why are the three buildings surrounding this particular changeset encircled by a footway? i've repeatedly removed these in the past but am in no state at present to mend this. these are not public footpaths but are on private property and just because it is possible to walk around a house does not make that a footpath worthy of osm, even were it to be composed of paving stones. there is no evidence of a manicured, or even maintained path across the lawn when a routing engine can just as effectively guide one from a random point a to point b by connecting the dots.

seems to me a logging road would be a multi-spur highway=track, unless it becomes overgrown to the point only a single spur can be usefully followed, whereupon it should then be downgraded to a path.

if off-limits i would not expect to see the path designation before it becomes too overgrown to map, unless a trail formed by wildlife follows the route.

they show no respect for access restrictions, the animals. ban them from these protected areas if they can't play by the rules.

are you going to move or delete this node? another town/city/country node spam appeared recently at this point, so naïve blind transfer from maps.me or onosm.org notes into osm data will result in a pile-up here, or in the other places.

i myself would nuke it, but due to android memory manglement i can't even bring up a javashite app without the browser being terminated, much less install flash.

oh yeah, i know that, mateusz ;-)

i was just brought here after i opened up an half-there area in potlatch1 to see why a reasonable-looking addition to the map had been deleted, whether vandalism or not, and the `more' button opened the way history hit by this changeset. all was suddenly clear.

there's also something i do which is to point the three tileserver cache ipv4 addresses to different nearby areas of the CDN, in order to identify performance problems, and the occasional serious differences between new additions and when they get rendered, where at least one points to apparently a different rendering instance that when debugging a province-wide building-in-error, helped to identify the extent by being at least six days behind the other two.

ideally i'd find a nearby CDN instance that caches a third renderer as two of them seem pretty much in lock-step.

this is of course something frowned upon by operations, but it's not much different than using a VPN so all three tileservers are from the same region different from my actual location, as i'm not using it to work around apparent poor performance to burden an out-of-area cache.

this is slightly off-topic, or totally off-topic as usual.

salut,

did you intend for way/158978069 to get the South of South Cleveland Avenue which joins it at both ends, but along its short length, it and the parallel opposite way have the name of Tamiami Trail that matches the rest of the tagging across route 865?

i can't tell if note/1262485 means it ends as cleveland north of the point where it switches as mapped back to cleveland, south of the above-mentioned way.

merci

yeah, wondering why some zoomlevels show remnants of green that disappeared when zooming in, leaving vast areas of white.

hoi,

when you made this edit, did you notice note/659256 from several years ago, and verify your change to way/175722958 here, now at v6 and not changed from any previous shop incarnation over even more years?

wondering if the note should be invalidated and resolved...

> The name ICI Paris XL means 'This is Paris'

`Ici' is a french word, not an acronym. if it is to be written in OSM in uppercase, it should be replaced with the words it represents written out in full. if there are no words to form this acronym, it should be written in lowercase to differentiate from an acronym - such as Man, a word, and MAN, a vehicle company - as when converting text to speech for vision-disabled users or navigation, they should imply an uppercase series of letters is to be spoken individually rather than attempting to form a single word.

it is the same when i read something uppercase, i attempt to read each letter separately. like the nearby HEMA which could also be a single-word name and spoken as such in adverts, and if so, it too should be given the mixed-case treatment for addition as an osm name.

i am assuming it is an acronym that like so many that are just as often treated as spoken words, based on
> Die Hema B.V. (für Hollandse Eenheidsprijzen Maatschappij Amsterdam,
that differentiating itself from
> hema electronic GmbH
which, like dm (drogerie markt) would appear in osm in lowercase.

Paris is a town name, and again, osm writes words and names in mixed case.

