[OSM-talk] Proposed bulk removal of service=driveway2

[john whelan]

I think we are in agreement that it should go but given it is only 10,000
occurrences the debate is more sledge hammer mechanical edit or at least
inspect each one before deleting and I tend to prefer the latter solution.

Cheerio John

On Sat, Sep 3, 2022, 11:03 Zeke Farwell <ezekielf at gmail.com> wrote:

> Given that the proponent of highway=service + service=driveway2 uses it to
> mean exactly the same thing as highway=service without a service= tag, I
> see no reason not to remove service=driveway2 from the database.  If they
> had used a different ATYL value that wasn't so easily confused with
> service=driveway then the value could be easily ignored or perhaps gain
> wider acceptance.  However, driveway2 is not a semantically useful value.
> It just causes confusion for mappers.  I say get rid of it.
>
> --
> Zeke
>
> On Sat, Sep 3, 2022 at 10:52 AM Andrew Hain <andrewhainosm at hotmail.co.uk>
> wrote:
>
>> The mapper who created service=driveway2 previously changed the
>> definition of service=driveway on the wiki, which was reverted. Uses of
>> service=driveway created by this user are also worth reviewing.
>>
>> --
>> Andrew
>>
>> ------------------------------
>> *From:* Simon Poole <simon at poole.ch>
>> *Sent:* 03 September 2022 11:52
>> *To:* talk at openstreetmap.org <talk at openstreetmap.org>
>> *Subject:* Re: [OSM-talk] Proposed bulk removal of service=driveway2
>>
>>
>> Am 03.09.2022 um 10:48 schrieb Minh Nguyen:
>> > Vào lúc 01:38 2022-09-03, Minh Nguyen đã viết:
>> >> In 2020, a mapper coined a new definition for service=driveway. In
>> >> the last two years, they've systematically applied it to about 10,000
>> >> ways in Japan and Singapore.
>> >
>> > This should read: a new definition for service=driveway2. Sorry for
>> > the confusion.
>>
>> You elicited a couple of WTFs from me :-)
>>
>> Anyway IMHO this would seem to make more sense as a maproulette
>> challenge or osmose warning than a bulk edit given the number is quite
>> manageable and at least some of the objects can probably be mapped to
>> one of the "proper" subtypes.
>>
>> Simon
>>
>> _______________________________________________
>> talk mailing list
>> talk at openstreetmap.org
>> https://lists.openstreetmap.org/listinfo/talk
>>
> _______________________________________________
> talk mailing list
> talk at openstreetmap.org
> https://lists.openstreetmap.org/listinfo/talk
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/talk/attachments/20220903/35f8a2a9/attachment.htm>