text/mode

Just FYI, for whoever is following this blog, my defense had to be rescheduled to next week: Thursday, June 12th, 2 PM. It’s all for the best — just more time to prepare.

Right now, I’m spending time putting together Gameology.org, version 2.0, finishing up an issue of <cite>ImageTexT</cite>, and learning about how to turn a dissertation into a book.

Also, I get to find out today whether Stacy and I are having a boy or a girl, so that’s exciting. We might also hear back about the house offer we’ve made, so this could be a momentous day indeed.

EDIT: We will be having a girl!

I’m not “done” done, but I now have what I feel confident calling a Final Draft. I still have to defend it, revise it, format for the grad school, etc. But this feels like an important milestone.

Some statistics:

• 5 chapters
• 2 appendices
• 314 pages
• 213 images
• 85,264 words

Somehow, that doesn’t seem like all that much. I have so much more I wanted to include but ran out of time. I originally had 8 chapters outlined, and even the 5 I ended up with were actually split chapters that each had originally been one of that initial eight.

When I was in college, I ran on the cross-country team, and for my first two years, I got a personal best time at nearly every race. In other words, I kept improving my times pretty consistently. At the conclusion of my sophomore season, for our all-conference meet, I beat my personal record by almost 90 seconds. Proportionally speaking, it was a pretty significant leap forward.

And afterward, I was completely spent — I think I needed help getting back to the van from the finish line, and I distinctly remember my legs giving out completely when I attempted to hop out back at my dorm — but I kept thinking I could have gone faster. I mean, the guy who beat me for 10th place was just ahead by half a stride. If I had dug just a little deeper, I could have beat him, could have dropped another 10 seconds off my time.

But I didn’t, and that was it. That race remains my PR over 8K (27:28), and I never really even got close to it again thanks to some injuries.

I guess that’s kind of how I feel now. I haven’t crossed the line on my dissertation just yet, but I’m at the point where it’s pretty clear that I’m not going to be able to add much more to it — like that chapter on heads-up display and holograph(em)ic emergence. I’m looking forward to the defense and the next steps, but I have to admit that I feel a little bit of loss as I think about the intensity of this project gradually fading over the next weeks. Of course, the choice I have to make is to keep the intensity up.

I guess I’ll just have to work on the book proposal.

First, though, some relaxing down time. I think I’ll overhaul Gameology.org!

This is the new title for my dissertation.

Also, I’ll be defending on June 6. Hopefuly this event will result in a nice birthday present for Stacy.

This has been bugging me for a while, but it’s come to the forefront as I’m working to wrap up chapter 3. Basically, the problem is that the title, “text/mode”, puns on textmode games (i.e. dungeon-crawlers), but I’m just not going to have time to deal with them here. I realized this is something I have to address because I just started writing a long footnote in ch. 3 explaining why I wasn’t talking about them in that chapter — and I realize there isn’t anywhere else I can put them in. (Though maybe they could fit obliquely in chapter 6, if I include all the holograph(em)ic stuff).

The problem with titles is that they invite you to conform to a cliched structure that always seems to define a lot more about what you’re writing than you mean to. Then there’s the whole <play on words> : <real, informative title> formula, which I’ve been trying to avoid. It has its place, and I think it can work well as long as the first part isn’t too cute and remains sufficiently relevant.

So the question is, what 2 or 3 word phrase captures the essence of my project in a pithy and memorable way? I’m pretty happy with my descriptive subtitle: “A Typographic Approach to Videogame Textuality,” but that’s a mouthful. It also employs at least two contestable terms.

Hmm, maybe I don’t need a pithy, punny first-half? At least, I could save worrying about that for the book proposal.

Well, it’s been even longer now between my latest post and this one than it was between that post and the one before that. It’s OK, though, I’ve been busy. Mainly, I’ve accepted a job offer with the University of Mary Washington. I’ll be starting as an Assistant Professor this Fall — so long, that is, as I defend my dissertation before August 15.

This idea of a very real deadline changes the dynamic of my writing process in interesting ways, but mostly I’ve found it kind of intimidating. I’m in a bit of a slump also because I just got an article rejected (or, rather, sent back with a suggestion to “revise and resubmit”). On that topic, I won’t say it doesn’t suck to get rejected, but in this case I kind of agree with the reviewer. The article as I sent it doesn’t do a good job of explaining my key points and, because this was originally an excerpt of a dissertation chapter, it proceeds assuming that some things are already defined, when in fact those things were defined in the parts of the chapter not included in the article.

So revising it shouldn’t be too hard, actually. The problem is I’m in a tight writing schedule as it is. Mid-terms will be coming up soon, so I’ll have lots of grading. Any day now, VUP should be getting in touch to ask me to create an index for Playing the Past. And the prospect of moving and actually starting my career is starting to loom large.

In more dissertation-specific news, I’m pleased to announce that Brian Slawson of UF’s department of graphic design has agreed to come on board as my outside reader. We had a good conversation yesterday, and it’s clear that he will bring a lot to the project with his knowledge of and interest in the histories of typeface design, as well as the impact of those histories on expression and communication.

