Puzzles and Games

There has been a drama of mis-attributions on the Internet lately, which, if you recall the anti-Wikipedia-style hysteria of years back, would seem forewarned. But the dramatic element is that the loose, crowd-sourced, volunteer aspect of the Network has been exposing and solving mis-attribution errors, not causing them.

There was a Martin Luther King Jr. quote falsely attributed, or perhaps better described as misinterpreted, and then corrected.

The last might be blamed on the haste of internet users to re-post a message without fully reading it and/or verifying it. However, in an unrelated intrigue, Wilko von Hardenberg and Tim Carmody got to the bottom of a falsely-attributed Decartes quote, that not only has been live on the Network for years, but printed in several books, going back to the 1970s.

And then yesterday, in consideration of these events, I decided to repost a quotation mystery that has baffled me for years: that in the University of Minnesota edition of Deleuze & Guattari’s Anti-Oedipus, there is a endnote left absolutely blank, with the quote unattributed. William Ball jumped on it, and solved the mystery by figuring out that the quote was oddly translated, the citation left blank, the punctuation misprinted so as to obscure the context, and perhaps the passage creatively-recalled to begin with. Like that, my mystery fell away, like scales from the eyes.

by Flickr user hans s

Perhaps it is apt that we can watch these little sessions unfold over Storify, because it is not really the work of one person that uncovered the puzzles of these mistakes. One person could perhaps fix a mistake, not unlike editing, if he or she had the knowledge to rectify an obviously perceived error. But it was because these were mistakes echoed wide and far throughout the network, or because one person’s doubt could be shared and extended by other interested parties, that there was a conclusion to these, and a narrative of the puzzle could be established. They are dialogs. One person has a doubt, and expresses it outwardly. From the topology of their Network, no doubt established by a previous acknowledgement of similar interests, comes the response: yes, I share your doubt. Then the synthesis: let’s see if we can’t manipulate the Networks to find our solution.

As I mentioned to @exstasis, who pretty much solved my mystery single-handedly as I watched in awe, it was amazing that he solved the issue using only online sources that were readily available. Search engines, scans of books uploaded (perhaps with dubious copyright conformity), and the various versioning that the wide duplication of resources on the Network can provide. While the puzzle could no doubt have been solvable with standard academic resources at hand, such as a good academic library, he didn’t need any of this. And considering the fact that we found no mention of the existence of the puzzle at all, it is entirely possible that no one has ever bothered to track down the solution. Therefore, the Network allowed a couple of “amateur” scholars satisfy their curiosity, without needed to avail themselves of the resources of the standard fact-checking institutions. Those institutions that through their mistake, created the puzzle to begin with; but we won’t blame them for that, as what all these cases make clear is that the Network is in fact equal to “higher” academia when it comes to creating these intellectual puzzles, as well as solving them. One wonders if in fact, the Network has further merit in that not only does it allow access to anyone with the basic ability to connect and the will to participate, but the pace of both mistake and correction is incredibly rapid (perhaps related to the scale of participation as compared to academia).

All of this being fairly apparent to anyone who is more than a casual user of Twitter, or some other tight-knit soft-network.

by Flickr user Randy Son of Robert

by Adam

But here’s what I wonder, and what I’d like to suggest. The theory of gamification states (my own generalization here) that a motivation strategy for behavior could be an assigning of points and reward structure so that a person can more easily visualize their progress towards a goal. But perhaps rather than gamification, we should be considering puzzlification as a strategy for utilizing these sorts of soft-networks.

The difference is this: a game is designed to structure goals via a definition of quantified points and winning conditions. A puzzle, on the other hand, is itself a structure of a qualitative and logical quandary. A game can be cheated when the points structure is manipulated to achieve the winning conditions without necessarily achieving the goal. But a puzzle can only be solved, or not. However, a proposed solution to a puzzle can be at first accepted as seemingly correct, and then later found to be incorrect. A puzzle can be be the goal, structured into part of game. And in a sense, the strategy for winning (or cheating) a game can be thought of as a puzzle. But the difference is quantitative/qualitative.

To more directly contrapose the two: the game structuring numerous small problems together via a generalized quantitative network, the puzzle an isolated network structure of specific logical quandary. Both are ways of structuring our assessment of reality, and so neither is more “real” than the other. The facts of gamification are not about the ability to cheat, so much as what that ability entails. A poorly-described puzzle is in no way superior to a well-designed game. Nor is a properly-apportioned game necessarily worse than a clever puzzle. They are merely alternate ways of describing a goal, so that the mind can attempt to guess what its move should be to satisfy that goal, so defined.

