I propose that denomination=latter-day_saint replace denomination=mormon as the standard recommended tag for this church. I believe this is a good balance between keeping a concise tag, while still following their recommendations of referring to the church by the full name on the first occasion (in name=), as in a very large majority of cases (every chapel used for regular Sunday worship), the full set of appropriate tags would be:
name=The Church of Jesus Christ of Latter-day Saints
amenity=place_of_worship
religion=christian
denomination=latter-day_saint
denomination:wikidata=Q42504
operator=The Church of Jesus Christ of Latter-day Saints
operator:wikidata=Q42504
From what I could find on taginfo, the following variants exist:
I recommend using a hypen instead of underscore between ‘latter’ and ‘day’ despite the latter currently being larger to be consistent with the full name of the church.
It was previously brought up on the talk page for denomination=mormon that there may be collisions with other splinters of this religion. Having no experience with any of them, I’m unaware if any also identify with the title for members of ‘Latter-day Saint’. Looking through the wikipedia page, it appears they handle splinters by appending the name of the one who first led the splinter to the name if they continue to use the same name after splintering.
Once a consensus has been reached, I would accept heading off a semi-automated project to perform this replacement.
I think the denomination should be singular as that’s consistent with denomination name. The specific church most of us are familiar with is plural, so leaving that for the Operator work for me.
A quick scroll through all values for denomination=link looks like no value with more than 50 uses is abbreviated, add on top of that general guidance on the wiki to avoid abbreviations, and I feel like =LDS would feel out of place
In the old days, there was a general consensus to limit the number of denomination=* values by sticking to generic values. At the time, apparently mormon was seen as more generic than latter-day_saint. (And, let’s be honest, some probably liked the idea of less typing.)
The general categories were always an overly simplistic view of Christianity, if not other faiths as well. It’s already been over a decade since catholic gave way to roman_catholic, protestant gave way to a multitude of specific values, and methodist to even more values besides. So it would be fair to use more specific values for the various Restorationist churches too.
Should denomination=latter-day_saint include the smaller churches as a generic value, using operator=* to clarify the specific church? Or should latter-day_saint always refer to The Church of Jesus Christ of Latter-day Saints, while we rely on other values such as strangite and true_and_living_church to refer to the smaller churches? If the latter, denomination:wikidata=* would help mappers make sense of the most obscure values.
A few years ago, when we did an import of places of worship in my area, we standardized on denomination=latter_day_saints (no strong opinion on punctuation) to refer to The Church of Jesus Christ of Latter-day Saints. We used operator=* to indicate more specific parts of the church, such as “Cambrian Park Ward of the Church of Jesus Christ of Latter-day Saints” and “The Church of Jesus Christ of Latter-Day Saints – San Jose South Stake”. I’m not entirely sure if it was necessary, but it seemed analogous to how operator=* was being used in other organized denominations, such as Catholic and Episcopal dioceses.
If larger religions have already broken down to be more specific, I say let the pattern continue and have denomination=latter-day_saint refer specifically to The Church of Jesus Christ of Latter-day Saints (I unknowingly previously did this by creating the one use of denomination=house_of_aaron). My first impression is that members of these other churches would not enjoy being categorized by their affiliation to the church they branched off of, either.
I worry about using operator= to refer to specific wards and stakes. New ones are created or old ones dissolved not very often, but often enough and quiet enough that you will only know about it if it is happening to the one you are a part of. If this leads to a full fledged import, it looks like it is possible on the current version to link directly to a specific building or ward (random examples), so I would prefer that to naming the wards and stakes in OSM.
This seems like a clear-cut case then. operator=* would be unnecessary, but not unreasonable either.
It was an import in only one county based on a one-off data collection program. It’s unlikely we’ll ever see an import quite like it again. In general, operator=* is often more difficult to determine in a field survey. While I wouldn’t necessarily advocate for a more widespread import, it sounds like it isn’t internal, behind-the-scenes information if it’s available on the public website. So an AllThePlaces scraper of that site would be able to keep OSM in sync with local changes as necessary.
My problem with changing the denomination to “LDS” is that it implies that they follow post-biblical teaching. Beyond that
it doesn’t say anything about the actual belief system or even it’s origin
I agree that we should deprecate mormon as a value.
For brevity, I think lds would work just fine. It’s widely used colloquially (I think also by members of the church?).
If we go with something longer, I would not want to mix dashes and underscores.
What do you mean by “semi-automated” @dknelson9876 ?
(I’ve lived in Utah for 13 years and come across mentions of the church almost daily.)
I mean that I wouldn’t blindly replace all current occurrences in one changeset, but I wouldn’t manually review every individual use either. Something smaller and more manageable (county scale is probably good enough outside of Utah) and looking at all of the tags combined in JOSM to make sure that nothing looks too out of place.
