Rachel Donahue: It’s too dangerous to go alone! Take this.

The opening from the game
The Legends of Zelda,

which was the first game to have
a battery pack for save games.

And it's too dangerous to go alone,
so I want everybody to take a sword.

(laughter)

And my presenter mode isn't working,

so it will be fun to see
how out of sync I get.

When I first starting talking about
video game preservation

I used to start off with a--

(man) That laptop
locks a lot, have you--

(Rachel Donahue) Let me see if I can--

(snapping sound)
(woman) Do you want a remote?

(Rachel) I have a remote,
but I need my speech notes.

Sorry, I'm short, what I really need
is a step stool.

So when I first started talking about
video game preservation

I used to have six or seven,
or twelve slides

with reasons why everybody should care
about video game preservation.

But four years later it feels like
it's realistic to skip that part

because video games have been
in a museum, not just a museum,

they've been in the Smithsonian,
and that's exciting.

Just to get one single definition
out of the way for this presentation,

I'm using video games to mean software
installed and played on a computer,

or games played on a dedicated console.

I am not talking about mobile games.

I am not talking about those
little LED slot handhelds,

and for the most part, I am not
talking about browser games.

So it worked on this,
and not on that, that's fascinating.

Let me restart it, hold on.
Sorry.

(woman) Can we have sword fight
while we're waiting?

(Rachel) Yes! This is what happened
when I did this foolish thing,

and I pressed "record presentation,"
and it fouled it all up.

My apologies.

So I'm going to start
by showing you my file system,

and then talking a little bit
about the history

of the Preserving Virtual Worlds
project, otherwise known as...

PVW.

[inaudible] and I record it.

PVW just ended its second phase last year,

ending sort of an era in my life,
and I feel empty.

(laughter)
If you have ever heard me speak before

you should read your email, or nap,
or play with your sword or something

because the next 10 minutes
it's going to be total recap.

If you don't recognize this,
this is what a tombstone looks like

in the not-quite original but
the colorized version of Oregon Trail.

So the first phase of PVW was funded
by that long thing that's up there,

and was a partnership between
the University of Maryland,

the University of Illinois
at Urbana-Champaign,

Stanford University,
and the Rochester Institute of Technology.

Those same partners
were also in the second phase.

The party line is that we were working
to investigate the preservation

of visual games in the context of
libraries, archives and museums.

The primary partners at Maryland
were myself, Kari Kraus,

and Matt Kirschenbaum,
who are all in the front row.

As a project dedicated
to preserving the actual games,

so the actual bits,
what's important about the game,

we looked at a couple of strategies

other than the one of sticking it in
a closet hoping the temperature was okay

and maybe we'd be able to play it later?

The strategies were creating surrogates,
migrating media and format,

emulation and adaptation.

Surrogates---you can see the
[inaudible] up there?---

namely video and screenshots,

they're really a valuable part
of a preservation package

because they help you show
if you got it right, essentially.

But they are lacking the most crucial
aspect of games, which is play,

other than the play button.

Migration, which is taking something
and putting it on something else,

whether that's a new digital format
or a new carrier,

in combination with emulation
is really the most realistic strategy,

we make a media-neutral copy of the game,

we'll be looking at
the Super Nintendo later,

and you play it through an emulator.

It's not quite a lossless process,
but in the long term, 50 or 100 years,

it's probably a lot easier
than maintaining

the original media and hardware.

The last possibility we looked at,
which is maybe a little bit controversial,

is the process of adaptation.

And that's basically just
preservation for remix.

It would be the idea that even if
you're not preserving the original game

in its whole,
or as it was originally intended,

you're preserving awareness of the game

by using its assets
and doing things with it.

And the two examples that we use
in the Preserving Virtual Worlds project

were the Mystery House Taken Over project,

which takes Mystery House,

the first text-adventure game
to include graphics,

and packages all of its assets up,

and teaches people how to make
their own game out of it,

and then fan translations.

There are a lot more games
released in Japan

than are released in The United States.

It is very common for fans to take
a media-neutral copy of the game,

and translate the Japanese into English
so that they can play them.

And then there's copyright, and patents,

and trade secrets, and that stuff.

And I could talk about it for years.

I think I have talked about it for years,
but I'm not going to.

Basically dealing with this is like
playing daemon met your mode,

if you've ever done that,

and if you haven't,
I recommend it...sort of.

It's complicated and I'm not going
to deal with it at all today.

The last thing we did was figure out
how to describe what we were preserving.

Slightly less complicated than
intellectual property, but not much more,

so I'm just going to let you bask in this
fuzzy image up there for a little bit,

and understand that that's what we did.

And so, PVW II.
We're very creative in our names.

I'm going to talk about the newer stuff.

This was funded by IMLS,
versus The Library of Congress,

and in this sequel we were
ostensibly studying

the significant properties of games
and how to figure them out,

because even in less esoteric areas of

digital preservation, like photos,
x-documents,

there's very little anything
that goes beyond theory,

not a lot of practical stuff.

There's probably a reason for that,
but first we tried.

So, what are significant properties?

They're the essence of the thing
you were trying to preserve.

The things that you need to keep,
the characteristics, whatever,

to preserve authenticity.

And authenticity, particularly
these two bolded definitions here.

And that's sort of saying that
we're preserving the thing,

the thing is what it says it is,

so I have a Super Mario cartridge or
a Super Mario media-neutral copy here,

and it is actually the Super Mario game,
and how do I prove it,

what factors do I have
that go into it being Super Mario.

