I must say that I share astarael7's feeling regarding this topic. On the official forums, BaconCatBug had the habit of continually reposting this idea, posting comments about it in unrelated threads, and generally just repeating the same points without taking in what other people were saying. This was the second-most annoying behavior on those forums, and I was hoping not to see it here.
I'm probably going to regret posting in this thread, but there are people here who I'm going to assume are arguing in good faith, so I'll just address some points.
The problem is that Magic is messy. There have been several major rules overhauls, and a whole lot of minor ones. Cards have been printed and reprinted at different times, under different errata policies. There have been misprints. There have been cards that were printed that were only later discovered to not actually work. And there have been cards that are just templated very poorly.
I'm a big fan of consistency. And I do find inconsistency jarring. However, because of the inconsistency in printed texts, this is really inevitable. If you try to find a single consistent algorithm for determining the Oracle text, you'll still get inconsistent end results, oftentimes undesirable ones.
As such, the policy they use allows for human judgment calls. There are a variety of factors that go into determining the Oracle text, but no one factor will always win, and the people determining the Oracle text can make the call for which ones should be prioritized in which cases. It's certainly desirable for cards with the same printed text to work the same way. It's
also desirable for a card to just work as printed, independently of other cards. It's
also desirable for them to have the possibility of fixing typos, misprints, or accidentally non-working cards. It's
also desirable for cards to keep working past rules changes. It's
also desirable for cards to just not be really weird. It's
also desirable for Oracle texts to stay constant over time. And these goals often pull cards in very different directions. So if you find a case where your favorite of these goals isn't met, that doesn't mean they're hypocritical or violating their policy. It just means a different one won out in that particular case.
Also, not all "power-level errata" is against their policy. Strictly speaking, all functional errata changes the power level to some degree, so forbidding that would be silly. Instead, they have a policy against "post-printing development". Neither just leaving the card as is nor issuing errata to restore how it was previously developed to work is a violation of their policy. In one case, they're not doing anything to the card, and in the other case, they're just reapplying a power level and functionality that was already decided for it. This is something very different from taking an existing card and intentionally making it into something new via errata.
If you want to know why
Lotus Vale and
Crosis's Catacombs work differently despite being worded the same, that's because the Vale is really old and was printed under different rules. That's also why
Waylay won't let you keep the creature past the current turn (even if you cast it in your end step), why
Carnivorous Plant can't attack, why
Chaos Orb's activation cost includes a
, and why
Wiitigo doesn't immediately go to the graveyard for having 0 toughness. In each of those cases, just reading the printed text and interpreting it as though it were a modern card will give the wrong result. In order to get the right result, you either need to know the old rule that caused it to work the way it does, or you need to recognize that it's an old card and check the Oracle text.
If you want to know why
Lotus Vale and
Phyrexian Dreadnought work differently despite being worded the same and both being old, it's because the Dreadnought didn't need the errata. That replacement effect thing is really weird, so they'll avoid it when possible. In the Dreadnought's case, the card itself works the same with or without the errata. It would take an interaction with additional cards for it to work differently. In the case of the Vale, however, no other cards are necessary; the card itself would be fundamentally different from what it was designed and printed as if it it didn't have that errata. Gaining the additional functionality of being a
Black Lotus is kind of a big deal. It's okay for them to fix what needs fixing, without the obligation to then also fix what
doesn't need fixing. That
would just be consistency for consistency's sake.
If you still think that
Lotus Vale should be changed, that's fine. There are certainly cards whose Oracle texts I dislike (e.g. I'd prefer if
Burning of Xinye,
Imperial Edict and
Wei Assassins had kept their "sacrifice" wording, rather than reading Portalified rules text as though it were normal Magicese, and I literally have a list of little templating inconsistencies I'd want to have standardized away). However, if you think it's objectively wrong for the Vale to have its current text, that it's a violation of their policies, or that they're obligated to change it, I think you're mistaken.