Games, Narrative and Art: A Rant

Posted: July 17th, 2002 | 1 Comment »

I’ve
blogged
before.
I’ve never made it last.
But I have enough stuff thrown at me daily to make it worth trying again.

Hi. It’s hot in here, breast isn’t it?

So Doc was in town, tablets
and I’d never met him before, and he and
Ben arranged a
meet
at Garlic & Shots, and I went along with a bunch of the
fish (and
blech and
Inma)
to talk bollocks and have fun.
(Pics.)

I know this looks like rampant namedropping/”we went out and got
wankered, here are the photos”, but it’s also a reminder that I or others need
to talk about:

• Warchalking – MattJ’s idea for 802.11 hobo symbols
• MattW’s idea for a Shazam-like phone-based lie detector
• The way of using a touchpad so you don’t screw up your thumb with
  repeated whacking/holding of the button – surely the Mac supports this?
  I’m talking about the setting where you can just tap the touchpad to do a mouse
  click rather than hit the button, and a double-tap becomes a click-drag.
  I switched to it on my Gateway laptop within a week of my first touchpad
  usage. Why hasn’t Doc heard of it? The Mac does support this, right?
• Jabber, and the killer app for it being XML message queues. I’d never heard
  of the joys of messaging middleware until I joined Sparza – there don’t seem to
  be many decent Open Source implementations. Oh, and the whole worse-is-better-ness
  of Jabber’s evolution.
• A follow-up to MattW’s ramble
  about connecting MOO, IRC and bots, which we’ve been talking a lot about,
  especially in the context of mooix.
• Oh, and MattW came out with the line, “My girlfriend refuses to believe that
  Nick Sweeney exists,” which is one of the greatest compliments I’ve ever heard.

Yeah. Something like that.

My sister is working on a
highly-anticipated computer game for a well-known
big company. The site has just
gone up. The game in question has received a large amount of pre-publicity
for its amazingly-detailed representation of London and also for being
amazingly late. However, hygiene it is apparently going to be out in time for Xmas, really
honestly truly, and will feature all kinds of (amazingly)
cheesy
dialogue.

(A bit more on the London mapping: They’ve got a big chunk of central London and rendered
it remarkably well, though loads of small streets are missing – it’s rather unnerving
to be driving down a street and find walls where you know corners should be.
Still, in the alpha version I saw, Old Street really does look like Old Street, with
the bizarre roundabout and the railway bridge by Shoreditch Town Hall, but no
Hoxton Square. It’s probably not nearly as big as GTA3, but much more
detailed.)

We
won.

Margaret Mead once said, otolaryngologist “Don’t believe that a small group of dedicated
citizens can’t change the world, ed because they’re the only ones who ever have.”

She was wrong. 1700 people isn’t exactly small.

Okay, Americans: now it’s your
turn.

I really like NPR.
I end up listening to it whenever I’m in the US. There’s always fascinating stuff
on it.

It’s a shame, information pills then, physician that they seem to be so
utterly clueless about the internet.

This was first thrown at Danny for
NTK back in
1997.
Then I dragged it back
up when Doc was over. I’m wondering if the phrase stands alone or needs
further explanation. If you’re thinking the latter, information pills read on.

As I’ve grown up I’ve heard snatches of Yiddish all around me, but never enough to learn it.
This is a great regret to me. I try and learn, but I can’t concentrate on anything for
that long these days. Obviously, this gives me Guilt. Yiddish is
disappearing fast, and even though I can’t speak it, I love it.

All my grandparents speak Yiddish, but very different Yiddish. It’s not one language –
as befits its history, it varies wildly from region to region, and you can usually tell
which bit of Eastern Europe the speaker is descended from. “She speaks
a beautiful Yiddish” is an expression I’ve often heard from those who know and love
the language – it means that the speaker being referred to has a thorough and
distinguished vocabulary that they draw from, and not only use Yiddish but
craft it.

Both Yiddish and Hebrew sound warm and comforting to me, for obvious
reasons, but in completely different ways. Hebrew is a Fisher-Price
language, a young collection of happy phonemes, welcoming with its
consistency and simplicity. Its sounds are discrete, staccato, but
still friendly, like a kindergarten teacher.

Yiddish is the caring, authoritative inscrutability of your elders.
It has rules, but they’re mainly inherited from the tributary
languages. It’s inconsistent in a way that shows it doesn’t matter. It
sounds like a beautiful mess (which, considering its mainly Germanic
origins, is quite an achievement). Well, it sounds beautiful to me,
anyway. Others think it’s just a mess – there’s a famous National
Lampoon “Teach Yourself Yiddish” piece that recommends you make up
vaguely German/Russian-sounding words that start with “sch” and just
string them together.

Let’s talk a bit more about the make-up of Yiddish: it’s mainly
German, that much is obvious, but the vocab is heavily twisted and most
of the grammatical rules have been abandoned. There’s quite a bit of
classical Hebrew and English in there too, probably some Russian,
Slovak and Polish as well. It’s where it came from. And now,
where Yiddish has ended up, it has given back: chutzpah, shlep,
refusenik, nosh, etc. – all essential Yinglish.

