Updating WikiProject Israel

I think icons would help locate the required POI quickly. In my opinion, pictures and text take longer to comprehend. For example, if I want to map ruins, it would be faster for me to locate the ruins icon than to go over a list of words and pictures and find the specific word used for that tag (it might as well be debris, rubble, wreck or wreckage) or to understand that one of the pictures depicts ruins. So even an outdated icon is preferable.
Links to OSM wiki are already there, because I’m using OSM wiki links for all tags and keys.
Regarding the legend tooltip - this would be great, but as @zstadler mentioned on the github thread, some icons require complex sets of tags, therefore the full info should be on the Israel wiki, while the tooltip will improve the workflow for simpler cases.
Regarding missing POIs - I will add all IHM legend entries to the wiki. Please feel free to suggest additional tags to be mentioned.

I will translate the entire wiki page to Hebrew once I finish the current set of changes.

Perhaps we should wait a bit before translating to see what the other editors think? Israel may be a special case where there’s a large amount of Arabic/Russian speaking people who may prefer English. I can’t tell how many of them are contributing, though…

The Hebrew translation will appear on a separate place (here).

I’d like to have consensus on the English version before I start translating to Hebrew. Unfortunately I speak only two languages…

Looks good!
I can’t currently think of anything that is missing.
In case I’ll find something that is missing, I’ll add it to both languages.

Looks good. We’ll have to make sure both versions are kept in sync.

Can I use the IHM legend rendering examples (licence wise and technically)?

Sure.

Right: we should prevent a situation where each translation of the page has different content. Maybe the English version of the page can be the master version?

Hey everyone,

I’ve finished a major edit of WikiProject Israel page. In terms of mapping guidelines and convention - there is supposed to be no change whatsoever.

The following steps would be:

  1. Adding rendering examples.
  2. Going over all the existing text and resolving all unclear and incorrect information.
  3. Translation to Hebrew.

Your help would be much appreciated.

Thanks

There’s potentially a minor issue with our track guidelines. Our conventions are “track is always dirt”. See this: https://www.openstreetmap.org/changeset/31346575.

Also, not really “incorrect” information, but a nice enhancement: We’ve previously agreed adding an photo example for every common road type would be a nice thing to have to help people orient themselves and choose the right tags. This is particularly relevant for highway tags that confuse new mapper like path, footway, cycleway, track.

No one has done this yet.

To spell out the issue: https://wiki.openstreetmap.org/wiki/WikiProject_Israel#Other_Road_Types specifies that highway=track is always unpaved, but https://wiki.openstreetmap.org/wiki/Tag:highway=track specifies that highway=track is “usually unpaved (unsealed) but may apply to paved tracks as well” (which can be tagged as surface=paved). Moreover, the Israeli guidelines specify that highway=unclassified includes agricultural and forest roads, but https://wiki.openstreetmap.org/wiki/Tag:highway=unclassified specifically excludes “agricultural tracks”.

An example is https://www.openstreetmap.org/way/346149552 (see also discussion in the linked changeset): it is currently tagged as highway=unclassified, correctly according to Israeli guidelines but incorrectly according to international ones, and the suggestion was to convert it to highway=track, correct according to international guidelines but incorrect according to Israeli ones.

How shall we resolve these conflicting guidelines?

Personally, I think the international guidelines should trump. The strength of OSM is that it uses just one data model for the entire world; if every country started using its own rules then the ecosystem wouldn’t be able to thrive.

The wiki for highway=unclassified clearly states that:

In other words, there is no internationsl guideline for unclassified roads that is in conflict with the local guidelines. On the contrary. There was an intentional lack of intent for international guidelines to take precedence over country or regional guidelines.

Do we have a special need that requires different rules?

In some African countries where nothing is paved, what we tag as “unclassified” could get tagged as “primary”, I can see the need for different rules there.

It appears the highway tag is solely for the road importance relative to the road grid. Different highway tags imply different paving statuses in different countries, and those paving statuses can be overridden explicitly, but the main message of the highway tag is the road importance.

Our guidelines imply: “track and unclassified are exactly the same importance-wise, but one is paved and the other is not”, thus we killed the importance aspect between the two tags.

So perhaps “Physically, the roads which should be tagged in OSM as highway=unclassified can vary greatly between countries” was interpreted too liberally.

Let me expand on my previous point with a practical example. Let’s assume renderers mark “unclassified” more prominently than “track”.

This unpaved road connects Daliyat el Carmel to Ein Hod. Internationally it’d be “unclassified” and would be rendered more prominently than the surrounding tracks because it’s more important in the road grid, but our guidelines make it look just like any nearby track.

Even worse, if there had been a paved but barely used forest track nearby, our guidelines mandate tagging it as “unclassified”, making it look more prominant on the map than the relatively important village-connecting road.

PS. the only reason the road looks special on Mapnik is because the forest nearby was mapped more precisely around it. The track line itself is identical to the nearby tracks

PS2. it’s a non-important road in the “grand scheme of things” but it is the most important in the area when it comes to the routing grid.

I think I’ll later open a dedicated thread for the unclassified vs track issue.

I’m still a tiny bit uncomfortable with adding lists of POIs that are not special or specific to Israel. It feels a bit like tagging/documenting for the IHM renderer. Although I don’t mind or oppose this in isolation, I think we should remember that the Wiki is about OSM and not IHM, and we shouldn’t over-fill it with too many IHM-specific sections in the future.

(Also, I think most of those are already present in the IHM מקרא list, and maybe we should add the OSM tags into that list)

“Tagging for the rendered” says “Don’t deliberately enter data incorrectly for the renderer”. [my emphasis]

IHM and its team are actively promoting correct mapping.

My emphasis. That’s renderer tagging.

Firstly, I appreciate your work and I love IHM.

As you pointed out, the list itself is not literally tagging for the renderer (hence I used the term “a bit like”), but it’s promoting and documenting a certain subset of tags solely because a certain renderer supports them, and “encouraging certain tags because of a certain renderer” is a similar concept to “tagging for the renderer”. The list lacks very common tags like bus stops and convenience stors, and has rare tags like quarries. So it’s not a list of common IL tags, it’s solely an IHM renderer documentation.

Also, Ron said it’s meant for quick lookups. If someone is encouraged to use “tourism=attraction” rather than the more specific leisure=horse_riding because IHM supports only the former and only that’s on the list, then that’s a textbook case of renderer mapping, encouraged by the list.

I am not trying to nitpick the list and start a discussion with little practical impact, I’m just pointing out that rendering and mapping separation is important long term, and hoping that future Wiki edits take this into account. It’s not an IHM documentation page.

Edited the post. Minor rephrasing only.