Remove all analyser_merge_(xxx).py

[Pecita]

Please remove all analysermerge(xxx).py that are useless and pollute Osmose with erroneous and unsolicited information.

[frodrigo]

This request seam a bit excessive.

Please explain more precise problems. We are open to fix and improve.

[Pecita]

This request just have sense that is obviously not the case of these analysers. The Corsican map need a significant effort to be improved and osmose is a necessary tool to identify errors. These scripts create artificial errors based on false data, are incomprehensible, etc. They do not need to be improved but simply deleted.

[frodrigo]

If some OpenData are too bad on your are we can disable it. But please be more specific. All OpenData data source at once cannot be bad on your area.

[Pecita]

Please disabled all your data : there are not errors.

[Pecita]

Please set the bug label. It is not a question.

[frodrigo]

Please disabled all your data : there are not errors.

Osmose reports issues, not errors, you have to review. You can use false-positive button on bad issues.

Please set the bug label. It is not a question.

You have no question. but I have. Please point what is wrong for you.

I cannot understand your point of view without examples of bad data. And since you want all to be be disabled, many various examples. Please provide arguments.

[Pecita]

There are no question. The fact that the data reported in this way is false and useless is not the main problem. The main problem is that this data are not appropriate on osmose. These are neither errors nor issues but an illegal copy of a copyrighted card. They are confusing without contributing anything. It could be a (false and useless) layer on JOSM if copying this data was not illegal: 4. Droit de propriété intellectuelle Les illustrations, les animations, les vidéos, le contenu éditorial, les divers éléments de la charte graphique ainsi que les données cartographiques issues de la consultation en ligne figurant sur le site Géoservices ou des lettres d’information envoyées par l’IGN sont des éléments dont l’IGN détient la propriété ou le droit d’exploitation. L'utilisation des documents ou éléments du site est soumise à l'accord préalable de notre établissement.

[frodrigo]

We care a lot about copyright.

Concerning the data from IGN, most of the IGN data is now available on the free licence "License Ouverte" compatible with the ODbL of OSM, since 2021. You can read eg https://www.zdnet.fr/blogs/l-esprit-libre/open-data-l-ign-libere-ses-donnees-39914791.htm

On the bad quality data, are you also including data from the IGN ?

[Pecita]

This is not the main problem. Osmose is not a mapping tool, it is a quality insurance tool. Please remove all analysermerge(xxx).py that are useless and pollute Osmose with erroneous and unsolicited information. Let the people work with survey, local knowledge, respect of the osm data conformity. Let osm tools be coherent. Don't cheat.

Notes :

• Ok for the opening of the BD TOPO database. So it can be made available on JOSM.
• Yes, the IGN data are in many cases obsolete and in this case uninteresting. A grave, abandoned or not, in a private property as there are hundreds in Corsica has no interest in osm data.

[Famlam]

You can disable analysers you don't like in the menu on the left ;)

[Pecita]

You can disable analysers you don't like in the menu on the left ;)

Please reply to the bug. Please do not speak about other things. Please solve the bug. This is not the main problem. Osmose is not a mapping tool, it is a quality insurance tool. Please remove all analysermerge(xxx).py that are useless and pollute Osmose with erroneous and unsolicited information. Notes:

• To like or dislike an analyzer you have to understand what it does there. In this case it's impossible because it has no sense being there. You work with conviction to provide an intelligent map and you use Osmose to check the consistency of the data you enter and you come across inconsistent alerts!

[Famlam]

But as frodrigo requested, you haven't mentioned what the bug is, just that you want to disable all analysers.

Osmose is not a mapping tool, it is a quality insurance tool.

I'd argue quality assurance also includes helping to provide output to spot missing items, i.e. making sure the data is complete. (Also, I like it that it gives suggestions of missing items, even though some items totally do not interest me. That's personal and those analysers I just disable :) )

[Pecita]

But as frodrigo requested, you haven't mentioned what the bug is, just that you want to disable all analysers.

The bug is the actual presence of these analyzers that make incomprehensible and false issues appear. I do not ask that they be disabled but removed, purged. Please don't distort my words.

