soccer instead of football, when there's a term conflict between British and American English, usually another term is found that's more accurate. Community garden is different in definition, both here and in the US (and elsewhere) but a useful comparison, as the ethos and values are usually similar. I still have yet to find a definition of lot. Can someone point me to one that is unabigious, from Wikipedia or a dictionary? Wikipedia's definition of lot is the same as my own: https://en.wikipedia.org/wiki/Land_lot (that is what comes up when you type land plot into wikipedia) And the term in usage: http://www.scoutingny.com/?p=3034 Despite searching the web, I can't find a definition to match your usage. In addition to the UK I've seen allotments in other places in Europe, but not in the US - does the concept even exist over there? We can't say until we know what the definition is, but my experience is that with a country that's as large and diverse as the US, it probably exists somewhere, whatever it is. Generally the OSM approach is map all the things! rather than map some of the things, making sure that everything is categorised absolutely correctly. That's not without its problems (as pointed out in the lack of concensus thread) but allowing people to add stuff local to them without necessarily worrying about correct tagging has got OSM to where it is now. Yes. That's the right way. It may well be that almost no-one has mapped allotment plots before**, which may mean that you get to pick some scheme that works for you. It'll almost certainly mean that there's no existing map that renders the data that you're interested in, so you'll get the chance to create that too. I've seen community gardens mapped. It may make sense, if they're similar, to hang off that tag. - Serge ___ Tagging mailing list Tagging@openstreetmap.org https://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] Proposal for new tag: landuse=plot
1. We do not map land lots in OSM, for reasons that have been discussed many times. 2. Even if we did, land lots do not talk about land use, which is what landuse is for. - Serge ___ Tagging mailing list Tagging@openstreetmap.org https://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] Ace Hardware
The difference between Ace Hardware and a company like McDonalds is that in McDonalds, you will have very strong branding associated with the franchise. A consumer goes into a store labeled McDonalds. But an Ace Hardware store is a complex beast, because the branding is all over the place, from being in the name (such as in the case of the store in Missouri), to being secondary to another brand- such as in the Maryland suburbs, where there are Strosnider's Hardware ( http://www.strosniders.com/ ). Each store prominently displays the Strosniders name and logo, employees wear the same uniforms, etc. They happen to be Ace Hardware storse, but as a consumer, I wouldn't know unless I was looking for it. - Serge ___ Tagging mailing list Tagging@openstreetmap.org http://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] Ace Hardware
Yeah, I think I like franchise too. cooperative is not really import from the consumer's standpoint. operator is good but separate, and yes there is overlap, just like there is with brand. Perhaps someone who likes to write will write this up as a proposal? - Serge ___ Tagging mailing list Tagging@openstreetmap.org http://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] Ace Hardware
On Fri, Aug 30, 2013 at 3:48 PM, Bryce Nesbitt bry...@obviously.com wrote: operator = Joe Smith, Proprietor. affiliation = ACE Hardware Corporation affiliation_2 = North East Hardware Cooperative This is exactly the kind of over-tagging I'd like to avoid. - Serge With ACE what's important (I suppose) is that the store probably carries a bunch of ACE products. Not to me. I don't think OSM is where we want to carry store inventories. We had that discussion on that list a year ago. What I care(d) about was that you can look up an ACE store by its ref id. - Serge ___ Tagging mailing list Tagging@openstreetmap.org http://lists.openstreetmap.org/listinfo/tagging
[Tagging] Ace Hardware
In the United States and Canada, we have a very large organization of hardware stores called Ace Hardware: https://en.wikipedia.org/wiki/Ace_Hardware Each Ace Hardware is independently owned and operated, and each of them uses their own name. But they all belong to a single organization called Ace Hardware, which provides them benefits in ordering, etc. Since each store is independently owned and operated, the operator tag does not fit, since the stores are not controlled by a central organization, such as would be the case with a franchise. And since each store is branded uniquely, and don't always use the name Ace Hardware in their name[1] or other material, I don't think brand fits either. But we may want to indicate that the relationship between store and co-op exists, so in this case, since they belong to a co-op, I've used co-op=Ace Hardware. If someone has an alternative suggestion for this type of arrangement, I'm interested in hearing it. - Serge [1] Sometimes they do, such as in the case of Springfield Ace Hardware (a real place), but two Ace Hardware stores near me are called Aquarius Hardware and AJO Home and Lumber, neither of which indicates any affiliation. ___ Tagging mailing list Tagging@openstreetmap.org http://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] Childcare Tag
On Fri, Jul 19, 2013 at 3:15 AM, Frederik Ramm frede...@remote.org wrote: On 07/19/2013 03:33 AM, Serge Wroclawski wrote: Of the solutions, I feel that calling it what it's called locally is preferable. Anyone who cares to compare across countries is going to have to parse the location first anyway. We've managed to handle creating definitions that we could use worldwide for pretty much everything else, including roads, sports, other schools, various amenities, etc. Have we really? I did say pretty much. :) (A quick count tells me that we have 7941 bars and 7209 cafes in Italy, while we have 3330 bars and 19319 cafes in Germany, this makes it seem likely that we do indeed use amenity=bar in Italy for things that would not be called a bar in Germany.) Personally I don't think that it would be terribly bad if amenity=kindergarten would mean something else in the US than in Germany, for example. (I *would* find it strange though if anything from elementary school upwards would be classed as childcare - in my mind, the focus in school is on teaching something, and the focus in childcare is on supervision. But maybe that's a cultural bias too?) Comments about the US education system aside, I think that you're touching on an important issue- which is that the tagging system is just there to provide a label. The human-readability aspect of it is nice, but it's not entirely necessary. Just as we don't really expect every single OSM user to understand the English that they're typing in as tags, we can't expect that these words mean the same thing- we just need to define the terms beforehand. And this seems like it's harder than it needs to be, since, as you say, at some point (most) children go to school- that's at around age 6. As for kindergarten, while the name may have an obvious German origin, my question is what the British definition of the word is, since it's British English that we use in OSM as our base language, and does that British definition differ from the US definition. - Serge ___ Tagging mailing list Tagging@openstreetmap.org http://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] Double and misfitting house numbers
On Thu, Jul 18, 2013 at 9:20 AM, Tobias cra_klinr...@gmx.de wrote: 1. Double house numbers: I want to tag the house number 101 of a shop. The building in which the romms of the shop are located has the house numbers 97,99,101,103,105,107,109. The house numbers of the building are either tagged on the entrances each or on the building-way (here: addr:housenumber=97,99,101,103,105,107,109) since the association to the entrances is unknown. Because there are several shops with the same house number, there would be several nodes or ways with the same house number. And only one is concerning the building/entrance. There are things I'm not sure I understand here. The example you show is a shopping mall, with the address tagged on the building as a set of housenumbers (as you say, 97, 99, 101...) But then you say all the stores share the same housenumber. I'm not sure I understand that, so maybe you an clarify it for me. Are you saying that all the stores are 97, 99, 101, 103, etc. or that one store, say the Garde bakery, is at 99, and the Schäfer's Backshop (you people sure do love your bakeries!) is at 103? If the stores have a housenumber- and you know that housenumber, include it in the store. If they're shared amongst all the stores equally, ie they all have 97, 99, 101, etc. then I'd tag the building. For those who would like to have an example from the real world - here it is: http://www.openstreetmap.org/?lat=53.101067lon=8.787526zoom=18layers=M Very useful to have an example... Another thing is that renderers are double-rendering those house numbers (which is better than the opposite) and since some rooms are added to a building (http://wiki.openstreetmap.org/wiki/Key:addr), house numbers are there not only twice. If each shop had one housenumber, then I'd remove it from the building as I added it onto the shop. The shop has the address 14-16 which means that none of the building addresses is fitting the shop address. Hence I would have to add addr:housenumber=14-16 which is like above in a way a double housenumber-tagging. It's okay if both objects have the address tag, because both are correct. The building I live in is mixed use. Some of the building contains retail shops, and then there's a large apartment complex (which is where I live). I've tagged the shops with the same addr:housenumber as the apartment complex, because that's the truth. Google is confused about this and thinks I live in a store, but Nominatim simply asks if I mean the apartment complex, or the shops.[1] - Serge [1] I'm anthropomorphizing a bit here. What really happens is that Google returns a single value for the address, while Nomatim returns multiple values. ___ Tagging mailing list Tagging@openstreetmap.org http://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] Double and misfitting house numbers
On Thu, Jul 18, 2013 at 1:05 PM, Tobias cra_klinr...@gmx.de wrote: It is kind of both. So in the end you will have the 101 on the building and the shop. That seems okay to me. We don't map to the renderer, so if it's both, it's both. If you feel strongly it should only be on one, let it be on the shop, and if two shops share the same housenumber, so be it, because it's true. The building I live in is mixed use. Some of the building contains retail shops, and then there's a large apartment complex (which is where I live). So there are two separate buildings? No, it's one building, but I live in New York city, and buildings here are often a mix of retail and residential, with retail on the bottom floors and residential on the top floors, and they share the same housenumber. I've tagged the shops with the same addr:housenumber as the apartment complex, because that's the truth. So you are preferring to double-tag each house number. In your example I would say that you do not need to add house numbers to the shops because they can be obtained from the building way. I don't like complex relations, either as a mapper or as a tool author. Working with them as a mapper is a pain, and writing tools that understand them is also hard. Google is confused about this and thinks I live in a store, but Nominatim simply asks if I mean the apartment complex, or the shops.[1] Does Google use OSM Data at you place? At my place I guess google bought the right to use those data from the local municipal authority or somebody else. Google does not use OSM data anywhere, but in the US (where I live), Google has shop data and when I put in my address, it geocodes it to a shop. - Serge ___ Tagging mailing list Tagging@openstreetmap.org http://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] Childcare Tag
