Decimals might not be used on US mileposts, but they are very commonly used on other countries' mileposts - and OSM is a global database.
While it's pretty save to say that foot traffic is forbidden on every motorway, worldwide.

Anyway, this obviously violates automated Edits code of conduct, so I reverted this.

How exactly is "operator:wikidata" helping to "mitigate this sort of data degradation in the future"?

It especially degrades the data when people forget to update both tags and they conflict with each other.

"Railroad features other than mileposts often have linear referencing positions in tenths and hundredths of miles"
Exactly, since many other railway:position's use decimal numbers, I'm apprehending .0 here to clarify that those mileposts haven't.
If data consumer's aren't intrested in that, they should remove it themselfes.

And where did you discuss this mass edit?

Where was it discussed to add those pointless wikidata duplications and replace "CSX" with a longer variant (and all other edits within this massive automated edit series)?

mi:100 means somewhere at mile 100, mi:100.0 means exactly mile 100

Uhmm ... where was this discussed?

This greatly reduces the data's accuracy.

We've gone over it, but we never "reached" anything.

The thing is: Those (razed) railroads are not compareable againest, for example, razed buildings or even landuses. Because the difference is that they are part of a large interconnected network. They are also still closely related to the existing network - because they often played an important role in how it looks today. Unfortunately, this is all a bit complicated to explain. But the point is that good railroad data also always includes some basic data on former routes.
I have 2 "professional" railroad map books at home, and they include every little abandoned branch line, even if it was only operated for a few years in the 1870s. (not from countries where I edit; don't worry)
And such data, which is just very strongly connected, should also be kept and maintained together.

If you look at the whole thing with e.g. buildings: A razed building is ...well a building, it is not directly connected to or dependent on other buildings. In addition, it has no relation to today's world.
And if one would start now to map tons of random demolished buildings in OSM, then one would lose the overview very quickly; tons of data would overlap each other. On the other side, with a razed track: it is a single way, which should hardly ever trouble anyone.

I do agree with you that the vast majority of historic data one can think of dosen't belong into OSM. But railroads are a special case, and not just because I'm personally intrested in them, and also not just because they're rendered on ORM.
I've listed a lot more reasons at osm.wiki/User:Hiausirg/Razed_RRs

Further, I also don't really see why "only for ORM" should be that much of a problem, assuming it would be the only reason. It's not tagging for the renderer - that would be to intententionaly misusing tags for certain result i.e. landuse=industrial for purple flowerbed or golf path for normal roads. railway=razed is explicitly intended for "gone without traces".
The number of people interested in such routes is huge, the number of people using ORM (also specifically for such routes) is also large. So the data is -contrary to much on OSM- very actively used and helps people. And that while the impact/conflicts to overall OSM data is/are minimal.

It is not my intention to "go againest everything OSM is tying to achive", and i'm sure i'm not doing that.
My view: The "no historic data" guideline dosen't exist because historic data is sh1t, there is a deeper reason. That is, historic data can quickly lead to massive clutter and can become unverifiable quickly. However, these two problems are simply not present here.

Greetings

Once again: Ways/locations like at osm.org/edit#map=18/28.18531/-82.72091 are still clearly visible - railway=abandoned is perfectly correct. And that's only one out of countless examples. I don't know what else this is supposed to be other than vandalism.

Further, how exactly are you improving overall OSM by such railway=razed deletions? Such mapping is not forbidden, the data is perfectly verifiable.
I already suggested a possible compromise, but you ignored it unfortunately.

GDOT projects website

And to make something clear here: It is not my intention to have mapped every track that was once somewhere.
For example here, an abandoned yard: osm.org/edit#map=18/33.80652/-84.45231
All side/yard tracks were removed, and a part was overbuilt. I deleted all ways on the overbuilt part, and left them were they aren't overbuilt. It would get too crowded otherwise.

Suggestion for a compromise: Leave abandoned main routes only as a single way (which shouldn't disrupt anyone), delete all overbuilt side (spur/yard) tracks.
I think this is a fair solution for both sides.

I didn't add new abandoned lines, I added primarly start_dates. The few abandoned ways i've added only close minor gaps within the old route, and are tagged as "razed" - this means "gone without traces", so no wrong tagging.

