Mijn naam mag er ook bij.
Ik ben er bij.
Ik zag dat je al bezig bent geweest Martin, en ik stel voor dat we voor deze reverts een gelijkluidend commentaar toevoegen en daarbij als source verwijzen naar dit topic.
Mijn voorstel voor het commentaar is:
“Changeset revert of inaccurate edits made by user GD”.
Op die manier kunnen we bij ons verzoek bij de DWG voor een ban, makkelijk refereren naar al dit werk.
Ik heb een link naar dit forum gemeld als motivatie; misschien toevoegen, dus: "Changeset revert of inaccurate edits made by user GD, ref. https://forum.openstreetmap.org/viewtopic.php?pid=616250"?
Een bewuste keuze, locatie en geschrevene. “Arbeid adelt”
Verzoek richting DWG wordt door mij ondersteund.
Zonde van tijd en energie van andere mappers om (blijkbaar) aanhoudende en onverbeterlijke fouten van één persoon te corrigeren.
Ik vraag me af of het zonder onderscheid verwijderen van alle edits van een account wel verstandig is. Daar kunnen ook goede edits tussen zitten (waarschijnlijk niet veel), en de DWG zou het daar wel eens niet mee eens kunnen zijn (eigenrichting). Is het niet verstandiger dat eerst met de DWG te overleggen?
Mee eens… Lijkt me ook verstandig om het op de formele manier op te lossen. Er is hier geen sprake van vernieling, maar van eigenzinnig mappen. Het is gewoon jammer dat Gerrit niet is te overtuigen.
In principe wellicht waar. Het probleem is echter dat de DWG geen Nederlands leest en dus de inhoud van deze forumdraad niet kent.
In mijn eerdere verzoek aan de DWG (zie ook deze draad) is teruggemeld dat er middels opmerkingen in de changesets van GD gewerkt zou moeten worden.
Punt is dat deze changeset-opmerkingen ook in het Nederlands zijn en de meeste opmerkingen over zijn gedrag niet daar maar juist in dit forum staan.
Maar het kan geen kwaad om namens ons allen een blockverzoek nogmaals in te dienen.
Overigens de voorlopig laatste changeset van GD geeft als cryptisch commentaar “Thanks for the perfect performance and architecture of OSM;>)” met een boerderij “Arbeid Adelt”.
Misschien is dit echt de finale changeset van Gerrit op de openbare OSM-server.
Maar mijn hoop zal denkelijk tevergeefs zijn.
Ik wacht even de ontwikkelingen nog een week af. Indien GD terugkomt op zijn eigen ‘niveau’ zal ik onder vermelding van alle users die hier ondersteuning gegeven hebben een formeel verzoek indienen bij de DWG. Indien iemand anders dit zelf, eerder wil doen: geen bezwaar.
Zet mij ook maar op de lijst.
De naam van de boerderij is door iemand anders aangebracht. De naam van de boerderij is inderdaad
Arbeid Adelt
Actually, I wouldn’t bet on that…
… but that bit’s certainly true. Although I’ve been vaguely aware of this thread since it started almost a year ago, I wouldn’t assume that anyone else is. There are 18 pages of discussions here, so a brief summary of what’s happened and what the continuing problems are would be really useful.
Thanks for the continuing changeset discussion comments - they do really help explain where the problems are.
Some of the “overlapping landuse” issues are to be expected though. The Netherlands had lots of landuse imports some years ago, making it much more difficult for new users to get started here than elsewhere, and what was imported doesn’t always match my recollection for places I’ve visited (though I’m not familiar at all with the Zwolle-Meppel area at all).
Take the area around https://www.openstreetmap.org/#map=19/52.63244/6.15755 for example. That came in 6 years ago through 3dshapes, and from looking at the imagery it would appear that it doesn’t match.
http://www.openstreetmap.org/way/451072665/history was an attempt to improve things by adding a “landuse=farmland” which doesn’t yet exist other than as a “hole” in the 3dshapes landuse=grass (see http://www.openstreetmap.org/node/1031108574#map=18/52.63207/6.15802 ).
So there’s clearly more to be mapped here, but it’s likely that some of the existing imported data will need to be deleted to allow that to take place. Is part of the problem here perhaps that users are unwilling to delete previously-imported data? In some countries that would indeed be a concern due to the overdependance on multipolygon’s but that’s not an issue with the 3dshapes data (see http://www.openstreetmap.org/changeset/6607083 ).
- Andy Townsend (DWG)
PS: Apologies for writing in English, but I suspect that people reading this will have far better knowledge of written English than I have of writing Dutch. If any clarification is needed please ask.
Hi Andy,
I hope I under estimate your fluency in reading Dutch. Further: the pictures as posted in this thread are illustrative in itself I think.