On Thu, Jul 18, 2013 at 9:11 PM, Bryce Nesbitt bry...@obviously.com wrote: Of the solutions, I feel that calling it what it's called locally is preferable. Anyone who cares to compare across countries is going to have to parse the location first anyway. We've managed to handle creating definitions that we could use worldwide for pretty much everything else, including roads, sports, other schools, various amenities, etc. One of the beauties of OSM is that the tags are relatively unified between nations, where they're not, it's usually just because a certain feature is highly localized. I'd be very sad if we threw away so many years of international cooperation and consensus and I don't understand why these tags can't be defined in the same way other tags are. - Serge ___ Tagging mailing list Tagging@openstreetmap.org http://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] Childcare Tag
On Tue, Jul 9, 2013 at 1:13 PM, alyssa wright alyssapwri...@gmail.com wrote: Yes, people keep saying that. But as a new editor, my inclination is to use approved tags and not ones that are in the proposal stage. This is consistent in what I've seen anecdotally with new members. This tag has yet to go up for a vote. How can it go up for a vote? But then why even have a voting process if you're saying extensibly saying it doesn't matter? Because some people like voting. Some people like bureaucracy, and rules of order, and all that, and so we have one for them. What kind of free-for-all would it be if we didn't have room for those whose idea of a good time is having a lot of structure? - Serge ___ Tagging mailing list Tagging@openstreetmap.org http://lists.openstreetmap.org/listinfo/tagging
[Tagging] The value of the list (was Observations of the use of the diet: tag)
On Wed, Jul 3, 2013 at 6:25 AM, Pieren pier...@gmail.com wrote: On Wed, Jul 3, 2013 at 5:14 AM, Serge Wroclawski emac...@gmail.com wrote: My experience with mapping has been that after talking on the tagging list, and being voted down, when I just went ahead and used my tags, they were adopted by the community, on more than one occassion. Yeah, you can just ignore the feedbacks. Then we end up with the mess of tags like power=substation or shop=bakery or designation=* which are completely misinterpreted in different countries. Then it takes years until we can clarify the situation (if it will happen ever), thanks for the guys ignoring the talks or votes (which are more opinion polls)... OSM data is a key/value store. What people decide to put in those key/value pairs is up to them. What I've found, over years of participating, is that: 1. This list is a small subset of OSMers. It doesn't represent many of the supermappers, and it doesn't include editor authors or renderer people either. In other words, it's a small, self-selected group of people who are spending a lot of time talking, or arguing, in an echo chamber. 2. This list's idea of good tags differs from the OSM community at large. Most OSMers dislike complex schemes, and will avoid relations when they can. But relations are quite common here. The same goes for colon tags, which are heavily proposed (such as in the diet proposal) but not often used by the public except in very limited circumstances (addr). 3. This list often ignores usage If two proposals are up for discussion, there seems to be little or no weight placed on existing usage vs this list's idea of correctness. There is value in having a place to discuss issues of a tagging question, or problem, but I fear that this list isn't it. If people on this list wanted to do more community work that wasn't mapping, there would be tremendous value in going in to the wiki, finding the tags that are in use but not documented well, and expanding, or translating those pages. That would be a useful exercise, and I would participate. But right now, my view, and my advice to others, is generally to go out and map.[1] - Serge [1] This is my advice to individuals doing individual mapping via manual survey. As it relates to imports or automated edits, I have very different views. ___ Tagging mailing list Tagging@openstreetmap.org http://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] Observations on use of the diet: tag
On Tue, Jul 2, 2013 at 1:59 PM, fly lowfligh...@googlemail.com wrote: Please, tell me the difference between gluten_free=* and diet:gluten_free=* and document them in the wiki. The difference is that one is the that was in higher usage than the other. * The wiki reflected *actual usage* of the gluten_free tag, of which there were over 2x (nearly 3x) more than there were diet:gluten_free, until your mass automated edit. We are talking about 50 objects. So are you saying you visited the 50 places, or are you saying that if I have a scheme I prefer, then I should go ahead and fix up any tagging I like? That is not how this project works. That means the wiki was accurate. No, as this pages should have never been created. As diet:gluten_free=* was approved over a year before the other pages where created. I did not find any discussion about these tags on this list nor any proposal. All I did was to move the pages to a proposal namespace and highlighted the approved tag to use. The wiki should reflect usage. And until your automated edit- gluten_free was in higher usage by nearly 3x. Please revert your edit, and undo your wiki changes. I did just add one tag to the objects with the same meaning. Thought that is the way to silently move from one tag to another. I did not delete anything. What is your problem with diet:gluten_free ? Did you not see my email when someone asked about transitioning tags? I went into depth about that process. It is not right to go around making automated edits. The fact it's a small number is relevant only in that it didn't effect many people- but I'm one of those people. Sometimes we use bots when there is large scale consensus, or when there is an obvious typo (if I had typed diet:gluten_freee, then that might be a good reason to change things). But this is not the same thing, and I'm asking you to either go ahead and check the objects yourself, or please remove your change. - Serge ___ Tagging mailing list Tagging@openstreetmap.org http://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] Observations on use of the diet: tag
On Tue, Jul 2, 2013 at 3:34 PM, fly lowfligh...@googlemail.com wrote: I have no problem in reverting my changes but please give me one reason why you can not live with diet:gluten_free=*. I do not want to revert my changeset now only to have a bot redo it in a week. I don't know how long you've been with the project, or in what form, but something that always concerns me is this question of the role of the tagging list, and wiki. Is the wiki descriptive or proscriptive- that is does the wiki reflect the mapping that's done, or some idea of what should be done? My experience with mapping has been that after talking on the tagging list, and being voted down, when I just went ahead and used my tags, they were adopted by the community, on more than one occassion. And eventually, when enough people used them in practice, the wiki reflects the usage of the tags, because by that time, editors already used them, despite the issue of voting. If someone chooses a different tagging scheme than you, or a different one than what's been voted on[1], they're free to do so. Time bears out those differences. You asked what I don't like about the diet tags, and my feeling about them is that they're complex. I don't see the value in using diet:gluten_free where gluten_free works just fine. And I'll continue to tag objects that way. And others will tag items the way they like, and eventually, one of those ways will have a clear majority in the OSM database, and then it will be obvious which one is accepted by the community. If people love diet:gluten_free, then I'll switch. But we're not there yet. - Serge [1] And it's important to realize that the votes on the wiki are a subset of the list, and the list is a subset of the OSM community. The real votes happen when people choose to use one set of tags vs another. In other words, I will follow taginfo before I will follow the wiki. ___ Tagging mailing list Tagging@openstreetmap.org http://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] Observations on use of the diet: tag
On Mon, Jul 1, 2013 at 11:32 AM, fly lowfligh...@googlemail.com wrote: * I did add diet:gluten_free= to all objects with gluten_free=. * I did add diet:lacto_free=* and diet:lactose_free=* * I did add links to diet from both pages and moved both under the proposal namespace and marked both for deletion * Maybe we could delete lactose_free=* right away as its usage is zero. Anything else todo ? Fly, I can't speak for others, but I'm deeply bothered by this set of changes you've made, and I'm bothered for several reasons: * This appears to me to be an automated edit (since I doubt you went around the world looking for every object). I n Unless you specifically went to New York City, and visited at least one of the places I tagged as gluten_free, then you're making assumptions. * The wiki reflected *actual usage* of the gluten_free tag, of which there were over 2x (nearly 3x) more than there were diet:gluten_free, until your mass automated edit. That means the wiki was accurate. Please revert your edit, and undo your wiki changes. Thank you. - Serge ___ Tagging mailing list Tagging@openstreetmap.org http://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] About url Key
On Mon, Jun 24, 2013 at 1:44 AM, Taichi Furuhashi tai...@osmf.jp wrote: Hi OSM tagging ML I'm working some Mapping projects in Tsunami affected area, JAPAN. In this week end, we will held a mapping party with local people. Just I have two simple questions about url key. http://wiki.openstreetmap.org/wiki/Key:url (I know website key for Official web site.) 1) Can we use url key for general(non ODbL/CC BY-SA) website? ex. Google+ Local url=https://plus.google.com/109010197644072797677 Usually the url tag is used for an organization's website. For example, many musems, libraries, restaurants, etc. have an official website. That's what the website (or url) tag is for (I think website is actually preferred but url is just as good). I'd generally not use a Google+ page for a site, just like I wouldn't use a Facebook page, or a MySpace page, or a Twitter account, unless that was the only site that exists, and is curated by the organization itself. Does Tokorozawa control that Google+ site? Doesn't the city have its own website, like http://www.city.tokorozawa.saitama.jp/ ? In this case, of a prefecture, I'd suggest using the wikipedia tag, and putting in en:Tokorozawa,_Saitama (obviously change the wikipedia link to the relevant one in Japanese). 2) And also, if that shop has other evaluation sites like the Michelin, tabelog and more. How to divide those information? Those generally aren't linked to. If we did, we'd be inundated with links for every site in the world who does reviews for nearly every POI on the map. - Serge ___ Tagging mailing list Tagging@openstreetmap.org http://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] About url Key
On Mon, Jun 24, 2013 at 9:03 AM, Taichi Furuhashi tai...@osmf.jp wrote: Hi Serge and all: Thanks for your opinion. I want to confirm about just one thing. Question 1 means, Can we use url key to non-ODbL/non-Compatible license website? Should i talk to legal-talk ML? Short answer: Yes, you can link to anything. But please try to link to the entity's own website in preference to its Google+ page, or its Facebook page, or MySpace, or other third party sites if possible. If it doesn't have one, but it has a Wikipedia page, use the Wikipedia page. - Serge ___ Tagging mailing list Tagging@openstreetmap.org http://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] How to tag apartments in a building that is multiuse
After discussing it on IRC, how does this sound as a potential for a proposal: A new tag called residential, where the value specifies the type of residential, such as: residential=apartment residential=condo residential=co-op residential=single_room_occupancy (these would be open for refinement and addition of course). Then a user could tag the apartments with all the normal tags one would expect- name, addr:*, phone, website, etc. - Serge ___ Tagging mailing list Tagging@openstreetmap.org http://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] How should I create a proposal to alter (not create a new) tag?