Also, by way of update, I’ve recently finished my draft of chapter 2, so I’m moving on with chapter 3 (I already have chapters 1, 4, and 5). First, though, I’m going to revisit the abstract to make sure my big picture argument is still supported by me new organization.

My apologies, dear reader, for not updating this blog in a while. I’ve been quite busy with dissertation stuff, which is a good thing to be busy with. I’ve also been dealing with phase 1 of the jobsearch (sending stuff out) and am impatiently waiting for the next phrase (interviews). I’ve gotten a fair amount written on chapter 2 — the histories one — and I’ve gotten ROMscrape working well enough that I’m able to use it for the main task I created it for. Specifically, I’ve used it to find instances of MICR-esque lettering in Atari Games. As I move into the next stage of my chapter, I’ll use it to make an even more complete map of typographic forms in Atari games.

It does have several annoying bugs which I haven’t had time to squash, but they’re mostly minor. The only slightly troubling ones have to do with finding certain patterns which I’m certain do exist. In a few cases, search queries which should find an exact result don’t.

I think this can be mitigated by amplifying the query pattern into more chunk-groups than I’m currently doing. This will require more server-processing time, but lately that hasn’t been quite the bottleneck it used to be, so maybe I don’t have to be as stingy as I have been up til now.

Lots of other things going on as well. My Refractory article has been accepted, my chapter in Music, Sound and Multimedia has just been published, and the first half of  Playing the Past came back from copy-editing.

Meanwhile, I’m teaching next semester, so I’ve had to spend a few days getting my syllabus ready and texts ordered. I now need to get the coursepack assembled.

Oh yeah, and I’m also hoping to get a few things submitted to journals since I don’t currently have anything under review. One thing I hope to dig back out is something I originally thought would be part of my dissertation — back when my dissertation was on cryptography.

The calculations have been complete for a while now, but just last night I finally got enough of the module coded to see if it might actually work. It’s hard to explain, but basically I got to the point where I can tell if a search query has returned results or not. I tried several searches which I knew should find matches, and one of them worked. That is, it returned results. I don’t know yet if those results are the correct ones, but it’s an encouraging sign. It’s also encouraging that it did so in about 1200 ms, which includes all the db queries and other page rendering stuff, as well as printing out the half-parsed, ginormous result arrays.

Still, some queries which should have returned results didn’t, so even if that one worked, I still have some work to do to find out what went wrong. Without going into too much detail, one possibility is to consider broadening the hash search, perhaps by filling it up to a certain limit of constituents.

Anyway, the news is good, and it’s a bright Monday morning.

I’ve just posted a new abstract/description of the whole project. I found it difficult to write for some reason, but it had to be done and it was a good exercise. I posted it here. As you’ll notice, the chapters are not the same as the ones on my “tentative outline,” which, as I’m sure you can guess, means that I’ll soon be posting a not-tentative outline. I’ll still leave the old stuff up, though, since it’s interesting (to me at least) to see how my ideas have mutated. Overall, I’m rather pleased with my argument. Some ideas are forming some interesting and productive overlaps and parallels with other ideas that I hadn’t thought of or noticed before writing this new abstract.

Here a couple of abstracts I’ve submitted to some upcoming conferences. The first, our department’s annual EGO conference, is coming up pretty soon. The second, for next year’s SCMS, is part of a panel submission with Bob Buerkle, Tracy Fullerton, and Jonathan Frome.

For Ego

“Quest for Saddam/Quest for Bush: Fantastic Abstractions in Videogame Representations of Ideology”

Zach Whalen (UF)

In 2003, a low-budget videogame called “Quest for Saddam” gave players the chance to hunt down and kill Saddam Hussein after fighting their way through hordes of Saddam-clones and racial jokes targeting Arabs . In 2006, a group affiliated with Al Qaeda released their own version, “Quest for Bush: A Night of Bush Capturing,” ostensibly giving young jihadis the opportunity to turn the tables on the coalition forces still occupying Iraq by taking on an army of George W. Bush clones. Quest for Bush is a straightforward hack or re-skinning of Quest for Saddam, where, for examples, source images of Hussein were simply exchanged for images of Bush. Taken together, these two games provide a useful point of analysis for exploring the rhetorical effects of representation within videogames. The two games are procedurally and formally identical (in terms of rules and mechanics), so the question this paper seeks to answer is whether their rhetorical meaning is also identical or whether the largely image-based opposition of ideologies overwhelms the procedural rhetoric of the gameplay itself. Furthermore, what can the simulation of conflict tell us about the actual conflict in ideological terms? As violent fantasies, do these games express the inherent violence of the opposed ideologies they supposedly represent, or is each, in the context of these games, simply a caricature of the other? This paper draws on current game studies scholarship, particularly the so-called “platform studies” approach, to address these and other questions as they inform an understanding of videogame expression.