"These should be removed from OSM since they no longer exist" -
I don't think so. Here are my reasons: osm.wiki/User:Hiausirg/Razed_RRs
"isn't permitted on OSM"
It is. There is a guideline which discourages it, but guideline =/= rule. And there are a LOT of reasons to map them like I did - they speak stronger than this guide (IMO).

"Where did you get the start dates for these items?"
From Wikipedia. A single fact from a text is certainly not copyrightable.

"Additionally, I suspect this was an import"
??? Import of what?

There are many locations like osm.org/edit#map=17/28.18454/-82.72033 where traces are obviously visible, yet you deleted them anyway. Therefore, I reverted your changeset.
(and stop blaming my manual review as bad or even as a lie, your's certainly not better)

-not import

you do realize that you're the one who fcked up reverting and deleted all ways, but not the nodes?
changeset/133150994

So stop blaming me for your mess.
And especially: clean it up, there are much more nodes whose outlines were deleted by you.

Hello,
actually, Esri Clarity is the oldest imagery. (usually from around 2011-2)
Bing or normal Esri, for example is far newer.

They rebuilt/expanded the substation, the original area which you added is now empty - except the old fence. I re-added it as way/1150970428
But nonetheless, thank you for adding the substation, even if the outline was slightly outdated.

Greetings

I would just use industrial=oil or maybe industrial=oil_storage as the main tag.

I always use an area for larger industrial facilities like this because it contains more information.

It might also make sense to only leave the address on the node, and move the other tags to the area. (And connect both using a type=site relation.) Then you'd have both the area of the tank farm captured and the point where routers should point at best. But that's just an idea.

Greetings

"We've been over this...Continuing to lie"
Repeating the same wrong information over and over again won't make it true. I know how many stuff I've fixed before uploading.
"Everyone" You're not "everyone", so don't speak for everyone.
"it's clear that you have no sense of shame" No, shame for what? For having a different view than you?
Yes, today, I would have done things differently, but there's nothing to be ashamed of.
"your newest lie about how "that's such a small percentage of the data!"." Now how exactly is this a lie? It's an quite obvious fact.

And you should calm down a bit. Statements like "I'm happy to give you a list.........I'll be working on the list of low-quality buildings; not for you" just contradict themselves.
And to make one thing clear: I absolutely HATE these never-ending discussions. I would much rather use the time to contribute something actually useful, but I can't leave this here like this.

- Which buildings are "not very well aligned"?
- Which buildings are "not squared"?

By the way, "some" buildings which are "not *very* well" aligned and "not squared" (unless veery bad) is not a good reason to delete someone's work. If you think this are good reasons, then sorry, but you can probably delete most HOT edits.

Where exactly are the bad buildings which justify to revert the edit (where I certainly reviewed most of it)?

Some questions...
... Why did you rename this to "Oil Reserve"? That's wrong, because they're always, without exception, called "Petroleum Reserve" (short "SPR").
... Why did you tag this as "man_made=works"? This is wrong, because "works" means that something is manufactured there (which is obviously not the case)
... Why did you convert the area to a node (large facilities should be mapped as area, that's more detail; consumers can extract the node if they only need a single coordinate automatically)

Hello, thank you for your landuse mapping!
However, please don't 'glue' landuses to pipelines or other unrelated features (like osm.org/edit?#map=22/29.63747/-95.32955) - this unnecessarily complicates editing.
Thanks!

"Because I shouldn’t have to preserve your vandalism to exclude valuable data from my import. "
My vandalism? Yeah, O B V I O U S L Y vandalism. How exactly is your data better, where did you document it?
...Nice transition to the next point, that HIFLD page is neither proper documentation nor discussion (which you want me to do). So where is it?

"We do have to fix your imports."
Repeating myself, what do you have to fix? I fixed a tiny problem, this was everything one could have fixed. Deleting everything is not fixing, stop lying to yourself.

"reverted in: "- Deleting perfect data only for "violating principles".... how far did it come. (pipelines/the data cable were even all reviewed and didn't violate anything)
Although I a b s o l u t e l y cannot imagine that this is the only reason for this drama. I can't get rid of the impression that a hurt ego plays a greater role here than the guideline.