XL is an acronym for eXtra Large and may be written in uppercase.

the use of ALL-CAPS in osm is unwanted, regardless of marketer wishes who want to stand out and out-SHOUT anything nearby. there are plenty of firms that have their logos in all-caps, or use all-caps for spoken words, but this is not the osm style which is to treat businesses impartially and not give any one an unfair advantage of using slogans, spam or advert copy to appear rendered more prominently.

i am repeatedly having this conversation with a mapper who is adding sponsor names to stadiums with their typographic style like WESERSTADION which only serves to make the map unbearably loud. if someone wants to see names in all-caps, there is no problem to convert lowercase as my garmin with a limited charset, but one cannot programmatically determine the proper casing of an all-caps entry, just as one cannot reliably expand an arbitrary abbreviation into its full name, hence the osm guidelines to write things out in full.

dunno about the AKO nearby, but if it's spoken with two syllables, it should be written mixed-case as well.

TUI nearby again gets impartial news article treatment as Tui, reflecting how i hear it spoken in adverts. i'm going to stop before ZARA because i can see while a lot of mappers do follow the lowercase rule, there appear to be a number who simply copy from signs.

adman, calm down and lay off the methanol. unlike the 20 to 30 per day fortified added-sugar beers i've resorted to for lack of good Reinheitsgebot beverages, i'm sure it can't be good for you.

TIGER roads probably do not have GNIS data. tiger was an import with a lot of questionably-done things and changesets that added a lot of useless metadata. tiger tags are partly automatically deleted, even in cases where i've found them useful for after-the-fact troubleshooting. and a lot of obsolete and irrelevant tiger tags don't get touched with updates, probably because like maps.me mishandling the `name' key, they are hidden from the casual editor.

There is not a single GNIS tag remaining after you have removed them, losing the origin of the data. there have been repeated cases where i have resorted to googling the gnis:feature_id value in order to check the data in osm, if it reflects the current reality or if something somehow went wrong with a node located offshore, several metres from its NADxx (versus WGS84) coörds but not making up for several dozens or hundred metres off from any logical position for an on-shore landmark.

the gnis:created field gives a further clue to the data, not reflecting the import time. if the object was created long ago but is absent in the oldest bing clarity images, it's reasonable to make further assumptions about it.

now the county_id and state_id, the two other fields you deleted, aren't that useful other than as a sanity check, and aren't too different from the tiger tags that i've seen be incorrect in the original data on occasion, in coded form.

to leave the tags present helps a mapper looking at the data after your edit to know the origin. if you've deleted them, i assume you've taken due care, but if other mappers delete the tags out of principle while armchair mapping, one has to look into the history to see whether a mapped feature is plausible, if it's likely to be mislocated, if someone nuked the `(historical)' label without understanding what it means to geonames, or what.

in short, the feature_id is unique to a gnis item, can be searched, and i feel it should be preserved for the reasons i've given and used repeatedly, pointing to its mechanical import origin and raising instant defences and wariness about blindly trusting the data.

i can't point out any guidelines and i don't feel it is worth searching, because it seems so blindingly obvious that it's useful metadata outside of an osm object property. even if you claim it's useless. let me point you at massgis tags that make navigating object properties a pain being uppercase and appearing early.

whatever, a lot of osm guidelines or rules have never been codified, or have been mealy-mouthed (use of `should' where due to abuse i'd interpret as a MUST in order to get along with the larger community, f'rinstance), that it sounds to me you're attacking the messenger rather than absorbing the message here.

i'll shut up and wade off to try to find some beer and watch the torrential water of the last three weeks continue to take away large parts of the ravaged hillsides.