Yes, it's true.

We selected our case set for this
to cover a range of genres,

such as platform games,
educational games, sports games, etc.,

with a preference for game series

or franchises with lots
and lots of titles.

So I've talked about Mario,
I'm going to continue to talk about Mario.

There are dozens of Mario games out there,

and they cover everything
from educational games,

to sports games, to puzzle games,
to party games, to... all of them.

Everybody worked on this fun game
called Typing of the Dead,

which is House of the Dead,
the arcade theme with the...

Except that instead of the...
you have a keyboard, and you have to

type the words that are over
the zombies' heads as they come by.

Here at Maryland we focused on Harpoon,
which is a naval simulation game

created by Larry Bond,
who is a local author,

The Oregon Trail, which was featured
in the headstone earlier,

and Mario.

(woman) Civilization, too.

(Rachel) And Civilization too, slightly,

we were the secondary partner
in Civilization.

So Sid Meier's Civilization, also local
there, up north towards Baltimore.

We looked at some of that.

And to "look at" the significant
properties, we played the games.

We played them a lot.

We recorded ourselves playing them,
we took notes playing them.

In my case I then looked at the videos,

and took notes on the videos
of myself playing them.

And we also interviewed players
and creators of the games

to sort of get a feeling for what
they thought was important about the game,

because what an archivist
thinks about the game,

or what an individual player
thinks about the game,

is probably going to be different from
what the developer thinks about the game.

And that proved to be extremely true.

So, just as an example of figuring out
some significant properties,

one of the games in our case set

is Paper Mario and The Thousand Year Door
for GameCube.

And as a staunch believer
in emulation for preservation,

you might remember I said,

and migration and emulation,
as I described to you earlier,

I used to be really, really kicking
and screaming against the idea

of hardware being important
at all for game preservation,

and this game changed my mind.

And the reason that it changed my mind

is that the controller plays
a really, really important part.

You know you have some games like Rampage,
where you're crushing buildings,

and the Rumble Pak adds
a nice sort of tone to the game,

but in Paper Mario it makes the game hard.

If it weren't for the fact that
you sometimes have to use your left toe

to hit a button on the bottom
of the controller while you're using

all of your fingers on
the top of the controller,

in order to turn into an airplane
and glide the right way,

it would probably be really easy.

If I could map that
to a set of keystrokes,

the game would suddenly
be half as much fun,

because it would be like, oh, ping
I'm doing that again.

So without that controller you lose
what a lot of people have said

is the essence of Mario, which is fun.

And then there are also conventions
that go along with controllers,

and while not necessarily important
to preserving an individual game,

it's very important to preserving
the history of a game system.

So in Mario games you pretty much know
that if you press A you're going to jump.

You can turn the game on, you can press A,
you know you're going to jump,

you know it's probably going to select
your menu items,

and things like that.
In Mario Kart it's the exception,

it's use an item to destroy other players.

B is usually attack, like using
your fire flower in the Mario games,

and down is to go down a tunnel,
which you want to do

if you want to skip all of those
hard levels and warp to level 8.

And then of course your
directional pads and things like that,

they make him move around.

Other properties, which is where
the difference between

players and developers
will sort of come in a little bit.

The data model--this comes into play
particularly with simulation games,

like Harpoon, like Civilization,
like Oregon Trail.

If you talk to a lot of people about
The Oregon Trail, and ask them,

"What do you most remember
about The Oregon Trail?

What do you think is most important
to The Oregon Trail?"

And they're going to say
things like dysentery,

trying to shoot squirrels,

making it to Independence Rock
before July 4th,

(woman) Fording the river!

(Rachel) Fording the river,

having enough axles in your pack,

having enough stuff in general
without weighing down

your oxen so much that they can't move.

Maybe if you're a little bit
more observant you might think

problematic portrayal
of native Americans?,

but you're not going to say
data model, I don't think,

I don't think anybody
thinks about the data model.

But if you talk to the creators
of The Oregon Trail

they are in fact going to say
the data model, the statistics,

those are the most important
part of the game.

And that was also true with Harpoon,

which is a naval warfare simulator,

so you have your radar pinging
for people coming in,

you're targeting your enemies,
you're waiting a very long time

for your missiles to hit your enemies,
and things like that.

And most people are probably the most
interested in going to war, right?

But, again, the creator said,
"No, it's the data model."

And they were really precise
about the data model.

They used every unclassified source
that they were allowed to use.

They had particular calculating wheels
to do trajectories and things like that.

So it's really like, I understand why
the data model is so important to them,

because they put so much effort into it,

whereas they just hired some other guys
to make the interface.

So if the interface... they don't care.

And it was based on a board game,
probably important fact there.

Surface characteristics--which we went
back and forth a lot about in them

in the group, about
whether they mattered or not.

Again this is sort of like
a player-developer dichotomy,

because as a player I care about
how things look and how things sound.

As a developer I might care about
physics algorithms and game engine,

and things like that, but not care
so much about the aesthetics,

because maybe I didn't work
on the aesthetics.

Iconic moves--there are some,
a lot of games, actually,

especially in first-person shooting,

that have things you can do

that were never intended
by the developer to be done,

such as bunny hopping in Quake,

which makes you move faster
than just walking.

And this is why, I don't know
about other people,

but any game that I play I just assume
that jumping makes me faster,

and I jump everywhere that I go.

And these become really an important part
of the player community

at the time that the game was played.

And because a game is a thing
that is made to be played,

it is very much defined
by its player community,

so I would argue
that preserving a game

without having some record
of these types of iconic moves

is doing an injustice
to the history of the game.

