Quote from: droqen on July 30, 2026, 03:43:54 PM

• gameplay (forever) is badly designed
• gameplay (forever) is badly designed on purpose
• there [is] no rhyme or reason to what blocks appear where in gameplay (forever)
• gameplay (forever) generates impossible situations to climb through
• gameplay (forever) is one of a number of "lackluster parodies of better games"
• gameplay (forever) is asking why anyone would bother with it

i linked it somewhere already i think, but in case you missed it, here's my 8490-point high score run:


by doing this and enjoying it (and not in a pretentious way, as in, my enjoyment was not experienced due to any pretense, i just like it -- but the act of doing it could be said to have been done with pretense), i hoped to show:

1,2 - that gameplay (forever) is not badly designed (for some definition of badly designed -- in this case i suppose the entirety of my argument would be 'i enjoy it', but i hope also that the video can be used to understand the interesting situations that can occur)

3 - that by virtue of the interesting situations encountered there must be some rhyme or reason to the tile placement (even if it is random),

4 - that it does not (meaningfully, in practice) generate impossible situations,

5 - that it is not merely a lackluster parody,

and 6 - that even if i do utter the seemingly imperative "kill gameplay" and literally ask "who could want gameplay?" that this is a deeply contextual and ephemeral perspective, and that the answer to the imperative can (and sometimes must) be "no" and that the answer to the question can (and sometimes must) be "me".

why would i attempt to get high score of 10000 points in gameplay (forever), an act that is difficult and time-consuming? (my initial estimate was that this would be an hour-long run, but my best score so far -- 8490 -- took an hour and two minutes (ish), so a bit longer than an hour for the winning run.)

in retrospect, i certainly wanted to show the falsehood of some of dot's claims. i knew that i would enjoy it, for example, which might address the question of whether it is 'badly designed' or not. but i think it's this final unaddressed claim that the process has revealed to me: i knew that it was possible.

dot's claim (#4): gameplay (forever) generates impossible situations to climb through

because of the deeply random nature of the level generator, it is of course possible that gameplay (forever) will generate a situation that cannot be climbed through. but i wanted to show that there was no way dot could know that, and i wanted to understand the precise shape of this possibility, and rather than solve this from afar -- mathematically -- i already had within me the (positive) sense that 'gameplay is volunteer work', and the desire to play around with the bunny's controls, and to climb!

why not see how high i can climb?

then, once i had reached almost three thousand points, i thought:

why not get to ten?

dot's claim (#6): gameplay (forever) is asking why anyone would bother with it

strictly speaking, Dot's full quote here is:

QuoteThe End of Gameplay gives the audience lackluster parodies of better games and asks why anyone would bother with them.

at the time he says this, he is not showing gameplay (forever), but instead the last screen of open field (another level in the end of gameplay), which visually mirrors the beginning of Mario 1-1 but without enemies, without the ability to bump your head into blocks, in which going to the right gives you points and loops around to the left side again, in which the player is contained within physical boundaries that are not present in the original level, and a line of text that reads, "who could want gameplay?"

(as an aside, i think the claim that this level is a 'parody' of a game because it visually recreates mario 1-1 (and quite poorly) doesn't have any legs to stand on, it's... obviously a different thing and not asking "who could want to play mario 1-1?" while presenting the player with a completely different experience. anyway, that's not relevant.)

while open field is definitely asking this question explicitly, gameplay (forever) does not explicitly ask why anyone would bother with it. open field is clearly a different beast -- it asks the question at the conclusion of a series of levels that move from chaos toward design, and only asks the question when we've gone from the wilderness to a fully designed world, while gameplay (forever) starts and ends in pure chaos. as dot puts it, there seems to be no rhyme or reason to how its tiles are placed.

dot's claim (#5): gameplay (forever) is one of a number of "lackluster parodies of better games"

i mentioned Probability 0 above because i think gameplay (forever) is better than it in a number of meaningful ways. my very first prototype of Probability 0 was a minimal infinite platformer in a gameplay (forever)-esque way.

there is a purity to the basic foundational action of quickly enough descending down an infinite field of randomly placed blocks while avoiding fall damage that is lost along the way -- Probability has a skill tree, 20-ish types of enemies, boss fights, biomes, etc. does Tetris need a skill tree?

(would like to note that i collaborated on Probability 0 with an awesome musician (jazz mickle, queen jazz) who created this shifting dynamic soundscape that breathes wonderful life into the world of that game, there is an album on bandcamp and also a remix album and they all rock, and none of this talk of minimalism is at all touching upon the collaborative multimedia component)

when held against the core repetition (positive) of Probability 0, gameplay (forever) is a triumph, not a parody. it does what Probability 0 was too afraid to do.