This changeset has been reverted fully or in part by changeset/77111240 where the changeset comment is: duplicate better mapped nearby

This changeset has been reverted fully or in part by changeset/77109502 where the changeset comment is: bogus maps.me castles that clearly are not

This changeset has been reverted fully or in part by changeset/77062103 where the changeset comment is: nuke bogus maps.me private residence castles with personal data

ciao gpserror,

i suspect that i view parking aisles as an enhancement on the service drive concept, with the additional info that in theory it would be possible to turn off into a parking space pretty much anywhere, taking into account granularity, but then with facepalm moments of the last week or two i suppose that i should consider the marked parking spaces advisory and in reality an anything-goes philosophy will better match the reality that i've facepalmed repeatedly.

satire meets real life. at least i've never failed to find the parked fiat panda taking up three normal parking spaces.

i see parking aisles here as not a means to get from point a to b, while in other layouts where parking exists to the side, i can see the primary/secondary function would be to get to point b.

i guess it's like private residential driveways. i suspect there is a USian language bias that uses the term driveway to refer to what osm maps as service drives, but i also interpret a private driveway to be available to traffic between a pavement (sidewalk) that pedestrians and underage cyclists are required to use, and the street accompanying, to a pavement reached by another private driveway. or a path the driveway connects with. to split a driveway into public-for-allowed-transport sections and private/destination to the actual home seems pointless in most cases and additional mapping work, the default should be off-limits unless you fall into a category that puts it on-limits.

i'm probably not expressing meself well. for your use case, i'm not sure if you can algorithmically lose parking aisles without losing access to the building itself, and i agree, a lot of what are parking aisles by my definition are also routes to get to point b.

to take yesterday's real world example, i wasted a large part of the day walking in one direction and ending up on what i would map as a private driveway, meself getting increasingly slower as it became clear i would not be able to reach the goal i'd not researched ahead of time, without treading on private property that is a relatively new concept to me the last month. as i found after-the-fact, the bing clarity imagery shows nowt in the part i strode in confidence, while the other imageries are offset from the mapped data so i'm unsure how to map reality as i didn't pack a gps or even view the map ahead of time.

why am i writing this drivel? obviously because i want to get the fact across i'm totally incompetent to get from point a to b.

and i can't navigate meself out of a wet paper bag to do the shopping i wanted.

i'm failing, with `lynx' browser, to see any difference between pre- and post-b-jazz.

i'm guessing the line break refers to one or more of the unix/dos/mac linefeed, carriage-return, or CRLF combination. to me these seem inappropriate for a value which i would expect should be limited to the printable unicode characters, and not the below 0x20 control characters of US-ASCII including bells and such. there's probably something at decimal 127 that i've not tracked into unicode, plus there are probably unicode non-printing bidirectional indications that may or may not be worthy of inclusion, seeing that i've given no thought to my random brane mutterings at this time of night.

and probably emojis or whatever pollutes the latest unicode updates should be met with fire and a nuke-on-sight policy. or am i firmly placing meself into the get-off-me-lawn category of senile fossils who prevent osm from being all it could be?

anyway, i've also observed the nightly dump from the abakus 'bot includes repeated changeset comments for select users, that i feel when transformed into e-mail could be seen as abusive, there should be no reason for multiple messages about a particular sin, whether during one nightly batch run, or over time.

but that's just me, as i've never paid much attention to the content of these automated messages.

now i'll go back outside and bay at the moon high in the sky, annoying neighbours and guests and cementing my role as unwelcome mischief-maker.

cheers.

i'm quite positive i used to shop at the netto (colour scheme dark-ish yellow and black) in several places in danmark, which, last i knew, was not yet conquered by the kingdom of east germany.

i've never heard anyone refer in everyday speech, or in adverts, to aldi nord or aldi süd except to differentiate between the product selection and offers, which again in danmark, the logo reflecting the connection with aldi nord despite the per-country product selection not being consistent over country borders, same as with penny, lidl, norma, or any discounter present in multiple countries.

personally, i fail to see the benefit of tagging such branding details; near me is an eni petrol station that remains with the previous agip tagging and wiki-whatever links. if i correct the name, then it seems the wiki-whatever links will continue to point to the outdated info that i, using potlatch, do not see, only because i had no choice but to fire up iD on someone else's computer due to missing flash, and saw it presents the petrol station logo, that i have no clue whether it is correct and current, or if there is a better choice, due to renaming.