Osmose is not a mapping tool, it is a quality insurance tool. I'd argue quality assurance also includes helping to provide output to spot missing items, i.e. making sure the data is complete. (Also, I like it that it gives suggestions of missing items, even though some items totally do not interest me. That's personal and those analysers I just disable :) )

Your clean map appears to be filled with false issues. To manage the display of these analysers you have to understand that they do not correspond to OSM data, and that they give false indications, which is a long and unnecessary waste of time.

[vinber]

Hi, for more than 10 years osmose has been proposing comparison with open data to improve OSM data.

I can totally understand that it does not suit you but to consider it as a bug is, it seems to me, a misunderstanding of the project.

If it doesn't suit you, you can not use it :) you probably have other quality tools that can better suit your uses https://wiki.openstreetmap.org/wiki/Quality_assurance#Error_detection_tools

[Pecita]

Hi, for more than 10 years osmose has been proposing comparison with open data to improve OSM data.

I can totally understand that it does not suit you but to consider it as a bug is, it seems to me, a misunderstanding of the project.

I said bug out of politeness not to say disinformation.

If it doesn't suit you, you can not use it :) you probably have other quality tools that can better suit your uses https://wiki.openstreetmap.org/wiki/Quality_assurance#Error_detection_tools

Thank you for this good advice. On the other hand, may I recommend you if you come to hike in Western South Corsica to rely on openstreetmap.org rather than on the topographic map of IGN if you want to avoid getting lost?

[Famlam]

Oh sure we will as osmose is a tool to improve osm ;) If any more specific issues are reported, we can also try to improve Osmose itself

[XandrexOSM]

Good evening !

I read thoroughly all the messages in this discussion. The original poster claims things, and does not provide any example. I do not see how the other people can be convinced without at least one proof.

---

Why was this inquiry opened by you, @pecita? I am not familiar of Corsica. I do not know you. What I read in your messages here look like what a educated troller would post.

If you really aim at making osmose better (I hope you do, I too wish to have good alerts in osmose and no useless noise), then the first thing for you do to is find a way to convince one person, and then two and so on. So far, I do not see any effort in that direction from the messages posted by you here. Examples are a very good method to convince people. The current method does not seem to be successful (but once again it depends on what your goal is).

[Pecita]

I read thoroughly all the messages in this discussion. The original poster claims things, and does not provide any example. I do not see how the other people can be convinced without at least one proof.

?! Il n'est pire sourd que celui qui ne veut pas entendre ! Why was this inquiry opened by you, @Pecita? I am not familiar of Corsica. I do not know you. What I read in your messages here look like what a educated troller would post.

If you really aim at making osmose better (I hope you do, I too wish to have good alerts in osmose and no useless noise), then the first thing for you do to is find a way to convince one person, and then two and so on. So far, I do not see any effort in that direction from the messages posted by you here. Examples are a very good method to convince people. The current method does not seem to be successful (but once again it depends on what your goal is).

First of all, if I claim for the deletion of these analyzers it is to preserve the relevance of Osmose. Who are the trolls?

• So be it, though. Let's take the example of "cimetière non intégré". These are graves, if they still exist, that are located on private property. What cartographic interest do they have? The code generated to "correct" them is also illegal: landuse:cemetery, without area.
• I've wasted too much time with this nonsense, but none of the analysis provides proper information. If you had examined them at all you would be convinced. They are only there to tell you "the good data you provided should be corrected by bad data". BD TOPO IGN provides mostly old, obsolete data. To give examples would be to list almost all of them. It would be rather the opposite that would be reasonable. You can for example, on my village, Coti-Chiavari, compare the layers BD TOPO IGN and OSM using geoportail.gouv.fr
• Once again, it is not Osmose role to impose mapping data, even if relevant, which is not the case.

[Famlam]

?! Il n'est pire sourd que celui qui ne veut pas entendre !

Please keep stuff polite

What cartographic interest do they have? The code generated to "correct" them is also illegal: landuse:cemetery, without area.

Cartographic interest depends per person. If it exists, it's mappable. Some people map individual street lights or trees. What you focus on depends per person. Also, landuse=* doesn't need an area tag

They are only there to tell you "the good data you provided should be corrected by bad data". BD TOPO IGN provides mostly old, obsolete data. To give examples would be to list almost all of them.

Start with 4 examples from Osmose :).

Once again, it is not Osmose role to impose mapping data, even if relevant, which is not the case.

Which is why one can disable any analyser (s)he doesn't like :). Also, you're the first and only person to report this (in an aggressive, non-constructive manner). Can you point us to a discussion where multiple people complain about this data source being completely inaccurate?

