Re: [Tagging] building typology vs usage
Am Fr., 13. Sept. 2019 um 21:38 Uhr schrieb Dave F via Tagging < tagging@openstreetmap.org>: > "OpenStreetMap is a place for mapping things that are both *real and > current*" > > > building!=yes = 65 221 930 > this is actually an encouraging number, given that it is around 20% while not so long ago, 98% of all building values were "yes", and most imports also used "yes" for buildings. It means that people care about the details and have started to add detail at this level. Cheers, Martin ___ Tagging mailing list Tagging@openstreetmap.org https://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] building typology vs usage
Am Fr., 13. Sept. 2019 um 15:20 Uhr schrieb Dave F via Tagging < tagging@openstreetmap.org>: > On 11/09/2019 14:50, Paul Allen wrote: > > > > I said that if it was a church and looks like a church then tag the > building as a church even if it now functions as something else. > > Buildings don't have a 'type'. There's no 'class', no standard > architectural style or size. A quick image search proves that. > maybe you should extend your search, and go beyond images ;-) The typology of buildings is for example a subject in architectural studies at the university ("Gebäudekunde"). You will find tens of thousands of books about building typology (usually each dealing with only a narrow topic, e.g. hotels, hospitals, office buildings, production buildings, specific types of apartment buildings, specific military buildings, etc.) A supermarket, prison church or townhall will typically by recognizable as such (with the exception of those that are built on purpose to not stand out), as will a hotel, an office or a residential building. Sure, you do not need an office building to set up an office, but this doesn't mean there aren't office buildings. buildings do have a type, but of course you're right, if you look at a very generic type like "residential" you will find all kind of dwellings and you won't recognize a common style or type. To recognize similarities, you'd have to go into more detail, e.g. terraced houses (that's clearly a kind of residential building type, with usually one unit per entrance (may be split now), a narrow garden to the back (usually), etc.). OSM "is a place for mapping things that are both real and current" > > 'building=*' is to indicate its current usage. > > no, its current building type. Cheers, Martin ___ Tagging mailing list Tagging@openstreetmap.org https://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] building typology vs usage
13 Sep 2019, 21:37 by tagging@openstreetmap.org: > On 13/09/2019 16:14, Wolfgang Zenker wrote: > > >> >> That would be kind of redundant, wouldn't it? We already use other tagsfor >> the current function of a building, >> > I'm repeating much of my of my previous comment, but no, the schemawhich > hijacked building=* to represent the original historicalfunction of a > building never took off*. The vast majority ofcontributors use it for > it's current purpose. OSM isn't for themapping of redundant historical > information. > (...) > *building:use = 628 167 > building!=yes = > 65 221 930 > That is because in vast majority current use is the same as suggested by how building looks like. I also often tag building=* about its structure without tagging building:use Note also that building tag is not about historical data. Industrial buildings with fast food inside is building=industrial Remodeled industrial building that lost indicators of its original use is not building=industrial -+-- Btw, can you link evidence that building tag was originally for current use, not for current appearance? ___ Tagging mailing list Tagging@openstreetmap.org https://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] building typology vs usage
* Dave F via Tagging [190913 21:37]: > On 13/09/2019 16:14, Wolfgang Zenker wrote: >> That would be kind of redundant, wouldn't it? We already use other tags >> for the current function of a building, > I'm repeating much of my of my previous comment, but no, the schema > which hijacked building=* to represent the original historical function > of a building never took off*. The vast majority of contributors use it > for it's current purpose. OSM isn't for the mapping of redundant > historical information. Well, I don't know of any hijacking. This thread is the first time I have seen people suggesting to tag the current function only in the building=* tag. But admittedly I'm only active in OSM since 2008, so that might have happened before that or I might have overlooked it. >> so building=* is mostly useful >> when the uilding does look like it was built for some other function >> than it's current one. > How do you know what it was originally used for just from your > interpretation of what a building of a certain function should look > like? It's just guesswork. How does tagging this perception add to, or > improve the quality of the OSM