On Sat, Jun 22, 2013 at 6:58 PM, Bryce Nesbitt bry...@obviously.com wrote: The proposal/voting instructions at: http://wiki.openstreetmap.org/wiki/Creating_a_proposal Are mostly about inventing new tags. What's the procedure for, and how can we document, a voting proposal for reorganizing things instead? This is quite a difficult process in OSM, and generally not done. Here's the general process: 1. First, see how the existing *use* of the tags are. In general[1], use wins out over anything else. So if you have a tagging that you think is better, use it, and better yet- show how the system is used by others already. 2. Use your new tagging system. If you have a lot of data, the best thing to do is to just use your own tagging system, and let the process be damned. 3. Get your scheme adopted by the renderers and editors This is where many tagging systems fail. There is a misconception on this list that being on the wiki makes a tag official and thus the renderers and editors must adopt it as documented. The reality is that the editors and the renderers use what's already present on the map. The key to getting your tagging system adopted is largely simplicity. Many of the proposed tags fail because they're too complex. If it involves a relation, it's generally overcomplicated. If it has multi-level tagging, ie: foo:bar:baz=alice, then it's generally overcomplicated. This process of changing a tag in the tools can be a long one. If you feel very strongly about this feature being shown on the OSM default stylesheet, you may have to patch it yourself. Same with the editors. Alternatively, in my experience, just wait for the adoption. 4. Mass-retagging Mass-retagging is generally frowned upon, strongly. This is something the DWG handles frequently. It's far better to use your own tagging system and not worry about someone else. They may be curating their own dataset for their neighborhood, city, country, etc. so if you just go in and wipe out their work, it can be very disruptive. Generally this is why tagging systems will be left around even when they've been deprecated. If you had to have a mass rename, and the tools all supported it, you might get some backing if you proposed it like you did an import (especially if you did it within a region like the US, vs worldwide). But generally, it's better to have two tags that mean the same thing, then try to go through the effort of retagging. - Serge ___ Tagging mailing list Tagging@openstreetmap.org http://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] How to tag apartments in a building that is multiuse
On Sun, Jun 23, 2013 at 1:25 PM, Bryce Nesbitt bry...@obviously.com wrote: Serge; So you mean for an existing building outline tagged: building=retail You might recognize the residential separately: building=retail residential=apartment;condo No. building=retail doesn't make sense for a vast majority of buildings in NYC. Most of New York City is multi-use buildings. And many buildings in large cities are the same way. building=office makes sense when you know the entire building is office space, but if the building is multi-use, then we can simply say: building=yes and then add appropriate nodes, which in this case, would include residential=[SOME VALUE] - Serge ___ Tagging mailing list Tagging@openstreetmap.org http://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] How to tag apartments in a building that is multiuse
On Sun, Jun 23, 2013 at 2:45 PM, Bryce Nesbitt bry...@obviously.com wrote: building=yes and then add appropriate nodes, which in this case, would include residential=[SOME VALUE] Then we need many or all of: building=yes commercial= residential= hotel= amenity=parking man_made=tower shop= No, you seem not to understand what I'm saying. I'm saying that we tag the building as building=yes Then we make nodes for the things inside it, *which is already what's done*, but I'm suggesting adding a new type of classification called residential. This *could be* used on a building directly, but does not have to be, and by not doing so, being more flexible with these multi-use buildings. - Serge ___ Tagging mailing list Tagging@openstreetmap.org http://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] How to tag apartments in a building that is multiuse
On Sat, Jun 22, 2013 at 5:31 PM, Clifford Snow cliff...@snowandsnow.us wrote: Like many large cities, we have apartments that are on the upper stories of multi-use buildings. The lower level(s) usually contain shops and business services. This is nearly every single building in New York City, so I'm quite familiar. This is a photo of the view from my apartment of buildings build similarly to mine: http://www.emacsen.net/view.jpg The buildings there are wonderful illustrations of the complexity I see here. On the ground (and basement) floor there are stores. The far storefront is a bank, the middle is a bakery, and the one closest (on the far building) is a multi-story pharmacy. Then you have some second story occupants- there's a school in one of the buildings, for example. Then the center (the tall part) are apartments, with the entrance in the center. And there's a rooftop terrace of the third floor, accessible by the apartment building. There really is no good OSM representation of this kind of building. My thoughts are 1) tag the building as an apartment building, and just add the businesses and 2) add a node as an apartment, but I'm not sure how to tag it in that case. I've struggled with this myself. Ultimately I think it depends on the main use of the building. But in a building where it's part retail, part school, part apartment- I think it's unclear. Even in buildings where it's 4 stories of apartment, and one story of retail, I'd say that it's unclear- so my advice is to simply tag the building as building=yes, and then place POI nodes in the building. Other OSM editors in NYC seem to be doing the same, since single use buildings are so rare here. As for how to tag an apartment, this is something I've struggled with myself. Ultimately I added an address with a name (not housename). Longer term, I think there are two easy solutions (and they're not exclusive): 1. We should have a POI apartments- that indicates apartment. 2. We could tag the building multiuse and find the term for retail on bottom, apartments on top, so a building would be tagged: building=foo - where foo is the term we come up with for that feature. - Serge ___ Tagging mailing list Tagging@openstreetmap.org http://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] pastry and confectionery
On Sun, Jun 2, 2013 at 1:45 PM, Martin Koppenhoefer dieterdre...@gmail.com wrote: any of the former. If I go to a telephone business book (yellow pages) or a book section in a book store, I expect bakery and baking books to cover breads, cakes, pastries, etc. To further complicate matters, in the US, many stores have opened up which only sell a single type of pastry. For example the chain Crumbs only sells cupcakes[1]. There are many such cupcake chains, or places that only sell cookies[2], or Cinnabon, which only sells Cinnamon buns. So would this new tag be for these places as well, or should we designate them shop=cupcake or shop=cinnamon_buns, etc? I'm certainly in favor or supporting tags that make local sense, but where we can generalize a solution, I think we should, and before I know if this is a generalized solution, I feel like I need to know how these examples would be handled. - Serge [1] They've recently started selling cakes too. [2] http://www.insomniacookies.com/ or a large chain like http://www.greatamericancookies.com/ ___ Tagging mailing list Tagging@openstreetmap.org http://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] Fast Food Restaurants
On Tue, May 28, 2013 at 11:26 PM, Paul Johnson ba...@ursamundi.org wrote: Well, I guess we need to back up. When is a name not a name to you? I don't always use brand, but when I'm tagging a store or restaurant that I know is part of a large megachain, then I do: http://wiki.openstreetmap.org/wiki/Key:brand It's not a big deal, as name is not wrong, brand is just more precise, but we were discussing what is the right way to tag this. Ideally you'd also include the website (http://wiki.openstreetmap.org/wiki/Key:website), their address (addr:city, addr:street and addr:housenumber), phone number for the store (http://wiki.openstreetmap.org/wiki/Phone), and store hours (http://wiki.openstreetmap.org/wiki/Opening_hours). Also if the location is certified kosher, you'd add that (http://wiki.openstreetmap.org/wiki/Proposed_features/kosher) as well as any dietary options for it (http://wiki.openstreetmap.org/wiki/Key:diet). Very few people add all this detail, especially all at once, but they do add it. - Serge ___ Tagging mailing list Tagging@openstreetmap.org http://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] Fast Food Restaurants
On Wed, May 29, 2013 at 6:01 PM, Tac Tacelosky tac...@gmail.com wrote: I see amenity:Restaurant and amenity:restaurant are both returned from the api. And I just discovered that postgres does case-sensitive LIKE's. Sigh. Those are the kinds of things that bots come along and fix (much like they do in Wikipedia). You seem to be looking up exceptions to the normal tagging... What are you actually trying to determine? - Serge ___ Tagging mailing list Tagging@openstreetmap.org http://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] Fast Food Restaurants
On Wed, May 29, 2013 at 7:33 PM, Tac Tacelosky tac...@gmail.com wrote: I'm looking at a picture of a restaurant, and asking the user what type of establishment it is. Because I'm not yet weighing tags, the user has choices like donut and donuts, and has no idea which to pick. I'm working on a way now to at least present the most likely tag (based on taginfo). The existing editors (JOSM, Potlatch 2, iD) have presets. All these editors are Free/Open Source and so you can take a look at what they do. You will find that a vast majority of users simply use the editor presets, so by following them, your application will most easily integrate with what's already being done. Also, the wiki has documentation on recommendations, and taginfo can show use. But if you follow the presets that the other editors use, I suspect a great deal of these questions will be answered definitively, and since Josm, Potlatch and iD constitute some 99.something% of all edits, your bases will be pretty well covered. - Serge ___ Tagging mailing list Tagging@openstreetmap.org http://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] Fast Food Restaurants
This discussion is best for t...@openstreetmap.org or for help.osm.org, but I'll bite. On Tue, May 28, 2013 at 6:52 PM, Richard Welty rwe...@averillpark.net wrote: name=Subway amenity=fast_food cuisine=sandwiches This is what the wiki would suggest, what the JOSM presets say, and what the renderer will accept. The tags are explicit values, so you cannot replace them with capitalized versions, or ignore the underscores if you want the tools to understand what you've done. In addition to above, though, you may want to consider that Subway is a chain restaurant, so you may want to consider brand=Subway instead of name=Subway - Serge ___ Tagging mailing list Tagging@openstreetmap.org http://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] Fast Food Restaurants