Furthermore, I would venture to say that in addition to simply quantifying the issue, a game’s rules are more generic and abstract from the actual tasks at hand, bridging beyond one issue to a whole set of issues, wired in series as it were. While a puzzle, in addition to quantifying the situation, is specific and concrete to the issue, considering everything holistically. Once the puzzle is solved, that is it. It may be intricate in its layout, but a puzzle is entirely self-contained.

Game: Quantitative, Bridging, Generic, Abstracting, Network-Extensive, Structural Assessment
Puzzle: Qualitative/Logical, Holistic, Specific, Concrete, Network-Inclusive, Structural Assessment

Because these are similar ways of assessing problems, both have merits which are no doubt applicable to different situations. But to a soft-network such as Twitter, I think we should look towards the puzzle. I am calling it a soft-network because Twitter is not meant to organize any particular process or activity. Sure, it is based around 140 character messages, but clearly the point of Twitter is not simply to create 140 messages. Twitter represents language itself, in a way. Language is for communicating, but that’s not all we do with it. We also grunt, express emotion, think, act, commune, organize, and many other things through language. With Twitter we send messages, but also network (in the verb sense), share, express, link, and a bunch of other things. It is that there is a very basic framework without explicit purpose that we are able to do so much with it, extending it outward from its premise. On the other hand, a hard-network is defined by specific tasks. The html structure of a website, for example, is designed to render information via a browser, and provide programmed functionality. There are different ways of doing this, and one can do an incredible number of things with such a structure. But it is a specifically-defined system and outside of its core task, has no other function.

Hard-Network: Specific, Network-Inclusive, Rigid, Concrete, Defined Structure
Soft-Network: Unspecific, Network-Extensive, Flexible, Abstract, Interpretable Structure

And this difference is ironic, when it comes to interacting with these structures. Problems with hard-networks, those structures that are very specific, are perhaps best solved by quantified assessment. HTML ought to render fast and error-free, and be coded simply and quickly. With a specific structure to act upon, we can take an abstract and generic method of assessing those actions and still assess very effectively. On the other hand, with a soft network like Twitter, it is very difficult to generically assess a “winner”. Rather, with such an open-ended structure, it is better to assess our actions within it logically, only according to a concrete and specific set of qualitative parameters. Do you “win” Twitter by tweeting the most, or fastest, or having the furthest reach? It all depends on the specifics of the particular puzzle you are trying to solve. You can see with the traits I’ve identified above, that a puzzle has certain attributes of a hard-network, which a game has attributes of a soft-network, and yet I suggest they should be oppositionally aligned.

These interaction pairings are antithetical to how we might think of them. Wouldn’t a well-defined hard-network structure benefit from an assessment system specific and concrete to its limited definition? And wouldn’t a more flexible soft-network require a general, far-reaching assessment? The answer is no–because assessment is not about mimicking what is being assessed. It is about control through overlap. Reality and our conceptual schema are always, in a sense, in opposition. We can’t think that our mental conceptions of the world will ever catch up with the detail of the mechanisms of the world. Instead, we need to model and simplify. The best model is one that overlaps the boundaries of what it attempts to model, rather than mimicking the subject. It looks at the difference between the object and field, rather than the undifferentiatedness of the middle of the object or the edge of the field. If a system is limited, a generic model will cover more of the extent of the system. If a system is more fluid, specific samples will gain a better sense of what needs to be observed. The model is part of the system it observes and assesses, and therefore it ought to fit in as component, rather than attempt to draw a map of each grain of sand.

by Adam

Consider again, these checks of attribution error conducted via soft-networks. Should we award each of these people who succeeded in correcting an error “points”? Why? What would these points mean tomorrow? What do they mean in terms of the errors themselves? The game, as it were, is not “winning the Internet”, as the joke often states. The puzzle was identifying and correcting an error that no one knew existed. These puzzles were each solved in their entirety, and no doubt many more lie out there waiting to be discovered. If we awarded these players a number of points, how would these points help them prepare for the next puzzle? On the other hand, if we congratulate them for solving a puzzle, we can trace the steps that went into the solution. We read back through the dialogic steps of the Storify, and see the moves they made. We don’t attempt to replicate these moves exactly, but we recall the strategy they imply: collect a network of intelligent, like-minded individuals; keep a sense of what search tools are more helpful; locate resources for finding illicit copies of otherwise un-retrievable texts; and when you think something is amiss, why not say so out loud, and see who responds? “TRY TO GET MORE POINTS” is not a helpful tactic here.