(although when gameplay (forever) did it, it was in another kind of ironic context ("kill gameplay"), but taken on its own... well, i'll get to that.)

the literal space that one navigates in gameplay (forever) -- the material of the 'level' -- is randomly generated, with very minimal designer intervention.

an example of the 'designer intervention' i speak of can be seen in these four lines of code:

if rows > 10:
    if randf() < 0.02: randomtile = 35 # poison
if rows > 50:
    if randf() < 0.08: randomtile = 35 # poison

the 'poison' tiles (which have various 'negative' effects -- the player loses control briefly, the music glitches out, and the player's score number goes down) do not appear uniformly across the entire level, but do not appear in the first 10 rows, then only appear in 2% of tiles in rows 10-50, then in 8% of tiles in all rows 51 and onward. all these particular numbers have been designed, and it could be imagined that a 'more designed' version of gameplay (forever) would contain more such interventions.

as a design process, this relatively undesigned level is a test for the quality of the bunny controller: if gameplay (forever) contains all possible level spaces with no design whatsoever, and the bunny controller can be used to enjoyably navigate this undesigned space, then the bunny controller is well-designed.

in theory, if gameplay (forever) was not fun, this process would lead to changing the bunny, rather than changing the level: thus the level will remain permanently 'badly designed' by design.

on the other hand, it may make no sense to even speak of the design of gameplay (forever) without speaking of the design of the bunny controller.

dot does comment on the player controls -- mostly negatively or neutrally -- so in that sense it may make sense to extend his judgement to the bunny controller: claim #1 would follow in this case ("gameplay (forever) is badly designed [because i don't like the controls]"), but not claim #2 ("gameplay forever is badly designed on purpose [because i'm not supposed to like the controls]").

but this is getting too speculative and i still have three claims to talk about!

however, if gameplay (forever) is undesigned on purpose (see #3 above), then if what dot means by "badly designed" is "undesigned", then the following claims do follow:

dot's claim (#1): gameplay (forever) is badly designed
dot's claim (#2): gameplay (forever) is badly designed on purpose

in designing the end of gameplay, the role of gameplay (forever) cannot be understated: if i can enjoy its gameplay (forever) given endless undesigned variation, then as a level designer i have freedom to "design" any shape that i desire, and know its gameplay will be enjoyable.

there were other things going on in my life, of course, but i'll stop my digressions there.

what if concern over 'games' and 'play' and 'gameplay' as words was no longer useful? i considered the idea of doing pleasant tasks via a computer game. a volunteer job. when it gets right down to it, i like to do things. it's nice to happen upon an occasion in my life when i get to solve a small math problem.

it was at this point in the timeline (of reality) that i really began thinking about what dot had said about gameplay (forever). this level's presence in the end of gameplay reflects a common element of my platformer character controller design practice --

in order to properly know whether a character controller is 'good', i.e. satisfying to me, one of the first things i will often do is place the character in a single screen where blocks are placed randomly at runtime. usually the ratio is something like 25% blocks to 75% empty tiles. if i'm feeling fancy, i'll add a floor so that i don't fall out of the bottom of the world.

dot's claim (#3): there seems to be no rhyme or reason to what blocks appear where in gameplay (forever)

here is the code that generates block placement in gameplay (forever); lines 113-123 from file stages/13GameplayForever/^sliding_maze.gd:

var randomtile : int = 11
if randf() < 0.02: randomtile = 25 # frut
if abs(x-xhole) > 1:
    if randf() < 0.30:
        randomtile = 90 # WALL
    else:
        if rows > 10:
            if randf() < 0.02: randomtile = 35 # poison
        if rows > 50:
            if randf() < 0.08: randomtile = 35 # poison
maze.set_cell_tid(Vector2i(x,0),randomtile)

so, there is a definite process by which blocks are placed, and this process has a very long history, but each individual block has not been placed manually by a person, nor by any sophisticated algorithm with any human-like intention (for example, there is no code that noticeably emphasizes some particular shape over another -- say, flat wide platforms). to dot, this may be a negative and he expresses a potential consequence of such rhymelessness and reasonlessness.

i have already made and released a game that uses a similar generator, and though i'm not sure if i've written about it, my absolute favourite part of "playing" Probability 0 (or in other words "starting a run and attempting to survive and descend") is interacting with a very similar (but slightly more sophisticated) "blocks placed randomly" space, via the platformer mechanical constraints: velocity, acceleration, collision, and fall damage.

weaving the avatar through such a landscape involves recognizing and navigating countless patterns to no end: forever.

in very rough terms, the impossible reversal explores four "types of play" and concludes by arguing that videogames are often moving in between all of these types of play (or moving the player, or enabling/inviting the player to move, between all these types of play -- whatever, i'm not concerned at the moment about the provenance or ownership of play), blending them together in such a way that it's difficult to see this complex interplay of modalities. modalities! that's a good word, too; modalities of play. anyway, it's quite a moment. as a collection of these concepts and a peek into the past the book is incredible, but in retrospect the conclusion is a very rare harmonious moment in these kinds of conceptual non-fictions; while there's no clean conclusion that wraps everything up, to see mcdonald show in the space of only a couple pages Inscryption dancing between all of these vast concepts is quite a moment, reminiscent of the lovely remix levels in the Rhythm Tengoku (Rhythm Heaven) games.

so i was partway through the impossible reversal and, quite serendipitously, the fourth modality -- Oblique Repetition -- came to me at the exact moment that i was already dwelling on golumbia's proposal that "what we do when we use [computer and video games]" is work.