yeah, i got out of the wrong side of bed this morning.

you have been told off by a fellow mapper in changeset/76837496 for failing to provide useful changeset descriptions. why are you continuing to fail to do so? `edits' or `Edits' or even `EDITS' is a meaningless description.

han6man, it is true that a detailed and accurate map will have a lot of useless info for specific end-users. however for a garmin, one can select to ignore or include select features, such as the residential driveways that have been simplified, with `mkgmap' rules, while still including generic service drives to businesses. even private homes if you see no reason to have them on your map, or smaller buildings like sheds i see mapped.

also, just as editors have a function to simplify ways, the utility used to pre-process the map data should be able to do the same, turning rounded curves traced from 10cm or better aerials, into much coarser curves for gps use without losing map data details for more critical navigation like pedestrian or blind routing.

a commandline revert of this changeset will lose the non-controversial parts of this changeset such as adding service=driveway or a residential classification. also, while way/136485122/history was deleted, the current map data includes two replacement ways with cruder curves, but perhaps correct in not being continuous. norway orthophoto shows an open gate between the two.

way/127365373/history adds a driveway classification i suspect is wrong as it terminates in a car park in a business/commercial area, if i am to trust the map data. the drawn carpark is badly outdated, covering decade-old bing overflights where maxar shows the buildings implied by the scattered address nodes as well as a now-grassy area, and rather than driveway, i'd want to classify the service drive as a parking aisle. which again could be ignored by a gps where the area is mapped, but not by an autonomous self-driving app where the aisles through a car park are important to reach a parking space.

amazon logistics has been adding private residential driveways at breakneck speed in certain countries (the us, uk, germany) for deliveries. norway orthophoto shows lots of missing details and in high enough resolution that any node savings resulting from simplifying residential driveways will be a drop in the bucket compared to newly added car parks and access drives or missing buildings, and it would be a shame to draw crude outlines to maxar satellite standards that are the best other parts of the world have available, when norway orthophoto gives native imagery to zoomlevel 21.

are bicycles really forbidden on that side of the glatt? i know the marked cycle path crosses to the other side just northwest of here, but i don't ever recall seeing any prohibition for cycles on any paths or tracks, particularly as the aerial imagery shows no obvious reason a non-road-bike should have problems.

that was decades ago, so of course things may well be different now.

you have been repeatedly asked to use meaningful and non-gibberish changeset comments, and you have received several DWG blocks for failing to do so.

yet you continue to make further map edits without explaining what you have done. your latest changes include a meaningless `edits' here, and
y662y
vqegf
reGGGA
f8y6
txkylf,hy
ygbrag
fbDfb
htjh
vad
[an off-topic sensible description]
ygcftrfff8f8fyaw98fyw8ru33hyrhw3ugttttttttttttttttttttttttttttttttttttttttttttttttttttttttttttttttttttttttttttttttttttttttttttttttfisufhruuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuu
(which caused another mapper to comment that you should take notice of)
qfeq
fr3r
4w5t5t
dvvwgffv
dfvrbvrefg
hj bh
fdrftgh
jhh

and that is all on the first page of your history.

i take it you have no interest in helping your fellow mappers understand what you are trying to do.

moin gpserror,

i find your interpretation of service v. parking aisle interesting, where you would use service to get from point a to b.

i've used parking aisle for sections of service drives where at any point one can turn off into a parking spot, even if it is possible to pass through the car park without parking, to get from point a to point b.

that is, in the absence of mapped parking areas like here (which seems depressingly common, where service drives have been added without bothering to add an area to show the extent of the car park), i would specifically use parking aisle to tell a navigator that in theory one should be able to pull off along that line and park, which somehow i think i explained already before being interrupted half an hour ago. but also that it could be possible to traverse that section in order to arrive elsewhere.

if that makes sense. anyway, i wanted to explain how i've mapped car parks, where a service drive designation alternates with parking aisle and back, often repeatedly.

whatever.