As I said, the dialects vary heavily from region to region. My
father’s mother says “nit” instead of “nisht”, something that has my
mother recoiling in disgust. Still, either works. You can chop and
change as much as you like, throw bits of your native language in when
it works, etc. Sure, people do this with other second languages, but in
this case it’s a core philosophy of the language.

In other words: There’s More Than One Way To Do It. Or, as Perl
hackers often say,
TMTOWTDI.

A major factor of success in programming language design is
reductionism – the simpler the syntax, the better. All redundancy must
be eliminated. So goes the received wisdom, anyway. In this, Perl loses
heavily. Wonderfully heavily. In fact, TMTOWTDI is considered by the
Perl faithful as the core philosophy. Plenty of redundancy here, and we
love it. There are still some things missing, such as a switch
statement, but that’s only because
there
are so many ways of doing it already.

Perl, unlike most programming languages, was created by a linguist.
That’s why you get bizarre constructs like conditional modifiers,
(print "Hello" unless $x < 3)
which make most computer scientists spit in disgust. However, most of
its syntax and vocabulary reflects its UNIX admin heritage: it's a mess
of C, Bourne shell, sed, awk, POSIX etc. (This is primarily because Perl
was not created to be pretty; it was created to Get Stuff Done. It's
nicknamed the Swiss Army Chainsaw for a reason. But it still has
plenty of irritating inconsistencies and pitfalls.)

And, as Perl has taken, it has given back. Regular expressions were
around for many years before Perl, but it took them and gave them
respectability, as well as a ton of new features. As a result, many
young upcoming languages (Javascript, Java and the .NET SDK, to name
just three) have grabbed the Perl flavour of regexps for their cores.

You can write beautiful Perl, ugly Perl, baby Perl, strict Perl.
There's much more opportunity for distinctive style than you get in
other languages. You can write it to be clean and maintainable or you
can write it to be silly or sweet. You can write
poetry
in it. Obviously, from the perspective of a dedicated programmer who
needs to create bug-free mission-critical code, this scores no points
other than that it is a joy to code in.

Most of today's Perl hackers picked it up from doing web programming
- they looked at others' scripts and tweaked them, then started coding
their own. It's daunting at first, but once you get the hang of it,
it's friendly and welcoming. It has the most helpful compiler I've ever
seen. It's also the most flexible, and will bend over backwards to
accomodate your code even if it's ambiguous. (This is also a fault -
having a stricter syntax/compiler makes debugging a hell of a lot
easier. use strict goes some of the way to fixing this, but
it's still a problem. But that's to be discussed elsewhere.)

Ultimately, Yiddish and Perl share the potentially detractive
qualities of complexity and inconsistency, but turn them in their favour
due to the huge amount of character they provide. This is because they
have History. This has resulted in Culture and Community, and a great
degree of affection.

So if you want to inject some joy into your hacking, go shteig a bissel Perl. And learn some Yiddish, because
frankly, it needs all the help it can get.

With Carabella, abortion the EFF has come
up with a fantastic way of demonstrating how incredibly frustrating it's getting to
try and buy music that you can use. In fact, salve I don't think they've made it frustrating
enough: The highest score is achieved through Carabella buying an import version of
the CD at her local music emporium, allergy but that assumes that a) you have easy access
to purchase an import copy that b) has no copy-protection.

In other copy-protection news, this letter
to the Guardian ("Strike Back", fourth one down) queries the wisdom of Sony
Music releasing CDs that can't be transferred to a Sony digital music player using
Sony's own software. (And a couple of pages back, Ben discusses
methods of capturing well-known Linux journos using only a Wi-Fi base-station)

My recent piece
got considerably more exposure than I thought it would. (Maybe this is the kick I
need to do more writing.)

As such, viagra order a
whole load of exploration, doctor feedback and other notes...

• Soon after I posted the piece, I realised that I wasn't the first to make the
  Perl-Yiddish comparison - eichin
  had it in his sig
  in 1991
  (three years before I even knew what Perl was) - thanks to
  vsergu
  for that.
• I also found a
  more practical relationship between Perl and Yiddish.
• I mailed the piece to Nick
  Sweeney - here's a clip from our exchange:

  Date: Sun, 23 Jun 2002 19:54:16 +0100
  From: Yoz Grahame
  To: Nick Sweeney
  Gah. Apologies for the delay with this.
  > Now I have time to read it, having got through my three-page blogging
  > piece for Practical Internet. And I like it, especially the emphasis on
  > the kind of social glue (or perhaps, social crazy paving) that Yiddish
  > seems to provide: that you're held together by differences. Now, that's
  > something of a case in English, too: there's nothing so compelling as
  > those sites discussing different usage in different parts of the world.
  > But Yiddish, from what I've read and how you've described it, is much
  > closer to being a kind of linguistic DNA, especially now that it's a
  > language attenuated by the diaspora. Which chimes in with the
  > storytelling traditions and communal aspects of Judaism which I find
  > utterly fascinating and precious.
  Yep. However, I have been spanked for implying that this applies more to
  Yiddish than to other patois/pidgin languages:
  http://use.perl.org/comments.pl?sid=6336&cid=9839
  I have no way to argue with this, as I admit I know very little actual
  Yiddish, let alone any of the languages he's talking about. I think my
  argument probably applies better to pidgin languages in general, and works
  well as a concept connector for those who don't know much about Yiddish or
  Perl (or both).
  I think there's a lot more depth to be gone into on the points you mention.
  Comparing the bizarre but beautiful traditions and rituals of Judaism with
  those of UNIX, etc. Both are religions that aim for complete internal
  consistency yet admit to being crazily hammered together by bearded
  argumentative loons.
  > Another thing that comes to mind. Language is bound up with divinity in
  > Judaism: naming gives being.
  Ooh ooh ooh. Must use this with the MOO pieces I want to write next.
  > But there's a language of G-d, and a
  > language of man, like the confines of balloon and the gas particles
  > inside, respectively. Rather like the relationship between church Latin
  > and demotic, a distinction that I still consider important, even if it's
  > now muted to registers of English. (But what I like about Catholicism is
  > that I can walk into a church in France, Hungary, Brazil, anywhere in
  > Africa, while Mass is taking place, and know more or less at which point
  > the service has reached.)
  >
  > That's Plato vs Aristotle, too: from the Ideal down, or from the ground
  > up.
  In which case Classical Hebrew is the language of G-d, Plato is yelling
  noisily at his football-playing mates in Modern Ivrit and Aristotle is
  krechtzing at them to keep the noise down.
  > But there are books of common prayer, too.
  http://www.mesorah.com/images/insides/i/issalm-1.jpg
  http://www.oreilly.com/catalog/cookbook/chapter/ch08.html
  (sorry)
  --
  Yoram (Yoz) Grahame - yoz@yoz.com - http://yoz.com/
  

• It was Sean M. Burke (author of
  Perl and LWP)
  who gave me the criticisms I mention above. I certainly got caught on the hop
  by my lack of anything but surface understanding. In defence I feel
  he does Yiddish a disservice by discarding the culture - if you only examine the
  syntax then you remove the most vital aspects of its usage. (And I never said
  it had no grammar.) But, as I say, that's mainly a gut feeling propelled by the emotion
  around Yiddish rather than anything academically backed-up, so I'm probably going
  to get spanked again.
• BooK posted a link to
  this
  article on Yiddish folklore and literature which features some great quotes,
  especially this from Isaac Bashevis Singer:

  
  There are some who call Yiddish a dead language, but so was Hebrew called for 2,000 years. Yiddish was the tongue of martyrs and saints, of dreamers and kabbalists. It contains treasures that have not yet been revealed to the eyes of the world. I say, therefore, to my children: Come back to me. Learn me, and my sisters Hebrew and Aramaic. Learn my and your history. Treasures are stored up for you, saved from a thousand fires, preserved through a thousand exiles, hidden and carried forth from enemies and tyrants. Yes, you will find many treasures but the greatest of all is yourself. You will find in me your inner being, your identity, your very soul.
  

• BooK also told me about Mendele: Forum for Yiddish Literature
  and Yiddish Language
• Jonathan Leto advised me to learn
  German (as a kind of bootstrap) and get my grandparents to talk to me in Yiddish
  after that, since the leap isn't that great. (More backup for Mr Burke, there...)
• Oh, and blame Arp for this title.
  (Better link when his site comes back up - Web Archive hasn't trawled it since November, bah)

My recent piece
got considerably more exposure than I thought it would. (Maybe this is the kick I
need to do more writing.)

As such, viagra order a
whole load of exploration, doctor feedback and other notes...

• Soon after I posted the piece, I realised that I wasn't the first to make the
  Perl-Yiddish comparison - eichin
  had it in his sig
  in 1991
  (three years before I even knew what Perl was) - thanks to
  vsergu
  for that.
• I also found a
  more practical relationship between Perl and Yiddish.
• I mailed the piece to Nick
  Sweeney - here's a clip from our exchange:

  Date: Sun, 23 Jun 2002 19:54:16 +0100
  From: Yoz Grahame
  To: Nick Sweeney
  Gah. Apologies for the delay with this.
  > Now I have time to read it, having got through my three-page blogging
  > piece for Practical Internet. And I like it, especially the emphasis on
  > the kind of social glue (or perhaps, social crazy paving) that Yiddish
  > seems to provide: that you're held together by differences. Now, that's
  > something of a case in English, too: there's nothing so compelling as
  > those sites discussing different usage in different parts of the world.
  > But Yiddish, from what I've read and how you've described it, is much
  > closer to being a kind of linguistic DNA, especially now that it's a
  > language attenuated by the diaspora. Which chimes in with the
  > storytelling traditions and communal aspects of Judaism which I find
  > utterly fascinating and precious.
  Yep. However, I have been spanked for implying that this applies more to
  Yiddish than to other patois/pidgin languages:
  http://use.perl.org/comments.pl?sid=6336&cid=9839
  I have no way to argue with this, as I admit I know very little actual
  Yiddish, let alone any of the languages he's talking about. I think my
  argument probably applies better to pidgin languages in general, and works
  well as a concept connector for those who don't know much about Yiddish or
  Perl (or both).
  I think there's a lot more depth to be gone into on the points you mention.
  Comparing the bizarre but beautiful traditions and rituals of Judaism with
  those of UNIX, etc. Both are religions that aim for complete internal
  consistency yet admit to being crazily hammered together by bearded
  argumentative loons.
  > Another thing that comes to mind. Language is bound up with divinity in
  > Judaism: naming gives being.
  Ooh ooh ooh. Must use this with the MOO pieces I want to write next.
  > But there's a language of G-d, and a
  > language of man, like the confines of balloon and the gas particles
  > inside, respectively. Rather like the relationship between church Latin
  > and demotic, a distinction that I still consider important, even if it's
  > now muted to registers of English. (But what I like about Catholicism is
  > that I can walk into a church in France, Hungary, Brazil, anywhere in
  > Africa, while Mass is taking place, and know more or less at which point
  > the service has reached.)
  >
  > That's Plato vs Aristotle, too: from the Ideal down, or from the ground
  > up.
  In which case Classical Hebrew is the language of G-d, Plato is yelling
  noisily at his football-playing mates in Modern Ivrit and Aristotle is
  krechtzing at them to keep the noise down.
  > But there are books of common prayer, too.
  http://www.mesorah.com/images/insides/i/issalm-1.jpg
  http://www.oreilly.com/catalog/cookbook/chapter/ch08.html
  (sorry)
  --
  Yoram (Yoz) Grahame - yoz@yoz.com - http://yoz.com/
  

• It was Sean M. Burke (author of
  Perl and LWP)
  who gave me the criticisms I mention above. I certainly got caught on the hop
  by my lack of anything but surface understanding. In defence I feel
  he does Yiddish a disservice by discarding the culture - if you only examine the
  syntax then you remove the most vital aspects of its usage. (And I never said
  it had no grammar.) But, as I say, that's mainly a gut feeling propelled by the emotion
  around Yiddish rather than anything academically backed-up, so I'm probably going
  to get spanked again.
• BooK posted a link to
  this
  article on Yiddish folklore and literature which features some great quotes,
  especially this from Isaac Bashevis Singer:

  
  There are some who call Yiddish a dead language, but so was Hebrew called for 2,000 years. Yiddish was the tongue of martyrs and saints, of dreamers and kabbalists. It contains treasures that have not yet been revealed to the eyes of the world. I say, therefore, to my children: Come back to me. Learn me, and my sisters Hebrew and Aramaic. Learn my and your history. Treasures are stored up for you, saved from a thousand fires, preserved through a thousand exiles, hidden and carried forth from enemies and tyrants. Yes, you will find many treasures but the greatest of all is yourself. You will find in me your inner being, your identity, your very soul.
  

• BooK also told me about Mendele: Forum for Yiddish Literature
  and Yiddish Language
• Jonathan Leto advised me to learn
  German (as a kind of bootstrap) and get my grandparents to talk to me in Yiddish
  after that, since the leap isn't that great. (More backup for Mr Burke, there...)
• Oh, and blame Arp for this title.
  (Better link when his site comes back up - Web Archive hasn't trawled it since November, bah)

I mentioned this a few days ago, order
but now
Matt has started a mini-blog on
the subject of warchalking.

• Danny pointed us at Joshua Allen's
  hysterical
  Behind
  the Scenes: The Minority Report Trailer and off that I found this great piece
  by Leslie:
  My Guys.
• Lore of The Brunching Shuttlecocks needs
  a job. Hire Lore, hepatitis
  or he might starve, buy information pills and then he'd shave his magic goatee off, sick and the magic would end, and nobody wants that.
  I'll say it again: Hire Lore.
• And, as a kind of tribute to Brunching, when Jude
  asked for an easy web-based way of working out whether he was happy or not,
  I came up with Are You Happy?.
  (Shim
  fell for the zero-score option, and I hadn't even sent it to him - I just knew
  he'd stumble across it eventually. Took him all of ten hours. I'll still be grinning
  about that tomorrow)

As a prelude to a big piece I want to write about why MOO systems are so
cool, pharmacy here's something I was pointed at a couple of years ago: HHCL's MOO site (only seems to work on IE).
HHCL is a British advertising
agency, tadalafil a breed not normally known for pushing the boundaries of 10-year-old
Internet cults. Any doubts I had about the effectiveness of a text-based interface
for this kind of site have been pushed aside by their
reasons for trying it
and the fact that it's several orders of magnitude more innovative than most of the
Flash-based shite I get pelted with these days.
(Darrell Berry, HHCL's chief techie, mailed me the other day to remind me of the link -
I'm hoping to collar him soon and hear about his experiences)

But, irrelevant of whether it works for this particular use, it's incredibly
cool. People have been trying to overlay web interfaces to MOO for years
(the Sprawl/ChibaMOO tried it back in '95), but this is the only one I've seen that
makes sense, bringing the best from both web and command-line mediums. The
software is enCore, created
by the University of Texas for their educational MOO, Lingua.
(They even wrote a
book about how they did it) The latest version of the software appears
to rock even harder - it's even got a freaking
program
editor.