database? I don't care about the buildings original function but about what it looks like. That might or might not match its original and/or current function. And if it looks like a run-of-the-mill nothing-special any- purpose-at-all type of building I tag it as building=yes. > "OpenStreetMap is a place for mapping things that are both /real and > current/" I agree. I recently mapped https://www.openstreetmap.org/way/722948688 "Howard School" in rural Montana. It is a historic building originally built as a public school, looks like a fine specimen of a schoolhouse of it's era, has a prominent sign saying "Howard School" on the front side and so I tagged it "building=school" even when it has not been used as a school since 1947 but is used since for community gatherings. So I also mapped it's current function by adding amenity=community_centre. Of course different mappers have different opinions about what would be the best way to tag something, but I don't see this as a weakness but a strength of OSM. By discussing what we individually think is best and learning from each other we collectively will arrive at better tagging by all over time. Wolfgang ( lyx @ osm ) ___ Tagging mailing list Tagging@openstreetmap.org https://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] building typology vs usage
On 13/09/2019 16:14, Wolfgang Zenker wrote: That would be kind of redundant, wouldn't it? We already use other tags for the current function of a building, I'm repeating much of my of my previous comment, but no, the schema which hijacked building=* to represent the original historical function of a building never took off*. The vast majority of contributors use it for it's current purpose. OSM isn't for the mapping of redundant historical information. so building=* is mostly useful when the uilding does look like it was built for some other function than it's current one. How do you know what it was originally used for just from your interpretation of what a building of a certain function should look like? It's just guesswork. How does tagging this perception add to, or improve the quality of the OSM database? "OpenStreetMap is a place for mapping things that are both /real and current/" *building:use = 628 167 building!=yes = 65 221 930 DaveF ___ Tagging mailing list Tagging@openstreetmap.org https://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] building typology vs usage
* Joseph Eisenberg [190913 16:45]: > I certainly recall reading about this in the wiki, but I agree that in > common use, the building=* tag appears to be used mostly for the > current function, rather than specifying a certain form. That would be kind of redundant, wouldn't it? We already use other tags for the current function of a building, so building=* is mostly useful when the building does look like it was built for some other function than it's current one. Wolfgang ( lyx @ osm ) ___ Tagging mailing list Tagging@openstreetmap.org https://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] building typology vs usage
I certainly recall reading about this in the wiki, but I agree that in common use, the building=* tag appears to be used mostly for the current function, rather than specifying a certain form. The most common values of building= are: 0) yes (non-specific) 1) =house - both a structural form and a function (residential) 2) =residential - function, not really a specific form of building 3) =garage - function and form 4) =apartments - function (multi-family residential) 5) =detacted - synonym for house but more specific 6) =hut - form of construction (crude/simple) 7) =industrial - function 8) =shed - form but also function ("used as storage or workshop") 9) =roof - form 10) =terrace - form but also function (residential) 11) =school - function mainly 12) =garages - form and function 13) =construction - lifecycle state 14) =retail - mainly function, because the form of retail buildings varies 15) =greenhouse - form = function here 16) =barn - form=function 17) =farm_auxiliary - function (no particular form, and this is only the general function) 18) =church - claimed to be a form? 19) =warehouse - function (but usually has the same general form) 20) =service - function Often the "form follows function" as they say, but it looks like tagging the function of a building is as common as tagging the form. - Joseph On 9/13/19, Dave F via Tagging wrote: > On 11/09/2019 14:50, Paul Allen wrote: >> >> I said that if it was a church and looks like a church then tag the >> building as a church even if it now functions as something else. > > Buildings don't have a 'type'. There's no 'class', no standard > architectural style or size. A quick image search proves that. > > OSM "is a place for mapping things that are both real and current" > > 'building=*' is to indicate its current usage. > > /If/ there's an insistence on recording it's original usage, if actually > *known*,not just observed, then an appropriate *clearly defined* tag > should be used. Something along the lines of 'original building use". > > Frederik suggests "Everyone may be confused about this.". It's been > evident for years that those who are perplexed are the ones who imagined > a 'typology'. I believe they've based their assumptions on anecdotal > observations around their own neighbourhoods. OSM is global. > > DaveF > > ___ > 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] building typology vs usage
