Is "Rogue Legacy" a rogue-like or literally a legacy to Rogue?
I mean, it has permadeath but then not really since your heir gets your money and stuff.

Imho the whole term "rogue-like" is problematic as a genre since it's too fixed on one game.
We don't call first person shooters "hovertank-likes", do we?

No, the problem is in a difference in what the terms refer to and in scope. Most genres have a very broad definition since they only refer to one or two qualities. First person shooter means just two things: it's experienced in first person view and you shoot things. If it's a third person view game, or a top view or if you don't shoot things, it's not an FPS. Would you call Dear Esther an FPS? Hardly, and if the dev did, the players would rightfully complain about false advertising. Same with RTS. Real time strategy. Says it's real time and it is a strategy game. If you call a Doom clone an RTS game - then you are lying. Or if you call some tank combat game a Flight-Simulator.

Now the term Roguelike says, by definition of the word, that the game is like Rogue. So if the game it refers to is not like Rogue, the term is misapplied. Same as the examples above. The best thing would be to create a new term. Call them procedural permadeath games (PPG for short) or whatever. But the problem is there were some successful indie Roguelikes (that actually were Rogue-like) an then some marketing oriented people wanted to cash in on the catch phrase and used it, no matter whether it actually described their game or not. And with that they softened the term until it became a catch-all, hollow phrase that doesn't mean anything any more. (Just like the term 'indie', which also was turned into a meaningless marketing term ... but that's a different topic)

Genre redefinition is not needed as there is lot potential for interesting ideas in classic roguelikes. The genre has some traits which define it.

What is genre maturity anyway? Ability to sell on Steam? Well that's one thing I didn't ask for.

Take bullet hell games. That is a niche genre with specific traits (shmup, lot's of incoming objects). As roguelikes you can call it stagnant. As roguelikes it got following of players who love them specifically for it's defining traits. Take the traits away and it is not really bullet-hell game. The players of the niche won't enjoy it then, even if you call it bullet-hell.

Roguelike genre did not need any saving. It haven't lost vitality and vanished, there were lot of new true roguelikes each month before the marketing happened. If something damaged the genre, it's calling everything roguelike.

The mark of a genre's growth is in the diversity of games that emerge - and the eventual scraping of the barrel once that experimental phase has ended. It is my belief that the roguelike is an anomaly. Where other genres got to stretch their wings creatively and commercially, the genre was essentially frozen in 90's.

Later on, the old guard sought to define the meaning of "roguelike" in 2008. Known as the Berlin Interpretation, it set down a straitjacket of rules that the genre was to follow. Other genres did not have such artificial restrictions placed upon them - the good games rose to define their meaning.

Organically developed genres are not defined by a panel of 'experts' - otherwise, we will still be stuck with strict Wolfenlikes and Pac-Man Clones.

Fully agree. It's not like anyone is saying that roguelike elements shouldn't be hybridised into new and interesting games, it's just not a roguelike any more.

Nobody is saying there should not be room for experimentation. Just don't stick a term to something where it doesn't apply. I don't really understand why you find that so hard to understand.

Or would you be OK with applying the term 'Wolfenlike' to a game like Civilization? Or call Skyrim a Pac-Man-Clone? Then why use genre names at all? If you strip a definition of all meaning, it becomes useless. Why not just program a game and release it on it's own merit, without sticking a mis-applied marketing term to it?

And the other genres did not grow beyond their definition. You are misrepresenting things there. Strategy games still contain strategy, otherwise they are not strategy games. In FPS you shoot things in first person view, otherwise they are not FPS. Yes, it is that easy. And Roguelikes are like Rogue, otherwise they are not Roguelike. Yes, that genre definition is much more narrow than others and that's exactly why it shouldn't be applied to so many terms.

Or a completely different example, that will perhaps help you to understand the problem: If you sell a beef-steak as vegetarian meal, because it has some herbs in it's sauce, people will complain. Yes, your meal contains elements that are also found in vegetarian meals, but that doesn't make it vegetarian. So if you sell a game, claiming that it is like Rogue, because it shares one property with Rogue while being completely different otherwise, then you are making a false claim.

Organically developed genres are typically defined by a popular example of the genre that other games tried to cash in on. Take FPS, they were originally called "Doom-likes" because Doom was the supreme example of the genre that everyone knew. Then came along some other games that changed the formula like Half-Life, and the genre was name changed to First Person Shooter, with "Doom-like" becoming just a subgenre.

But nobody would call Half-Life or Bioshock "Doom-likes", because there are plenty of things that make them different from the original Doom. By the same token, creating a new genre name for games that share some gameplay elements with Rogue is fine, calling them "Roguelike" when they aren't isn't.

As things stand now, there are plenty of games that have almost nothing in common with each other and yet use the term "roguelike" to define what genre they belong to, which makes said term entirely useless in terms of describing what a game is like. What's the point of that, and how is it better than the narrow, traditional definition of "roguelike"?

Perhaps it is because the definition of Roguelike IS NOT "Like Rogue" anymore? Meanings change, especially when the masses adopt them.

The typical gamer would recognize the genre as two defining traits:

1 - Procedural randomization.
2 - Enforced loss of progress.

Spelunky meets that criteria, as would FTL, Dwarf Fortress, Shiren, Isaac, Necrodancer, and Rogue itself.

Wow. This is quite possibly the dumbest thing I've ever read. I actually feel dumber for having read it.
No, sorry, this is even dumber. In any case, it is obvious that long-winded, reasoned responses to this garbage aren't getting through, so I won't bother adding to what others have said.

What's this? You are actually capable of saying things I agree with! Flame wars have been fought over minor differences in interpretation, and I don't recognize this "Berlin" group as a final authority. Then you continue writing, and prove that it was just a fluke. I'm guessing that some of the words in that statement mean different things to you, anyway, so I probably don't even agree with you after all.

The problem is that I've played Spelunky, FTL and Dwarf Fortress, and I personally would say that they have very little to do with each other and certainly don't belong to the same genre. As such, a "genre" that would include all those games is entirely meaningless and pointless.

So if I have to choose between a definition so wide as to be pointless and an overly narrow one that refers to a game that very few people have actually played, I'll pick the latter. I can understand if you feel differently, but personally when I look at what genres a game is identified with in its description, I want said genres to have a clear definition and to give me a rough idea of what the game is actually like.

Yeah, I couldn't agree more.

Whatever elements Spelunky and FTL share, they certainly do not belong in the same genre, unless we're counting 'game' as a genre now too.
Dwarf Fortress is a little complicated as it is infact two games in one, one of which is a pure roguelike (as far as I'm aware anyway, it's been a while since I played Adventure Mode) and the other a blend of roguelike and city builder. As such it's impossible to classify it in a way that makes sense for both parts of the game.

To me the point of a genre is to be narrow. FPS covers a wide range of games, but what you can pretty much guarantee is a first person perspective, guns or other ranged weapons and real time action.
Then there are sub genres, linear FPS or open world FPS which narrow it further. Or hybrid genres like FPSs with RPG elements.
I don't want to go looking for a first person, action oriented game and end up buying Myst because it is in first person perspective.

Equally if I want a game like Rogue, I don't want to end up picking up Spelunky, because it's not.
Roguelike is a very narrow genre, and should have remained so. Say your game is inspired by roguelikes or has roguelike elements, but don't say it's a roguelike if it's not.