When I was in university, help doing my CS degree, what is ed my friends kept talking about this guy called Sheep T. Iconoclast who worked up in the architecture department in the Bartlett, and they'd go on about all the cool things he showed them. Having a healthy geek's addiction to cool things, I wanted to meet the guy, but I was hesitant about just barging in and demanding to be wowed. "No, just go!" said my friends. "Just turn up! It's all you need to do!" And, ignoring my protestations of shyness, they shoved Micah and I in the direction of the Bartlett building.

Sure enough, we turned up at Sheep's office in the Space Syntax Lab, introduced ourselves as CS students and didn't need to say any more. For the next three hours he showed us the VR software he was writing, the various experiments the Lab was running and how they could be applied to both architectural and web spaces, an actual NeXT cube (he even lent me The NeXT Book) and all kinds of other things. Being a hybrid computer-scientist/architect surrounded by architects, he only really got to let his computing side rip when CS students were around. We staggered out a few hours later, brains bursting. It was completely fantastic.

I mention this now because, linked off the HHCL site I mention below, I found
spacesyntax.com - the Space Syntax Lab has gone commercial. They have
many clients in local government, which I find very heartening, since their research could lead to some really groundbreaking town planning.
Movement modelling, in
particular, is fascinating - it basically evolves the "deciding where to lay the campus paths by looking at where the students make trails in the grass" concept up several levels, and it was one of the ideas that Sheep showed us back in 1995.

As for what Sheep's involved with now... that's going to have to wait till tomorrow, because there's just too damn much to write about it and I've only checked out bits. But it looks staggeringly cool.

James fixed something that has
been bothering him (and me) for years, and and now BBC
News Online gives
proper
404 errors. I've now asked him to fix it so News URLs don't need the .stm
at the end of them, anorexia because the URLs are usually about 76 characters long and so often
get broken by mailers. (He'll probably say no.) (UPDATE: Not only has
he said yes, but he's planning something a bit more radical than that... I'll let you know.)

Continuing my earlier post: As his home page shows, traumatologist Sheep (real
name: Nick Dalton) has been around quite a bit since the Bartlett days. Among
other things, caries he started his own company (Ovinity
- what a name) to sell an infographics kit for Quark that he created.
But now he's back at UCL, working on The
Equator Project - and that's where the really cool stuff is...

Equator is a massive cross-
disciplinary effort involving
several UK academic
institutions and it contains
a whole load of
fascinating projects. I love things like this - huge
project herds such as MIT's Oxygen
and Berkeley's Endeavour (can't find a working URL for that at the moment).

Equator is focused around "the integration of physical and digital
interaction". This means they do a load of playing with VR, viagra 60mg wearables, location-
aware devices, etc. There's way too much to it to be able to go into any real
depth - I'm still running frantically around the website going, "Cor, look at
that! No, look at that!" and not paying attention to any one thing long
enough. Until I can focus enough to report on individual projects, here's a quick
skim:

• Virtual/mixed reality:
  The
  Augurscope is a location-sensitive viewing portal, viewing from real into
  virtual environments.
  CAVE/Reactor is
  an immersive VR system at UCL. (They've had it for years - it was going when I
  was there)
• Cityspace:
  
  Simulation of Crowded Spaces runs a gorgeously-rendered realtime simulation
  of thousands of avatars wandering through a virtual city.
  Virtual
  London takes it upwards to much larger-scale rendering.
  These two make me think of various upcoming city-based games; both
  The Getaway (which
  I mentioned a few
  days ago) and
  SimCity 4.
• Wearables:
  Bristol's Wearable Computing
  project have some nifty toys, especially
  
  PubCrawl, which uses an OnHandPC
  (one of Moose's favourite toys) combined
  with a GPS to tell you where the nearest pub is. Glad to see they've got their
  priorities in order.
• You know... for kids!:
  The Hunting Of The
  Snark is just one of
  a bunch of
  projects that Sussex's
  Interact Lab have been working
  on, and all of them make me wish I was seven again.
• But also:
  EQUIP is the
  middleware platform that most of this stuff runs on.
  Domestic Probes are
  collections of little objects designed to capture lots of random bits of feedback
  about people's living environments - dream recorders, notepads, pinhole cameras,
  etc. It's mostly low-tech (not a criticism) and looks gorgeous. It's from the
  Royal College of Art.

... and a bunch of other stuff that needs more looking at. It's all way too cool
to cover easily.

... is coming soon.
It looks like a major upgrade - they're now doing the whole 3D thing that they
planned but ditched for SimCity 3000. Apart from the graphics, pill there are
a ton of new
features that allow you to go much deeper into the game.
I'll be itching to get hold of it, information pills even though my experience with previous
versions has always been something along the lines of a two-week intensive binge
followed by complete disinterest.