On Fri, Sep 13, 2019 at 9:20 AM Dave F via Tagging wrote: > On 11/09/2019 14:50, Paul Allen wrote: > > > > I said that if it was a church and looks like a church then tag the > > building as a church even if it now functions as something else. > > Buildings don't have a 'type'. There's no 'class', no standard > architectural style or size. A quick image search proves that. > > OSM "is a place for mapping things that are both real and current" > > 'building=*' is to indicate its current usage. > > /If/ there's an insistence on recording it's original usage, if actually > *known*,not just observed, then an appropriate *clearly defined* tag > should be used. Something along the lines of 'original building use". > > Frederik suggests "Everyone may be confused about this.". It's been > evident for years that those who are perplexed are the ones who imagined > a 'typology'. I believe they've based their assumptions on anecdotal > observations around their own neighbourhoods. OSM is global. In the part of the country where I live, the vernacular architecture is based on an idea of hardline Protestantism that rejected trappings. The older buildings tend to be symmetric boxes (albeit with more-or-less steeply pitched roofs; it *snows* here) that give no hint to their purpose. There's one listed historic building in my township that in its history served as a school, a social center, and a private house, and is now subdivided into office space. The only real indicator of its current purpose is that the front door has a sandstone lintel reading, 'District School Nº 4'. Likewise, buildings may reveal obviously their complex history. Consider the Imam al-Khoei Foundation building in Jamaica, Queens, New York. https://en.wikipedia.org/wiki/Imam_Al-Khoei_Benevolent_Foundation#/media/File:Imam_Al-Khoei_Foundation_8989_Van_Wyck_jeh.jpg https://www.flickr.com/photos/imjustwalkin/29799850223 . It's obviously a converted factory - and just as obviously a mosque. At what point does the former usage become obscured enough that the building acquires a new type? The example that everyone loves to cite is 'building=church'. That appears to come about because people imagine very likely a building with a tall steeple or campanile, stained glass windows, perhaps built in a Gothic or Romanesque style. But a couple of centuries ago in stern, Calvinist, North America, churches were plain affairs, with no stained glass, no iconography, not even a cross atop the steeple: https://www.flickr.com/photos/steveguttman/2814490383 is fairly typical of a church of the denomination and period. Is that obviously of the "church" type? If so, can you say what features in particular distinguish it from https://www.oldhousedreams.com/wp-content/uploads/2018/07/10-21-Haskell.jpg, which is pretty typical of a primary school of the same period? Many of these buildings also started out their lives as government buildings - the "meetinghouse" of a village would have been its seat of government as well as its church, in an era before the separation of church and state was a familiar idea. Meetinghouses were often even plainer than the examples that I've given so far. https://commons.wikimedia.org/wiki/File:Exterior,_Sandown_Meetinghouse.jpg was in fact the town's meetinghouse, simultaneously its place of worship and seat of government, but from the exterior could just have easily have been a workshop, a school, or a boardinghouse. If you have a high-Gothic building with twin campaniles, a magnificent rose window, and similar trappings, that's now a banquet hall or has been subdivided into flats, go ahead and tag it as "building=church" if you like. I really don't care. But don't expect that every building will fit an imagined typology. Frederik and others have told me repeatedly, "if it still looks like a church, tag it building=church, if it still looks like a school, tag it building=school, and so on." But that doesn't inform me about the historic buildings that I'm most interested in tagging. For the most part their history is complicated, and their appearance is either likewise complicated, or else undistinguished. What does a church, or a school, or a government building, look like? -- 73 de ke9tv/2, Kevin ___ Tagging mailing list Tagging@openstreetmap.org https://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] building typology vs usage