On Tue, May 28, 2013 at 8:29 PM, Paul Johnson ba...@ursamundi.org wrote: On May 28, 2013 6:08 PM, Serge Wroclawski emac...@gmail.com wrote: brand=Subway instead of name=Subway Very, very few Subway locations have names other than Subway. Not really sure that's a valid complaint except in particularly well-branded edge cases (like Batman's Subway, which is in Batman Fuels (Shell) and related to Batman's Auto Sales). I'm sorry but I don't understand this comment at all. Are you saying that because of the brand tag, that there would some consequence? This is a matter of the data consumer, such as the renderer, knowing what to do with the tag. When I write software to get the label of an object, I always use brand or operator over name. - Serge ___ Tagging mailing list Tagging@openstreetmap.org http://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] Juice restaurants
I agree with specifically tagging them as juice bars (amenity=juice_bar). They also don't all qualify as cafes, since they don't all offer seating, for example, if they're in a food court, or just a street vendor with no seating. - Serge ___ Tagging mailing list Tagging@openstreetmap.org http://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] Grades for obstacle=vegetation
On Thu, Feb 7, 2013 at 12:47 PM, Konfrare Albert lakonfrariadelav...@gmail.com wrote: Hello, Some people has suggested me to include a gradient for the key obstacle=vegetation. One suggestion was something like vegetation=light|medium|dense. I think we don't need so much detail about the obstacle. If it's light vegetation, then it's not much of an obstacle, but this proposal doesn't bother me as long as it's optional. Another private proposal was: obstacle=vegetation_grade1 (vegetation bother when walking) obstacle=vegetation_grade2 (the vegetation must be moved with the hands to continue on the path) obstacle=vegetation_grade3 (vegetation requires me to move my body to get through) obstacle=vegetation_grade4 (the vegetation must be cut by manual tools to get through) obstacle=vegetation_grade5 (the vegetation must be cut by mechanical tools to get through) Or to use obstacle=vegetation (for grades 1,2,3) and obstacle=fallen_tree (for grades 4,5) I think that in this case, could be better use the key vegetation=grade1|grade2|grade3|grade4|grade5 This kind of scheme never gets used other than by extreme mappers (usually just by the person who made the proposal). I say forget it. At first, I rejected the proposals because verifiability is not guaranteed, but perhaps it is necessary to have a gradient for obstacle=vegetation like in tracktype=* (where the verifiability isn't guarenteed). I don't think we need it, but if we have it, the first proposal sounds ok. I pose this question because I'd like to hear your opinions about it. I like to keep tagging as simple as possible, and vegetation as an obstacle has a lot of potential to change frequently. Ever seen a corn maze? - Serge ___ Tagging mailing list Tagging@openstreetmap.org http://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] Follow-up on Time Domains
On Wed, Jan 23, 2013 at 4:32 AM, Janko Mihelić jan...@gmail.com wrote: It shouldn't be too hard to make a JOSM add-on that converts 3 letters into 2. So that's no problem. You seem to be not seeing the point. Two letter days of the week (DOW) may be standard in German, and that's fine. But the tags we use in OSM are in English. They aren't in an abstracted system which we then render- we use English and then codify from there. It's what many software projects do, and it's what we do. So then we must ask What is the standard way of representing a day of the week in English?. The way is to look at a standard, such as the locale (http://en.wikipedia.org/wiki/Locale) So if you look at your locale from a *nix system- you will see the abday, and you will see unicode encoded strings that show the day of the week. Since that is a pain to look at, we can use Python to help us: import time time.strftime(%a) 'Wed' If you aren't familiar with Python (or the C it borrows from), strftime prints out the time, and I've given it the parameter to display the shortened day of the week, according the locale (in my case, en_US). I'm not about to say that whether we use three letters or two is the end of the world, but I will say that we should strive to use things that are standard- things that are defined elsewhere. Doing so will make it easier for folks to use the software, but also easier for programmers to have something they expect. - Serge ___ Tagging mailing list Tagging@openstreetmap.org http://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] Follow-up on Time Domains
On Tue, Jan 22, 2013 at 5:32 AM, Andrew Errington erringt...@gmail.com wrote: Surely this is 'internal'. That is, it's nice that some people can read Mo, Tu, We, etc., but for others, they are just 'coded' days of the week. Date producers need to understand the meaning of Mo, Tu, etc. so that they can record them properly in the database, but data consumers need to translate these to something that can be displayed nicely. Except why use abbreviations that no one uses elsewhere? I've never seen two letter abbreviations for days of the week outside OSM, in any computer system. So why use a codification that no one else uses, to save a byte? - Serge ___ Tagging mailing list Tagging@openstreetmap.org http://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] Follow-up on Time Domains
On Tue, Jan 22, 2013 at 11:43 AM, Eckhart Wörner ewoer...@kde.org wrote: nobody else? Windows Vista: http://www.askdavetaylor.com/1-blog-pics/vista-date-time-pop-up.png Windows 7: http://www.homeandlearn.co.uk/bc/win7/taskbar/changeDateTime.gif Windows 8: http://www.liberiangeek.net/wp-content/uploads/2012/12/date_time_windows8_2_thumb.png Calendars are different than written. In calendars, one does things they don't do elsewhere. Many physical calendars use a single letter, but in computer systems, two letters are common because it makes it easier to come up with a column display of the dates (ie 01-31). That's quite different than written. - Serge ___ Tagging mailing list Tagging@openstreetmap.org http://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] Dispute concerning shrimp pond dikes
On Sat, Jan 19, 2013 at 10:43 AM, doug brown dougc...@hotmail.com wrote: In the past couple of years I have spent several tens of hours digitizing the shrimp pond dikes in San Blas Municipio, Nayarit, Mexico. They are a significant feature on the local landscape and of importance because of the degradation they are causing in the mangrove forest ecosystems in which they are located. I was much dismayed this past summer when all of these features were deleted from the OSM data base (changeset 11807195), with the comment Deleting vandalism, those are not roads. The tags I used on these features are as follows: access = private embankment = yes highway = unclassified man_made = dike name = shrimp pond dike My reasoning being that all of the dikes have, at a minimum, a footpath on top of them, with most having a motor vehicle accessible road. While they are all private roads, most of them are open to public access and I frequently ride my bicycle on them. Doug, As others have said, some of the tags you used to make these edits were a bit off. First, highway=unclassified probably doesn't mean what you assumed it means. It does not mean I do not know the classification of this road. Unclassified roads are roads which do not have a classification, or have a classification of unknown. They are drivable roads. The value of highway where you do not know its classification is road. That is very confusing, but you can blame our British friends, who have unclassified as a classification in their official road system, which OSM borrows from. Others have given you examples of highway classifications you can use instead. But the real trouble may have been, name=shrimp pond dyke. name is the name of the object. In the example or a road, it's the name of the road. In the example of a restaurant, it's the name of the restaurant. The result of the two would be that someone looking at the area you made would see lots of roads all named shrimp pond dyke. If the embankment roads have no name, that's better than them all being named the same. Usually when someone sees lots of the sane feature, it's either vandalism or a newbie making errors that look like vandalism, and the person simply took the steps to remove what looked obviously wrong. What he should have done is, as you did, reach out to you and discuss these edits. So what's to be done now? My suggestion is that you use Josm (if you are using potlatch, there will be a bit of a learning curve), to open up that changeset and revert it, but you'll want to clean up the embankments, and possibly make a new tag for the ponds themselves, such as waterway=shrimp_pond or something else that can be used to identify the water feature. - Serge ___ Tagging mailing list Tagging@openstreetmap.org http://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] Source tag - deprecated for use on objects?
On Mon, Jan 7, 2013 at 2:10 PM, Dudley Ibbett dudleyibb...@hotmail.com wrote: As a relatively new mapper I am rather confused about this discussion about source. To quote the wiki for the tag, source: The wiki is not an authoritative source, it's edited and re-edited by people with an oninion. On this very topic, I edited the wiki and someone removed my edits, and isn't even bothering to respond to email. In other words, the wiki is quite a thuggish place. - Serge ___ Tagging mailing list Tagging@openstreetmap.org http://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] Source tag - deprecated for use on objects?
On Thu, Jan 3, 2013 at 7:23 AM, Jason Cunningham jamicu...@googlemail.com wrote: From my experience most mapping sessions involve changes and additions made to a wide variety of objects with a several sources used. Effectively requiring a new changeset for each object edit makes complicated process significantly worse. I never said anything about requiring a new changeset for each object edit. What I said was, and I'll reiterate it: 1. source on objects is silly because it fails to capture to the complexity of the editing process, including the multiple sources used for geometry and tags. 2. A better way to capture that information /if you're interested in capturing it at all/ is to use source on the changeset, which is what most mappers do if they use source. I then gave a lot of reasons why that was better. 3. A person on the list complained that source on the changeset doesn't allow for multiple sources. I suggested that they use a semicolon, such as survey;bing or bing;tiger. 4. Then someone said that this method (using a semicolon) would not allow for differentiating which changes came from which source. Notice that at this point, that you're still better off than you would have been with source on the object. You're already at a net gain. But I suggested that /if/ you weren't happy with semicolons, and you wanted to tag not only the sources, but the specifics of the source of each and every change, then you could do so using the changesets, using one changeset per source. Personally, I just use semicolons on the source on the changeset, if I use source at all. - Serge ___ Tagging mailing list Tagging@openstreetmap.org http://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] Source tag - deprecated for use on objects?
On Wed, Jan 2, 2013 at 11:16 AM, Andre Engels andreeng...@gmail.com wrote: Secondly, any changeset can have multiple sources so adding the source tag to a changeset is not flexible enough. We use semicolons for all tags as a separator. But it's less precise Then you'd make a new changeset. It's the same order of complexity as source tags, but without all the aforementioned issues described in the email you omitted. - Serge ___ Tagging mailing list Tagging@openstreetmap.org http://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] Multiple purposes for buildings
On Tue, Jan 1, 2013 at 8:05 AM, Svavar Kjarrval sva...@kjarrval.is wrote: Hi. How does one tag buildings which are both commercial and residential? There isn't a good solution for this at the moment. It's a pain for high density areas like NYC where most buildings are multi-use. It might make sense to have a multiuse tag for these situations 2) The ground floor is commercial but the residential part is in the floors above. The housenumber is the same. Do I mark the building with commercial;residential and leave it at that? And there are some buildings where part of the ground floor is residential, while other parts are commercial, such as the neighborhood where I live. The basement and ground floor of parts of the building are commercial, while other parts are residential, and then the high part is residential. Or do I just mark buildings with multiple purposes with building=yes and wait for a tagging scheme which can handle such situations? That's what I do. - Serge ___ Tagging mailing list Tagging@openstreetmap.org http://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] New tag proposal Amenity=meditation centre
Christian Science Reading rooms have very little, if anything to do with medication centers. - Serge ___ Tagging mailing list Tagging@openstreetmap.org http://lists.openstreetmap.org/listinfo/tagging
[Tagging] Shark tagging
I took this photo of a building across the street. How do you propose I tag it? http://www.emacsen.net/shark-bldg.jpg :) - Serge ___ Tagging mailing list Tagging@openstreetmap.org http://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] any tag you like, but why create parallel systems for established tags? DCGIS
On Sat, Mar 10, 2012 at 6:38 AM, Martin Koppenhoefer dieterdre...@gmail.com wrote: I recently stumbled upon an import in the US prefixed with dcgis. While this tagging makes it possible to have these data inserted parallely to other OSM data I still wonder why someone would do that. In particular I am refering to this: http://taginfo.openstreetmap.org/keys/dcgis%3Aaddress I don't see the point why this is not the usual addr:street etc. and there seem to be also other issues: http://taginfo.openstreetmap.org/search?q=dcgis The most used tag in this namespace is dcgis:captureyear http://taginfo.openstreetmap.org/keys/dcgis%3Acaptureyear which can be found 93 022 times in the db but isn't documented in the wiki [1] It seems that you didn't ask any of the users who were involved in this import. This import was done four years ago, with many users and a variety of processes, but I don't get the impression you checked with any of us. I don't see any actual questions in your mail. If you have specific questions, feel free to ask them. - Serge ___ Tagging mailing list Tagging@openstreetmap.org http://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] Health and other stories
parking=* requires amenity=parking. That seems entirely redundant to me, and if I were king, I'd strip out amenity=parking. (the whole parking scheme needs re-design to integrate the various schemes currently in place). - Serge ___ Tagging mailing list Tagging@openstreetmap.org http://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] Tagging a club's meeting location
On Tue, Nov 8, 2011 at 2:21 PM, John F. Eldredge j...@jfeldredge.com wrote: In addition to being a mapper, I am the secretary of the Nashville Linux Users Group. I have been debating whether the club's monthly meeting location should be tagged in OSM, and, if so, what tag to use. The question arises because the club doesn't have its own dedicated space; instead, it meets once a month in a particular lecture hall at Vanderbilt University. For the rest of the month, there is no sign or marker designating the lecture hall as NLUG's meeting space. How should such a meeting location be tagged? It shouldn't. We mark locations and just like you don't mark every class on a university campus, you don't mark your monthly LUG meeting's location. - Serge ___ Tagging mailing list Tagging@openstreetmap.org http://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] [Imports] Bus data for Fairfax Connector, Fairfax County, Virginia, USA
On Sat, Aug 27, 2011 at 6:32 AM, Mike N nice...@att.net wrote: On 8/27/2011 3:09 AM, Bryce Nesbitt wrote: Skip the name for a bus stop. If rendered it would create clutter. I'd say the opposite; the stop name is very useful to anyone using the Public Transport JOSM plugin to check and organize stops so that stops can be recognized, rather than just working with a column of anonymous stops. It also assists riders following a printed set of directions. The name doesn't currently render on Mapnik or Osmarender. A bus stop name or ID number is generally useful for navigation. The issue of if rendered' is one where the right answer is always We don't map to the renderer. - Serge ___ Tagging mailing list Tagging@openstreetmap.org http://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] How to tag destroyed stuff?
On Mon, Aug 8, 2011 at 10:18 PM, Richard Welty rwe...@averillpark.net wrote: is their removal permanent, or are they to be rebuilt? if they are going to be rebuilt in substantially the same place, i'd just set access=no with a README=destroyed by storm 201y-mm-dd, to be rebuilt, ETA 201y-mm-dd It sounds like we just need a new tag to add to the HOT tags: http://wiki.openstreetmap.org/wiki/Humanitarian_OSM_Tags/Humanitarian_Data_Background bridge=damaged access=no And the changeset seems like the right place for the metadata about the damage, just add it to the comment. - Serge ___ Tagging mailing list Tagging@openstreetmap.org http://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] Feature Proposal - RFC - Directional node
On Tue, Aug 9, 2011 at 8:44 AM, Ilya Zverev zve...@textual.ru wrote: How about just using a tag to indicate direction? direction= There are a couple of such proposals in the wiki, see links at the bottom of the discussed proposal. The purpose of relation is not only to state the direction of a point feature, but also sometimes link it to the surrounding features: for example, a direction of a sign is determined by the road segment leading to it. Also, this is much easier for mappers and consumers than to calculate degrees. Fundamentally, I think we need to remember that OSM is not intended to be a true representation of the world, but rather a logical representation. The direction that a bench, or a sign faces is not significant from the perspective of a map. When you get to the object, you'll see its direction, and you don't need to know the object's orientation to get there. The problem with a direction tag is that then suddenly every object has a direction relation, and we make editing the map far more complex. A simple tag seems the right solution in the rare circumstance where that's needed. - Serge ___ Tagging mailing list Tagging@openstreetmap.org http://lists.openstreetmap.org/listinfo/tagging
[Tagging] Relations (was directions)
On Tue, Aug 9, 2011 at 10:11 AM, Ilya Zverev zve...@textual.ru wrote: Of course. When you get to the road you'll see its surface and lane count, Lane count and surface are useful for many things: For routing, for rendering, for analyzing traffic and capacity. Tell me what direction is useful for other than 3d rendering. OSM offers many perspectives. From the perspective of routing the direction of road sign is important. How? From the perspective of 3D rendering physical attributes and rotation are important. OSM has stopped being just a map the moment someone specified building levels count. OSM isn't a 3d representation of the earth. It's a 2d representation, and a rough one at that. The problem with a direction tag is that then suddenly every object has a direction relation, and we make editing the map far more complex. The problem with a name attribute is that then suddenly every object has a name tag, and we make editing the map far more complex. This is not a proposal's problem, but with relations in general, I guess? YES! Relations are a solution that should only be used as a last resort. They cause many many problems. We have to use them but before we do we should ask ourselves: 1. Is this data representable some other way? Through tagging for example? 2. What is the relationship I'm building my relation on? A relation describes relationships. What are my objects and their role to each other? 3. Is this something people will find useful other than me? 4. How would I code support for this? Since relations are complex, maybe spend some time in the PL2/Josm code and think about how you'd build a UI for it, and maybe think about how a renderer would support it, not just in theory but in practice. Should we reduce their count to mininum, creating alternative ways to map turn restrictions, destination signs, surveillance cameras, public transport routes? Yes!!! Relations are overused in OSM, and it causes a huge amount of difficulty in spreading mapping. I am often editing relations that other people make; they're often broken. They're broken because it's hard to make them correctly, and fixing a relation isn't easy- you have to break the objects apart, check their roles, check their construction, and reassemble the objects. Relations make the map hard to work with. Let's take your example. Let's say I find a road need to be split and fixed because of construction. Now I have to worry about the relations on that road, and check each and every segment that's created. I'll tell you that many mappers won't do that, which means that the relations won't be right. What's left is bad data. It's like an abandoned wiki. The more I fix other's data, the more clear these problems appear. - Serge ___ Tagging mailing list Tagging@openstreetmap.org http://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] Pet supplies store but doesn't sell animals
On Fri, Jun 10, 2011 at 6:56 PM, John Smith deltafoxtrot...@gmail.com wrote: On 11 June 2011 06:16, Serge Wroclawski emac...@gmail.com wrote: On Fri, Jun 10, 2011 at 3:43 PM, Dave F. dave...@madasafish.com wrote: Hi Is there a specific tag for pet supplies (food, collars, chew toys etc) I'm used to pet stores being ambigious, and don' have a problem with that. But I'd say shop=pet_supplies is better than animals=no. Still, I think that the distinction is fairly narrow. Lots of pet stores here now no longer sell animals, but they still call themselves pet stores since they still sell products for pets, still listed in the yellow pages like that etc, I'd be inclined to still tag them as a pet shop, and use your animals=no suggestion, because then you can have animal:fish=yes as well if they sell one type of animal. The problem with these types of proposals, of N levels of depth of a tag, is that they quickly become complex, and thus get unused. You, Dr. Who, are proposing changing shop=pets to now: shop=pets animals:fish=yes and shop=pets animals=no The logical conclusion is: shop=pets animals:cats=yes animals:reptiles=gecko;snakes supplies:cat_food=dry;canned supplies:fish=block;flakes;filters;nets supplies:fish:treasure_chest=no ... Going back to the original point: Is there some minor ambiguity between a pet store that sells animals and one that doesn't? Sure, but it's a minor. I tend not to like to frequent pet shops that sell pets when I can, but it's easy to find out which ones those are when you need to. But for OSM, lat namespaces are a good thing. They're easy to explain to our users, and easy to code for. So if we need to distinguish, let's use something simple and flat to do so. - Serge ___ Tagging mailing list Tagging@openstreetmap.org http://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] Pet supplies store but doesn't sell animals
