« Kerry on non-discrimination | Main | on "Forgetting the First Amendment" »

The Grokster appeal

News from the oral argument in the Grokster case yesterday is quite good. The argument on our side of the debate was excellent. Fred was extraordinary, and the judges seemed perfectly tuned to the issues. EFF's got a recording of the argument here, and Ed Lee's got a nice write up on his blog as well.

The tide is turning.

| | technorati

TrackBack

TrackBack URL for this entry:
http://lessig.org/mt/mt-tb.cgi/1162

Listed below are links to weblogs that reference The Grokster appeal:

» The Grokster appeal from datacloud
I’ll ditch the interpretation here in order to let Lessig’s fast-breaking news just run: News from the oral argument in the Grokster case yesterday is quite good. The argument on our side of the debate was excellent. Fred was extraordinary,... [Read More]

Comments (23)

Congratulations. Insisting that the choice of words is important and that copyright infringement is not theft, is showing its fruits.

February 5, 2004 7:38 PM Matthew Saroff:

That's why I insist on referring to IP as intellectual PRODUCT, not intellectual PROPERTY.

It's no more property than a liqueur license.

Indeed. Lawrence Lessig, Fred Von Lohman, and Eben Moglen have made me refrain from all those funny anti-lawyer jokes (well....most of them. ;)

I'm really glad that this case is being heard in the 9th Circuit because when it comes to technological innovation, a liberal court is a good thing these days.

That leads me to wonder: In an age of instant gratification and incessant innovation, is it time to re-evaluate the lifetime terms of the federal judiciary? Further, can our established legislative and judicial branches sustain that pace?

--Jason

February 5, 2004 11:48 PM Nathanael Nerode:

New York doesn't have life terms for its judges. It does have *very long* terms -- 14 years for Court of Appeals members, for instance.

This seems to be sufficient to insulate the judiciary from the pressures of politics, without excessively insulating them from the public. It's definitely preferable. Unfortunately, I don't think changing the Constitution to allow that for federal judges would have enough support to do any time soon.

February 6, 2004 12:15 AM three blind mice:

Judge John Noonan: "So 10% is noninfringing? That sounds like a lot of noninfringing files to me."

Well, gee, that sounds line a lot of INFRINGING files to us. Last time we did the math 90% is NINE times as much as 10% your honor.

We don't have a problem with the 10%, it's the 90% which brings us to the bench, your honor. Perhaps you would favor us with a discussion about this.

"You don't solve it by calling it 'theft.' You have to show why this court should extend a statutory monopoly to cover the new thing. That's your problem. Address that if you would."

What do YOU call it, your honor, when someone takes something of value and doesn’t pay for it? Is it theft when someone sneaks into a movie theater, or a concert, without paying? What’s that? Trepass you say... Well, isn’t trespass a crime? Isn’t the effect on the movie owner's lawful revenue the same as theft?

Would you consider it theft if someone plays a round of golf at your country club without paying the greens fee? What’s that? We’re having trouble hearing you? Trespass again? Still a crime? Isn’t the effect on the country club owner's lawful revenue the same as theft?

Let's call it a banana your honor so you can get past your semantic difficultites and get to the real issue here. You ask, "why should this court extend a statutory monopoly to cover a new thing?"

You must be a Republican, your honor. It's called the Constitution. Article 1, Section 8, if you are unfamiliar with it. We understand it’s probably been a while since you looked at it. Gets in the way of all that free downloading your daughter’s been doing and that enemy combatant stuff you've been turning a blind eye to.

Just for old times’ sake maybe you could have a quick glance. We have a pre-John Ashcroft copy right here, if you've misplaced yours. We won’t tell Mr. Bush. Honest.

"Well, gee, that sounds line a lot of INFRINGING files to us. Last time we did the math 90% is NINE times as much as 10% your honor.

We don’t have a problem with the 10%, it’s the 90% which brings us to the bench, your honor."

Just as 93% is umpteen times greater than 7% in the case of VCRs. The proportion argument holds no weight if we are to acknowledge "significant non-infringing uses" as a burden for the plaintiffs to overcome. If, as you suggest, that burden ought to be the reverse, then I guess you'd care to bring whatever % of Internet traffic might constitute illegality against the ISPs, if indeed not computer and stereo manufacturers. Off to the races!