And then with DOOM,

and again, this is something that's
much more common

for first-person shooters, although
Civilization has something similar,

the use of demo files and scripts,

we had it here as
using it as an audit tool,

I'll talk about auditing a little later,

But basically this is the fact that
you can create essentially a text file,

recording all of your actions in a game,

and send it to somebody else who has
the exact same version of that game,

and they will be able to play it
and see what you did.

Nintendo ran-- I didn't interview
Nintendo, I wish I did--

but they ran a really great
series of interviews

for the 25th anniversary of Mario,

I recommend looking it up, it's part of
the "Iwata Asks" series on their website,

and the developers were
really, really frequently asked

what made Mario Mario,
what was the essence of the game,

what differentiated it
from other platformers.

And their answer invariably,
every time, was "fun".

And this isn't really the type of

significant property
that we're hoping for.

How do you quantify fun?
How do you describe fun?

Fun is kind of an individual thing,

even though I know very few people
who don't like some Mario game.

And this sort of turns out
to be the reason

that there are very few practical
resources on significant properties.

They're really hard to figure out.

So in the second phase of PVW,

where we had intended to do
more quantitative experiments

towards significant properties,
that was obviously impossible.

So we instead turned to practical ways

that we could help professional
preservers of video games do their jobs.

And I specifically say professional
preservers of video games

because the amateurs have it
completely figured out,

and nothing that follows is something
that I would have been able to do

without the game community.

So this is the exciting part.

You'll probably want to stay awake.

Unless you're an electrical engineer,
in which case I encourage you to sleep,

so I don't embarrass myself.

All of these things happened.

Especially Burnination.

(laughter)

So I talked about auditing
a couple of times.

And this is sort of a small step
in the preservation workflow,

where you're looking at
the file that you have,

and you're making sure that the bitstream
that you have in your repository

is the same as the bitstream
on the original media,

or the bitstream
that your donor gave to you.

It is what it says that it is.

Or down the line, making sure that
the bitstream in your repository

is still the bitstream you originally
put in your repository.

And one thing that we looked at
for doing that was those DOOM

demo files, or lump files,
as they're called, LMP files.

Those are very specific
to a version of DOOM.

Not just like DOOM 1,
but like DOOM 1.13,

so one method of auditing that
yes, this is actually version 1.13 is,

"Does this demo file
which I know is from version 1.13

work in this version of DOOM?"

It's a very small thing that you can do,
but it's something,

and something
is always better than nothing.

When we migrate console games
to the media-neutral format

we're taking code that was originally
burned onto a read-only chip,

and creating a digital file.

We then access that file
through to the software emulator,

instead of the proprietary hardware,

assigned to do nothing
but read the cartridges.

So, how do we know that this ROM

is actually representative
of what's on the cartridge?

How do we know that the emulator
is correctly interpreting the ROM?

And, again, a small thing
that you can do for that,

much like the lump files,

is to take the save game
from an original cartridge

and see if the ROM that you've burned
from the original cartridge,

that you were playing in an emulator,
works in the emulator.

(woman) I feel like I need
a diagram for that one.

(laughter)

(Rachel) I'll talk about the diagram
while I'm doing this.

So, you had your console
and your cartridge,

and you are putting the--

I don't have a game here,
but let's pretend this is the game.

You're putting the original game
on the computer,

and you're playing it on the emulator.

How do you know that it works?

You take the save game,
which is stored in a different part

of the game cartridge and in fact
requires a watch battery to exist.

And you see, will the emulator
open that same game

in the copy of the game
that I made in the emulator?

And then another way to test
the accuracy of the emulator

is to delete that same game,
to play the game in the emulator,

and to take the save game
generated within the emulator,

and inject that back into the cartridge,

and then put that cartridge into
the original console and see if

the original console recognizes
the original save game.

Theory going again if this is playing

in the space that it's originally
playing in, then this emulator is probably

a reasonably accurate facsimile
of how game play occurred.

I hope that makes sense.

And so for the Super Nintendo we did this
with a device called the Retro 2,

which I really wanted to demo for you,
but it refused today.

You can go and look at what it looks like
on the spline on your way out.

This is a really nifty device,
made by a dude in Germany,

game community, amateur,
named Matthias Hullin,

and it allows you to play Super Nintendo,

and Sega Genesis games on your computer,

using the original cartridge
and the original controller,

and you play it through the emulator.

So this is a case of instead of making
a media-neutral copy of the game,

you are emulating the original
hardware for the game,

and trying to preserve only the cartridge,

so you get really a very
authentic experience,

even though you're playing it
through an emulator.

You know that this is the game,
there's no doubt about it.

Coincidentally, as a way of functioning,
this also means that you can copy the ROM

and the saved file from the cartridge,
which is kind of handy.

And so this allows you to do
that whole process

that I originally talked about,
and it's very easy to use.

It uses a USB cable.
There's a little bit of finicky things

you might need to do to update
firmware when you first get it,

but it's fairly well documented.

And then here, this is the config file.

There are a couple of things
that you need to do in config files

to make it work, so here we have
write-protect SRAM.

You want that value to be zero or you will
not be able to inject your save game.

Unless you were just playing a save game
and then you want it to be one

so that it doesn't get overwritten.

checksum and filename you could do,

that provides you another way
to validate your file,

it's just the MD5 that's generated

when you run an MD5 algorithm against it.

But it makes a really ugly file name,
so it's up to you.

And this is what it looks like.

