I don't see that noted. I'll be honest, when I first saw that table, I assumed that you intended for it to be read to say, "The DC for the same lock is different for people of different levels."
Based on the rest of your post, that is not what was intended, and fair enough; but that is how I honestly read it.
Then obviously it needs to be communicated better. I think the concept is communicated--such as, in the quote from the DMG I provided--but it may be worth including elsewhere down the line.
I don't like the assumption that the PCs are always going to be up against something vaguely their level.
I like the idea of high-level PCs having to deal with things that are significantly lower-level than they are, because that sort of thing happens sometimes - and so does the converse.
Oh, I don't disagree with you at all. However, the improvisation rules make that assumption because the
majority of the challenges the heroes will face are going to be level-appropriate. Facing vastly lower-level challenges, in many cases, aren't actually challenges at all. In the example of the 26th level rogue breaking into the mayor's house, at that point there's likely no roll even involved--it's auto success. That's fine, but it's also not something that the DM is likely to need to improvise in the first place. It's either something he already improvised (at low levels, and thus has those numbers for his reference already), or it's something that isn't intended to stand in the players' way and the DM just narrates it rather than requiring a roll.
The problem is that, by assuming the PCs are always going up against something appropriate, you have made the game feel as though it is all about the PCs, and just about the PCs. And from the game perspective, that's all well and good - but for someone like me, who isn't interested in just that perspective, I find it irksome. The PCs aren't the center of the universe, and assumptions like this make it feel like they are.
Well, maybe it's just a philosophical difference between you and I, but the game is all about the PCs. Your world, as the DM, only actually matters when the players are interacting with it. I'm not saying that other details about the world--such as, what is the economy of this city based on--don't matter unless the PCs are going to interact with them. They do, because they help you as the DM have a more clear picture of the world, and that, in turn, makes it easier for you to both design adventures that used logical consistency as well as improvise more consistently. However, I personally think that the campaign world exists for the PCs to adventure in.
This is not level-based, this is world-based. Give me world-based DCs, and I will determine what I should throw at my PCs based upon their level.
The end result is most likely the same, sure, whether we use your method or mine. But I value processes as much as I do results.
I think that the two are inextricably linked, actually. When you design your world, you are probably creating it with some idea of what level the heroes are going to interact with what element. We assume that players will be interacting with the City of Brass at epic level, so when we create the adventure and other elements of the city we do so with that in mind.
World-based DCs absolutely, 100% do exist in 4th Edition, but they're in a different place. Now, they're in adventures. What the DMG gives you is the guidelines to build your own locations, with their own DCs. Instead of giving you six locks, we've given you every possible lock in existence (or, close enough). When you buy or write an adventure, you're using those tools to create the DCs of your world. You're not choosing a pre-made lock and inserting it into your world; you're creating the lock for your world, and assigning the right DC to it.
It really does come down to a matter of why the DCs are different. You are trying to tell me that it's due to the PCs being higher-level, and that that means that the DCs should reflect that and be more difficult. And I would agree, because otherwise the game would be a cakewalk, and that wouldn't be fun.
The problem is the reason. I want tables with varying DCs based upon the world; I like the 3.5 open locks table, which describes the difficulty based on the kind of lock (admittedly not very well, but it's the principle behind the lackluster execution). The 4e open locks table is based on level, which just rubs me the wrong way.
The truth is, though, that lock DCs are different because you need them to be different. The reason why those different locks exist is to provide you with different challenges for your heroes. The idea that there's some academic reason why the locks in the City of Brass are better is really just an illusion; what it boils down to is that the City of Brass is a place where epic adventures take place, and so it needs to be able to provide epic challenges. Part of that, of course, is that it makes sense. After all, if the City of Brass didn't have epic locks, then the epic inhabitants would just barge through any door.
I feel like what you're talking about and what I'm talking about are almost the same thing. However, I think it's like you're approaching the same endpoint from the same direction. It's like I say that locks in the City of Brass are DC 35, which makes sense because it's guarding against epic threats like the PCs; you're saying that the locks in the City of Brass are designed to guard against epic threats like PCs, so they are DC 35.