"Isn’t the effect on the movie owner’s lawful revenue the same as theft? . . . Isn’t the effect on the country club owner’s lawful revenue the same as theft?"

Theft involves material loss, not hypothetical loss. The examples you cited of infringement on uninterrupted resources have never and will never be called theft, any more than eavesdropping is "the same as" interrogation, murder is "the same as" criminal neglect, and on and on. If you're bothered by games of semantics try using the language as it was plainly intended.

February 6, 2004 3:28 AM three blind mice:

Theft involves material loss, not hypothetical loss. The examples you cited of infringement on uninterrupted resources have never and will never be called theft,

strider, arguing over the semantics of the word "theft" obfuscates the real issue here which is copyright INFRINGEMENT and the people who erroneously conclude that since it is not THEFT there is nothing wrong with it.

playing golf on a course (even if no one is using it) and not paying for the privilege is depriving the owner of her legal right of ownership – even if you take no divots.

theft, trespassing, infringement, whatever you want to call it, the EFFECT on the owner is the same, you have taken something of value from the owner without paying her for it.

look at it this way, when you download over grokster a copy of your favorite justin timberlake song – which justin has the constitutional right to control the distribution of - the money which should be in justin’s pocket (to pay for his legal defense for that unfortunate wardrobe malfunction) is still in your pocket.

yes, justin still has his original copy, but have his money.

how is this not theft?

it is simply intellectually dishonest to advance the argument that copyright infringement does not equate with theft.

Just as 93% is umpteen times greater than 7% in the case of VCRs.

comparing a VCR which makes single, real time, IMPERFECT copies of an analog signal for personal use to a P2P network which distributes thousands of EXACT digital copies anonymously is patently absurd.

the EFF’s argument reflects a staggering ignorance of the technology involved and the result its use, or, more likely, it is simply just another bold faced EFF misrepresentation to advance their nonsensical, unconstitutional, innovation stifling agenda that anything which gets in the way of all that free downloading is a bad thing.

we wish that the anti-copyright crowd would argue these issues honestly instead of playing semantic games.

or maybe not... then it would be a short argument and an easy victory.

and what's the fun in that.

three blind mice

strider, arguing over the semantics of the word “theft” obfuscates the real issue here which is copyright INFRINGEMENT and the people who erroneously conclude that since it is not THEFT there is nothing wrong with it.

I think that most people here agree that copyright infringement is illegal. I even agree that it is wrong to commit copyright infringement (excluding violations of a few recent changes). Despite this, it is still worth using the word 'theft' properly. Of course its not being theft doesn't mean that it isn't wrong or that copyright holders aren't losing out (though I remain to be convinced that copyright holders actually are losing out).

it is simply intellectually dishonest to advance the argument that copyright infringement does not equate with theft.

<raises eyebrows> The anti-copyright crowd (strangely including many who support copyright) didn't start the semantic games by bringing up this talk of theft. If we must discuss copyright infringement in terms of theft, then at least we could use the proper definition of theft.

comparing a VCR which makes single, real time, IMPERFECT copies of an analog signal for personal use to a P2P network which distributes thousands of EXACT digital copies anonymously is patently absurd.

That may be, but it is difficult to get 16bit 44kHz PCM representations of the original masters from filesharing networks.

Do your objections apply to the internet as a whole, or only to filesharing networks? What opportunity for infringement is provided by filesharing that is not otherwise provided by the internet. Not only would making the internet illegal prevent much legal distribution, it would remove the only method of distribution (or at least the only efficient method) of some material.

I would first like to convey my compliments to Fred Von Lohmann for an excellent oral argument. He didn't seem to hesitate at all during the entire 15 minutes (as opposed to the defense in the Aimster trial).

Also, I would like to point out a glaring flaw that I kept noticing in the "proportionality" argument submitted by the plaintiff. It seems to me that the "substantial noninfringing use" standard is relative at best.

