There was that one Doom roguelike cell phone game, wasn't there? I honestly can't recall any more if the dungeons were actually randomized or not, which I guess is the defining characteristic...

I haven't played it so I wouldn't know. (The screenshot above, however, probably comes from it.)

Sorry if this seems a little light by the way, I hope to have something a lot more interesting next time out.

Yah but the thing about puzzles is that they are so unlifelike. So if you are accepting the premise that you are a character, to then impose the premise that the world is going to behave like a puzzle - with a very finite number of solutions - is kind of like cheating. If the word we're in is a puzzle, then I should be a puzzle-like character. I can't tell you the number of times a game that imposes one or two solutions frustrated me when I could obvious solutions (like, hey man, can't I just climb that obviously climbable wall over there to get past this door?).

GRGM: I remember that the whole point of Deus Ex was to enable those kinds of alternate solutions, which was cool. But in general, that kind of thing is the province of roguelikes in general, and Nethack and ADOM in particular.

Wow. This is the first installment of @Play that I've actively hated.

For one thing, I can't figure out what your point is. Do you not like the fact that adventure games became "safer" in order to be more fun to play? Calling Phoenix Wright "nothing more than [a] sophisticated version of a 'next page' prompt" is nine shades of petty. Especially since I just finished case four last night, the one you oh-so-hated, and thought it was a fantastic gaming experience. Because, in the end, I did have to figure out what happened. I did have to find the contradictions in the testimony of the witnesses, and I did have to closely observe the evidence I'd been given to solve the case.

And in the end, wasn't this merely a concession to how players were playing the game? Please don't tell me that you played through an Infocom adventure without saving before every potentially dangerous action. What's the difference between picking, getting it wrong, putting the game into an unwinnable state, restoring your game and trying again and picking, getting it wrong, and being allowed BY THE GAME to try again? Except for reduced frustration for the user?

And you've neatly sidestepped the real concern here. When you allow your game to branch - truly branch - you're now making two games. This is why Deus Ex didn't branch until the very end - the designers realized that if they branched early, which was their original intent, they were actually going to have to make three games.

Now, I could be just as petty and complain about how stupid it is that I have to consult a reference chart to find out what is going on at any point when I'm playing a roguelike ("Huh...what's a lowercase l again? Do I fight it, eat it, or wear it? Oh, crap...this is the NetHack chart and I'm playing Crawl."). But I won't.

Because that is one of the strengths of the roguelike. A roguelike's cost of content is very low - a symbol on the screen, a text description, a bit of code. The DevTeam can think of everything because they don't have to DRAW any of it!

Of course, this is also the very thing that prevents roguelikes from becoming more mainstream. And I can't help but think that that's the REAL reason for today's rant...it rankles you that prettier games with fewer choices are more popular than roguelikes, where you can wear hats on your feet!

And that's a perfectly valid opinion...I think you just presented it in a disingenuous way.

You know, I've liked your column a lot, but I have to disagree with you on this one.

The correct comparison w.r.t. Phoenix Wright would be the old Infocom text adventures (Zork et al). In those, it was all too easy to get into a situation where you couldn't possibly win, and you would have to restart the game.

An example that is particularly evil (and intentionally so) is Douglas Adam's Hitchhiker's Guide to the Galaxy. To solve the final puzzle of that game required you to have kept every single item you get throughout the game; if you didn't, Marvin would ask for the item you were lacking, and you'd have to start over.

Now, starting over from turn 1 in an Infocom game wasn't so painful; the puzzles (and the commands to solve them) were usually always the same and once you knew the solution, it was only a few minutes to re-enter them. So the punishment to the player of making a mistake that required starting over was not so harsh (although it encouraged people to make good maps and take good notes!)

By contrast, having to start over a Roguelike can take you tens of hours of attentive play, especially if you were close to the end. For most people, having to endure that kind of setback every time you died is just too much, and they give up in frustration.

While there is somewhat of a market for that (witness Hardcore mode in Diablo II) most people are simply not that masochistic. So game designers (needing to make money) make games that are reasonably possible to defeat without being too hard.

Thanks for the column... it makes me want to go back and play Angband again.

"Calling Phoenix Wright 'nothing more than [a] sophisticated version of a 'next page' prompt' is nine shades of petty."

Well it wasn't intended to be negative in that way. I recognize that there is room for many kinds of games, I was just offering it from a certain point of view.

Heavens knows I've played and enjoyed enough Phoenix Wright, but the fact is, there really isn't a lot of true interactivity there, and interactivity is what makes a game a game. I submit that what makes Phoenix Wright entertaining is its non-game attributes: its writing, it artwork, and its characterization. The puzzle-solving aspect of figuring out cases is game-like, but there are no resources to manage.

