November 2004

« October 2004 | Main Index | Archives | December 2004 »

26
Nov

We need a new term for "computer game".

There have been two major stories in the electronic entertainment news over the last couple of days. Or at least two big ones that Google has pointed me to.

The first is the oh-so-shocking announcement that, what with the release of GTA San Andreas, Doom 3, Half Life 2 and a new Leisure Suit Larry game, this is a bumper year for sex and violence on the computer screen. This is a priori a bad thing, of course, and a sign of the evil things our kids are getting up to.

Children play games, you see.

Except that buried two thirds of the way down some of these articles is the embarrassed admission that the average age of the computer game player is something like 29. Computer games are adult entertainment1 now, and we shouldn't exactly be surprised that adults want the same sort of things out of games as they do out of movies.

The second piece of news was the release of JFK Reloaded an attempt to simulate the precise circumstances of the assassination of John F. Kennedy by Lee Harvey Oswald (the game is based on the assumption that Oswald did it). The hue and cry over the game's release has been predictable.

Traffic, the Glaswegian company behind the game, did a lot of things wrong PR-wise, but their biggest mistake was giving their simulation a hokey name, and calling it a "computer game".

If a major TV network were to air an advertisement tomorrow for a new documentary on Kennedy's assassination, nary an eyelid would be blinked. Ditto if they promoted that "through state of the art technology, we have been able to simulate the precise conditions of that fateful day, and prove that Oswald could have been the killer." Ditto if said simulation involved a 3d fly-through of the trajectory of each bullet.

You can't package that simulation for everyone, though. Democratise the documentary so that anyone can attempt the reconstruction themselves (offering a prize for the person who comes closest to proving Oswald could have done it), and pundits will be calling for your heads.

Because it's a computer game. Kids play games.

If you think about it, it's not much of a game, is it. There's one level. You get to fire at most three bullets, and then all you can do next is try again. The victory condition (making the shots exactly as described in the Warren Commission report) doesn't even make any sense from a gaming point of view. It's like writing a golf game where the aim is not to make the lowest possible score, but to reproduce a particular Greg Norman game down to the last stroke.

Its only real value, then, is as a simulation, an investigation of a particularly controversial historical event from a new angle.

(Then again, given the amount of time I've spent pushing simulated people down stairs, or clubbing penguins long distances, maybe the tolerance for such shallow and repetitive gameplay is greater than i think)

So we need a new word for "computer game". Preferably something short and unpretentious -- I don't want "simulated" this or "immersive electronic experience" that -- but without the baggage of the mental image of a thirteen year old kid playing Super Mario Brothers.

1 Admit it. You read "adult entertainment" and thought "porn". I dislike the way the word "adult" has been co-opted. Just once, I'd like the "Adult themes" disclaimer at the start of a TV show to mean that the program explores the nature of reality and the tragedy of existence, rather than just being a warning that someone might mention sex.

What I love about TheServerSide's occasional cartoons is not just that they're all remarkably un-funny, but that the kind editors of TSS go to such great lengths to explain the joke in advance in the sidebar.

Inspired by this article about the working conditions of an Electronic Arts employee, here is my list of things a company can provide to get more work out of a computer programmer:

  • An office with a door
  • ...and no phone
  • A culture of asynchronous communication
  • A fast workstation
  • ...and two monitors. You wouldn't believe how much difference a second monitor makes
  • ...and their operating system of choice
  • Good development tools.
  • A fast Internet connection1
  • Snacks and drinks they don't need to leave the office for
  • A good-natured working environment
  • Flexible working hours
  • Tasks appropriate to their ability
  • ... and if at all possible, that they find interesting
  • Investment (emotional or financial) in the end-product

The list could be much longer, but it boils down to:

  • Give them the tools to do their job efficiently
  • Remove potential interruptions or distractions
  • Make sure they're motivated

Mandatory overtime isn't on this list, believe it or not. Not just because it sucks (which it does), but because it's counter-productive.

Programmers who have good working conditions and a personal investment in the end result will often volunteer overtime at crunch periods, or just when they have a particularly thorny problem to overcome and don't want to go home until it's done. Also, there will always be the occasional deadline which leaves everyone putting in a tough couple of weeks to push the project over the line on time.

If, however, you find you're working weekends before every deadline, or in the case of the EA employee, for the lifetime of the project, there's something seriously wrong. The project is being mismanaged, and the programmers are being whipped harder to compensate.

Even too much voluntary overtime should be viewed with caution, because people who work too long get tired. Speaking for myself (although there are various reports suggesting this applies generally), the more overtime I work, the less I find myself able to concentrate or apply myself to the task at hand. I'm fine for short bursts of extra work, but over prolonged periods the extra hours I'm working are overcome by the productivity I lose while I'm working them.

Mandatory overtime is even worse, because it will eat through any developer enthusiasm you may have fostered until you are left with a team of programmers who just want to go home. A programmer who wants to go home will cut corners, ignore bugs, write spaghetti code, find devious ways to pass work on to other members of the team, anything so that they can declare their part of the program finished and go somewhere that isn't the office.

1 Some would put the Internet under the "distractions and interruptions" column. It can be. However in this world of postmodern programming, it's vital that a coder has fast access to Google, online documentation, library downloads and so on.

A Fond Farewell

  • 11:19 AM

I heard on the radio that a tenth anniversary "special edition" of Jeff Buckley's Grace is going to be released. I would like to say, from the bottom of my heart, STOP IT. Stop digging up the poor man's corpse and parading it around on holidays. We all have Grace already, and after ten years the bottom of the barrel of unreleased material from an incredibly short career has been scraped raw.

Posthumous albums can be powerful eulogies. They can be unfortunate, but well-meaning collections of tracks the artist wasn't happy enough with to release when he was alive. They can be blatant record-company cash-ins.

Elliot Smith's swan-song, From a Basement on a Hill, on the other hand, is just annoying. I was expecting some kind of closure, but what I got was just a bloody good CD.

To fill in the back-story: after the patchy Figure 8, Smith retreated to his own studio, getting permission from Dreamworks to release his next album independently. Three years later, he was dead. If you take the official reports at face value (many are skeptical), stabbing yourself in the heart twice shows a certain commitment to ending it all.

I think I assumed that this retreat would mean a step back to the more acoustic sound of the pre-Dreamworks albums, and that From a Basement would be a revisiting of Either/Or, a full-circle, a full-stop.

It was a return to form, but not a return to the past. Smith took advantage of his freedom: using it to flirt with a heavier sound, experiment with songwriting and studio tricks, and do things with producing his music that were more daring, and more raw than Dreamworks may have allowed.

The characteristic Smith melodies are still there. The songwriting that has been compared so often with Lennon/McCartney to become a cliché. The multi-tracked vocals. The skilful picking of acoustic guitars. The melancholy lyrics. The occasional track that strips everything back to voice, guitar and tune. But From a Basement on a Hill was Smith progressing as an artist.

Despite the emphasis on lines like "I can't prepare for death more than I already have", and "Give me one reason not to do it", this CD shows me a Smith moving forwards musically, not one about to abruptly stop. And that's why it's annoying.

I feel cheated.

I want to hear what he does next, damnit.

I've mentioned before the irony of Atom being touted as better because it is "clearly specified", while at the same time certain large organisations are trying to force the widespread adoption of a spec that is still, in fact, being vehemently argued over, and is full of "To be decided" sections.

I would, however, like to thank the tireless participants in Atom's standardisation process. Whilst testing mail archiving, my 30Mb atom-syntax folder, with its sprawling, multi-forked threads, made great test data.

dummy.gif

  • 11:18 AM

We use a lot of icons in Confluence. Nice icons can make the UI friendlier, and generally squash a lot more information into a 16×16 block of pixels than you could with text.

One problem, though, is that you inevitably have the "stop the world" moment in the middle of coding a new feature, when you're plugging in the UI and need to choose an icon to match. For me, this generally means annoying Mike, because he can draw and I can't.

The solution, of course, is to have a standard placeholder graphic. We call ours dummy.gif. The placeholder allows you to defer your choice of icon to late in the release-cycle. When all the features are done, the more graphically inclined team-members can grep the source-code for mentions of the filename, and replace them with consistent, happier-looking icons1.

It was Nick who found the perfect placeholder:

It's perfect, partly because the angry red animated gif makes it very obvious that this isn't the final graphic, but mostly because it's funny when you've got ten or fifteen of him on a page, all lined up and glaring out at you.

1 Needless to say, at least one Confluence development release has shipped with the placeholder still quite obviously in place, but the zip was only up for about half an hour before we noticed.

So it looks like the US election is going to Bush.

Even though it hurt me deeply, there was a part of me that could understand the Australian election. Howard may be a fuckwit, but it's still hard to get rid of a government with a good economic record, and deservedly or not it's on that record he swung the voters.

Bush, on the other hand... I simply can't understand how a rational human being would vote for Bush:

  • He's presided over a record budget deficit, so any claim for economic credibilty is shot
  • He's failed utterly at the war on terror: placing rhetoric over action at every step, and sowing careful misinformation to redirect the real threat of Al Quaeda to the invented threat of Iraq
  • Even if you agree with the invasion of Iraq despite the constant flip-flopping on why it's happening, the Bush administration has run the war incompetently, constantly pushing for some kind of “quick victory” at the expense of the safety of both Iraqi civilians and US troops.
  • He's presided over a constant and brutal attack on civil liberties and individual rights
  • He's presided over the most blatantly dishonest US government on record: even Nixon aides have come out to say "Hey, we weren't as bad as this."

I simply don't understand how anyone in their right mind could vote for Bush. I'm sure a lot of really smart, rational human beings have done so, but I just don't understand the degree of self-deception that has to go on to say "Hey, this lunatic is a good thing for the future of our country."