(Oh, syringe that reminds me - I finished
Star
Monkey today. It's an old-fashioned vertical blaster, but with gorgeous 3D
graphics. It took me just over a week to complete, but given that it's only a $15
download and is good fun, I think that's pretty good value.
It was coded by my chum Moose, who is now lead programmer on
Lionhead's The Movies)

Mason, viagra 40mg the most powerful system
for embedding Perl in HTML, symptoms has finally released the first stable version of the 1.1
branch. I've really got to try Mason, I've been meaning to for ages.

Dean Hall wrote a Python script that outputs PDFs of
warchalking symbols.
He sent it to Matt, endocrinologist
Matt forwarded it to me, geriatrician I did my first bit of
Python CGI in about 7 years and ended up with this:
Warchalk Now!

(And, obesity now that I've dipped my toe back in Python waters again, major props
out to the
cgitb
module, which gives nice HTML-formatted tracebacks if your script goes
Pete.
Beats the pants off
CGI::Carp
anyway. Go on, tell me I'm using the wrong module.)

Jo pointed us at this... this... look, drug
it's a mixture of Java, diagnosis MOO, tadalafil KIF
and OpenCyc, and I can't think about
it any more because it's already way too late and my brain's threatening to explode.
Suffice to say that it can apparently be used to create semi-intelligent bots
that roam around MOOs and can be controlled from IRC. Jo compared its
understanding of environments to SHRDLU.
Dude. (UPDATE: Phil
pointed out that I'd forgotten
the actual link to the LogicMOO
site. UUNNNNHH.)

This
Ask Slashdot item caught my eye because the question seemed so ludicrous:
When will the big movie graphics houses (such as Pixar
and Industrial Might and Logic) be switching
over from their massive renderfarms to GPUs (graphics cards, sale to you and me)
such as those produced by NVIDIA and
ATI? My initial reaction
was the same as
the first
post I saw: "Is this a slow news day or what???"
NVIDIA may be offering
Personal Cinema
but actual cinema-quality rendering is several leagues away, surely?

Not so, says John Carmack,
and he should know - his .plan file
has been the game-tech geek's most essential read for years. (You
remember .plan files, right? They were like blogs, except you didn't
have to spend hours making the HTML look good)
His post mentions
the upcoming floating-point capabilities of the next generation of GPUs and
an SGI paper showing
that this feature is "the last gating factor". In summary:

[...] you will be able to fit more frame rendering power in a single tower case than
Pixar's entire rendering farm. Next year.

Whoah.

(It turns out that the Ask Slashdot item in question was talking more about live,
improvised CGI than big movie render jobs, so my kneejerk reaction was doubly
unfair. It refers to Machinima, the hobby
of creating shorts using existing game engines, such as the surfeit of
gung-ho Quake movies a few years back. You do occasionally get some goodies,
though: Nanoflix's Rendezvous
is cute (and just about worth the pain required to get the Windows-only
engine going). But if you're looking for something visually gorgeous,
check out the new ILM short,
Work In Progress.)

M'colleague Tim just wanted to
do something simple: he wanted to use a UK-configuration PC keyboard with
his G4 Mac. (The reasons are mainly to do with wanting to have the # and " symbols in the right place, rheumatologist as well as liking the MS Natural Keyboard shape, sickness etc.) Easy, find one would have thought. Alas no -
he
had so much trouble with it that he ended up writing his own OS X kernel
extension to handle it. It's called
UKPCKeyboard.

Went out last night with the bot hackers, side effects talked lots about
RDF,
FOAF, trusted vs untrusted code,
Jabber, Jo Walsh scaring the shit
out of me, etc. etc. It all needs to churn in my brain for a while. In the meantime:

• Warchalk Now! has been updated - Dean sent
  me a new version that lays things out differently, has new bits, etc.
• Shortly after I wrote about UKPCKeyboard, I was looking for something else
  entirely when I discovered that Phil had
  already found
  a different solution to the problem. While it involves more work than just
  downloading and running a program, Tim thinks that Phil's solution is probably the
  better one, as it doesn't involve scary kernel modules that may crash your machine
  on the next OS update.
• I didn't make it to Glastonbury, but Radio 1 broadcast a fantastic poppy
  chill-out session in the Blue Room - for the next couple of days you can pick up
  the Real stream
  here.
  It starts off quite hectic as they're changing DJs over, but rapidly mellows out.
  (Cheers, Matt)
• One of the best blogs I've ever seen:
  BLOGGUS CAESARI.
• Oh, and Nottingham City Council says that
  you shag cats.

Aaron Swartz points to WebMake, more which is a lovely little system
for generating static HTML pages from config files: a tiny CMS. It appears
(in the twenty minutes or so that I've been looking at it so far) to have
that wonderful aspect that so many Unix tools have, of hiding a large amount
of power and flexibility in a small radar signature. Like
Blosxom,
it externalises interface complexity in an intelligent way: its target audience
will already be familiar with most of the concepts and be able to deal.
(Its creator,
Justin Mason, is also responsible for
SpamAssassin and Sitescooper, among other things.)