For SCMS

An Archaeology of the Videogame Image

Summary:

The study of videogames within the disciplines of visual and rhetoric and media studies must take into account the status and structure of the videogame object in order to understand its textuality or the ways in which videogames construct meaning and express ideas. Dominant conversations in videogame studies tend to focus either on videogames in terms of their reception (i.e., as interactive narratives) or as deeply formal systems (i.e., as simulations or games). Both of these approaches take for granted the material artifact of the game itself, so new and better approaches, such as those undertaken by Nick Montfort and Ian Bogost, are beginning to unpack the stratigraphical, cultural discourse which make up or contribute to videogame imagery, logic, and rhetoric. Montfort has provided one such context for textual analysis in his article “Combat in Context,” where he proposes a 5-level model for the videogame object: Platform, Game Code, Game Form, Interface, and Reception and Operation. The challenge to putting this sort of analysis in practice lies in part with the difficulty of gaining access to and contextualizing the actual platform or machine code in question. In order to address this challenge and advance this line of thinking, this paper describes a software-based technique for extracting and comparing visual information embedded in Montfort’s Game Code level of textuality. “ROMscrape” is a web-based search application which will allow users to search a database of Atari 2600 games for specific byte sequences and patterns. The significance of ROMscrape lies in what it can reveal about relationships between low-level inscription like machine code and high-level reception like an interface display. For example, when used in exploring videogame typography, ROMscrape reveals constraints introduced at the platform and code levels which manifest in certain visual features that influence interface design in other contexts which lack the original constraints. In other words, the expressive domain of videogames is always already inflected by the material origins of their inscription and tranmission. These and other findings demonstrate the value of a layered, textual architecture to videogame objects and reveals the discourses of influence which obtain in the relationships among literal levels of inscription.

Sample Bibliography:

Bogost, Ian. Persuasive Games: The Expressive Power of Videogames. Cambridge, MA: The MIT Press, 2006.

Galloway, Alexander R. Gaming: Essays on Algorithmic Culture. Minneapolis, MN: University of Minnesota Press, 2006.

Juul, Jesper. Half-Real: Video Games between Real Rules and Fictional Worlds. Cambridge, MA: The MIT Press, 2005.

Kirschenbaum, Matthew G. Mechanisms: New Media and the Forensic Imagination. The MIT Press, 2008.

Montfort, Nick. “Combat in Context.” Game Studies. 6.1 (2006) <http://gamestudies.org/0601/articles/montfort>.

Plaisant, Catherine et al. “Exploring erotics in Emily Dickinson’s correspondence with text mining and visual interfaces. Proceedings of the 6th ACM/IEEE-CS joint conference on Digital libraries. Chapel Hill, NC, USA : ACM Press, 2006.

I’ve been working pretty hard on ROMscrape, but the frustrating part of that is that it’s a lot of waiting: Assess - Tweak - Run - Wait - Assess - Tweak - Run, etc.

I am happy to report, though, that my tweaks have become increasingly minor and I’ve seen fairly positive results so far with my current approach.

What I’ve figured out is that for my search to work well in PHP and on this particular server, I need to keep it’s comparisons to no more than around 3000 at a time. I can do a few of those in a row, so I can still search very large sets. The challenge lies in organizing the data set so I can pass a query first to a list of 3000 or so indexes, select one or two, then search each of those indexes. Doing this successfully means creating indexes that accurately correspond to a set of similar data, and that similarity in this regard is relevant to the types of queries I’ll be passing to it.

I don’t want to go into my whole solution, but the bottom line is that I’ve got a program running on my home computer that is generating the best set of 3000 indexes that it can come up with. By my calculation, this requires doing around 1.5 million compare-and-sort operations. That sounds like a lot, but I’ve got it optimized to the point that it can do 100 in about 15 seconds. When I left it this morning, it had 1 million left to go, so I think that means it will be done some time tomorrow.

Working on this has been really interesting, and I’ve learned a great deal about the kinds of logic that must go into big search engines like Google. Assuming they work on systems at least remotely similar to what I’ve come up with, these algorithms probably have plenty of influence from humans making value judgments, despite their apparent mechanical objectivity. I know Google has emphatically denied manipulating their indexes to “hide” certain sites or objectionable material (the infamous “jews” case being a rare exception), but even the algorithm must operate on some values and assumptions that are pre-determined by human choices.

In my case, ROMscrape’s indexer is flexible enough that I could easily make several tweaks that would dramatically change its findings. Deciding which metric for determining similarity is the most obvious one, but even deciding how to store similarity information (I’ve attempted what I think is a “fuzzy” approach, incidentally) might make a big difference down the line. For example, one scale I use arrives at a calculation along a hexadecimal scale of 0 - F, thereby retaining 16 levels of similarity for that slot. If I wanted to, I could double the digits and store that information on a scale of 0 - FF (0 - 255). This would be a finer degree of similarity, but it would cost more in terms of computation. And until I get a result set and try some searching on it, I won’t be able to tell if the two-digit method will be worth it.

At any rate, I’ve had several starts and stops (and restarts) on this, but I feel confident enough in the present indexer that I’m working in earnest on chapter 5 now. Hopefully it will turn out well and I’ll be able to post some good news soon.