This, again, is not necessarily bad in itself, but I wouldn't have devoted a column to this, if it didn't seem that computer games were slowly becoming all about this kind of play, and less about interactivity.

Also Viridan: don't hold back in complaining about roguelikes! I'm here to try to see them as they are, not to blindly appreciate them. The fact that I had to devote half the second column to freaking vi should indicate that I know they can be horribly difficult to pick up.

Bilge: Yeah, I know about those instances too. I'm not arguing that it's good to be able to get the game into an unsolvable state, just that the need to prevent that can have profound implications for gameplay, which sometimes reduce interactivity, and sometimes actively give away puzzle solutions. Sometimes you can even "Scooby" Phoenix Wright cases.

Roguelikes, beyond most other games, avoid falling pray to this--but Nethack DOES absolutely prevent destruction of four objects in the game, which could be taken as a hint for how to win it. So it's not black-and-white.

As for this week's column being substandard, well, it's been a weird weekend. I almost took the week off to be honest, but figured I shouldn't wreck a 17-column streak.

Don't get me wrong, I've really enjoyed reading @Play so far...I think you're right about this particular column. I'm certainly not going to stop reading your stuff over it.

As for criticisms of roguelikes...

I have a real love/hate relationship with them. I adore the concept - small, playable, portable, cross-platform RPGs that focus on gameplay rather than graphics.

What I hate about them are the terrible control systems, the impregnable interfaces that often hide vital information (like when you're starving to death) from the user, and the fact that the vast majority of roguelikes delight in killing the player.

And even when you know everything there is to know about the game, there's still a very good chance that your attempt will be cut short on level 1 by a goblin with a wand of magic missiles. That's not game design to me. Heck, it's pretty close to ANTI-game design.

"What I hate about them are the terrible control systems, the impregnable interfaces that often hide vital information (like when you're starving to death) from the user, and the fact that the vast majority of roguelikes delight in killing the player."

The control systems, again, come from vi, which is a venerable text editor that has stymied generations of computer science students. But once you learn the keys they become very easy to use, it's actually a very elegant interface once you're initiated. Long-time roguelike players tend feel the same way about these games.

The possibility of dying on level one, actually, can be taken as a point in favor of roguelikes; there's always something unexpected that can take the player out if not properly defended against.

And it seeming like "anti-game design", actually, is allied to my point above: traditional games try to guarentee success to a sufficently skillful player, while roguelikes try to make the odds of winning acceptable but do not guarentee a win. That one difference can make for wildly varying types of games.

I also, for the love of God, cant figure out - what is your point ? I just cant see any, in this particular article.

When people says roguelikes have terrible control systems I can't help but point them out to the *NEW* roguelikes? Somehow they never get the attention they deserve.

Some people have already adressed the unnecesary complexity of user input in games such as netHack, which is ancient and right, from its age comes its complexity and charm, but it also carries out much cumbersomeness over the years :P

Complementary to the topic of how roguelike games retain their challenge from one play to another, it's interesting to note how the various challenges grow or diminish in importance throughout the course of a single game.

Rogue is short enough that one rarely identifies all the items in the game in even a fairly successful run; fast-paced in the extreme, such there is always a very real chance of starving to death if one does not keep moving, moving, moving; and the dungeon gets tough faster than you do: centaurs have only just become reliably meleeable when trolls show up, and if you last that long, you're likely to be still needing significant rest after each troll when black unicorns, medusae, and the other nightmares of the deep come on the scene, and then you just run, run, run to the bottom, and maybe survive.

In contrast, in most of the longer games, there come points when the character has essentially solved a certain type of survival problem, and then can move on to cautiously addressing their remaining survival issues, such that throughout the midgame, a sufficiently conservative player need not really be in danger. The experienced or spoiled player essentially has a checklist of things they must accomplish or acquire to be insulated from harm in more dangerous environs, and mechanisms to force players to take on threatening challenges are sometimes lacking.

Diablo not a roguelike? Then what is this? :P :
http://diablo.chaosforge.org/

The point of the article, simply, was that games have evolved more in the direction of story, so much that they have lost something along the way, and that is the primary reason roguelikes are still popular, maybe even gaining in popularity. For they are not afraid to be games first.

Nice comment Cyrus, for a moment I had to look twice to make sure I hadn't written it, heh. Good points.

Epyon: Heh heh heh.

@Slash: What are these new Roguelikes you speak of, with good user interfaces? Other than DoomRL, that is :). I'd love to see some links, if you have them.

Upon further consideration, I vote that Civilization IS a roguelike. After all, it does have permadeath.