The plaintiffs claimed that roughly 700 million files were traded, and only ten percent were (possibly) noninfringing. As expressed as a percentage, 90 percent as opposed to 10 percent certainly causes one to lean away from a finding of "substantial" noninfringing use. However, if one assumes that all infringing use was suddenly eliminated, that leaves over 70 million cases of noninfringing use, quite a substantial (and arguably "commercially significant") use. I see no reason why the court should find for the plaintiff on this issue.

Tim has the right idea - there is nothing intrinsically innovative about File-Sharing networks. These networks simply integrate various technologies the Internet infrastructure provides (I develop a File-Sharing network so I know). If you want an absolute end to copyright infringement, pull the plug on the Internet and all will be well ;) .

Another interesting point that I thought was lacking in the Grokster appeal oral arguments : what about substantial, POTENTIAL non-infringing uses? Who is to say that development of File-Sharing networks won't lead to innovations we can't even envision right now? There are reasons to stifle technology, but doing so in the name of a business model is wrong.

February 7, 2004 2:37 AM three blind mice:

The anti-copyright crowd (strangely including many who support copyright) didn’t start the semantic games by bringing up this talk of theft. If we must discuss copyright infringement in terms of theft, then at least we could use the proper definition of theft.

reproduced here: theft - when one "dishonestly appropriates property belonging to another with the intention to permanently deprive the other of it".

ok tim ivorson, let’s ignore our argument that money is property and focus on the “copy” itself and see if your definition offers you any shelter from the storm.

there is a tangible and an intangible way of looking at property. downloaders prefer to advance the tangible argument we will slay this dragon first.

on a quantum level, there is no such thing as a copy. copying involves the non-destructive transfer of energy between a source and a destination. in other words, using a thermodynamic argument, one can show that each digital “copy” includes an infinitesimal part of the original signal energy. if one believes that energy, as einstein observed, is tangible matter in a different form, then each copy includes tangible matter taken from the original. so when you make a digital copy, you take a part of this matter. thus if you take a copy without license from the owner, you have "dishonestly and permanently" deprived the owner of his property. ergo it is theft.

the thermodynamic argument is a complicated one and people who cannot see the difference between an analog VCR and a digital P2P network are unlikely to be able to grasp this so let’s look at the property argument from an intangible perspective. let’s assume that intellectual property is intangible property - not real property, but a legally created fiction of property which is undiminishable. using this construction, copyright entitles the author to some limited control over ALL of the copies of his work. not just some copies, but all copies. the law may allow you under fair use to make a copy for personal use, but this copy remains under the author’s control. taking one of these copies – without his permission and outside any legally accepted definition of fair use, "dishonestly and permanently" deprives the author of his intangible property. ergo it is theft.

and here is one more spear for good measure, because dragons are hard to kill. copyright protects expression not content. when you buy a CD, you are not buying the content, you are buying the expression of this content on the plastic medium. separating the content from the medium and placing the content onto a P2P network creates a new expression which dishonestly, and permanently deprives the copyright owner of his original expression. ergo theft.

finally, tim ivorson, it is rarely stated that copyright infringement equals theft, the argument most often forwarded is that copyright infringement is tantamount to theft, that it equates with theft, or that it has the same effect as theft. these arguments (which are far easier to understand than the three which we three blind mice set forth above) are also rejected by the anti-copyright-when-it-stands-in-the-way-of-all-that-free-downloading crowd who advance the non-sensical argument that copyright infringement is nothing at all like theft, that it has no harmful effect on the copyright owner, and that it is a GOOD thing.

we respectfully disagree.

if you wish to argue that copyright infringement is literally not theft, or that it does not have the same effect as theft, the mice stand at your service, ready to clear up your confusion.

February 7, 2004 2:54 AM three blind mice:

if you want an absolute end to copyright infringement, pull the plug on the Internet and all will be well ;)

susheel, there is also the alternative of promoting the responsible use of this technology. if the internet community would adopt the position that illegal file sharing is wrong and it is not cool, or tolerated, or accepted, by the peer community, this would go a long way towards diminishing the problem and keeping the architecture open.

the continued flagrant abuse of the technology for illegal purposes, promoted and enabled by services such as grokster and supported by many of the kind and thoughtful people who post to this blog, is what is going to pull the plug.

if you truly and honestly want services available for “non-infringing” uses, then you should be using all means at your disposal to speak and work against the “infringing” uses.

unfortunately, it seems to us that “infringing” is what this IS all about and all the EFF and those who support them are using the thin patina of legitimacy as cover so that use of the internet to plunder the wealth society may continue unabated.


February 7, 2004 4:24 AM Joseph Pietro Riolo:

In response to comment made by "three blind mice"
on Feb. 7, 2004 at 2:54am:

While you have a point that it is very disturbing
that many people have no intention to abide by
the copyright law and they should stop violating
it routinely, I want to point out that those who
are copyright maximalists are very obsessive with
control that they want to expand it more and more
and more to the point that they want to eliminate
the non-infringing uses. Based on your comments
in this blog, you are very close to copyright
maximalist. That is where we differ.

Regarding the comment made by "three blind mice" on
Feb. 7, 2004 at 2:37am:

Readers should know that the statements about
quantum level are not correct.

To keep the example as simple as possible, I
am using paper and ink as an example.

An author uses ink to write a large letter "A"
on a paper.

A copier sees the author's paper and remembers
the letter as written on the paper. The copier
goes to a store and buys a bottle of ink and
a piece of paper. He then uses that ink to
write a large letter "A" on the paper that he
bought from the store.

Here, we have two different papers with different
inks. On the quantum level, both are made of
different energies. Yet, both show the same
letter "A". How is this possible?

For those who are hard-core materialists like
"three blind mice", they will produce all
different theories on how two different things
can have the same letter. "three blind
mice"'s explanation about signal energy is
an example.

For those who are influenced by Charles Peirce's
semiotics like myself, there is an explanation:
The expressions that we see are the result of
the triadic relationships that all material
objects have. At the material level, there
are several streaks of ink on a piece of
paper. At the triadic level, these same
streaks of ink represent the letter "A".

The hard-core materialists won't recognize
the triadic level and so, there is no
point in arguing with them on what copy
really involves. On the other side of
the view, those who are of Peirce's
influence can easily recognize that
copy is simply an action of changing
the form of matter to look like the
original form of entirely different
matter.

To conclude the example, the copyright law
essentially forbids you to change the
shape of ink (that you personally
and lawfully buy from a store) to
look like the shape of ink in the works
whose copyright is still valid.

Joseph Pietro Riolo
<riolo@voicenet.com>

Public domain notice: I put all of my expressions
in this comment in the public domain.

February 7, 2004 11:23 AM three blind mice:

Joseph Pietro Riolo thank you for your thoughtful comments.

despite its simple appearance, your example of ink and paper is actually impossibly complicated if one is to analyze it using a rigorous scientific approach. the analysis associated digital copying is itself very complex, but with certain assumptions it can be proved using thermodynamics. philosophical arguments such as yours are somewhat more difficult.

copyright maximalist.... sticks and stones may break our bones, but names will never hurt us.

we're not sure that this this, or any, label fits us. our views as expressed in this blog come from the perspective of the rights of the individual in which we are unashamedly maximalist.

we believe very simply that original writings and inventions are the property of those whose genius created them and that this property should not be subject to capricious confiscation by the masses. if this makes us copyright maximalists, then so be it.

the wholesale public confiscation of private property has a long and dark history which speaks for itself. opposing this needs no defending.

it is those such as yourself who support the public confiscation of private property through the weakening of intellectual property laws who should be tasked with defending it against the backdrop of this history.

three blind mice

reproduced here: theft - when one “dishonestly appropriates property belonging to another with the intention to permanently deprive the other of it”.

ok tim ivorson, let’s ignore our argument that money is property

I am content with the idea that money is property. I thought that the fact that the copyright infringer does not receive the copyright holder's money caused the copyright infringement to not be theft.

I'm not sure that I understand your next argument. I don't that it matters whether copies are strictly possible. There are events that are near enough to copying for me. I think that the tangible matter taken from the original is not (directly) what copyright holders are interested in. The copyright holder has the creative work and the money that they had before.

the thermodynamic argument is a complicated one and people who cannot see the difference between an analog VCR and a digital P2P network are unlikely to be able to grasp this

I do understand that VCRs are analogue copying devices and that P2P is a digital copying device. I was not aware that it is important to the legality of VCRs that they do not make perfect copies and I think that it should not be. There is not significant use made of P2P that preserves the audio or video quality of the original. As long as we are talking about significant uses, P2P does not deliver perfect copies of the original, but perfect copies of poor copies.

If we get our perfect copying digital VCR and connect it to the internet, then we have better P2P copying. If Grokster should be banned or restricted (beyond the existing illegality of copyright infringement), then shouldn't this too? What should be banned or restricted? The internet or the digital VCR? Shouldn't we have the right to use the internet and the digital VCR for non-infringing purposes?

let’s assume that intellectual property is intangible property - not real property, but a legally created fiction of property which is undiminishable.
...
the law may allow you under fair use to make a copy for personal use, but this copy remains under the author’s control. taking one of these copies – without his permission and outside any legally accepted definition of fair use, “dishonestly and permanently” deprives the author of his intangible property. ergo it is theft.

This is plausible. I'll give it some more thought later. I am inclined to believe that copright is not even intangible property, but it behaves like it, so you could be right.

copyright protects expression not content. when you buy a CD, you are not buying the content, you are buying the expression of this content on the plastic medium. separating the content from the medium and placing the content onto a P2P network creates a new expression which dishonestly, and permanently deprives the copyright owner of his original expression.

Why does it matter whether the same expression or a derivative is distributed? I agree that the copyright holder should retain control (less fair-use/fair-dealing and expiry). Why should the copyright holder also control non-infringing copying?

it is rarely stated that copyright infringement equals theft, the argument most often forwarded is that copyright infringement is tantamount to theft, that it equates with theft, or that it has the same effect as theft.

Sorry, I misunderstood your point the first time. I thought that by equation you ment equality. Yes copyright infringement is like theft.

February 7, 2004 8:26 PM Joseph Pietro Riolo:

In response to comment made by "three blind mice"
on Feb. 7, 2004 at 11:23am:

I think that you should understand that I am
pro-copyright, just like you. (I used to be
anti-copyright but no more.) We differ in
how we want the copyright to work.

Take p2p file sharing software as an example.
Your complaint about this new technology
is a bit overdone. Over the time, violators
of copyright law using the software will be
caught. The amount of non-infringing uses on
p2p file sharing software will increase. Give
it time. It is a wonderful tool. You three
blind mice can even use it to share the notes
about farming with your friends across long
distance.


Joseph Pietro Riolo
<riolo@voicenet.com>

Public domain notice: I put all of my expressions
in this comment in the public domain.

February 8, 2004 2:06 AM three blind mice:

I was not aware that it is important to the legality of VCRs that they do not make perfect copies and I think that it should not be. There is not significant use made of P2P that preserves the audio or video quality of the original. As long as we are talking about significant uses, P2P does not deliver perfect copies of the original, but perfect copies of poor copies.

Take p2p file sharing software as an example. Your complaint about this new technology is a bit overdone.

the difference is the potential effect of this technology on the copyright owner.

tim ivorson, you earlier made made a valid point that “it is difficult to get 16bit 44kHz PCM representations of the original masters from filesharing networks.” indeed it is. it is similarly difficult to get this from the CD recordings which are available to consumers. MP3 encoding is an even less perfect representation of what is available, but in making a comparison to the Sony precedent what has to be considered is:

1) that exceedingly easy to get these “near perfect” MP3 copies over P2P networks
2) that there is no degredation with further copying,
3) thousands of copies can be distributed anonymously without the owner having to surrender his original MP3.

comparing this to VCR technology where:

1) analog copies on tape must be physically shipped,
2) there is substantial degredation as subsequent copies are made,
3) the owner of a copy cannot “share” his tape without surrendering his original.

shows that there is a vast difference between the two technologies and in this regard the Sony is a quite irrelevant precedent.