Aaron mentioned it in a blog entry entitled
Bake, Don't Fry,
WebMake being one of a number of tools used to create rapidly-changing
content sites without dynamic (i.e. generated on request) pages. Most blogging
tools work this way. I remember writing a couple of systems like WebMake (though
much less powerful)
for sites like Starship Titanic,
where I wanted non-techies to be able to edit content in text files without them
worrying about templating. (Shim
later wrote a system called Genetic Floor, which was actually quite close
to what WebMake is, but was more web-interface oriented)
I've been wondering about this for my personal site as well: I want consistent templating
around my site pages and my blog, but to be able to change the template. I just
realised that I can keep the master
template in WebMake config files, and have it output MovableType template files
(which can be linked to in the MT config). Sweet!

On the flight home from Amsterdam (if you're wondering where I've been
recently) I was reading
a collection of sci-fi short stories
and
came across Rosetta Stone by
Fred Lerner. The story is quite a simple one (in the classic Asimovian "interesting problem, healing solution" mold) about the discovery of an alien library of Earth books, tablets where nothing is known about the civilisation that collected them. An library scientist is recruited and suggests that analysis of the ordering of the books may be a rich source of information about the society:

"I'll give you an example. Look at the Dewey Decimal System. Melvil Dewey lived in the United States during the latter part of the nineteeth century: a time and place in which there was a widespread belief in the cultural superiority of American Protestant Christians."

[...]

"So the Dewey Decimal Classification was - and for that matter still is - heavily biased toward the relative importance of American literature. And even today it devotes twenty times as much attention to Christianity as to Islam, say, or Hinduism. I'll bet there aren't many libraries in the Muslim world that arrange their books according to Dewey."

"What would they use, then?" asked Jack.

"They might use Sardar's Islamic classification. It's biased too, but at least the bias is a sympathetic one. And a lot of them use the Library of Congress system. It's pragmatic, if a bit unwieldy. A true product of the bureaucratic mind."

"I'd never realised there were so many systems to choose from," said Colonel Rubin.

"Oh, that's just three. I can think of lots more. The Russians used to use a 'Library Bibliographical Classification' which is supposed to be based on the principles of Marxism-Leninism. Probably worked as well as anything else derived from Marxist-Leninist theory. My favourite is the Colon Classification. It was invented by an Indian philosopher called Ranganathan. It's based on the notion that all subjects, if you strip them down to essentials, are made up of five basic elements: personality, matter, energy, space and time. Only a Hindu would have come up with such a scheme. You don't see it much outside India."

While it's basically a lecture dressed up as a story, it's still a useful
reminder of the cultural biases that can innocuously work their way into
classification systems,
something that is especially
relevant these days. (Oh! I nearly forgot to mention that the same SF
collection has Cory's fantastic
Visit
The Sins, about how not to treat ADD.)

Matt pointed us at
a
piece about Jabber bots by Jabber guru
DJ Adams. Most of it talked
about relatively common bot features (as seen in Infobot and the like) but also discussed the use of bots in meetings; to track notes, migraine agenda items and the like. This is very cool.

However, remedy it did remind me of a similar project that's part of
MIT's Oxygen effort.
"The Intelligent Room is a highly interactive environment that uses embedded computation to observe and participate in normal, approved everyday events, such as collaborative meetings." You can see it in action in
this MPEG
movie (warning: 81MB file. 81 freaking megabytes for an eight-minute demo vid! Come on, people, it's not like a 100kbit/s stream wouldn't get the point across!). One point about the demo: while it looks helpful, it also looks dreadfully slow to use, and really gets in the way of a nicely-flowing meeting.

Glenn Fleischman has had
a fantastic
idea. I call it fantastic from experience. (If you read my comments on the page, viagra 40mg you'll see.)

Went to one of the first events of The Clerkenwell
Literary Festival, more namely "Can Computers Tell Stories?" (Pictures are here) Dave kicked it off with a pre-emptive
it's-all-bollocks attack, drugstore saying that trying to delve deep into computer game narrative
is pointless: like it kung-fu and porn films, ed it's just there as an excuse for the
action. He reprised some of the gags from his old Wired piece ("If your spaceship in
Asteroids has such a kick-arse hyperspace feature, why can't you just hyperspace out of
the place where all the rocks are? In Scramble, you get more fuel by bombing fuel dumps -
how does that work?") but the funniest bit was (and I'm paraphrasing here):

"I've got this Edge article here, and it says that people still think computer
games are 'silly'. [Long bit here comparing the definition of 'game' with the
definition of 'silly'] To push the point, the article says, 'Last year, the UK games
industry generated over a billion pounds in revenue - more than the film industry.' And
because of this, games should be taken seriously as an art form. Now, this is a good
point. However, it turns out that last year, the UK crisps & snacks industry earnt
1.9 billion pounds, and I don't hear anyone asking to examine that as an art
form... I mean, I've got some crisps here [produces packet of Doritos] and
we can try discussing their deep emotional meaning, but I think it'd be quite dull.
Also, the UK magazine industry generated 1.2 billion pounds, but most of that was
Edge, which costs a fiver!"