On 11/09/2019 14:50, Paul Allen wrote: I said that if it was a church and looks like a church then tag the building as a church even if it now functions as something else. Buildings don't have a 'type'. There's no 'class', no standard architectural style or size. A quick image search proves that. OSM "is a place for mapping things that are both real and current" 'building=*' is to indicate its current usage. /If/ there's an insistence on recording it's original usage, if actually *known*,not just observed, then an appropriate *clearly defined* tag should be used. Something along the lines of 'original building use". Frederik suggests "Everyone may be confused about this.". It's been evident for years that those who are perplexed are the ones who imagined a 'typology'. I believe they've based their assumptions on anecdotal observations around their own neighbourhoods. OSM is global. DaveF ___ Tagging mailing list Tagging@openstreetmap.org https://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] building typology vs usage
By the way, there are currently 5 objects with the tagging on the same object. { "type": "way", "id": 59218539, "tags": { "addr:housenumber": "8", "addr:street": "Pier Place", "building": "church", "leisure": "sports_centre", "name": "Alien Rock", "note": "former St Andrew's Church", "phone": "+44 131 552 7211", "source": "NLS_OS_Edinburgh_map_1940s;Bing;survey", "sport": "climbing", "url": "http://www.alienrock.co.uk/; } }, { "type": "way", "id": 93180076, "tags": { "building": "church", "leisure": "sports_centre", "name": "Kletterkirche", "old_name": "St. Peter", "sport": "climbing", "wheelchair": "no", "wikidata": "Q20181755" } }, { "type": "way", "id": 202350995, "tags": { "building": "church", "leisure": "fitness_station", "name": "Three Wise Monkeys", "source": "OS_OpenData_StreetView", "source:location": "Bing", "sport": "climbing" } }, { "type": "way", "id": 275133732, "tags": { "HE_ref": "1025007", "alt_name": "The Bristol Climbing Centre", "building": "church", "listed_status": "Grade II*", "name": "Undercover Rock", "old_name": "Saint Werburgh's Church", "sport": "climbing", "wikidata": "Q7595628", "wikipedia": "en:St Werburgh's Church, Bristol" } }, { "type": "way", "id": 381773790, "tags": { "addr:city": "Newcastle Upon Tyne", "addr:street": "Shields Road", "building": "church", "building:material": "stone", "leisure": "sports_centre", "name": "Newcastle Climbing Centre", "note": "formerly St Marks Church", "phone": "+44 191 265 6060", "sport": "climbing", "website": "www.newcastleclimbingcentre.co.uk" } } ] } ___ Tagging mailing list Tagging@openstreetmap.org https://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] building typology vs usage
On Wed, 11 Sep 2019 at 14:38, Martin Koppenhoefer wrote: > Am So., 8. Sept. 2019 um 15:13 Uhr schrieb Paul Allen >: > >> Good idea. A better idea might be to add it to the description, since it >> is information that >> may be useful to non-mappers: data consumers may suppress notes but allow >> the >> display of descriptions. It's useful to know that the art studio you're >> looking for is in a >> church... >> > > > these descriptions could be autogenerated from semantically detailed > tagging, localized for every language. You can add a lot of useful > information to the descriptions, but it shouldn't substitute good tagging. > Semantic tagging makes it possible to find church buildings where you can > climb, descriptions don't. > I think you took my paragraph in isolation and missed the point. I said that if it was a church and looks like a church then tag the building as a church even if it now functions as something else. Somebody said he added a note to the effect that it was once a church, I said that note might be better as a description. The note or description are there to clarify the tagging in order to minimize the risk of confusion in somebody who has trouble reconciling building=church with shop=convenience. -- Paul ___ Tagging mailing list Tagging@openstreetmap.org https://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] building typology vs usage
Am So., 8. Sept. 2019 um 15:13 Uhr schrieb Paul Allen : > Good idea. A better idea might be to add it to the description, since it > is information that > may be useful to non-mappers: data consumers may suppress notes but allow > the > display of descriptions. It's useful to know that the art studio you're > looking for is in a > church... > these descriptions could be autogenerated from semantically detailed tagging, localized for every language. You can add a lot of useful information to the descriptions, but it shouldn't substitute good tagging. Semantic tagging makes it possible to find church buildings where you can climb, descriptions don't. Cheers, Martin ___ Tagging mailing list Tagging@openstreetmap.org https://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] building typology vs usage