So this is the Retrode,
this is the monitor at my desk,

this is a Super Mario cartridge.

This is the original save game
that was on the Mario Kart,

loaded in the snes9x emulator,

and this is the new save game data
that I created within the emulator,

which I then was able to run
in the original Super Nintendo.

And sometimes,
somewhere in doing this

you do something really terrible

that makes the emulator
refuse to read the cartridge anymore.

Luckily, putting it in the original
console fixes it, so that's happy.

(laughter)

And there are a couple of
command line things you do,

you copy the same file over,
but it's really not that difficult,

and you can do
the command line in any OS,

you can do it in cmd
windows and it's fine,

it's literally one line of command.

And the neat thing about the Retrode

is that it works with SNES
and Genesis natively,

but it also includes plans on the website
for a variety of other formats,

like Game Boy, N64,
Master System and so on.

And he isn't currently selling
the plugin kits themselves,

but the instructions
are pretty easy to follow.

And, moving shortly into
our burnination segment,

I am fairly handy.
I can splice a wire.

I can replace a capacitor.
I can snip a pin.

But if I want to do this same
procedure with NES games,

I cannot desolder a chip.

When I try to desolder a chip
I get third-degree burns!

It's really hard!

Although I should qualify that with,

it's really hard without
the right equipment.

Remember that the whole goal
of this was to come up with

procedures and tools
that would help the average archivist.

The average archivist is not going
to have a resoldering station.

They are like $500.

Maybe they're lucky and they would be able
to pair up with their local hackerspace,

and that's something that
I think everybody should do,

but I tried to do this
with regular tools, and it was hard.

So I'll pass this around, you can see
I have labeled the badness,

and the badness that happened here
is that there's little metal sleeves

in the holes that the pins go into,
and those are really important,

and I lost five of them.

I'm not sure what functions those five do,

but I didn't really want to plug
the chip back in and find out.

But the goal of this was
to remove the CPU,

to replace it in the copy NES board
which then plugs into this,

and I did a good job of that,
so I have a perfectly functional CPU,

so if I get myself another NES board,
and kill the CPU to get it out,

because if I had been able
to cut the pins,

desoldering, no problem,
that'll be fine.

So you can do it if you have two NESs,
not so much if you have one.

And be careful because
the back is a little poke-y.

(laughter)

Also for that, I discovered that NESs

sometimes have a #1
Phillips size screw in them,

sometimes have a #2
Phillips size screw in them,

sometimes have a rusty
#2 Phillips screwdriver head in them.

There were a number of screws
that I was only able to remove

by virtue of Matt having
the perfect tool for the job,

which is this really kinda cruddy,
sorry, screwdriver.

(man) In evidence bag.

(Rachel) Yes, in an evidence bag,
I should have put a label on it.

I tried, there's a good trick where
if you take a wide rubber band

and stick it in the hole
of a stripped screw it will work,

and you'll be able to get it undone.

No.

Rusty screwdriver? 10 seconds.

(laughter)

This is kind of an old one.

(woman) Same set of boards?

(Rachel) This actually comes from our
Deena Larsen collection, right there.

When we first got this,
all of the Mac Classics were tested

to see whether they worked or not.

And being a curious person,
I wanted to see what you do if they don't.

And it turned out for a set of errors,
probably like three of them,

certain things that "happened"
on the screen,

like a sad disk,
or a checkerboard pattern,

the answer was to pull out
the logic board and wash it.

So I put it in the dishwasher,
with no soap, and I washed it.

And the reason that this works
is something else

other than desoldering
that I could actually do,

which is just that the capacitors
start leaking corrosive fluid

on the board, and the corrosive fluid

interferes with all the important
little connections there.

So you could also have scrubbed it
with a toothbrush,

or replaced the capacitors
that were leaking,

and you would be good,
but that's easier.

And so one of the last things that we did

which didn't entirely make it to fruition

was we wanted to create,
I'll call it a strategy guide,

for curators working
with video game collections.

And this is a set of much less exciting

tools that helps them with things
like identifying what they have,

getting the input from creators
and developers like we got

getting our significant
properties' information.

Knowing what tools they need,
after knowing what they have.

And so these were
a series of questionnaires,

and the questionnaires
themselves do exist

and if you would like to see
the work files you can see them.

We were hoping, but you know... time.

Time always exceeds,
or imagination always exceeds time.

So to make this a fully realized website
where it would have guides

helping you how to identify things,
this is where the metadata is

on a cartridge, and so on and so forth.

And even for some things as easy as
a Nintendo cartridge you're like...

"Oh that's easy!

It's one of these green things, right?"

This is Zelda.

Except for when it's one of these

black things that doesn't even say

Nintendo on it, or similarly,

one of these blue things,

that also doesn't say Nintendo on it.

And the same thing, a Super Nintendo
cartridge is generally also

a grey rectangle, except when it's not.

(laughter)

So I'm just going to show a few examples

of some of the questions that we had,
that people could ask.

These are examples of questions that we

did actually ask creators when we were

talking to them about
the games in our case set.

What was your idea? Where did the game
come from, what were the key concepts?

Genre is a really awful
question to ask in gaming.

It's very, very contentious,
and you might need your sword.

But I figure if the creator
wants to assign a genre to it,

they're kind of authoritative, right?

If the game was part of a series,

what makes for continuity
within that series?

So why does a Mario game continue
to be a Mario game,

even if it's about soccer?

When working on a series there's
the tension between

sticking to the...

honoring the original game,
and adapting to technologies.

The original Mario had
a moustache and a cap

because mouths and hair were hard.

So how do you adapt to that,
the change of technology,

where we can have
almost photorealistic sprites?

These are questions for curator--
the curator survey is really

the longest one, because it's really
designed to help them

identify and preserve themselves.

Do you have the hardware necessary
to play the game natively?

This would be a spot where I would like
to eventually have a guide saying,

"Here are these different types
of game cartridges.

Here's what you are playing with."

Same thing for migrating it to
a media-neutral format.

And, just real quick, I mentioned a lot
of Nintendo during this,

but the community has also been
really active, well everywhere,

but the Atari 2600, or Atari VCS
community is really active,

although they can make it difficult
for identifying counterfeits,

but they, in addition to being able
to read cartridges, and write cartridges,

there is the ability
to make new cartridges.

So this is The Empire Strikes Back
for the Atari 2600.

Officially released, copy licensed by
Lucas Film and Parker Brothers.

This, however, is Return of the Jedi,
which was never released

for the Atari 2600 in its lifetime,

but somebody found the ROM
"in the wild", as they say.

So I worked with a couple of people
who made a bunch of cartridges,

that have the same case as that.

We found what may have been
the original art,

and worked on upping the quality of it
to actually be okay for print,

made an instruction manual,
packed it all together,

and sold them, for 35 bucks.

You want to compare that.

(woman) Can we get an order form for that?
(Laughter)

(Rachel) No more, sorry.

And then of course in order to do this
audit procedure that I discussed,

you have to know if it has the same game,

and how that same game works,
so questions about that.

And then there are player questions,

and this first question
really we ask everyone.

What is the core heart of the game?

What is the most meaningful
part of the game to you?

We have similar questions
to the developers as well.

What contributes to the continuity
of franchise?

If a game is multiplayer, how important
is the multiplayer aspect of it?

Have you ever played it that way?

What new things has the game
introduced to you?

Because maybe if it's introducing
a new technology,

that's an important part of history,

like Mystery House was the first
game to have graphics.

Whether or not it's a good game,
that makes it a really keystone game,

and something that would be
important to have in a game history.

(man) First piece of
interactive fiction graphics.

Oh, sorry, first piece of
interactive fiction graphics,

not first game, first text adventure
in graphic fiction.

- (woman) Yes.
- (man) Important distinction.

(Rachel) Yes, my bad.

(woman) Although what counts
is graphics in that case is--

(laughter)

(Rachel) It's so wonderful, ascii art.
(laughter)

And then knowing that migration
is never a lossless process,

what features would you
be willing to sacrifice

if you had to sacrifice something?

And to sort of concretize this,
I'm going to talk a little bit about

how I took these questions
and modified them for a specific game

for my dissertation research.

And that game is the game Glitch,
by Tiny Speck, which is no more.

I said I was mostly not going
to talk about browser games,

this was a flash-based game,
which is kind of why it's no more.

And it was a massively online multiplayer,
2D, side-scrolling, no fighting,

really imaginative game, started by
the people who started Flickr.

And it closed in December 2012,
and in a reaction to that

I looked at all of these questions that
we had been creating,

and thought how do I make these for Glitch

so that we can remember
everything about Glitch.

So these are some questions
specifically for Glitch creators.

Glitch had an open API,
very well-documented open API,

that the players utilized so well
in creating their own tools,

that oftentimes the developers would then

incorporate the player-created
API functionality back into the game,

which was great on the one hand,
but on the other hand,

"Well, now my tool doesn't matter."

An example of this is just
there are certain...

storefronts we'll call them,

they're character vendors,
characters that sell things,

and only an animal vendor
sells animal goods,

and it can be kind of hard to search
the map trying to find where they are,

so there was something
that you could just click on,

on a separate web page,
and it would say,

"There's an animal vendor
three streets away from you!

Click here to create a path to it."

And they actually built that
right into the built-in encyclopedia

within the game so that you could do that.

Perhaps the most contentious portion
of the game came in the form of nerfs.

A nerf is when you take a previously
very powerful game mechanic,

and make it a lot harder.

And typically, as a developer you do this
for game balance, right?

Because if... one of the first was mining.

Mining was a way to build up a lot
of money really, really, really quickly,

totally out of proportion with
any other type of ingame action.

And what this leads to is players who
are going to do nothing but mine all day,

and that's not really fun,
even if that's what they want to do.

And it sort of punishes players
who hate mining--I hated mining.

I liked to cook, that was my thing.

So you want to balance that so that
players of all types can play it,

and there's lots of discussion about this,
because the developers were very

in touch with the community,
and would actually discuss the reasons

behind a nerf, and things like that,
and sometimes even warn us ahead of time,

so that we could deal with that,
and obviously making that decision

is kind of important to
the development of the game,

so probably something significant there.

I mentioned that there was no fighting,

a distinguishing feature is that
there's no player-player conflict,

you do do things like eat meat,
but you get meat by petting a piggie,

you don't kill anything in the game.

It's very relaxing and peaceful.

The closest you get is
there are some races.

That's the only built-in
competitive element.

And those are all without
the player vs. player conflict,

or the raids, or the whatever.

So how did this affect
the development process?

Why did you do it?
Why was it important to you?

Player questions? We mentioned nerfs,
there were lots of other changes.

Glitch was first released
in a very small alpha release,

and it looked really funny then.

Then they had a very closed beta,
which was invite-only.

Then they had a slightly more public beta.

Then they launched publicly as,
"We're a game! You can sign up!

Have fun! This is great! Yay!"

and then they said, "Crack, we really
screwed up," and it went back to beta.

And most of us applauded that as being
a really brave, great decision,

to go back and rework
some of the core development

to be more attractive to a wider audience,
but in the end maybe it wasn't great.

So the player question for that is,

"If you were present during these
transitions, how did they affect you?"

"Did knowing that the game
was going to launch or unlaunch

affect your ingame behavior?"
things like that.

I mentioned API tools, but there
were also a lot of player-run activities

within the game, because Glitch was
what's considered a sandbox game,

which means you sort of
make your own fun,

the most popular example of this
is probably Minecraft,

where you're making your own fun.

And so there was a sandbox group,
which created player quests,

ghost tours, because there's an ingame
quest where you have to find ghosts,

and it's really hard,
and things like that,

so what did you participate,
and there I'd like to see

what player-created games
are the most popular,

because those are probably some
of the more important to talk about.

And then because the game was closing,

and we had six weeks', I think, notice,

what did people personally do
to help remember Glitch?

And lots of things were done.
There were lots of really great websites.

Down to people creating a way so that
you couldn't play the game,

but you could have sort of an interactive
video of your home street,

so you have a street that's your own,
that you can plant on and stuff like that,

do whatever you want,
that other players can visit,

and this made it so you could run, and you
could jump through your home street,

just like you could otherwise.

There was a player who in her Etsy shop

started crafting dolls
of people's butlers,

which is a character on your home street
that could greet other people.

They were mix and match,

there were thousands of different ways
you could make your butler.

And they were really cute, we have one.

Just like lots of fan art, and all
the things you might expect

to remember a favorite place,
or a favorite club,

or a favorite group activity,
anything like that,

because the game
was defined by the players,

the players needed to remember playing,

and for me as a game-preserver,
sort of a life preserver,

knowing what people themselves
were doing to preserve

is a way to help preserve
that feeling of the players,

because having the world itself
is meaningless in an MOL,

you can wander around and look, right?

So that's about it for the Glitch stuff.

I'm always really bad
at writing concluding lines,

so this is my last slide.

Questions?

(man) Thank you, Rachel.
(applause)

Questions, comments, for Rachel?

(woman) Oh, I have so many questions,
but I'll just limit them to two, for now,

which is, you mentioned finding some of
these games "in the wild,"

and I'm kind of interested
in where all the materials

for the case sets came from,

because what happens when you burn out
your Super Nintendo board,

and then the last one in the world
costs however much,

I'm just wondering how you went about--

(Rachel) None of the games in the
case sets were particularly rare,

nor were they for
a particularly finicky hardware,

so in terms of MITH, most of what we got
was sourced either from eBay,

or in the cases of console games,
a website called JJ Games,

or Jay's, yeah JJ Games,
there's a couple,

basically second-hand shops,

it's still very easy to get them,
and Nintendo hardware is extremely robust.

The biggest problem is with the original

toaster-style Nintendo
entertainment system,

but the lovely fan community
came out with a fix for that,

which is to replace the 72-pin connector,
which unlike desoldering a CPU,

is extraordinarily easy,
all you need to be able to do

is use a screwdriver,
and you could do it.

So in that case, that was fine.

"In the wild" refers more to games
that were never released,

games that were
in some phase of development,

people had maybe heard rumors about them,

or maybe there was
an extremely limited release,

or they were a recall,
or something like that,

and somebody just finds them.

And the reason somebody can just find them

is because, especially prior to the 90s,

when [inaudible] were relatively small,

whoever wanted to took
the development materials home.

So these things can end up
in somebody's garage sale,

and that's basically
what happened in this case,

was somebody found the development copy
of the cartridge at a yard sale,

and was like, "Oh, isn't this neat,

I would like to share this
with the community," and they did.

(woman) And then just, well, also we're
very dependent about the...

Once these things do
get archived and preserved,

I'm thinking about ways of displaying them
or having someone interact with them

in the setting, I just don't know where...

I'm wondering about where you see this
project ending up.

(Rachel) Did you see
The Art of Games exhibit?

(woman) No, I haven't.

(Rachel) So they did a really good job,
I think, other than having it all

on projectors which...
Pac Man on a projector is a little weird.

But, they used, wherever possible,
the original hardware,

which in some case involved
a little bit of watchful engineering,

and they hid it where
nobody could get at it,

and just had the original
controller coming out,

and they had it set on a timer,

so that the power basically
shut off after a little bit,

you could only play for a little while.

That's one way to do it.

The Strong Museum of Play,
in Rochester, New York

rotates what type of video game
materials that they have out,

and most of what
they have out for actual play

is standup arcade machines and stuff,

but they very much believe in
having the original experience.

Someplace like the Computer History
Museum, in Sunnyvale, California,

they're much more traditionally
museum-oriented,

and in a lot of cases care more about
how the artifact itself looks

than whether it's playable,

and that's sort of completely
the opposite end of the scale.

The difficulty right now with offering
something via emulation is,

you know I avoided the whole
intellectual property thing,

it's completely illegal.

So we had, from 2006-2009
there was an exemption

that made it not completely illegal
for educational institutions,

but it is currently
still completely illegal,

so if you're at some
small non-profit organization

getting government funding,
not going to do it for the public.

Kari?

(Kari) Okay, so you made some gestures
toward tools for the average archivist,

but the more I watch you,
and the more I work in this area,

the more convinced I am that we need
a kind of MacGyveresque approach

to digital preservation in this area,
and really at large,

not just video games and 3D virtual worlds.