For brevity, I think lds would work just fine. It’s widely used colloquially (I thnk also by members of the church?).
While a style guide (linked above) is not law, it does include in bullet points 2 and 3 to not abbreviate “Latter-day Saint” to “LDS”. It does still get abbreviated in conversation, simply because decades of habit is hard to break. The church’s website did use to be lds.org (they still own it, it just redirects).
Abbreviation is also not that important to me, when in most cases thanks to our tools won’t it turn into lat<tab> for autocompleting?
If we go with something longer, I would not want to mix dashes and underscores.
I don’t have strong feelings either way. If more people would rather have just underscores, I’m fine with it. I just used the name of the church as an almost tie breaker between the two versions.
This seems a bit too brief? There is some precedent for using TLAs as tag values, such as cycleway=asl, but if we’re optimizing for mappers working with raw tags, then a mapper is liable to misread the lowercase L as an uppercase I…
If the LDS (sorry) style guide discourages it, I don’t see why we should go against that. I don’t think there’d be a big issue with data size, it’s not like there are millions and millions of wards and temples.
The church is Saints, the denomination is Saint. The are other churches under the Saint denomination. In the same way there are different Catholic churches or different Anglican churches. All being Christian.
I’m not sure that distinction is normally made based on pluralization. A denomination=latter-day_saint tag, however punctuated, would refer to either the Latter Day Saint movement or the Latter-day Saint Church, more commonly referred to as the “LDS Church” or “The Church of Jesus Christ of Latter-day Saints”. If it refers to the movement generically, then in fairness there should be a denomination=church_of_jesus_christ_of_latter-day_saints or more practically denomination=lds_church to distinguish the specific church.
To elaborate, the wiki documentation for denomination=* nods to the old consensus on the tagging mailing list, that denomination=* is for general denominational categories, while operator=* is for the specific church. In other words, denomination=* is only for the -isms on the right side of this phylogenetic tree:
But so many denominations are difficult to categorize into a traditional hierarchy like that. Wikipedia classifies Christian denominations on multiple axes. For example, the world’s largest Methodist denomination, the UMC, is classified as:
We could say a clearly-marked UMC church should be denomination=methodistoperator=United Methodist Church, ignoring all these historical and theological nuances, but over 2,400 occurrences of denomination=united_methodist would suggest otherwise. It only gets harder from there. Is the AME Churchmethodist, episcopal, or anglican? Is the Church of God (Anderson, Indiana)wesleyan, holiness, or restoration?[1]
In some localities, a mapper on a field survey would hardly be able to tag denomination=* at all. Even if so, you can imagine how difficult it would be to use the POI data for finding compatible Sunday services. The tags representing general categories are fine as a first pass, but they aren’t a replacement for more specific values, and the database reflects that reality.
In hindsight, we should’ve made denomination=* a freeform text key like operator=* and name=*, using denomination:wikidata=* to distinguish any naming conflicts. Then denomination=The Church of Jesus Christ of Latter-day Saints would be a no-brainer, and it would be possible to recreate this fantastically complex tree of denominations. But that’s not what we ended up with.
For what it’s worth, my local phone book has no qualms about spelling out “Church Of Jesus Christ Of Latter-Day Saints” in full. Then again, look closely at the second entry under “Church Of God, Anderson, Indiana”.
The headquarters location isn’t a part of the denomination’s official name, but phone books always include it to distinguish it from the other Churches of God. ↩︎
This phone book landed on my doorstep back in 2017. Its alphabetical listing of landlines was much scrawnier, malnourished, compared to the ones I remembered from my childhood. I took pity on it and decided to not immediately toss it out. Instead, I used it to gauge OSM’s level of completeness in my area, counting the entries by hand and comparing the tally to an Overpass query. It sounds ridiculous, but not as ridiculous as many of the alternative methods for assessing OSM’s coverage. At least it gave me something to say about the state of OSM in my community.
Then I told a coworker about my little project, and they immediately ran home and gifted me their copy of the phone book for their area. Thanks.
These days I still occasionally turn to this phone book to gut-check whether the classification decisions we’re making in OSM align with the sort of classifications that people use in the real world day to day. For example, in this discussion: does it make sense to distinguish Christian denominations more specifically than a few broad categories? Or is it adequate to treat denominational groups as mere organizations?
To me, the phone book’s detailed breakdown suggests that a churchgoer wouldn’t want to show up at a Methodist church on Sunday, only to discover that it’s the wrong kind of Methodist. By contrast, the “Restaurants” section is broken down by cuisines such as “Pizza”, but it doesn’t devote a subheading to each franchisee of a pizza chain. operator=* suffices for that sort of detail.
Additionally, old phone books are still used quite heavily in the fields of legal defense (for crafting an alibi) and cultural heritage management (for preparing a historic context statement). I hope someday OpenHistoricalMap will also have an important role for these use cases.