On Sun, 8 Sep 2019 at 10:09, Tom Pfeifer wrote: > > In cases for usage apparently contradicting the building type it often > helps the fellow mapper to > tag a note that this school building was converted into a hostel, or this > church building is used > for climbing now. > Good idea. A better idea might be to add it to the description, since it is information that may be useful to non-mappers: data consumers may suppress notes but allow the display of descriptions. It's useful to know that the art studio you're looking for is in a church... -- Paul ___ Tagging mailing list Tagging@openstreetmap.org https://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] building typology vs usage
On 07.09.2019 11:00, Frederik Ramm wrote: It is true that this is the canonical way of dealing with things, however it would be interesting to check how mappers and editing tools actually use this. We might well find that everyone is confused about this. [...] I think we cannot simply throw the distinction over board and therefore I do not agree with Josh, but I also think the distinction is not really well thought out/well implemented in OSM and needs clarification. In cases for usage apparently contradicting the building type it often helps the fellow mapper to tag a note that this school building was converted into a hostel, or this church building is used for climbing now. tom ___ Tagging mailing list Tagging@openstreetmap.org https://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] building typology vs usage
On Sat, 7 Sep 2019 at 10:03, Frederik Ramm wrote: > > When we say "a cafe in an old church" we think of a building that has > certain properties that make it discernible as a church even long after > it ceased to be one; however, depending on location and denomination, > you might also build a church using a blueprint for a plain community > centre. In that case would it still be building=church becasue that was > the original, intended use? What if apartments are put into an old > factory building - building=industrial and ...? > I think ducks are important. Most people know what a traditional church or chapel look like. Navigational instructions might be "Carry on up that road until you see a church on your left, take the next turn to the right." This church https://goo.gl/maps/yyXYZcucuWwpyu7z9 quacks like a church. This chapel https://goo.gl/maps/tJ7XDt6tCM1xcyR89 quacks like a chapel. And this church https://goo.gl/maps/w5ce112JVP5C7cCE9 honks like your five-year-old found your stash of vodka, got hammered, and then started playing with his Lego. Some buildings are recognizable for what they are (or were). Others are not. We live in an imperfect world, so we use our judgement (however flawed that might be). -- Paul ___ Tagging mailing list Tagging@openstreetmap.org https://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] building typology vs usage
sent from a phone > On 7. Sep 2019, at 11:00, Frederik Ramm wrote: > > however, depending on location and denomination, > you might also build a church using a blueprint for a plain community > centre. In that case would it still be building=church becasue that was > the original, intended use? from my understanding you would be building a community centre with the intent of using it also for divine services. Ciao Martin ___ Tagging mailing list Tagging@openstreetmap.org https://lists.openstreetmap.org/listinfo/tagging
Re: [Tagging] building typology vs usage
Hi, On 07.09.2019 09:16, Joseph Eisenberg wrote: >> While in theory building=school could be reused as a hotel/pub (See >> https://www.mcmenamins.com/kennedy-school) in that case the building >> will be inside of a tourism=hotel polygon Why would it - a standalone former school in a city that now houses something else doesn't necessarily have to acquire a surrounding polygon. On 9/7/19 10:40, Tom Pfeifer wrote: > Please understand that the building typology is orthogonal to the usage > of the building. > Thus having both a building=X and leisure/amenity=X on the same polygon > is not redundant. It is true that this is the canonical way of dealing with things, however it would be interesting to check how mappers and editing tools actually use this. We might well find that everyone is confused about this. When we say "a cafe in an old church" we think of a building that has certain properties that make it discernible as a church even long after it ceased to be one; however, depending on location and denomination, you might also build a church using a blueprint for a plain community centre. In that case would it still be building=church becasue that was the original, intended use? What if apartments are put into an old factory building - building=industrial and ...? I think we cannot simply throw the distinction over board and therefore I do not agree with Josh, but I also think the distinction is not really well thought out/well implemented in OSM and needs clarification. Bye Frederik -- Frederik Ramm ## eMail frede...@remote.org ## N49°00'09" E008°23'33" ___ Tagging mailing list Tagging@openstreetmap.org https://lists.openstreetmap.org/listinfo/tagging