On Sat, Jun 11, 2011 at 10:31 PM, John Smith deltafoxtrot...@gmail.com wrote: The logical conclusion is: shop=pets animals:cats=yes animals:reptiles=gecko;snakes supplies:cat_food=dry;canned supplies:fish=block;flakes;filters;nets supplies:fish:treasure_chest=no ... I don't see a problem with this except to prove your point you made things excessively specific, but I don't see a problem with more general forms, how is this any different than tagging the types of fuel sold at amenity=fuel? I'll elaborate on why this is a bad idea: 1. It's a lot of tags The problem with over specificity is that in my experience, it scares mappers off. We want users to come in and be able to map quickly and easily; but what we see from a subset of our community is the desire to design complex hierarchies of tags. These tags intimidate users. 2. It won't get used in real life The second problem with these complex scheme is that the tagging list is an entity unto itself. People on this list love specificity, they love the idea of mapping in extreme depth. The problem is that it's rare that these tags get used. The original mapper uses them, maybe a few others, but overall they're just noise in the data, but appear prominently in the documentation. See #1. 3. It's nested In the Python programming world, there's a saying Flat is better than nested.- I feel the same applies to tags. In this example the question was Can we distinguish between pet shops that sell animals and those who don't. My suggestion was that the term is ambiguous, but if you wanted to make them less ambiguous, you could distinguish a pet supply store by simply having a tag: shop=pet_supplies That's simple, easy, elegant, and solves the problem. if you wanted to have an animals=yes tag, I'd /almost/ be okay with that. But when you start getting into these nests and subnests, I think this is an exercise in complexity and futility. 4. It's apt to change animals:fish = yes is you listing a store's inventory. It would be the same as store=clothing men:bowties=yes Inventory changes, and this leads to increasingly bizare tagging of individual items. After all, why stop at fish animals:fish:neon_tetras=yes animals:fish:angel_fish=no ? 5. It's outside the reasonable scope of the project. There's value in knowing what store carries what product, but it's not OSM's job to do that tracking. In some cases we bend the rules a bit, as you mention in your fuel example, or as I sometimes do with cuisine= on a restaurant. That's why I'm okay with the distinction between pets and pet supplies, but where I'm not okay with listing individual items of the inventory any more than I would be if we had: cuisine=thai menu:pad_see_ew=yes menu:pineapple_fried_rice=no We should be striving for simplicity wherever we can. - Serge ___ Tagging mailing list Tagging@openstreetmap.org http://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] Pet supplies store but doesn't sell animals
On Fri, Jun 10, 2011 at 3:43 PM, Dave F. dave...@madasafish.com wrote: Hi Is there a specific tag for pet supplies (food, collars, chew toys etc) I'm used to pet stores being ambigious, and don' have a problem with that. But I'd say shop=pet_supplies is better than animals=no. Still, I think that the distinction is fairly narrow. - Serge ___ Tagging mailing list Tagging@openstreetmap.org http://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] Tags for neighborhoods / subdivisions
On Wed, May 11, 2011 at 3:00 PM, Colin Smale colin.sm...@xs4all.nl wrote: Tagging something wilfully and deliberately wrongly in order to obtain the desired visible results is called tagging for the renderer and is almost universally frowned upon - see [1]... If Nominatim doesn't know to look at other objects than boundary=admin etc, then wouldn't it be better to fix Nominatim? +1 - Serge ___ Tagging mailing list Tagging@openstreetmap.org http://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] Tags for neighborhoods / subdivisions
On Wed, May 11, 2011 at 3:28 PM, Josh Doe j...@joshdoe.com wrote: I'm all for creating something else, however I didn't tag it this way, this is how the TIGER places import was done, so this affects at least the entire US. Saying it's the way it was done in the single worst import we've done in the project doesn't help make it any more appealing. The right solution isn't to look to the US Census or other organizations, but to go back to basics and try to find a scheme that works for OSM. Can we begin discussion of this? A place_level that allows for unincorporated areas, neighborhoods, and the like. I don't think that's the right solution because that's precisely what brought about a discussion a few months back where NE2 wanted to delete Silver Spring, Bethesda, and a whole lot of other cities and towns in the US. Political boundaries aren't the most important thing, certainly they're less important than actual usage. - Serge ___ Tagging mailing list Tagging@openstreetmap.org http://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] Fwd: Feature Proposal - RFC - automated tasks
I think the number of tags in this kind of proposal would balloon quickly, so let me suggest instead that tags of this nature be put on the changeset, with a then request to bot authors to check object history and changesets, rather than more tags on the object. In other words, use the changeset to hold this (what I think of as) metadata, rather than the object itself. - Serge ___ Tagging mailing list Tagging@openstreetmap.org http://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] landuse:illegal and illegal:yes/no
Pieren In general, I'm also against tagging what is not physical and Pieren immediatly verifiable attributes. But hey, the idea of indicating the Pieren legal status is not worst than many other tags already widely used. I disagree. Pieren And it is verifiable, Pieren not easy but it is (not less, for instance, than the power lines voltage). The power line voltage is gleaned from signs, which is how we get many of our features. I only know what street I'm on based on the sign that tells me so. If there was an official sign that said Illegal dumping, 300m, I'd be more inclined to listen to this proposal. Peter First, it's not simple and measurable, at least I really wonder how Peter many of us measured an incline which isn't signed (I have tried, takes Peter time and a bit of math, and I don't even play it for most of the Peter cases). How many of us know what a street is other than by official signs? Peter Second, both, but especially smoothness (and all subjective tags, Peter since I personally find smoothness very useful and informative) Peter depends on the judgement of the mapper. Your argument about smoothness is why people prefer to know the road material than the subjective smoothness. Peter For example if I see a concrete pipe which clearly 100 years Peter old I wouldn't, but when I see one hiding under bushes and clearly Peterbuilt this spring I'd say it's hardly legal, This case is perfect, since it illustrates that you're basically deciding something you feel is suspicious is illegal. You have no idea if that pipe is there legally or not. Serge illegal use is not as easily measurable in the same way.It's similar Serge to proposals to classify places as dangerous. Peter Indeed similar. My point was that dangerous was rejected as a tag. It's just too problematic to use. Serge That's very subjective- it's why we have courts! Peter OSM will never go there and force the people to remove the pipe. Peter That's for the courts. But we can tell other walkersby about it. No one's arguing that there may not be value in the data you want to collect, only that OSM may not be the place for it, just as there's a lot of useful information that's not appropriate for Wikipedia. Serge I think there are so many reasons why this tag is a bad idea it's Serge almost not worth bringing any individual reasons. Peter Such blocking comments are not really very cooperative. If you cite Peter those so many reasons they might be accepted or rejected, but Peter nobody can help you with those theoretical so many reasons This is a fair criticism. The problems with this tag are that it's: 1. Far more subjective than other accepted tags 2. Carries a lot of weight. It's a serious thing to accuse someone/something of being illegal. 3. Libelous, in that this is an accusation. 4. Outside the scope of the project, unlike any other tag we have. 5. This tag seems to beg for an edit war. 6. Apt to change very frequently. 7. Activism/advocate. OSM is not advocating positions. That's why even in the case of political borders, there's sensitivity in how it's handled. I think this data belongs in another dataset- not OSM. - Serge ___ Tagging mailing list Tagging@openstreetmap.org http://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] landuse:illegal and illegal:yes/no
On Tue, Mar 29, 2011 at 10:46 AM, Peter Gervai grin...@gmail.com wrote: On Tue, Mar 29, 2011 at 15:21, Bill Ricker bill.n1...@gmail.com wrote: Under the usual rule-of-thumb, to map what's visible on the ground (signed or built), Like smoothness=, incline=, etc? Those are simple, measurable things. illegal use is not as easily measurable in the same way.It's similar to proposals to classify places as dangerous. Illegal use, in those words or similar. (Or tagged so by a suitably free Govt GIS file.) Incidentally sometimes that's the case, as it turned out. I think there are so many reasons why this tag is a bad idea it's almost not worth bringing any individual reasons. Anything else, an OSM member is making a value judgment and OSM is publishing it as a fact, which has legal consequences in most countries. OSM has a legal entity in UK which is democracy most favorable to libel tourists, where Truth is NOT a defense. (The new coalition gov't is looking at reform but don't bet you assets on it.) But then you can be sued on virtually anything, like stating there is a road when the owner thinks otherwise, or state its smoothness as horrible which is clearly offensive, etc. Obviously it's quite acceptable if you request the addendum for this tag not to be used in the UK. :-) Let's not resort to hyberbole. If you saw something illegal, presumably you'd report it to the proper authority. For example, if I see waste water dumping, is this now illegal=yes? What if someone else doesn't think it's illegal? Now it has to go to court. And that's where it belongs. You can say There is a pipe here with water coming out- but illegal? That's very subjective- it's why we have courts! But then again you are worried about the word and not its use, so since you seem to be a genuine englishman, please utilise your native vocabulary to suggest an alternative _word_ to use which conveys the same meaning without suggesting that this status is defined by law instead of common sense or otherwise. I'm no native English speaker so you may be better suited to pick the proper word. (Prohibited and debated was two suggestion which I didn't really like.) It's not the word that's the problem, it's the concept that's trying to be conveyed. - Serge ___ Tagging mailing list Tagging@openstreetmap.org http://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] Feature Proposal - RFC - Sidewalks as separate ways
On Fri, Mar 25, 2011 at 4:37 AM, David Paleino da...@debian.org wrote:
One can take exactly the opposite stance, which is that in order to
help the blind, we should make it as easy as possible to map things
that they care about. Therefore a sidewalk=yes tag would be the
fastest way to get the maximum data into the map.
That's not the maximum data, you know?
The blind-argument was the one that made me re-think the whole scheme.
This is especially important when talking about rendering and routing,
which I think are the main use cases of this tag.
Routing, not rendering. We don't care about rendering, do you?
And there are plugins drawing parallel ways, you know?
1) add name=... to the sidewalk, but it's redundant, even if simpler;
2) use a relation.
This is really a myth IMHO.
They're hard to work with in Potlatch, maybe, but I see it as a bug of the
editor. In JOSM, for example, they're correctly handled, and are rather easy
to
work with IMVHO.
http://wiki.openstreetmap.org/wiki/Proposed_features/Sidewalk_as_separate_way#Relation
The proposed relation is associatedStreet -- or, my favourite, the proposed
street (which should IMHO replace associatedStreet, but that's another
story)
If you want some examples:
http://www.openstreetmap.org/browse/relation/254299
http://www.openstreetmap.org/browse/relation/1198910
Do you want me to link these from the proposal page?