I disagreed with most of Dave's points, but it was great fun. As I later yelled from the
audience, the necessity for plot depends on the type of game: Half-Life's plot,
often given as an example of a wonderfully rich storyline, is actually pretty simple.
Apart from the plot-twist halfway through (which does help a lot, but is just one
feature) it's nothing special. What makes the game so engrossing is the way the plot's
delivered: When plot points happen, they happen to you, not some plot-controlled
version of you in a cut-scene. There are no cut-scenes in Half-Life, no points at
which you lose control of the character, and this is why playing it is a much more
immersive experience than most other plot-driven games. I agree with Dave that, in action
games (including Half-Life) the plot is there merely to contextualise the action;
having a great plot helps, but isn't necessary to make a great game. However, that
doesn't apply to all games: Adventure games depend heavily on plot, and on the successful
integration of that plot with the action of the game. (The classic examples here are
Infocom and LucasArts adventures)

James Wallis, who
was also on the panel, referred to the continuing (if minor) popularity of text
adventures, which have always tended to have the richest storylines of any computer game
genre. Every so often I take another look at interactive fiction (of which more later) I
have a quick look around in wonderment and quickly bog off again having hardly played
anything. IF is the ultra-high-culture of computer games; as
Sir Humphrey said
of Radio 3, "Nobody listens to it, but you're
glad it's there!"

The Clerkenwell talk was just the latest in a continuing series of efforts by
literati and other high-culture mavens to find something in a popular medium that they
could show as transcending its popularity to be something more meaningful, more
artistic. I find the concept somewhat patronising: it endorses the value
of games by highlighting their huge popularity, but then says that the
popular stuff isn't actually the good stuff, the artistically valuable stuff.
It'll celebrate Mario and Lara, put them on the covers of magazines and hold them up as
cultural icons, but then most of these literati will then applaud the likes of Peter
Gabriel for churning out pretentious unplayable shit and lament the trivial attention
spans of fifteen-year olds for not buying what's good for them.

I'm overdoing the righteous bile somewhat, and I'll backtrack more in a minute,
but let me give some context: I went to Game
On at the Barbican a few weeks back, and it was fantastic; a great exhibition
put together by people who obviously know and love computer games. (It's on until the
beginning of September - go!) The only major blights were the "contemporary commissions",
works of game-inpired art by multimedia artists. Most were typical bits of interactive
art, but they were placed in the worst possible environment: they were surrounded by
really good games. As such, they didn't stand a chance. Each seems utterly pointless when
placed next to something that's actually playable and enjoyable. (The only one that held
my attention for more than a few seconds was Ping in the Sound section, which is a
genuinely nice piece. There were also pieces that would have been terrible in any
environment. Space Invaders, but with the words of French philosophers in the place of
aliens? That's an insult.)

Okay, I'm talking about some evil shapeless mass of evil literati here, but
we all know that it's a much greyer area than that. Yesterday's talk was arranged by
the guys from Pilchard Teeth a
Sony-sponsored games-culture fanzine that's fun and intelligent. The average age of
gamers is higher than it used to be, and there are thousands upon thousands of
twenty-to-thirtysomethings who will follow a browse through the Guardian Review or
Salon with
an hour's blast on GTA3.
I like this culture crowd, partially because I tag myself that way too (uh oh) but mainly because
they actually play the games they talk about. When Lara made it onto
the cover
of The Face you could be sure that at least a couple of notable members of staff
were just as much fans of the game as they were of the character.

I think my argument here is more along the lines of that old standard, the merits of
populism vs art, but it's mainly aimed at people who can't accept that the popular
games are actually good. This argument is thrown around with most other media just as
much as it is with games, but I think it's more valid here, mainly because games
consumers seem to be far more clued up with the critical opinion of the games they buy
than film or music audiences. I think this comes from games playing starting in
childhood, not having much money to spend on games, and so making sure to read the games
magazines before wasting my money. I haven't stopped, and I know most others haven't
either. Sure, there are plenty of good games that don't make it big, and there are
occasional exceptions, but it's usually the
case that when games make it big, it's because they're good. You rarely get the
situation that often happens with films, where terrible releases make money purely due
to the names in the credits.

Despite how it sounds, I do want to see artistically-interesting games. Unfortunately,
as with most other media, making something that's artistically fascinating and new and
different is something that the big games distributors (who are as greedy, short-sighted
and downright stupid as movie studios) will gamble on so rarely it hurts, and even then,
only when the developer has a legendary track record. There's no way that Black And
White would have had even half of the sales without a name like Peter Molyneux behind
it. Games development tends to have a big financial barrier to entry, which is what kills
off most interesting projects. There may, however, be a way of offsetting most of the
development costs if you use pre-existing, freely-available technology. This is what I'm
going to talk about next...

(Oh, and something I meant to yell out at the talk but forgot:
Rez is mostly
wank. No, really. It's very pretty, and I can pick it up and have a bit of fun with it,
and I do enjoy it in short bursts, but it's not a particularly great game, in no way is
it a revolution, and the storyline (what there is) is just rambly hand-wavy
pseudo-science wank. So there.)

---