[Bibi56]

?! Il n'est pire sourd que celui qui ne veut pas entendre !

Still not a single example, closing this fake bug report is probably the best way to move on.

[Pecita]

?! Il n'est pire sourd que celui qui ne veut pas entendre !

Still not a single example, closing this fake bug report is probably the best way to move on.

Fake news is when you quote the answer without the question. It is when you present as errors the information provided by the IGN which is mostly outdated, false or without interest. It is when the intelligence of cooperative work is flouted.

[Famlam]

I support closing it as the information requested is consistently not provided

[Bibi56]

Fake news is when you quote the answer without the question.

My assumption was that people were able to read a few lines upper. I don't see any fake news here. On my side at least.

So just for you, as exception proving my thumb rule:

I read thoroughly all the messages in this discussion. The original poster claims things, and does not provide any example. I do not see how the other people can be convinced without at least one proof.

?! Il n'est pire sourd que celui qui ne veut pas entendre !

Still not a single example, closing this fake bug report is probably the best way to move on.

If a specific OD source seems to be outdated at a specific place, you should indicate which analyzer (which can't be analyser_merge_(xxx).py of course, therefore the "fake bug report") and where, giving examples.

Please remove all analysermerge(xxx).py that are useless and pollute Osmose with erroneous and unsolicited information. This is fake news: many analysermerge(xxx).py provide Osmose with useful information, tips, etc...

As you can't give any example after it has been with good right requested, KISS: if in your area a specific source is inaccurate, just filter it out and use this filtered URL, that's it.

[Pecita]

Compare https://www.geoportail.gouv.fr/carte?c=8.79757000805735,41.7886428088602&z=16&l0=GEOGRAPHICALGRIDSYSTEMS.MAPS.SCAN25TOUR.CV::GEOPORTAIL:OGC:WMTS(1)&permalink=yes and https://www.openstreetmap.org/#map=16/41.7881/8.7995. It doesn't have much to do with each other. They are not the same place names, not the same paths layout, not the same information. (etc. etc. etc...) Prescribing to copy IGN topographic maps into OSM is a big mistake.

[Bibi56]

(Sorry I changed the language saw that where rereading the text)

Deux semaines pour ne pas avoir de lien potable ! Bon allez enfin un lien mais pas avec Osmose... https://osmose.openstreetmap.fr/fr/map/#zoom=16&lat=41.7882&lon=8.80309&item=xxxx&level=1%2C2%2C3&tags=&fixable=

Et donc on voit que vous ne respectez pas les conventions. Avant de critiquer, commencez par respecter les conventions. En France (et jusqu'à nouvel ordre la Corse en fait partie), la convention est, comme dans les autres pays, de mettre le ou les langues officielles dans le champ name. Et comme vous ne respectez pas cette convention, la règle renforçant le respect de la toponymie à la française dans name fait tilter Osmose. De même la différence entre le name et name:fr.

https://tile.openstreetmap.fr/?zoom=16&lat=41.79092&lon=8.80573&layers=B00000000 propose des cartes en breton, occitan, basque, pas en corse. Si les Corses veulent une carte en corse, elle y sera bienvenue (et il en existe peut-être, il faudrait demander dans ce cas à OpenStreetMap.fr de l'ajouter afin de la valoriser). Des débutants ont commencé comme vous en breton, basque, occitan. Maintenant name=name:fr dans les régions en question avec des name:br, name:eu, name:oc qui vont bien et de bons progrès sur les cartes en langues régionales.

Autres erreurs :

• Intersection entre la voirie et une petite surface d’eau. Effectivement il faut indiquer que c'est un gué (je suppose).

• Voie sans issue sur un chemin avec de multiples sorties : Comme indiqué il faut préciser que c'est effectivement sans issue.

• Point de vue non intégré : si ce n'est pas un point de vue, mettez un faux positif. Usuellement les points de vue de l'IGN ne sont pas déconnants, mais pourquoi pas.

• Sommet non intégré Punta di Gradello : là non plus je ne suis pas du coin, si c'est un faux positif, noter comme tel.

[BalooUriza]

I personally find the analysers for traffic signs and street objects incredibly valuable for locating missing details and it's trivial to flag incorrect detections as such or just disable the category when not relevant to the task.