[Tagging] I am reverting an edit to the "Names" page on the Wiki

[bgo_eiu (OSM mailing list email)]

Sorry, but your response contains no information other than just, "no I don't like this." Punjabi is a more widely spoken language than Polish or German and it is not particularly exceptional in this. Serbian, Belarusian, Brahui, Mongolian, Kazakh, Kurdish, Azeri, are all major languages which have multiple orthographies just to give a few examples. The wiki is for public editing and discussion, you did not even bring this up on the wiki talk page, and I don't think many people have ever agreed with your whining on here. I notice people complain about your Wiki edits on Slack regularly.


--
https://www.openstreetmap.org/user/bgo_eiu
https://wiki.openstreetmap.org/wiki/User:Bgo_eiu

Sent with Proton Mail secure email.

------- Original Message -------
On Thursday, August 11th, 2022 at 10:00 AM, Niels Elgaard Larsen <elgaard at agol.dk> wrote:


> bgo_eiu (OSM mailing list email) via Tagging:
>
> > Hello,
> >
> > I am the user that made these changes and would request that these be restored. My intent was not to cause disruption, but rather to document the current situation on OSM more accurately.
>
>
> That is not the current situation in most of OSM, but maybe for some areas.
>
> > I can provide some examples which may be informative.
> >
> > I have been editing place names in Punjabi, which is a language which is written with two writing systems, Shahmukhi and Gurmukhi. Most Punjabi speakers are only able to read one or the other, so it is common practice to transliterate to have strings for both scripts. If we are saying transliteration is a problem as a rule, that would mean we are discouraging transliterations which allow people to read text in their own language which is clearly not something the OSM community would encourage.
>
>
> Or Punjabi is the exception to the rule.
>
> > Further, it may not be obvious what "duplicating" a translation is if they are unfamiliar with the scripts being used. Each Arabic-based script uses different rules, and what may look like a duplicate to some readers is actually a necessary clarification.
>
>
> But that is reason for the wikitext discouraging automatic transliteration.
> They bring no clarification, but they can convey false information.
>
> > Take for example Punjabi and Saraiki which have a character ݨ. In Urdu, this character is written the same as ن, which is also used in Punjabi and Saraiki. So the word پانی in Urdu may be represented by پاݨي in Punjabi Shahmukhi and Saraiki Shahmukhi because it includes a nasalized sound which has its own character in those languages. Now let's say we have کارنا. We have to write this as کرنا in Urdu, Punjabi, and Saraiki. This is because in this word, the sound is not the one which gets a different character in some other languages. The only way to record differences like this is with multiple strings, but if editors think these are "duplicates" instead of indicating different information about how the word is pronounced for each language, then we are losing information. It would not be appropriate to encourage this without a warning that unless you know about the writing systems of the other languages, you cannot know that the same letters mean the same thing just because they look the same.
> >
> > Hopefully that clarifies why these updates are necessary. Cheers.
>
>
> Maybe you can get support for changing the wiki to better describe how to handle this
> issue. But do not just change the wiki.
>
>
> --
> Niels Elgaard Larsen
>
> _______________________________________________
> Tagging mailing list
> Tagging at openstreetmap.org
> https://lists.openstreetmap.org/listinfo/tagging