Let me start by indicating here the most important issues about Gerrit that do disqualify him as a valuable OSM editor; I hope other Dutch forum readers can add or detail here as well
Imports: Gerrit seems to work at a municipal (Staphorst) department where he has access to GIS services, which he uses to make imports from, mainly in his living area (Staphorst, Lemele and around); when importing (without proper source reference) he does not follow the OSM tagging scheme and imports unknown tag/value combinations, many times even in Dutch (e.g.: “attribution=testomgeving”). Gerrit has been informed and warned on that many times by forum users here and does not react
Tag/values: in addition to the imported data Gerrit also creates his own data, often at micro-scale, where he creates and applies OSM unknown ‘fantasy’ tags, also often in Dutch. He has been informed about that as well, with no structural results
Mapping for the renderer: in his micro-mapping efforts Gerrit uses tags/values improperly for the visual effect on the standard Mapnik rendering: e.g. regional admin boundaries to visualize individual house-parcel boundaries; he has been informed. Also he cuts up residential areas in individual parcels to make boundaries visual.
Overlapping areas: Gerrit has been many times informed about his practice to draw new landuses over, in or partially over existing landuse; not only old landuses but also over his own mapped landuse. Also he maps separate lines for adjacent landuses, without sharing the lines and nodes; keeping a ‘vacuum’ in between. He has been informed.
When editing Gerrit totally ignores the JOSM validator errors and warnings; checking his edits generates many, many errors and warnings.
If all above mentioned issues would be solved, there would be no reason to ban Gerrit. However it is his behaviour that causes us to ask for a ban:
- he does not improve; he keeps sticking in a vicious circle of repeating errors
- he makes many edits (also as a result of imports) and we, the other mappers, are tired and frustrated of correcting him again and again
- he very seldom reacts to the contents of messages posted in this forum, but circumvents the issue by answering a meaningless message or no message at all
- he does not (fully) repair the errors he made and leaves that to us
- his behaviour is permanent and pathologic; we, experienced Dutch mappers, have done everything to get Gerrit on the right track, without success.
Martin
========
[edit]: typo
Added to Martin Borsje’s excellent summary:
Apparant from his edits, GD imports data from (probably) his employer’s GIS system. We have no clear answer yet of the licence and attribution conditions of this GIS, i.e. whether the updates aren’t under copyright and/or database right, causing a potential liability problem (for OSM, and also for GD himself)
Tijmen
Earlier he mentioned that he took the data from.
There is mentioned, BGT downloads worden beschikbaar gesteld via de CC-BY-4.0 licentie.
BGT = Basisregistratie Grootschalige Topografie , Goverment name of this dataset.
Later he called it on the changeset “Basisregistratie Gemeentelijke Topografie” a twist, G for g gemeentlijke=municipality, bgt.
Maybe he still used the above download method or, as IIVQ mentioned, used data from (probably) his employer’s GIS system.
No answer from him to this point, here on the forum or at changesets notes.
Die boerderij met de naam “Arbeid Adelt” bestaat echt. Je kunt hem op Google Maps opzoeken.
Het changeset commentaar illustreert wat mij betreft echter weer wat ik altijd al dacht: dat er vooral “gratis” meegelift wordt op de OSM infrastructuur voor eigen doeleinden.
Andy,
As others stated, Martin Borsje’s summary is accurate and to the point. The user is violating about every written or unwritten rules regarding usage of OSM infrastructure.
There is not much I can add to Martin’s remarks, except maybe that the user’s behaviour also sets a very unwanted precedent: that official government agencies can attempt to take over parts of OSM that they consider under their personal jurisdiction. This is highly undersirable from multiple points of view. Besides the obvious negative consequences of “control” by such an agency, it is likely that OSM is being used as a no-cost alternative to internal mapping servers. While the latter is probably to some extent allowed usage under fair usage policies, I doubt the OpenStreetMap community would like to see all professional GIS users of the globe to start micro-map their jurisdictions and more or less implicitly claim ownership.
As to the specific remark about “Is part of the problem here perhaps that users are unwilling to delete previously-imported data?”:
While it is certainly true that there may be some “reluctance” at times to delete previously imported data, as it often requires extensive re-digitization to fill in the “gaps” with current data, it is only a secondary issue related to this request towards the DWG. Also, there are enough users in the Dutch community, that do make the modifications if and when required to update areas of the map. Of course, with high coverage of data in the Netherlands, noticing missing or outdated data is less obvious than in some other countries.
But again: these are secondary considerations regarding this request in my opinion.
Marco
Gerrit is still active.
If we see, as an example, changeset 43420815, he maps a path as follows:
highway=path
area:highway=yes
Apart from senseless area tagging of a (rather small) footpath (area mapping of highways is still proposed) it should then be:
area:highway=path instead of ‘yes’
If have not reverted it. Waiting other comments.
Gerrit is nu druk bezig met een serie bewerkingen: “Staphorst, foutjes hersteld” waarbij hij osmose als source opvoert!?
Hmmm…, dat kan toch in zekere zin?
Osmose als bron voor de lokatie van fouten, zodat je weet waar je wat moet fixen? Dat vergt vervolgens natuurlijk wel een andere bron van data of veld-inspectie + gezond verstand en editor tools om het probleem vervolgens te kunnen oplossen.
De kwaliteitscontrole tool “Osmose” natuurlijk niet te verwarren met de conversietool “Osmosis”…
De naamgevingen willen nog wel eens verwarrend zijn: