0:07

It's really hard for people to

0:10

fathom that all of these forces,

0:12

political, judicial, legislative, funding, are all

0:14

going against a tradition that built

0:17

America out of a set of

0:19

immigrants and brought us together to

0:21

be able to know how to work as a populace.

0:24

And all of that is in, well,

0:27

some of it is going to be

0:29

judged again by an appellate group in

0:31

lower Manhattan, and we'll see

0:33

what it is they decide towards the fate

0:35

of libraries as we know them. It's

0:38

a slippery slope into a licensed-only

0:40

culture in which libraries cannot exist

0:43

and mission this challenge. I do agree. A

0:45

lot of people say, oh, this is just

0:47

about the Internet Archives Open Libraries program. It's

0:50

a narrow focus. I don't think that's true.

0:52

It's about libraries entering and being

0:54

supported in the digital space. And

0:57

through the use of either licensing or

1:00

litigation, that's a one-two punch, really

1:02

will threaten the library mission now and

1:05

in the future. Hello,

1:11

and welcome to Why Is This Happening with me, your host, Chris

1:13

Hayes. There

1:20

is this news recently that

1:22

I saw that just brought

1:25

me up short, put my heart in my throat,

1:27

which was nothing with

1:30

life or death stakes, just that the mtvnews.com

1:33

archives of 20 years

1:35

had disappeared off the

1:37

Internet. It had been removed by

1:40

MTV's parent company Paramount for reasons that

1:42

were a little unclear, I think some

1:45

financial decision or maybe site reorganization. But

1:48

for a freelance, a one-time freelance writer like

1:50

myself, the notion of your work being on

1:52

the Internet and then being taken off is

1:54

really unnerving. I remember back when I started

1:56

freelancing, I was writing for the Chicago Reel.

1:58

which didn't really even have an online version.

2:00

It was just paper. I would go and

2:03

pick up the paper version and I got

2:05

myself this kind of like leather book that

2:08

you put the, you know, as like the cellophane and

2:10

you put the clippings in. And this was a big

2:12

ritual for me when I published a new piece. Go

2:14

get the Chicago Reader, bring it home, clip it out,

2:16

put it in my clips book. And

2:18

then at a certain point, you know, everything was online. And

2:21

there was some notion that we had back then that the

2:23

internet is forever, people would say, but the internet is not

2:25

forever. So one

2:28

of the only ways the internet is forever

2:30

is an institution known as the Internet Archive,

2:32

which does a bunch of stuff, which we'll

2:34

talk about in this hour, but

2:37

runs the Wayback Machine you might

2:39

have seen, which basically archives the

2:41

internet in real time. And so

2:43

the Wayback Machine presumably will have

2:45

the archives of mtvnews.com. And

2:47

the Internet Archive is fascinating and important,

2:49

I think, because it's one

2:52

of two pillars of what I

2:54

would call the old pre-commercial and

2:56

now non-commercial internet. Wikipedia and the

2:58

Internet Archive, neither of

3:00

which are for-profit enterprises, they

3:02

embody the notion of a

3:05

non-commercial, collaborative, civic vision of

3:07

the internet as a storehouse

3:09

of free and publicly accessible

3:11

knowledge that people collaborate on.

3:14

And that pre-commercial vision of the Internet Archive,

3:16

which runs the Wayback Machine, but also does

3:18

a bunch of other stuff we'll talk about,

3:21

has come into conflict with commercial

3:23

visions of knowledge and intellectual

3:26

property in the form of

3:28

a lawsuit, which has been brought by a bunch of

3:30

publishers, Hachette, HarperCollins, John Wiley

3:32

& Sons, Peg & Random House, who is

3:34

gonna publish my next book, I should note,

3:36

Full Disclosure. They sued the

3:38

Internet Archives in June of 2020, because

3:41

Internet Archive digitizes physical books and

3:44

makes them available to people to

3:47

loan. We'll talk about how that loaning works.

3:49

And the publisher said, look, you can't do

3:52

this. This is, they said it was willful

3:54

digital piracy on an industrial scale with the

3:56

program that the Internet Archive runs called the

3:58

Open Library. And the publisher... included

4:00

in their complaint that the Internet Archives does not

4:02

have licensing agreements with them, nor does it pay

4:04

its authors. So a federal

4:06

district judge found in favor of the publishers,

4:09

in summary judgment ruling last year, and said

4:11

that the use of copyrighted material that merely

4:13

repackages or republishes the original is unlikely to

4:15

be deemed fair use. Fair use, of course,

4:17

is the carve-out for copyrighted material. And the

4:20

Internet Archive has built a collection of more

4:22

than 3 million books. It's purchased them in

4:24

print or received donations, and patrons of the

4:26

digital library could borrow up to 10 books

4:29

simultaneously for two weeks each, just the way

4:31

you borrow a physical book. And

4:33

the Internet Archive says, look, this

4:35

free ebook lending program is fair

4:37

use, and it's a controlled

4:40

digital lending practice. The

4:42

publishers say, no, this thing you're doing,

4:44

controlled digital lending is a frontal assault,

4:46

I'm quoting here, on the foundational copyright

4:49

principle that right-shoulders exclusively control the terms

4:51

of sale for every different format of

4:53

their work. A principle has spawned the

4:56

broad diversity in formats of books, movies,

4:58

television music the consumers enjoy today. Okay.

5:02

So that's part of the topic of today. We're

5:04

going to talk to some folks from the Internet

5:06

Archive, but it's a deeper conversation about the commercial

5:08

versus non-commercial Internet, about what fair use is, about

5:10

the notion of a public intellectual commons. And we

5:12

should note that this conversation was recorded on June

5:14

26, 2024, which was actually just

5:17

a few days before there was

5:19

the oral arguments in the Internet Archive's

5:21

appeals before three judge panels to defend

5:23

its open library control digital lending practice.

5:27

Again, I want to be sort of bending

5:29

over backwards to be fair here, because we're

5:31

only listening to one side of the litigants

5:33

defendants actually in this lawsuit. So just to

5:35

lay this out, but I'm really, really fascinated

5:37

and obsessed with the notion of reviving,

5:40

expanding non-commercial models for

5:43

digital life. So it was really

5:45

enlightening and energizing to have this conversation about

5:47

a case study in how that's playing out

5:50

right now with Brewster Kale, who's a digital

5:52

librarian at the Internet Archive, intimately

5:54

involved in the creation of the Wayback Machine, a

5:56

founder of the enterprise, and Kyle

5:58

Courtney, who is a lawyer librarian. and

6:00

director of copyright information policy for Harvard

6:02

Library. He's co-founder of Library Futures, which

6:05

aims to empower the digital future for

6:07

America's libraries. Here's our conversation. ["The

6:16

Internet Archive"] I

6:18

mean, you, this is, you're one of

6:21

the founders, creators, right? Yeah, I'm the founder of the

6:23

Internet Archive. You're the guy, right. Yeah, we started in

6:25

1996, and it took us that long because

6:29

we had to get a lot of the rest

6:31

of the Internet. Like, it was the ARPANET when

6:33

I got going, so we had to actually get

6:35

the Internet going. We had to have computers and

6:37

search engines. You get the publishers online. But once

6:40

I helped, I mean, I didn't do all of

6:42

that, but I helped go and get, you know,

6:44

New York Times' first website up, the first Wall

6:46

Street Journal website up, then we could build the

6:48

library. So in 1996, we

6:50

turned to build the library. And

6:53

it's been great. We've been archiving

6:55

web pages, but also television, books,

6:57

music, video, all your old

6:59

flash games, all of that stuff. So just

7:01

tell me about yourself, Percy, like how you

7:03

found your way to this work. You're obviously

7:05

tech savvy, early adopter kind of

7:08

person. Like, what was your first

7:10

connection to the Internet? Oh, I'm a

7:12

geek. So I started to have

7:15

an MIT, but I'm sort of

7:17

in the tail end of the hippies, right? The sort of

7:19

like, what can you do with your life that would be

7:21

a good thing to do? And the idea

7:23

of building the digital library of Alexandria seemed

7:25

like a great idea in 1980. We've

7:28

been promising the library of Congress on your desktop

7:30

for 35 years before that, starting

7:34

with Vannevar Bush and Ted Nelson. We've been

7:36

promising this. It's like, how hard could

7:38

it be? And technically, it turns out to

7:40

not be that hard, but boy, it's

7:43

been a long road. We're not even there yet.

7:46

I like this idea, the Alexandria library of the

7:48

Internet. Like, where did the idea come from and

7:50

how do you start actually doing it? It

7:53

started with the technical conversation. We

7:55

knew that digital publishing was

7:57

going to happen. I mean, this is before the Mac

8:00

and... So it's like, we knew all this stuff was

8:02

gonna go online. What years are those? We're talking in

8:04

1980. Okay. So it goes back that

8:06

far that you were thinking about this. Yeah, I'm kind

8:08

of old. But it

8:10

wasn't new to me. The idea

8:12

of building access to

8:14

all the information in the world, having that

8:17

all at your fingertips has been

8:19

in the air since the Library of Alexandria.

8:21

The idea of having all the published works

8:23

of humankind available to you was the promise

8:25

when you walked into a library. If they're

8:27

gonna slog your bones, when I was growing

8:29

up, you slogged your bones in the library,

8:31

they said, we have everything. And if we

8:33

don't have it here, we'll get it for

8:36

you from interlibrary loan. And

8:38

that was how the American

8:40

sort of Carnegie Library ethos

8:43

of America, and it had so important

8:45

for educating people, for getting people ready for

8:47

elections. A lot of it, we came

8:49

out of the progressive era to

8:52

go and try to get it so that when

8:54

people are doing direct elections, they're educated citizenry. All

8:57

of that momentum was the world that I

8:59

grew up in. And I

9:01

had a new trick, I knew technology.

9:03

And I also knew how long it

9:05

would be before we would be able

9:08

to have all of the books in

9:10

the Library of Congress online. And then

9:12

all the movies online, and all lectures

9:14

at Harvard and anywhere else online. And

9:16

you could just plot it out. And

9:18

we just knew how, so it was

9:20

conversations with Richard Feynman and Danny Hillis,

9:22

Marvin Minsky, Steven Wolfram, we just charted

9:24

it out. And it was like, this

9:27

isn't that hard, let's just go build

9:29

it. And so that is, I

9:31

don't know, some of the early ethos of

9:33

the internet. It wasn't uncommon, but

9:35

it also sort of led some

9:38

of those things. By 1996, it's

9:40

still before Wikipedia, it was before

9:42

Google, it was way before YouTube.

9:45

We knew that we needed to go and

9:47

take the wonderful things that people were going

9:49

and putting onto their websites. They were offering

9:52

and sharing and putting on unbelievable things. And

9:54

it was all getting lost. The average life

9:56

of a webpage is only a hundred days

10:00

changed or deleted. And

10:02

in the old days, publishers would go and publish

10:04

things and they'd be bought by libraries and put

10:06

in many places. So even when the publisher goes

10:08

away, because they always do, there's

10:10

libraries that will be there, that

10:12

will store and keep the real

10:15

precious things, which is the works

10:17

of humankind alive. And that's

10:19

what we went to try to build

10:21

with the Wayback Machine and the Internet

10:23

Archive. And that incorporated like officially

10:25

in 1996? Yes. And

10:28

how- That's when we started crawling the web with the Smithsonian

10:30

Institution. Yeah, so how did you build it institutionally? Like

10:32

what was it, a 501c3? How

10:34

did it work as an institution? Non-profit

10:36

from the very beginning, because this is

10:39

everybody else's stuff, right? We're

10:41

custodians of everyone else's materials,

10:43

starting with the presidential election

10:45

of 1996. You

10:48

should see these old websites, man. They're

10:50

really pretty crafty, but they're like the

10:52

first bumper stickers in the early 1950s

10:55

is the way the Smithsonian saw it.

10:57

And we saw it as the beginning

11:00

of the library of not just the

11:02

elite that can publish books, but

11:04

everyone's words, everyone's blogs. I

11:06

know there weren't blogs yet,

11:08

but everybody's- Web diaries already.

11:11

Web diaries, all of these things that

11:13

were happening at that time as people

11:15

were sharing up a storm because they

11:18

could. So walk me through technically

11:20

what the project was. Like you said that

11:22

you figured it out technically before you started

11:24

doing it as an institution. Before

11:26

we get to that, were you raising money? Like how does it, you

11:29

have donors, you have members, did the Smithsonian pay you

11:31

for the service? Like how did it work financially? No,

11:34

they didn't pay us. I got lucky because along

11:36

this route, I was helping building the publishing systems

11:38

to get the publishers on board so they could

11:41

start making money by publishing on the net. And

11:44

that was bought by AOL. It was

11:46

called Waze. It was the system before the World

11:48

Wide Web. So that's, I guess why I'm probably

11:50

in the internet hall of fame is I did

11:52

the earliest systems for publishers. And

11:54

then that was put on AOL

11:56

to help publishers. Then once that

11:58

was done And I laughed, then

12:00

the idea was like, what's

12:03

the most ephemeral of media? And

12:05

we said, let's do the web. And

12:07

so he's made little crawlers and

12:09

these crawlers are little robots. They're

12:11

programs that basically click every link

12:13

on every webpage. And it

12:16

just downloads the page and remembers it and

12:18

then makes the file out of it and then finds the links

12:20

and adds it to the queue. And then it

12:22

clicks on that one and then finds the links on it and

12:25

it's a queue. And it just goes round and

12:27

round and round and round. And we collected

12:29

a copy of the web

12:31

every two months. A snapshot, snapshot,

12:33

snapshot, snapshot starting then. It's

12:36

gotten a much more complicated now, but we're

12:38

collecting about 300 million

12:41

web pages every day, about a billion URLs

12:43

every day now. I have to say back

12:45

in 2002 or 2003, I started a blog.

12:49

This was pre WordPress, so I had

12:51

to hand code it on HTML. Yep,

12:53

God bless. And put it up. And

12:56

I had a thought a while ago, like, I wonder

12:58

if that exists anywhere and sure enough, on

13:00

the Wayback Machine. I'm not

13:03

going to tell you the name of this blog because I don't want to

13:05

go read when I was writing when I was 23. It's

13:08

fine. It's nothing offensive. It's all perfectly

13:10

defensible, but I was

13:12

just a little baby. Well you started out

13:15

with the MTV news and we got that

13:17

for you. So MTV, we think we have

13:19

it completely, but we also have the newspapers

13:21

in Turkey and Russia and all

13:23

these others that have been basically taken

13:25

down with no warning. From a technical

13:28

standpoint, it seems like a ton of

13:30

data and then you have, this is

13:32

all very text based, mostly text based.

13:35

It texts an image, still image when you start doing

13:37

this in 96. Then

13:39

you have the dawn of video and

13:41

video comes to dominate huge portions of

13:43

the internet. I think it's probably, if

13:46

you think about social media, it's the dominant

13:48

media now. And that seems

13:50

like an order of magnitude change in how you

13:53

would do things and how you would save

13:55

it. Do you start to

13:58

have constraints on what you could... whole.

14:01

Thank God that these guys are working so

14:03

hard to make hard drives bigger and bigger

14:05

and bigger and bigger. So we have to

14:08

replace all the hard drives every five years.

14:10

We have to reel in a new rack

14:12

every few weeks to be able

14:15

to keep up with the unbelievable

14:17

amount of material. So we collect

14:19

about a billion URLs every day.

14:21

And a lot of

14:23

those URLs now are images. It's

14:26

nuts. People love this stuff. They are sharing

14:28

up a storm. The idea that people will only

14:31

do things for money is just not true.

14:33

No. People will do things because they can get

14:35

some positive feedback from each other and share

14:37

where they can as long as it doesn't cost

14:39

them any money. So why don't we make it

14:41

so that giving things away doesn't cost you.

14:43

And that was sort of the premise of libraries

14:46

in the Internet Archive. This point, just to

14:48

go back to my, I'd like to hear some

14:50

of your thoughts just to sync up with

14:52

the monologue about the Internet I fell in

14:54

love with. And again, I always sound like an

14:56

old man talking about kids these days and

14:58

in my youth, whatever. But

15:01

the Internet I fell in love with is largely

15:03

non-commercial enterprise in which people were motivated by all

15:05

kinds of things, some of which were noble and

15:07

some of which were base. I mean, humans

15:10

are humans. But almost none

15:12

of it was pecuniary. So people might

15:14

want to get into a flame war

15:16

because they wanted to assert domination or

15:18

they were incorrigible contrarians and like to

15:20

argue where they might want to share

15:23

really cool stuff because creating cool stuff

15:25

and showing it off as fun, the

15:27

whole sort of spectrum of human motivation.

15:29

But it was not commercialized and it

15:31

was not done explicitly for profit. And

15:33

it wasn't done for profit at scale

15:35

in which engineers are working

15:37

very hard to figure out how to hack people's

15:40

attention such that you can maximize monetization.

15:42

And I would like you

15:44

to talk a little bit about why you think that's

15:46

an important ethos to preserve because that is part of

15:48

what you're doing. And if it is an important ethos

15:50

to preserve. There's a

15:52

role on all sides of this. And

15:55

so most people just want to be

15:57

heard. They want to have some recognition.

16:00

want to feel like they've been taken advantage of.

16:02

And we developed search engines to help people guide

16:04

people to these even obscure topics to go and

16:06

find your people no matter where they are in

16:08

the world. Fabulous. But

16:11

you also have people that wanted to make money

16:13

by publishing on the net. And that's

16:15

not bad. No. As long as they don't, if

16:17

they're not exclusionary about it, or if they're not

16:20

sort of this platform that's trying to control a

16:22

whole media type, that that's

16:24

the controlling aspect that is really

16:26

causing so much trouble out

16:28

there. It's not so much that a

16:31

kid wants to sell his song off

16:33

of his garage band. Actually, it's really

16:35

hard to do that. And that's, I

16:37

think, one of the faults is we

16:40

made peer-to-peer selling so difficult. And

16:42

we basically have to sign up to go and

16:44

put on iTunes or Amazon or

16:46

the like. We're trying to make a more

16:49

decentralized web so that you can go and

16:51

have people go and participate in their communities,

16:53

sell a little bit if they can make

16:55

some money off of it. And most people

16:58

aren't going to buy your rock and roll

17:00

song, sorry. But at least it's a possibility.

17:02

And let's keep it so that there's no

17:04

central points of control. And we have got

17:07

people vying for monopoly control now in such

17:09

a way that is really debilitating. So while

17:12

we're trying to keep up with the social

17:14

media, some of those are very locked down.

17:16

So you can't even keep a record of

17:18

it. Even that's sometimes the only way politicians

17:21

are going and communicating their points of view.

17:23

That's not good. Oh, that's interesting. So there

17:25

are parts of the internet on platforms that

17:27

are essentially closed to the archiving that you're

17:29

doing. Yes. But some of

17:31

them are opening up. Reddit just yesterday

17:34

went and said, yeah, we're closing up

17:36

to the AI guys, but the internet

17:38

archives, A-OK, which we're kind

17:40

of happy about, and other research

17:42

organizations. So I think that going

17:44

and showing that there is a

17:46

role for libraries that's different from

17:48

Babylon is a good thing. But

17:50

the control aspects, and I wouldn't

17:52

say it's just the platforms that

17:54

you're normally talking about, the internet

17:56

platforms, but the publisher platforms and

17:59

the bad ones. the back ends

18:01

of these things, the television networks

18:03

that are very, very closed, the

18:05

book publishers, the magazine, the academic

18:07

journals, just so closed. So

18:10

all of you're saying, when you say closed, what do you mean

18:12

by that technically? That it doesn't

18:14

operate the same way it used to. That

18:16

basically libraries and people would buy things

18:19

in the old days, they preserve them

18:21

and then they would make them available

18:23

to themselves or their communities through lending

18:25

and they'd interoperate. But what we've got

18:28

now is organizations going and saying, no,

18:31

we're going to want to watch every page

18:33

turn. We're going to want to be

18:35

able to take that thing down. We want to be able to

18:37

change that ebook. We want to

18:39

go and make it so libraries don't

18:41

own anything in the digital world. That's

18:44

the problem. And there's been

18:46

lots of talk of the social media networks

18:48

and the issues around that, issues.

18:52

But I don't think we're really looking

18:54

at some of the back end organizations

18:57

that are exerting a level of control

18:59

because they can in this digital world

19:01

for surveillance and ultimately being

19:03

able to change and delete things without

19:05

anybody having a record of it. No

19:08

libraries are allowed. Right. So if

19:10

we think broadly, not just, I mean, so when we started

19:12

talking about this and when the way I think about the

19:15

way back machine specifically, which is just one part of the

19:17

Internet Archive, right, is this all

19:19

this sort of digital ephemera, you

19:21

know, someone's live journal, right?

19:23

Someone's MySpace page, someone's

19:25

political blog. When

19:27

you start moving up to entities that are

19:30

big corporate interests that are producing intellectual property

19:32

that they want to charge people for, which

19:34

again, as you said, is all well and

19:36

good. There's a different

19:38

set of questions, right? In terms of

19:40

so like when let's talk about a

19:42

show that streamed. This is a great

19:44

example. The archivist

19:46

in you, the librarian in you says,

19:49

what should be possible is that the

19:51

library buys that show. It

19:54

then owns it the way the library owns

19:56

a copy of Shakespeare's completed works or

19:59

library. the best-selling novel

20:01

that just came out, you know, by Maranja

20:03

De Luy and then can lend it to

20:05

people, right? And

20:07

what I'm hearing from you is that the entities are

20:09

like, no, we're not gonna sell you

20:11

our show for you to

20:14

just have archived and lend to people. We

20:16

want to control if anyone sees it.

20:18

We want complete control of that. That model,

20:20

we're not gonna abide, basically. You got it.

20:23

That's exactly what's going on now. So libraries

20:25

are not allowed from the big publishers. The

20:27

small publishers are still, you know, in there

20:29

trying to be, you know, publishers

20:32

that we love. But the big

20:35

corporate monopoly companies are basically making

20:37

it so that libraries cannot own

20:40

a digital ebook. So when

20:42

you borrow an ebook from your local library, you

20:44

think you are, it's actually

20:46

going and just being redirected

20:48

into the publisher's databases. And

20:51

they are surveilling everything about it. And

20:53

they're changing and deleting those books at

20:55

will. And so this is a

20:58

future that is absolutely

21:00

dystopian and counter to

21:02

what it is we should be able to do now.

21:05

We just want these people to sell. And

21:07

it's not happening. Right, you're not saying give

21:09

it to us. You're saying we want to

21:12

buy these things from you, the way that

21:14

a library buys things and then make it

21:16

available. Absolutely. And what libraries do is they

21:18

often, not just by the blockbusters, but they

21:20

buy the local content from the local authors

21:22

with the local and all of that's being

21:24

shut down and shuttled into these very few

21:27

platforms. And with the

21:29

surveillance thing, there's a

21:32

long history of libraries getting surveilled and then

21:34

their patrons getting rounded up and bad things

21:36

happening to them based on the books they've

21:38

read. This isn't good. So

21:41

the libraries have always been a bastion of

21:43

privacy. It's part of our ethics. And

21:45

so how do we go and move into this

21:48

digital world in such a way we have

21:50

many winners? So we have

21:52

many authors, many publishers, many

21:54

bookstores, many libraries and everyone

21:56

a reader. That's the

21:58

world that we can live in. But

22:00

it's not the thing that we're seeing

22:02

happen at the tops of these mega

22:04

corporations. That I think brings us to

22:07

some of the legal issues the Internet Archive is facing.

22:09

And maybe I'll bring in Kyle Courtney here, who I

22:11

know is also sort of read in on this. So

22:13

a bunch of publishers have sued the Internet

22:16

Archive. And Kyle, what is the

22:18

lawsuit about? What are they content in? So

22:20

ultimately, I think the lawsuit is about to

22:22

prevent libraries from being able to loan in

22:24

the digital space now and in the future.

22:28

So if we imagine that

22:30

loaning has always happened at the circulation desk,

22:32

when it moves into digital space, this is

22:34

what Brewster was referring to. They say, no,

22:36

we haven't sold you anything. We're

22:39

going to force you into renting your collections with

22:42

these terms that basically

22:44

eviscerate the library mission. So

22:46

the case is about a methodology of

22:48

lending called controlled digital lending, in which

22:50

I co-wrote in a paper in 2018.

22:54

And it's been endorsed by national library

22:57

organizations, regional library consortiums, specific library systems,

22:59

individual librarians, legal experts. Ultimately,

23:02

though, it comes down to this.

23:05

We are being told libraries, the royal we

23:07

are being told, we're not capable

23:09

of entering the digital space with our

23:11

same mission that we had. Now,

23:14

of course, the library mission relies

23:16

on the ability to acquire creative

23:18

works, which by the way,

23:20

serves the economic purpose of copyright, right? We buy

23:23

the stuff, right? This is not

23:25

free, we're spending money, millions and millions of dollars,

23:27

and then distribute those works to the public,

23:29

right? And that's, I always think this is

23:32

a constitutional narrative, right? We're promoting the progress

23:34

of science and the useful arts by

23:36

distributing those materials to this. And

23:39

so that, I think that the library loaning

23:41

programs are part of this copyright cycle. Now,

23:44

this method called controlled digital lending,

23:46

which again, digitally replicates what happens

23:49

at the circulation desk, is

23:51

more efficient, it's less ableist, it's good

23:53

for folks that are remote, that can't

23:56

get to the library, senior citizens, etc.,

23:58

especially during our pandemic closures. But

24:02

it's through the lens of something called fair

24:04

use, which is a copyright exception. And

24:07

I'm sure that your show has heard

24:09

about fair use before, or you have

24:11

heard about fair use before, but this

24:13

is basically saying, hey, copyright doesn't mean

24:15

total control of the

24:18

copyrighted work. It does not. It

24:20

is some control, certainly, but

24:23

there are generous exceptions granted to

24:25

users, to patrons, to libraries, to

24:27

be able to circulate those materials

24:30

in service of that promoting the

24:33

progress of science and the arts. And

24:35

again, I want to be very

24:37

clear, libraries and their systems represent

24:40

the commitment to both the economic

24:42

and access parts of the copyright

24:44

cycle. More of our conversation

24:46

after this quick break. If

24:55

you are struggling to get cool, quality

24:57

sleep, the ChiliPad Bed Cooling System is

24:59

your new bedtime solution. The

25:01

ChiliPad by SleepMe is a water-based mattress topper

25:04

that works with your existing mattress to continually

25:06

control your bed temperature from 55 to 115

25:08

degrees, allowing your body to

25:12

rest and recover. Your regular

25:14

mattress cannot do the job of keeping

25:16

you cool, but the ChiliPad funnels cool

25:18

water consistently under you all night long,

25:21

allowing you to customize your sleeping environment

25:23

and wake up refreshed. Visit

25:25

www.sleep.me-y to get your ChiliPad

25:27

and save up to $315

25:29

worth of code Y. This

25:34

offer is available exclusively for Why Is This

25:36

Happening listeners and only for a limited time.

25:38

Order ChiliPad today with free shipping and try it out

25:40

for 30 days. If you don't

25:43

like it, you can return it for free. Visit

25:45

sleep.me-y to get your sleep. Because

25:51

you're not investing in better sleep, you're investing

25:53

in a better life. All

26:01

right, so just tell me at a more brat,

26:03

how does digital lending work under the scheme that

26:05

you would like it to work and

26:07

what don't they like about? Like I still

26:09

am having a hard time with this. What

26:12

do you want to do and what do they not

26:14

want you to do? Yeah, so here's imagine that. So

26:16

libraries spent millions and millions of dollars of collections that

26:18

are in print that are on our shelves, right? We

26:20

have those books. We have paid for them,

26:22

we've controlled them. You own them. We own

26:25

them, we should be able to lend them without permission. Right.

26:27

Right. You don't have

26:30

to check with the publisher. You can lend

26:32

a book, yes, a thousand times and you

26:34

never pay the rights holder again. Right? You're

26:36

lending that book and you're making sure it stays,

26:38

you know, you bind the glue on

26:41

the board, you know, you make sure that book

26:43

lasts. That's the way public libraries especially get a

26:45

return on investment in their communities and you're holding

26:47

onto those books. Many of

26:49

those books never made the

26:51

jump to digital. Right? Maybe

26:54

they're not popular. Maybe they didn't do

26:56

well. They were important for

26:58

preservation and access purposes. So

27:01

the controlled digital lending methodology says, okay, we

27:03

take that book we bought and

27:05

we digitize it and then we hide it away. So

27:07

it cannot be loaned. It's not on the shelf. It's

27:09

not in a room. We hide it away. Then

27:12

we use that surrogate digital version

27:14

to loan to one person at a time,

27:16

just like we do at the CERC desk,

27:18

except it's digital. And then

27:20

that person reads it, keeps it for two weeks, three hours,

27:22

whatever the loaning period is. And then

27:25

it automatically returns back to the library.

27:27

So you've just created a digital version,

27:29

a digital token of the

27:31

physical book, but you've been the one

27:33

that digitized it. Yes. And

27:35

more importantly, when the person's reading it and looking at it,

27:37

they can't download it, they can't copy it, they can't access

27:39

it. It prevents them from spreading.

27:42

It's protected. It's protected using the same software that

27:44

the publishers use for their e-books. I want to

27:46

be very clear. Brewster has

27:48

this system tight and perfect. And

27:51

that's what they're suing about. Okay. So

27:53

Brewster, the Internet Archive uses this method.

27:56

Yeah, we own a lot of books. We've

27:59

been digitizing. these and making them

28:01

available for ourselves. But also hundreds of other

28:03

libraries have been doing this all since 2011.

28:06

This is common library practice. And

28:08

we even stay away from the

28:10

most recent five years just to

28:12

knock off the commercial angle

28:15

of these. And then we weave them

28:17

back into Wikipedia. We basically fulfill a

28:19

couple million links in Wikipedia so you

28:21

can flick to open to the right

28:23

page. What's interesting is people only look

28:25

at these books usually for

28:28

a couple minutes. They're checking a

28:30

fact. It's not beach reading.

28:32

It's basically you're using it like you're standing

28:37

in a library. And why

28:39

are they suing on this? Because

28:41

they feel that the licensed ebooks,

28:44

again, the forced rentals they've put

28:46

upon libraries, should be the only

28:48

way that the only standard

28:50

that a patron can access at work. Licensing

28:52

culture is out of control. You used Hulu

28:54

and Netflix as an example, right? We don't

28:56

own any of that, nor are we likely

28:58

to. My fear is as a

29:00

result of this case, libraries become

29:02

the next Hulu or Netflix. And that says,

29:05

you know, last day to read this book,

29:07

August 31st, because that's what happens. These

29:10

books expire and they

29:12

leave our communities unless we pay more.

29:15

So the standard that you're using, this CDL standard,

29:17

which is we digitize books, we have tight controls

29:19

on them so we could lend them one at

29:21

a time. They can exist in digital space such

29:23

that they can be used to check a fact.

29:25

They can add to our store of knowledge. We

29:29

don't like that protocol because we don't

29:31

control it. We want to control the

29:34

digital versions of our intellectual properties such

29:36

that we kind of hold always a

29:39

sort of veto that

29:41

it can expire. Like we're licensing it, which means

29:43

we can always take it back. Ultimately,

29:46

you don't own it. You just sort of

29:49

are a custodian of it for the time

29:51

being. You're a renter. They're the landlords and

29:53

we're the renters. Libraries are being turned into

29:55

renters. Right. So that's actually very useful. So

29:57

the library model is an ownership model. because

30:00

you guys buy the books and you own

30:02

your collections. That's the asset you have. And

30:04

that's what you make in public. And they

30:06

wanna convert it to a rental. Absolutely, because

30:08

I understand why they would want us not

30:10

to loan a book a thousand times and

30:12

only pay once. They want us to get

30:14

a thousand licenses, which is a

30:16

revenue capital stream that looks much better

30:19

to them than the- Oh,

30:21

so they wanna charge you for each rental?

30:23

Oh God, yes, there's checkouts per price checkouts.

30:25

There's 26 checkouts and you have to buy

30:27

the book again. There's in a year or

30:30

two years, these works expire. There's many methods

30:32

they use to prevent us from owning. I

30:34

just wanna make sure I understand this. You're

30:36

saying that's the current market model for

30:39

digital books from the large publishers?

30:42

26 times and then it's gone. It costs hundreds

30:45

of dollars just to get those 26 potential lends.

30:49

Okay, I know. Okay, so I'm gonna publish

30:51

a book next year through

30:53

a major publisher, Penguin Random House. And

30:56

so walk me through. So I'm

30:58

gonna publish this book from Penguin Random

31:00

House. Let's say Kyle or Brewster or

31:03

any librarian wants to get a digital

31:05

version of that book to lend

31:07

digitally from their library. How does that work?

31:10

So the publisher says, okay,

31:13

here's the terms by which this book is

31:15

lent. Now, sometimes they say, no libraries for

31:17

the first two months. They've done that before.

31:19

They embargo it, right? Or sometimes they

31:22

say, okay, the average person gets it for 1999, library's

31:25

path to pay 199. And

31:28

they only get it, yeah. Oh, $199. Dollars,

31:30

yeah. But for multiple rentals,

31:32

presumably. No, not necessarily. Libraries

31:35

pay five to 10 times as much for

31:37

access to eBooks than the average consumer does

31:39

off of a regular eBooks site. Really? Absolutely.

31:43

And so we're being punished for our mission,

31:45

right? I call it a tax on libraries

31:47

for being able to loan or provide access

31:50

books. And Chris, this is what I'm

31:52

talking about. The convenience of eBooks comes

31:54

at the expense of the library

31:56

and mission. And I don't think that's

31:58

a trade off, but your book. will be

32:01

offered to libraries for under a 26

32:03

checkout model or under a multiple seat

32:05

model or a pay-for-loan model or You

32:08

know, you can load it as many times you

32:10

can in a year, but then it goes away

32:12

it expires and disappears So there

32:14

there it depends on the publisher depends on

32:17

the vendor. Yeah, okay So this is clarifying.

32:19

So the point is that there is a

32:21

straightforward market logic here, which is that they

32:23

don't you know, I mean So

32:26

no one said to me and it's always stuck

32:28

with me that if libraries didn't exist. There's no

32:30

way we could create them now I believe that

32:32

I think Brewster might like you couldn't go to

32:34

the publishers and all of them say hey, we're

32:37

gonna create a Nonprofit

32:39

non-commercial, you know in some cases subsidized

32:41

or municipally owned system that just gives

32:43

away your product for free Now it's

32:45

not afraid. This is gonna be very

32:47

clear libraries pay through the notes, right?

32:51

To the consumers. Yes, right the

32:54

point being right now I can From

32:56

where is it right now, right? I can walk to the

32:58

library and borrow Trust by

33:00

Hernán Diaz or I can

33:02

walk three blocks the other direction to my local

33:04

bookstore and I can buy her non-trust

33:08

Her ideas is trust and there's

33:10

not a ton of products about which

33:12

that is true. That is true Right.

33:14

Like I if I want like, you

33:16

know, if I want You

33:18

know a new basketball, you know, I'm gonna buy a

33:20

new basketball There's nowhere I could borrow a basketball and

33:22

go to the Y But like

33:24

the point is it and I think the reason that

33:27

you would say and I mean I could take your

33:29

side of the argument Here is that like information

33:31

is just different than other products because

33:33

it has a social civic cultural role,

33:35

right? I'll go one further Libraries

33:38

are supposed to be immune from market

33:40

forces like that to

33:43

provide information to their communities, right?

33:45

So I agree there is that that marketization

33:47

of this in business model But libraries are

33:49

a special slice of the pie and that's

33:52

what we're trying to get done here and

33:54

they're widely supported So it's about

33:56

20% of all trade publishing

33:58

is the revenue comes from libraries.

34:01

Wow. And what libraries are

34:03

getting is less and less and less for

34:05

that. So for instance, the diary of Anne

34:07

Frank cost $27 per student for a 12

34:09

month subscription. Wait,

34:12

but isn't that, that's not public domain?

34:14

No. Should be. Oh,

34:16

right, right, because the copyright keeps getting

34:18

extended. Right, right, right, right. Yeah, uh-huh.

34:20

Or you take Mary Grove College, a

34:23

college that was focused on social justice

34:25

and they announced their closure and they

34:27

had this unbelievably great library. They donated

34:29

the Enrid Archive and we digitized

34:31

70,000 books. But

34:33

now these publishers demanded that we

34:35

take 14,000 of those social

34:37

justice books. Most of them are

34:39

decades old off the library shelves

34:41

so no one can really effectively

34:43

have access to these. These

34:46

are not commercially, this is not

34:48

the Chris's newest book. This is

34:50

the long tail of books. Libraries

34:53

support a broad range of authors.

34:55

And what we're being told is

34:57

you can only have bestsellers and

35:00

you can't even have them. You

35:02

can just have them make it so that your

35:04

patrons can go and give over their private information

35:06

to us, the publisher. Just to be clear to loop

35:09

back around because I think it's important when I

35:11

was using the thought experiment in my book, I

35:13

mean, it's not a thought experiment, it's gonna come out

35:15

next year. And hopefully you guys can go get it

35:17

at your local library, I would

35:19

hope, that that

35:21

isn't what's at issue in the suit

35:23

because that's not the CDL stuff. The

35:26

stuff that's like the new to

35:29

the market eBooks is already being

35:31

operated under these licenses that you're

35:33

talking about. The thing they're suing

35:36

about are the digitized books. Well, we

35:38

would want to go and buy one

35:40

copy, just one copy so

35:42

that it could be put into Wikipedia

35:44

so people could fact check out of

35:46

your book and be able to refer

35:48

to it. Not going through these checkout

35:51

systems that often cost several dollars per

35:53

ding to go and even go into

35:55

a fact check on. That's what

35:57

we'd like to be able to do.

35:59

The Penguin Ram room house is... repeatedly

36:01

refused to sell ebooks. This could all

36:03

be solved by they're just selling ebooks.

36:05

They don't sell them, they license them.

36:07

Right. They license them in these draconian

36:09

terms. So we end

36:11

up buying hardcopper books that we

36:13

digitize, and then we make one

36:15

copy available. But through this

36:17

lawsuit, 500,000 books have now disappeared off

36:22

of the digital bookshelves. This is

36:24

a devastating blow for people that

36:27

would never have bought these books.

36:29

The substitute of these books, and

36:31

also page numbers, if you had

36:33

a page number reference, I

36:35

have never seen any book with a page

36:37

number that you could refer to. So citations

36:39

fail. We have the equivalent of what we

36:41

were trying to fix in the web, a

36:44

fix in the physical book

36:46

world. And they're basically saying,

36:48

nope, that people cannot have

36:51

digital access to the

36:54

published works in book

36:56

form. So Kyle, this

36:58

lawsuit, the publisher's lawsuit was

37:00

successful at the district court level, right? So the

37:02

district court found it not to be a fair

37:04

use. So

37:07

we lost. Well, meaning they do a success. I

37:09

mean, it's on appeal right now, right? Well,

37:12

it was successful for the plaintiffs. I mean, the plaintiffs

37:14

won. I agree. So they won at the lower court.

37:16

I'm not saying that they should have. It's not a

37:18

normative question. It's a descriptive question about reality. Sorry, I

37:21

get very careful with my answers. The

37:23

publishing, they sued. They said that the

37:26

method you're using was not fair use.

37:28

They sued and they were found to

37:30

be. Infringement. It was copyright infringement. And

37:32

now that's on appeal, right? Yes. So

37:35

on Friday, the 28th of

37:37

June, there was the oral

37:39

argument. All the briefs have been

37:41

filed in the second circuit. And this

37:44

appeal is focusing on a few things, right? That

37:46

through the control digital lending process,

37:48

there are legally and economically significant

37:50

aspects of physical lending that we

37:53

want the court to review through

37:55

the lens of fair use, that

37:58

it continues to preserve the power of a print. A

38:01

library has significant legal usage rights and

38:03

great fiscal value in its collections, and

38:06

public library systems have spent millions of

38:08

dollars in building these collections. And so

38:10

the appeal is about a

38:12

proper reading of the fair use

38:14

analysis to control digital lending. But

38:17

it's also to reverse something that was stunningly

38:20

disruptive, I think could be if it continues,

38:22

which is anyone that has a donate now

38:24

button, or donate books or

38:26

donate things if you're a nonprofit, but

38:29

suddenly that's a commercial action according to

38:31

the district court that may weigh

38:33

against your fair use and that cannot stand either.

38:35

Oh, that's interesting because fair use, right, because part

38:37

of fair use, right, is that we're not trying

38:40

to monetize this thing. It's in the public domain.

38:42

And so one of the findings of the district court

38:44

level is that the donate now button is

38:47

essentially commercial activity that might, it might vitiate

38:49

your claim. Yeah, which is just absolutely

38:51

wrong on every level. I mean, it ignores

38:53

major precedent in the second circuit. And so

38:56

we're hoping the second circuit court of appeals

38:58

will hear that as well. Are

39:00

there broader implications just for like the

39:02

knowledge universe that we're entering into? Yeah,

39:06

yes. So obviously, so I'm also a fan

39:08

of open access and

39:10

the knowledge economy is very important. I

39:12

think certainly because we're talking about control digital

39:14

lending, you know, the books that we have

39:16

on our shelves for 20th

39:18

century works that are in libraries, especially,

39:20

you know, we refer to this

39:23

as a 20th century black hole, right? They're

39:25

not available for purchase. No new copies are

39:27

made. They're unlikely to make the

39:29

jump to digital, but they're important works that need

39:31

to be read, accessed and utilized

39:33

by our modern patron. And by the

39:35

way, our modern patron is younger and

39:38

younger. And if something's not digital,

39:40

it's as if it does not exist. So

39:42

I would hate to see the generations

39:46

be unable to access all of these works

39:49

that we have preserved. And then the

39:51

future for them is also bleak because

39:53

it's all licensing, whether it's iTunes or

39:55

Amazon or Netflix. We'll be

39:57

right back after we take this quick break. A

40:07

U.S. Senator destroyed by blackmail. He

40:10

was not bound by the truth

40:12

or by facts. The country's most

40:14

outrageous political demagogue ascending toward the

40:17

peak of American power. Millions

40:19

upon millions of devoted followers. This is

40:21

a story of heroes willing to face

40:23

down tyranny and the risk to the

40:26

country if they fail. Rachel

40:28

Maddow presents Ultra, season two

40:30

of the chart-topping original podcast.

40:32

Listen now on Amazon Music.

40:37

Stay up to date on the biggest

40:39

issues of the day with the MSNBC

40:41

daily newsletter. Each morning you'll get analysis

40:43

by experts you trust, video highlights from

40:45

your favorite shows. 2024

40:48

is now truly the most important

40:50

election in the history of our

40:52

country. Previews of our podcasts and

40:54

documentaries, plus written perspectives from the

40:56

newsmakers themselves, all sent directly to

40:58

your inbox each morning. Get the

41:01

best of MSNBC all in one

41:03

place. Sign up for MSNBC daily

41:05

at msnbc.com. So

41:12

one of the great ironies here, right, to go back to

41:14

where this conversation started Brewster with the

41:17

notion of the Library of Alexandria, a

41:19

single repository for all human knowledge and

41:22

cultural production, and, you know, all captured

41:24

in one place, archived, stored, made accessible.

41:27

At one level, right, I have more

41:29

access. I mean, it's a crazy thing

41:31

that at any moment I have access to anything. And I think

41:33

about this all the time. You know, when

41:35

you sit down, I have a long

41:38

riff on this in my forthcoming book

41:40

about, you know, the attention forcing mechanism

41:42

of the video store, which

41:44

is that you had to go to the video store and you had to figure

41:46

out what you're going to watch that night. And then

41:48

you were making a committed decision that you come back, you explain

41:50

to everyone at the house what you've gotten, maybe they like it,

41:53

maybe they don't, but that's what you're going to watch. Right.

41:56

And at any given moment now, so at one

41:58

level, we have this sort of overwhelming. I have

42:00

access to so much of the world's knowledge. I

42:02

can watch anything at any time. But

42:05

then there are these black holes emerging,

42:08

a huge swath of things that are

42:10

winking out of existence. And

42:12

I remember, you know, there are movies that

42:14

you can no longer get. There

42:16

are people, you know, a variety of places

42:19

have taken shows off their service. And

42:22

I know showrunners and actors who were on

42:24

a show that's now, it just

42:26

doesn't exist anymore or anywhere. When

42:30

Turner Classic movies, there was some discussion in the

42:32

corporate reordering that that might be taken

42:34

off. And there was just like rebellion of folks,

42:36

including, I think, like Martin Scorsese and everyone, because

42:39

it's like, well, this is the only way we

42:41

can see these films. So there's

42:43

a weird paradox booster here, which is that we have

42:45

access to more than anything ever before. But

42:48

the notion, the possibility of total

42:50

death, of black hole, of

42:52

things winking out of existence is also

42:54

as present as it's ever been. Absolutely.

42:56

It is so easy to get something

42:59

on everything. And thank God for Wikipedia,

43:01

another pillar of the open world. But

43:03

if it's not, I love

43:06

the line from the headline, current affairs

43:08

said the truth is paywalled, but the

43:10

lies are free. That

43:12

we basically have this world that we're coming

43:14

to where things are more and more promoted.

43:18

And if we're going to

43:20

have people be educated, be

43:22

ready for elections, we

43:24

need to be able to think critically. We need to be

43:26

able to quote what people have said. We need to be

43:28

able to compare and contrast it. We

43:31

need to be able to say whether historical

43:33

events actually happen. This requires

43:35

getting access. It's the

43:37

wonder of the Internet. It's so freaking

43:39

easy. I mean, all of the words

43:42

in the Library of Congress would fit

43:44

in two hard drives that cost less

43:46

than a month's rent. Our

43:48

opportunity for educating people on a

43:51

global scale is with us. And

43:53

we have money. There's a

43:55

support for all this libraries and things. It's not going away.

43:57

But we also have this sort of... Oh

44:00

my God, I can control and clamp down

44:02

and sell less. Four

44:04

more is an ethos that we're

44:07

seeing. The St. Charles City

44:09

County Library in Missouri is probably

44:11

going to close this summer because

44:13

of the ebook budget problem. So

44:16

we're having libraries hit with bannings

44:18

and the like. We're having them

44:20

hit by budget shortages. We're

44:22

hitting them with license problems that we've covered

44:24

for the last half an hour here extensively.

44:27

But we're now having also

44:29

the judiciary starting to judge

44:32

against libraries in ways that we

44:34

haven't seen in 100 years. The

44:37

Carnegie libraries exist because we

44:39

have legislature and judicial support

44:41

for educating a broad swath

44:43

of people. And

44:45

I don't know, the United States that I

44:47

grew up in is such a steeped in

44:50

that tradition of the Carnegie thing. It's

44:52

really hard for people to fathom

44:55

that all of these forces, political,

44:57

judicial, legislative funding are all going

44:59

against a tradition that built America

45:02

out of a set of immigrants

45:05

and brought us together to be able to know how

45:07

to work as a populace. And

45:09

all of that is in, well,

45:11

some of it is going to be

45:14

judged again by an appellate group in

45:16

lower Manhattan. And we'll see

45:18

what it is they decide towards the fate of

45:20

libraries as we know them. So that's a stark

45:22

way to talk about it. I mean, is that

45:24

how you see this case, Kyle, that like this

45:27

case will determine the fate of libraries as

45:29

we know them? I mean, it's a slippery

45:31

slope into a licensed only culture in which

45:34

libraries cannot exist and mission

45:36

this challenge. I do agree. A lot of

45:38

people say, oh, this is just about the

45:40

Internet Archives open libraries program. It's a narrow

45:42

focus. I don't think that's true. It's

45:45

about libraries entering and being supported

45:47

in the digital space. And

45:49

through the use of either licensing or

45:52

litigation, that's a one, two punch, really

45:55

will threaten the library mission now and

45:57

in the future. That's why I, you

45:59

know, I have from

50:00

patrons and folks that are aware of

50:02

this case, general stuff are

50:04

interesting. But the answer from the publishers

50:07

and the rights holders is always more licensing.

50:09

That's the answer to everything, right? No fair

50:11

use, get a license. No text data mining,

50:13

get a license. No AI training, get a

50:16

license. Everything is licensing. And this

50:18

answer of licensing threatens the

50:20

purpose values and missions of all

50:22

libraries. It undermines the ability

50:24

of the public, right? The taxpayers to

50:26

access the materials that are purchased with their

50:29

money, right? For use in public libraries and

50:31

state institutions. And further, I think

50:33

it's short sighted is what

50:35

you're saying, right? You're like, how come I

50:37

can't get this stuff anymore? How can I

50:40

have access to stuff? We have a 20th

50:42

century black hole. We have swaths of information

50:44

that just will never be released again unless

50:46

you pay a license. So it's not in

50:48

the best interests of library patrons, the public

50:50

at large, but that kind of knowledge economy

50:52

that you're discussing that we need to be

50:54

able to provide these materials to everyone. It's

50:58

funny too, as you talk about the licensing,

51:00

I mean, on the flip side, right? On

51:02

the other side of this question about fair

51:04

use is AI, where now you've got this

51:06

insane situation where everyone

51:08

put everything on the internet for

51:11

public use. And then

51:13

a bunch of companies sent

51:15

their large language learning models and

51:17

fed it all that stuff, not for the

51:20

purpose that the internet archive exists, which is

51:22

to make it publicly accessible, but

51:25

to train large learning models

51:27

to talk like people in

51:30

a chatbot basically, or

51:32

do whatever they're gonna do, and then monetize

51:35

it. So there's this whole other set of,

51:39

I realize that the publishers are the, quote

51:41

unquote, bad guys in this lawsuit from

51:43

your perspective, but they're now suing,

51:45

there's gonna be a bunch of lawsuits on the

51:47

other side, on exactly

51:49

this in which you're already seeing publishers

51:52

and I think record label, a whole bunch

51:54

of, you know, entities that

51:56

publish or produce or own the

51:58

intellectual property on content. suing the

52:01

AI people to say, you

52:03

can't just take this stuff to pick your

52:06

models and be okay with.

52:08

Yeah. I would suggest what

52:10

we need is a rule of law, not

52:12

rule of contract. What happens when you have

52:14

a rule of contract is you have very

52:16

big people just bludging it out with lawyers.

52:18

Yes. And nobody that has lots of lawyers

52:21

are going to be at that table. So

52:23

we'll end up with a deal between these

52:25

publishers and the big AI companies. But the

52:27

idea I'd love to see, we have

52:29

a new tool in town. It's kind

52:31

of cool. Let's go and use it for

52:34

some big, hard problems like healthcare or climate

52:37

inventions. There are

52:39

things that we could go as

52:41

libraries. It could be the Research

52:43

Libraries Day. The research libraries have

52:45

been basically holding onto these materials.

52:47

Why would you read agricultural records

52:49

of Argentina from the 1930s? We

52:52

happen to have them. So

52:55

why would you go and no

52:58

human's going to do that, but machines

53:00

could. Right. Unless these lawsuits

53:02

make it so that it's taken out

53:04

of the possibility of the public sphere

53:06

and only lands with the very, very,

53:08

very wealthy. And people usually

53:10

don't think that next step through. But

53:13

we've seen the United States the last

53:15

time that AI came around. It was

53:17

Google. They basically hovered up other people's

53:20

websites and made

53:22

a search engine out of it. And they were allowed

53:24

by law in the United States to do that. And

53:26

it was important they did, because if they weren't, then

53:28

we wouldn't have to enter. Absolutely.

53:31

the.com boom happened in the

53:33

United States because of law

53:35

and judiciary allowing that

53:37

to happen. It did not happen in

53:39

Europe or Japan. You notice there's no

53:41

search engine in Europe. There's no search

53:43

engine in Japan. It's not because they're

53:45

stupid. It's because their laws

53:48

sided with the publishers. And

53:50

if we go and do that in

53:52

this AI round, and I'm not just

53:54

trying to chill for the big guys.

53:56

I want the little guys. I want

53:58

every little. library to

54:00

go and take their collections or

54:02

take the ornithology information from Cornell

54:05

and around all of the ornithologists

54:07

and start training models. If we

54:09

make that illegal based on this,

54:12

we will lose out on a major

54:14

opportunity in the United States. And I'll

54:16

tell you, Europe has seen around this

54:19

corner and they've already made regulations allowing

54:21

cultural heritage institutions and research organizations, nonprofits

54:23

to train AIs. Japan went further and

54:25

said, we're not going to

54:28

step out of this next level of innovation.

54:31

We're going to let robots read. In

54:33

the United States, it's just a free

54:35

for all. We don't have a legislature

54:38

that's really guiding things very well. And

54:40

we'll see what happens in the judiciary.

54:42

So it's a new game. And innovation,

54:44

is it going to be grassroots or

54:47

is it only going to be a

54:49

few gigantic players that have enough lawyers

54:51

to sue themselves into a deal with

54:54

these gigantic publishers? That's where

54:56

I see this is going. And I'm a big

54:58

fan of a game with many winners.

55:00

I want to see libraries and archives be

55:03

able to bring their materials to bear under

55:06

the right circumstances. Maybe it's nonprofit

55:08

works. Maybe it's these nonprofit models to

55:10

be able to do great things

55:12

with these tools to help solve some

55:14

of the big problems we've got.

55:16

That's interesting. That is very interesting. All

55:19

my feelings about AI are

55:21

sort of, I would say

55:23

instinctual, impulsive, visceral, and

55:26

not- Careful. Yeah. I mean, they are.

55:28

I mean, I'm thinking about it a lot, but I don't feel like I

55:30

have a fixed view on things. I'm sort

55:32

of open to persuasion of a bunch of things. I find there's

55:35

a bunch of it that I find weird and

55:37

crass and I'm being- It's early

55:39

days. They're weird. Yeah. And there's a

55:42

lot of shoving a solution to me

55:44

without a problem. Yeah. You need to

55:46

buy this solution. It's like, what is

55:48

this solving for? It's like, well, okay.

55:51

Well, I don't know if I need it. You're

55:53

talking to archivists. So you remember the.com boom?

55:55

Do you remember when you got warm cookies

55:57

and milk delivered to your door? Just

56:00

our dog food shipped through the mail. It

56:02

was just, there was this weird, wonderful things

56:04

that came about. Most of it made no

56:07

sense. But it was kind of cool. And

56:09

people believed and they put their dreams into

56:11

this new technology. And right now

56:13

people are putting their dreams and their fears

56:16

into this AI thing. Let's shape it to

56:18

be something that serves us. Let's

56:20

go and actually have some rules

56:22

and guard rails on it by

56:24

legislatures, not just corporate. Yes,

56:27

that line rule of law, not rule of contract

56:30

is a really important one. Because you're right that so much

56:32

of the way this all gets settled in the US is

56:34

large litigation between powerful entities fighting each

56:37

other and working out some

56:39

settlement between them. As opposed to- The people

56:41

are not at that table. Yeah, we're

56:44

a culture that litigates rather than regulates

56:46

generally. That's how things work here. Oh,

56:49

well, that's why there's over 30

56:51

lawsuits currently in the courts about

56:53

just, and it's by the way, it's mostly books.

56:55

Now there's some music on there too. Just

56:58

last week, the major

57:00

music labels sued some AI

57:02

startup music companies. But

57:05

again, the answer, I just want to say, it's

57:07

the same answer we're seeing in the CDL case

57:09

to bring this together for a moment. The answer

57:11

is licensing. So I've been

57:14

in many AI discussions room where they're like, you want

57:16

to use our stuff? You want to train? Get

57:18

a license, get permission. That's the only

57:20

way to do it. And thankfully

57:23

the rule of law is not that. We

57:26

have text to data mining. We have Google

57:28

books. We have Google images. We have all

57:30

these things where you can copy the entire

57:32

thing, the whole thing, use

57:34

it for something new and different, and allow

57:36

that technology to exist. So I think we

57:39

have our answers through the rule of

57:41

law already, but all

57:43

these cases, all this litigation is

57:45

going to slow down the entry

57:47

for nonprofit research-based.

57:50

And this is, Bruce and I are waving

57:52

our arms saying, we're not mega companies. We're

57:55

not for-profit things. We're researchers

57:57

that want to do text to data mining. school

58:00

students, I want to experiment with AI.

58:02

All of these things should be able

58:04

to be done in the

58:06

nonprofit educational standpoint where libraries have

58:09

been planting their flag for forever.

58:12

Brewster Kale is a digital librarian at

58:14

the Internet Archive. He's perhaps best known

58:16

for the Wayback Machine. Kyle Courtney is

58:18

a lawyer, librarian, director of copyright and

58:20

information policy for Harvard Library, co-founder of

58:22

Library Futures, which aims to empower the

58:24

digital future of America's libraries. That was

58:26

a totally fascinating conversation about something I

58:28

knew literally nothing about. It's one of

58:30

my favorite kinds of conversations. Thank

58:33

you very much, Chris. Once

58:42

again, my great thanks to Brewster Kale and Kyle Courtney.

58:45

No decision has been made in this appeal yet. A

58:47

decision may come this fall. We will keep you updated.

58:49

We'd love to hear your feedback on all of this.

58:51

As always, I love reading your emails.

58:53

I should write back to more of them.

58:55

We do read each one we get. You can

58:57

email us at whithpod@gmail.com. Get in touch with us

58:59

using the hashtag WhithPod. You can follow us

59:01

on TikTok by searching for WhithPod. You can follow

59:04

me on threads at Crystal Hayes and on Blue

59:06

Sky at Crystal Hayes and on what used

59:08

to be called Twitter as Crystal Hayes. Why

59:11

is this happening is presented by MSNBC

59:13

and NBC News produced by Donny Holloway

59:15

and Brendan O'Melia. This episode was engineered

59:17

by Katie Lau and Bob Mallory featuring

59:19

music by Eddie Cooper. Aisha Turner is

59:21

the executive producer of MSNBC Audio. You

59:23

can see more of our work including

59:25

links to things we mentioned here by

59:27

going to NBC.