Generally, I think this proposal seems to be more of a reaction.
I agree, but sidewalk={yes|left|both|right|none} is just not enough. It
could be used as a first-pass mapping, yes. But, as written in the proposal
page, I'd regard it as highway=road for streets. Yes, there's a sidewalk
somewhere, but that's it.
I think it tells me just enough.
You're proposing a new relation type, a set of associated tags, etc.
in support of the sidewalk data.
I'd like to suggest you should sit down and work out some mapnik rules
for this, and work out a way for PL2 users to enter the relation and
associations you've created.
At the very least, this could help the downstream tool folks
understand your proposals, but I think it'd also help you refine your
proposal by helping you step back and see how tools would need to
interact with it.
The only thing I out and out disagree with entirely is your suggestion
to tag sidewalks with a name.
I'm concerned this will confuse folks. And by folks, I mean mappers,
routers, renderers and editors.. That's why I think this scheme needs
more work, because, when you map sidewalks as separate ways, you have
to use a relation, and how exactly that's to work isn't fully figured
out.
- Serge
___
Tagging mailing list
Tagging@openstreetmap.org
http://lists.openstreetmap.org/listinfo/tagging
[Tagging] Feature Proposal - RFC - Sidewalk
As per the discussion last week about Sidewalks, I'm re-opening the sidewalk proposal as per: http://wiki.openstreetmap.org/wiki/Proposed_features/Sidewalk We've already had some preliminary discussion on this tag and there's been very minimal disagreement, which is a good sign for its adoption. I do want to emphasize that this proposal does not in any way replace the existing highway=footway tag. This will remain separate and untouched. Also, if accepted, there will be a tag link from Pavement to Sidewalk made in the Wiki, and a link from the highway=footway tag to indicate the possible use of sidwalk instead. If there are any other issues with this tag, let's discuss them now so we can get into the voting process! - Serge ___ Tagging mailing list Tagging@openstreetmap.org http://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] Feature Proposal - RFC - Sidewalk
On Mon, Mar 21, 2011 at 12:47 PM, David Paleino da...@debian.org wrote: Since my proposal was the one most agreed on, why can't you just start using the tags/way-of-mapping in my page? :) David, I suggested re-opening the sidewalk proposal because I have a specific need for this data and would like to use official, rather than unofficial tags. What specifically is your concern? You've used the word my proposal several times. If your concern it's credit, I'll happily put your name on the proposal. If it's not credit, then use the outlined tagging RFC process. If you have specific issues with the proposal, please bring them up for discussion. The issue is: your proposal is the opposite of what was agreed upon last week. I'll be able to get some free time on Wednesday/Thursday: can you wait up to then? I'll make a definitive proposal :) You're free to make recommendations to change it, and you're free to vote against it, but the process is now in place. - Serge ___ Tagging mailing list Tagging@openstreetmap.org http://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] Feature Proposal - RFC - Sidewalk
On Mon, Mar 21, 2011 at 12:57 PM, Josh Doe j...@joshdoe.com wrote: Serge, I think we're really talking about two proposals here, both of which have merit. The linked proposal has been around for a while, and involves tagging the road to indicate the presence of a sidewalk walk on one or both sides of the road. David refers to this proposal as deprecated [1]. He can claim that, but it's not depricated; the proposal is 3 years old and wasn't ever voted on. What David proposed, and what I'm interested in, is mapping the sidewalk as a separate way from the road. This rough proposal is on his page here [2]. That's entirely separate. I have views about that, but they're not relevant in this discussion. I think both schemes can coexist. In cities, especially ones with regular grids, the first scheme might be preferable for simplicity. However David's scheme has advantages when sidewalks don't have a constant offset from the road, and if there is a desire for precise navigation. In my area of suburbia, I plan to use David's scheme. David should follow the tagging RFC process and make a proposal. I don't want to overload one simple tag with something else. We should certainly link both together though, so users can determine what level of detail is appropriate. I disagree. As mentioned in the Sidewalk tag, we already have highway=footway, which is what David's proposal would largely change, rather than an additional tag on roads. In other words, this is a tag about roads, not footways. - Serge ___ Tagging mailing list Tagging@openstreetmap.org http://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] Feature Proposal - RFC - Sidewalk
On Mon, Mar 21, 2011 at 2:18 PM, M∡rtin Koppenhoefer dieterdre...@gmail.com wrote: In all cases where the sidewalk is not perfectly parallel, or where you want to add barriers between the sidewalk and the road, you will in any case have to move to the other scheme. IMHO explicit sidewalks have the huge benefit of simplicity. I know that many of you are computer scientists eager to create the most elegant way to model all kind of complex situations with pure mathematics, but don't forget that we are building the people's map: the easier it is to contribute (and to understand the model) the more people can actually map. +1 I've mapped sidewalks as highway=footway. It makes sense to do that when you're mapping a park, when mapping a walking area in a community (eg townhouses), but most areas which are paved and designated pedestrian areas in an urban or suburban area are directly associated with a road. In my case, I see a city full of street data, where most streets have a sidewalk. I want users to be able to map those sidewalks quickly and easily, as Martin says. But beyond that, I don't see a good solution to routing without the sidewalks tags on the road way. Are there other ways to accomplish the same task? Sure. You could create a separate way which is the sidewalk, tag it with all the appropriate tags, then create a relation which has both the road and the sidewalk. The problem is that while relations are more correct, it's hard to execute. I can explain to a new mapper that if a sidewalk exists, tag the street sidewalk=yes, it's much harder to show them how to create the second way, tag that way, then to create a new relation, move the appropriate tags (like name) from the road way to the relation, etc. sidewalk=yes is a single step, and I have the code to make it a dropdown in PL2 (JOSM is forthcoming). To create the relation in order to get the routing right is a multistep process and it's hard to make smooth. My problem with your answer is, that tagging at the main street is hard to expand to explicit mapping later. Yes, most likely we would remove these tags and draw explicit ways later. You could do this, or let sidewalks be rendered, but not explicit ways. - Serge ___ Tagging mailing list Tagging@openstreetmap.org http://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] Feature Proposal - RFC - Sidewalk
On Mon, Mar 21, 2011 at 3:50 PM, Chris Hill o...@raggedred.net wrote: On 21/03/11 19:21, Serge Wroclawski wrote: [...] I suggested re-opening the sidewalk proposal because I have a specific need for this data and would like to use official, rather than unofficial tags. [deliberately off list to keep the noise down] Serge, There are, never have been and probably never will be official tags. Anyone can use any tag at any time for any purpose. It follows that there are no unofficial tags either. That's partially true- it's true we have no official tags, but it's also true that the tags which are approved are the ones which get rendered, and the ones who people write software against. The proposal process uses a voting system which makes people feel that they have some kind of official status and this is fundamentally wrong. It leads to people running bots to 'correct' tagging to the 'standard' or 'official' version when no such thing exists. Also just because 6 people voted for a tag means they think it MUST be rendered or MUST be supported by editors or MUST be used buy routers. I'm of two minds on this, so please tell me why this is a bad thing in your mind? Many people just don't get this but it is the greatest strength of OSM. If you need 'official' tags then you can never something novel to OSM until a new tag has been authorised. BTW, just who should police these tags? Please discuss tags, please document the tags that are used, but please don't assume that the stupid, broken voting system somehow makes tags official and more useful. I'll admit I'm a bit frustrated because I have a very pragmatic problem to solve and have spent several weeks trying to make practical tools to solve it, and I feel David wants something else entirely and is suffering from a bit of NIH syndrome, and Josh (who I think is actually local to me) doesn't see what I'm doing because I haven't announced it yet. But maybe you'll have some feedback... Go to http://mappingdc.emacsen.net/pl2/ and what you /should/ (hopefully) see is a map of DC, and you can select DC SIdewalks as a background. That's real DC sidewalk polygon data. But it's too complex. I'd like to attach those as information about the streets. Also if you look at the stylesheets, you'll see Pedestrian Wireframe- that's a style I've created that highlights sidewalk data, missing sidewalk data, and highway=footway. In other words, I'm trying to make it as simple as possible to get this data in OSM, without an import. - Serge ___ Tagging mailing list Tagging@openstreetmap.org http://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] Feature Proposal - RFC - Sidewalk
Blerg. It turns out Chris's mail wasn't off list and neither was mine. Apologies to all. - Serge ___ Tagging mailing list Tagging@openstreetmap.org http://lists.openstreetmap.org/listinfo/tagging
[Tagging] Sidewalks vs Footways
There's an abandoned tag for sidewalks along the side of the road that apparently has some use in the UK: http://wiki.openstreetmap.org/wiki/Proposed_features/Footway http://tagwatch.stoecker.eu/Great_britain/En/tags.html And there's a nearly identical tag proposal called Sidewalk: http://wiki.openstreetmap.org/wiki/Key:sidewalk I'd like to suggest collapsing these two tags into a unified tag and making a final vote, and then fixing tags as necessary. Thoughts? - Serge ___ Tagging mailing list Tagging@openstreetmap.org http://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] Sidewalks vs Footways
On Thu, Mar 17, 2011 at 10:42 AM, j...@jfeldredge.com wrote: Collapsing the two tags into one seems reasonable, but there should continue to be a wiki page for whichever tag is discontinued, in order to direct people to the preferred tag. That's what wiki redirects are for. :) - Serge ___ Tagging mailing list Tagging@openstreetmap.org http://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] Sidewalks vs Footways