Chewing gum and duct tape save bits,
and so much of what we see

is a kind of approach that involves

tracking down rare, obsolete cables,

and hooking up modern PCs
with obsolete hardware,

and having to troubleshoot in
really creative kinds of ways,

and even The Art of Video Game
exhibition, as you pointed out,

involved some electrical engineering,
so I'm wondering if you have any thoughts,

and let's set aside the IP issues,

any thoughts on what a kind of

digital preservation course
centered on hardware might look like?,

what might a module be like?,
what might a lesson plan be like?

(Rachel) So just preservation specifically
focused on hardware.

(Kari) Specifically focussed
on hardware, yeah.

(woman) And where can we take one?
(laughter)

(both women talk at same time)
(Kari) Yeah, go ahead, Rachel.

(Rachel) So that's difficult,
in a sense, for me,

because I work in a pretty esoteric area,

now there are like 20 of us
in the Western world

doing this type of work,
specifically in archives and museums,

so what I'd teach people how to do that?
I would probably have one day

where I talked about the things
that I went through

to try and understand how to do it.

One thing that I would definitely do
is teach them how to read

circuit diagrams, and schematics.

(man) This is coming up for [inaudible],

right outside, waiting,
it's the next step.

Well, yeah, and we have those,
what we don't have is,

how do you disengage at the tape?

And basic soldering and wire splicing,
these are things that are really easy,

you need a $10 soldering iron
and a $2 thing of solder.

(woman) And a $10 glove to protect
your finger next time.

(Kari) Yeah, there we go.
(laughter)

(Rachel) That might have been
when I got frustrated

and turned it up to 50 watts.
(laughter)

(woman) We'll buy you a glove.

(Rachel) Just some really
basic stuff of that nature

so that if you have a Mac Classic
you can replace that capacitor,

and I'm not just saying
that's easy,

if you have a certain amount
of hand-eye coordination.

If you can build with legos,
you can replace a capacitor,

It's not that hard.

So that sort of thing that people
tend to be really scared of,

they're like, "Oh, the cord is broken,

I have an Atari 7800, I would love
to play it, but the cord is broken",

which happens all the time
with that machine.

You just splice it.

Those are core skills that
a lot of archivists don't have

that would be really helpful,

along with being able to read
the circuit diagrams.

And then I would probably
talk just a little bit

about some of the less
esoteric media, like floppy disks.

All archivists deal with floppy disks,
and knowing how to do so is good.

Knowing how to image a hard drive,
probably also good,

and what your options are for that,
vs. having a crazy dedicated machine,

like the FRED, which you should all
look up, acronym FRED,

or a cable--I have a cable
that reads hard drives, it's much easier,

and cheaper, $20 vs. $5,000.

(Kari) Yeah, see I like this, but what
I think what you just painted a picture of

is hardware in the context of archives,

it's not about display items or artifacts.

It is not just about
access devices for games,

but this more media-archaeological
approach that is about

actually dissecting a plane
and hacking with them somewhat.

(woman) Have you heard any headway
in your dialogues with the creators,

have you gotten any sense that they are

interested in taking over this function?

Not at all? Okay.

(Rachel) There is sort of a cycle
that occurs with all new media,

starting from the book,

where this new thing,
this new-fangled thing, comes out,

and very few people with a lot of money
are really, really interested in it,

and everybody else is like, "Whatever."

It happened with books,

it even more closely in parallel
happened with movies,

and eventually people are like,
"Oh, this is pretty cool.

This is a great way to spend my time,
but it's not something that my kids

should be learning in school."

and then eventually they're going,

"Maybe this is something my kids
should be learning in school."

And then finally at that point
higher-education institutions

and cultural heritage institutions
like museums and libraries start to think,

"Maybe we should be worrying about this.
People are going to need access

to this old stuff if they want
to learn about it in school."

And then 50 years after that,
then the media creators

start to be interested in it,
and I don't really blame them,

because archives are always,
always, always a cost center,

and you're talking about
commercial enterprises

that are interested in the bottom line

and in video games they're not just
interested in the bottom line,

they're interested in, "I need to get this
done in time for Christmas."

Which means I'm going to work everybody
like 16 hours a day until October,

and they just don't have the time
to really deal with this,

nor do they have the understanding of it.

They're just starting to see,

with things like Playstation Network
and Xbox Live,

and virtual console and stuff,

the value in resurrecting
some of their older titles.

In almost all cases they've
had to completely [inaudible] them,

because they didn't have the old assets,

or they didn't know
how to convert the old assets.

So it really is very much how to
convince the--what is the value for them?

The value obviously is not monetary.

There is no way to prove
that it's monetary.

So it's either proving that,
or alternately proving to them that

archivists and curators are not the enemy,
and that the things that you want to do

with their materials are not going to
infringe on their intellectual property

in a way that will cost them money,
but instead would be like free marketing.

(woman) So, can I ask a
sideways question to this?

Which is so much in the last two years
has been made about teaching scholars

how to build games,
or to consider gaming

as an avenue towards student education,
K-12 education, things like that.

Where does that intersect with
the archival preservation, or whatever,

because we see so many of those games
being built in some scale or another

that may be good or bad--
What do you recommend,

or what do you address in terms of
what these people need to know

before they even start investing
their time in something that seems

so cost intensive, when you can't
necessarily guarantee the back side?

(Rachel) This again sort of comes against

who you think is going to
use them the longer,

and I would go beyond K-12 and say

undergraduate and graduate programs
of game design, which now exist,

and this is where I always make
the parallel to art history.

If you're taking art courses, unless
you're me and purposefully avoid them,

you're going to take art history courses,

you're not just going to learn
how to do the art,

and part of the reason we can have
great art history courses,

is that we have great museums
dedicated to art,

where we have the history,
where we have the concept sketches

and things like that that went into it,
where they're available.

And artists have really big egos
and keep that stuff a lot of the time,

and like to share it when they're dead.

But we don't have that for games,
and so I look at it like

education to me is a really big aspect.

Don't we want to train
the game developers of tomorrow

on the lessons learned in the past?

Wouldn't it be really interesting to see
what are the developments

that went into being able to have
photorealistic 3D environments?

What are the elements that went into
having really accurate bounce physics?

Things like that, because knowing how
one person had their Eureka moment

and how the development evolved

can help you do your own evolution
and development somewhere down the line,

just like learning stippling, or whatever,
crosshatching, or whatever does

when you're learning drawing.

Matt.

(Matt) I wonder what the sword's for?
(laughter)

(man) He always asks
the hardest questions.

(laughter)

(Rachel) It's dangerous to go alone!
(laughter)

When you first start playing
Legends of Zelda, Link talks to an old man

who says it's dangerous to go alone,
take this, and that's the sprite

for the sword that you could use here.

(man) Is this also a sort of living
demonstration of significant properties?

(Rachel) Sure.
(laughter)

It is also a living demonstration,
although I made it myself,

of what I allude to on twitter
called feelies which is particularly

Infograme, InfoComm
with their interactive fiction,

and in the 1980s often packed
other stuff in with their games,

and my favorite example of this is
Pocket Link was packed in within

The Hitchhiker's Guide to the Galaxy,

so my talk comes
you only trust it with the talk,

which you all knew you were getting.

But also an item
to protect you from danger.

(laughter)

(man) So I expect this to be
the back of your business card.

(laughter)
(Rachel) But I have a stamp [inaudible].

(laughter)
Peter.

(Peter) Could you talk a little bit more
about conflict-free gaming?

That was an issue that you reached
with Glitch, just in light of all

of the questions that occur to you,
I'm wondering...

The history of gaming, has it become
more conflict-intensive or less?

For the archivist, is it harder,
does it raise different issues

that the degree of conflict
that is built in,

and I was also thinking about
your comment about Mario,

and how people like Mario games
because they're fun.

I'm wondering whether there's
a correlation between

conflict or lack of conflict
somehow, and fun.

(Rachel) I would say that games have
neither become more conflictual or less.

The graphics depicting the conflict
have become better,

the mechanics of the conflict
have become better.

Pong, what is that? That's a competition.
That's conflict, right?

(man) Spacewar!
(Rachel) Spacewar!, competition.

Oregon Trail and [inaudible],
you hunted.

So it's always been there, and game
balance doesn't create real balance,

I'm just going to say that right now,
that's ridiculous.

I think that the--
that conflict-free gaming,

other than puzzle gaming, has been
greatly improved by modern technologies.

Because the thing with
conflict-free gaming

is you have to have something else, right?

Like most games,
they are essentially conflict-driven,

whether it's a game of chess
or a game of Call of Duty.

You're driven by conflict,
you're driven by competition.

With a conflict-free game
what do you do, right?

And with some things like Minecraft
or Legos, they're physical-analog,

it's kind of ob-- you build, right?

But with more complexish things,
you really need other people.

Other people are what helps make
a conflict-free game fun.

And that was certainly
the case with Glitch.

I don't know if that really answers
your question or not?

(man) I was just wondering, it seemed
to me that your interests,

you're interested in Glitch,
among these millions of other things,

and you raise the issue of conflict,

and it actually just seemed to me
that gaming and thinking about

the preserving of games could be
an interesting way of thinking about

conflict over time, and actually
cultivating skills of evading conflict

or redirect-- and I take your point too
that conflict, what is conflict?,

is a very complicated
question to begin with.

(Rachel) Though I have to be honest

and say I had no deep thoughts
when it came to that.

I said, "Oh my God, my favorite game
is closing, what am I going to do?"

It was nothing about
what example of whatever,

which was to me it's a great
example of a startup,

it's a great example of trying to do
something really ambitious with Flash,

it's a great example of
a sandbox game,

a great example of 2D art in a world
overwhelmed with 3D,

but mostly I just loved it,
that's why I'm using it,

and had access to the developers,
because they're so from the community.

(man) I'm wondering if you think whether
walkthroughs and game guides

have a role in preservation, for example
with video games reach stories,

how RPGs can be.

I've been through several walkthroughs
and some of them tend to be

very technical,
focusing on different things,

some others will tell the story, and
they will have [inaudible]

telling you what is happening
in the game until you can see it,

and definitely that is a form
of preservation in a way,

especially if you at some point
lose access to the game,

because [inaudible] disappears.

(Rachel) Sure. So there is a concept
in archives called context information,

and that's the things that build
the context around the game,

that help provide a meaning for that game.

And one of those things is walkthroughs
and strategy guides,

so in some cases, in PVW I,
we would find old walkthroughs

from bulletin board systems,
that people were posting,

and actually when we submitted the game

to the dark archive that
it sits in at Stanford,

we included that type of information,

and also the Strong Museum of Play
in Rochester we talked about earlier

has a complete collection of
the print guides, for everything.

So yes, absolutely, they're very
important context information.

(man) With that, let's give Rachel a hand.
(applause)