Based on this thread, there seems to be general consensus that the term sidewalk is less linguistically ambiguous than footway. Where folks are concerned about dual meaning, we can ensure that's resolved via fixing the wiki, checking JOSM presets, and checking Potlatch/PL2. Data consumers who are using either tag should be aware that neither has been accepted, so they're using it at their own risk. On Thu, Mar 17, 2011 at 11:34 AM, Josh Doe j...@joshdoe.com wrote: Are we just talking about using sidewalk as indicated in the linked proposals? I've been mapping sidewalks in residential areas near me as individual ways; they usually run parallel to the road, but oftentimes are offset by a variable distance, plus I'd like to have very accurate routing. I've been tagging these as highway=footway, however when a way no longer parallels a road (and for my local area the surface almost always changes from concrete to asphalt) and goes into wooded areas, I start tagging the ways as highway=path. Josh, There's nothing wrong, pe se, with that. I've gone back and forth on whether that's the right way to go or not, and I think it's up to individual mappers, but after considering it, I've decided I prefer the road to be tagged for the following reasons: 1) It's less ways overall. That's not normally a consideration I'd consider important, but in this case, I'd say that the sidewalk is a feature of the road, and therefore makes the map more useful. 2) Sidewalks are sometimes interrupted, but still logically connected. An example of this would be a crosswalk. The sidewalk ends, but roads continues, and the governance of whether a pedestrian can cross can be placed at the intersection. 3) Making sidewalks a road feature aids the router. It lets the router say things like Walk along Main Street. If the term sidewalk is used for this proposal would it make sense to also apply it to individual ways? Individual ways which are unassociated with a road are footways/paths, and not sidewalks. - Serge ___ Tagging mailing list Tagging@openstreetmap.org http://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] Sidewalks vs Footways
On Thu, Mar 17, 2011 at 1:30 PM, David Paleino da...@debian.org wrote: On Thu, 17 Mar 2011 10:15:48 -0400, Serge Wroclawski wrote: There's an abandoned tag for sidewalks along the side of the road that apparently has some use in the UK: http://wiki.openstreetmap.org/wiki/Proposed_features/Footway http://tagwatch.stoecker.eu/Great_britain/En/tags.html And there's a nearly identical tag proposal called Sidewalk: http://wiki.openstreetmap.org/wiki/Key:sidewalk ...and I tried to make a unified proposal some time ago (which I have been following for the few sidewalks I mapped). It has been written down with the help of some osm-it(aly) folks. http://wiki.openstreetmap.org/wiki/User:Hanska/Sidewalk Comments? :) I think that in the area of sidewaks, it mainly mirrors the discussion here, so it sounds like we have largely reached consensus. It's great that we've all come to similar conclusions on our own- it means we can move forward. David, it's great you've thought so much on these issues. The link you sent has proposals for several features, and normal OSM procedure is to vote on only one at a time, so let's focus the discussion on sidewalks, and then when that's done, we can go through the others. I think that way we're less likely to get bogged down and make iterative, incremental progress quickly, which is ultimately what we both want, I think. - Serge ___ Tagging mailing list Tagging@openstreetmap.org http://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] Draft - Vegetarian/Vegan
MHO, a vegetarian tag would be best expanded to any dietary guidelines. For example, I can't have gluten, and some places specialize in gluten free options. Big chains will have an allergy menu and include gluten, but some places are actually specialty gluten free, just like some places are pure vegetarian or vegan. I'd suggest making the tag more expansive for this reason. - Serge ___ Tagging mailing list Tagging@openstreetmap.org http://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] Airport subtypes
On Tue, Jan 4, 2011 at 12:45 PM, M∡rtin Koppenhoefer dieterdre...@gmail.com wrote: Also, I definitely think we should try and align to external standards. +1 Look at the ICAO classifications. http://en.wikipedia.org/wiki/Airspace_class Nonetheless I think the OSM classifications are good because they handle things that ICAO doesn't, like international flights. - Serge ___ Tagging mailing list Tagging@openstreetmap.org http://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] Why addr:state rather than is_in:state? (response to 2010-12-26 05:29:46)
On Fri, Dec 31, 2010 at 7:57 AM, Robert Elsenaar rob...@elsenaar.info wrote: Totally agree. More over, every time you use the symbol : in your tag, you mean that the subtag is telling something more specific about the maintag thats in front of it. I have no idea what subtag and maintag are, but the symbol tag shouldn't be there. - Serge ___ Tagging mailing list Tagging@openstreetmap.org http://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] self-storage facilities
On Sat, Dec 11, 2010 at 6:32 PM, Ed Hillsman ehills...@tampabay.rr.com wrote: On Sun, 12 Dec 2010 10:04:11 +1100, Steve Bennett stevag...@gmail.com wrote: I think storage and self-storage imply different things. The former would be warehousing etc for business customers, and the latter for the general public. That said, I'd be inclined to go with amenity=storage for both, with some subtag like self_storage=yes where appropriate. We should have less top-level tags. I had been inclined to avoid amenity= because of the volume of past comments on other proposals, concerned about the number of amenity values. But amenity=storage makes sense. Because so many self-storage facilities are businesses, owned by different companies, landuse= is less appropriate. Amenity=storage would lie within landuse=commercial or, in some instances landuse=industrial, with operator= or name= for the firm, and self_storage=yes. Internal driveways, parking, buildings, and external fences and gates could be mapped if desired. I agree with most of what's been said- I'm fine with either amenity=storage or amenity=self-storage, so long as with the former there's a subtag. I'd also suggest that since self-storage is usually commercial, we encourage the standard set of tags for businesses, including name (or operator), hours, website, etc.) - Serge ___ Tagging mailing list Tagging@openstreetmap.org http://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] self-storage facilities
A bit OT, but the minute I started reading this thread, I couldn't help but think of this: http://www.youtube.com/watch?v=MvgN5gCuLac - Serge ___ Tagging mailing list Tagging@openstreetmap.org http://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] Announce: New #osm-tagging dedicated IRC chat on oftc.net
On Sun, Oct 24, 2010 at 7:26 PM, Sam Vekemans acrosscanadatra...@gmail.com wrote: Would calling it 'Open Map Features' make it a bit more clear for everyone? As this IRC chat room can cover more general chat (about map features) so it will attract the wider geo community to participate. Sam, Will you just be honest with people? You want a new list because you want to create a tagging system for the forks. Whatever my personal feelings on the forks are, I think you owe it to the people to just be honest with your intentions. If you want to fork, fork, and be decisive and proud of your decision. What irks me is how you and some of the other forkers are creating little staw men arguments Wouldn't it be nice if OSM provided a way to automatically sync all the usernames and passwords onto another system? and your Wouldn't it be nice if all the tags were 'centralized'? (ie you want the tools and mapnik stylesheets OSM developed to work on your maps). Please just start being honest with people. - Serge ___ Tagging mailing list Tagging@openstreetmap.org http://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] atms with names?
On Fri, Oct 22, 2010 at 2:21 PM, Richard Welty rwe...@averillpark.net wrote: On 10/22/10 1:45 PM, M∡rtin Koppenhoefer wrote: 2010/10/22 Richard Weltyrwe...@averillpark.net: for ATMs, brand and operator are likely to be the same, as what the user really wants to know is if it's his bank, or one that will hit him up with extra charges. so i'd say use the street name of the financial institution. Don't know in your country, in Germany we have different situations: there is cash group with 5 or so private mayor players in banking (you don't pay fees on any of them if you're account holder on one). the situation is different in the US, and hard to map with current OSM tagging conventions. ATMs can serve multiple networks, and cards can work in multiple networks, but no-fee is restricted to ATMs belonging to card holder's financial institution. If we want to get precise (or pedantic), that isn't quite accurate. It's your bank's ATMs, plus any other bank your bank has an agreement with. For example, as a credit union member, I've never payed a fee at another credit union ATM. I don't know if that's universal or just a feature of the credit unions I choose to go to. But then there's also another bank I belong to which doesn't have any branded ATMs, but gives me a list of ATMs in my area where I won't be charged a fee. An accurate map of ATMs to determine fees would be very hard for us to make. - Serge ___ Tagging mailing list Tagging@openstreetmap.org http://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] tagging single trees
On Fri, Sep 10, 2010 at 7:13 AM, David Groom revi...@pacific-rim.net wrote: Maybe I'm missing something in this discussion, but what exactly is so important about the fact that the tree is standing alone that it needs to specifically be tagged as standing (or not standing) alone? David, Maybe you missed the beginning of this painful thread. The issue is this (and I'll try to be as neutral as possible): * Nop points out that the wiki definition of trees says a lone tree and interprets this as a prominent tree (a landmark, etc.). * He then concludes that trees in OSM which are not prominent should be tagged to indicate that. * Me, Martin and others say that the wiki definition is wrong, that people aren't using it, that it's ignored in imports, etc. and landmark trees are the special ones and should be retagged. * Nop says that this is unfair because he's already been doing the right thing (ie following the WIki guidelines) and so it's everyone else that's wrong. * Nop then points out stats from Germany which he says support his point. I think I understand where Nop is coming from. This doesn't appear to be a tagging issue as much as it is about doing the right thing. I think he feels that he and others who followed the Wiki definition are being punished by needing to retag their data. The position of the rest of us is that: 1) We don't tag normal things as normal, we tag special things as special. 2) The wiki is, more or less, supposed to reflect actual usage. (I'll elaborate more on this point later on in the mail) 3) The definition makes common sense if it's any tree, rather than this complex definition of a special tree, having to do with space or landmark, or any of that. Now, I want to also bring to the table an extract I did this week of all the trees in the world: http://www.emacsen.net/trees.osm.gz People, feel free to download and examine the trees. To elaborate on #2: This is a big difference between languages. In French, for example, there's a society which determines what can be considered official French. In English, it's quite different, especially in the US. Dictionaries document words in their current usage. They're descriptive rather than prescriptive, but of course all schoolteachers teach children to look words up in the dictionary and use the words properly. That is the constant tension that exists when you define terms, and is similarly the tension that exists in our wiki regarding definitions of features. Are we describing tags in OSM as they're used, or explaining how to use the tags? A bit of both, I'd say. In this case, it's clear to me there's a disconnect between the actual usage and the wiki definition, so it's the wiki which should change. - Serge ___ Tagging mailing list Tagging@openstreetmap.org http://lists.openstreetmap.org/listinfo/tagging