the argument advanced by many proponents of file-sharing that the availability of “near perfect” MP3 versions encourages sales of the “more perfect” CD has no economic basis. in fact the copyright owners are arguing that it has the opposite effect and there seems to be some support for this position. judging from the popularity of MP3 players, it would seem that the MP3 version is good enough for most people to be a suitable commercial equivalent for a CD (the smaller file size also makes it more attractive.)

it also seems a reasonable assumption is that if the widespread availability of free MP3 copies encourages CD sales, the sales driven RIAA would be among the first to embrace the technology, instead of going to draconian lengths to discourage it. it seems silly to assume that profit driven corporations are driven by any other ideology than profits.

perhaps there is a compromise, perhaps it makes sense for copyright owners to make available lower quality MPx recordings, or good quality shortened versions of songs available over P2P networks, i.e., something that would attract interest in purchasing a legal CD without being a replacement thereof.

what we three mice are saying is that the P2P status quo is not good, and for the EFF to argue that it is, or that greater harm to 10% of legitimate use will be harmed by addressing the 90% of their illegitimate use is simply not being honest about the situation. we believe that a successful effort by the EFF in this matter will not lead to a solution, but rather a continuation of the problem.

we also feel that the EFF’s actions reveal that they are more interested in the redistribution of wealth than in the distribution of information.

February 9, 2004 1:59 AM Joseph Pietro Riolo:

In response to comment made by "three blind mice"
on Feb. 8, 2004 at 2:06am:

Your characterization of EFF is not quite correct.
You may wish that EFF shouts down on infringers,
curses them, and damns them to hell. But, that is
not what EFF is. Allow me to use yang-yin idea.
EFF is more like yin while RIAA is like yang.
RIAA is bossy telling people what to do and scares
them with lawsuit - the masculine thing. EFF is more
like protecting new technologies for non-infringing
uses - the feminine thing. (It just happens that
liberty is represented by woman on some U.S.
coins.) Thanks to them, the copyright world is
well balanced instead of being lopsided.

Okay, I admit that yang-yin analogy does not work
too well. :-)

Regarding your claim that p2p file-sharing technology
is very different from VCR that Sony ruling does not
apply to p2p file-sharing technology, you need to go
back to the time when Sony ruling was made. At that
time, what was the best technology available that
will copy movies? Film. If you compare film with
VCR, you will find that VCR is superior to film in
many ways. Yet, VCR is allowed by Sony ruling.
Fast-forward to the current time, p2p file-sharing
technology is to VCR as VCR is to film.

The U.S. courts and Congress should not let RIAA
and others to hijack p2p file-sharing technology.
If they do, RIAA and others will not just stop
at p2p file-sharing technology. They may go
after not only the new technologies that will
occur in the future but also the current
technologies. That will be a very bad precedent
that is much worse than Sony ruling. That will
not look too good for the future for the common
people.


Joseph Pietro Riolo
<riolo@voicenet.com>

Public domain notice: I put all of my expressions
in this comment in the public domain.

I'll start with a question to everyone.

How do you think that this case will square with the Aimster copyright litigation (334 F.3d 643), were the Seventh Circuit affirmed an injunction shutting Aimster down? There are two things that bother me about this decision:

1. The court rejected the proposition that Aimster was not a contributory infringer because its service was capable of substantial noninfringing uses because there was no proof that Aimster was actually used for anything noninfringing.

2. The fact that Aimster had no actual knowledge of copyright infringement (it encrypted its service so that it could not monitor its users) did not insulate it from potential liability because, as the court put it "willful blindness is knowledge."

Granted, the Aimster opinion is full of law and economics language, but it sets itself up in pretty much diametric opposition to the district court's opinion in Grokster.

And now onto the discussion.

The industry's problem with P2P isn't the system itself. It's with the content. The problem is not the distribution, but the quality of the copies. TBM pretty much stated as such - high quality digital copies are scarier from a content owner's perspective than analog copies that degrade over successive generations.

But the fact that people are using P2P networks to distribute these files is no reason to shut them down. The industry would be better off if it spent more of its time refining the methods it uses to catch infringers on P2P networks (like digital watermarking) and less time trying to sue the network.

The problem is that the RIAA seems bound and determined to have something in cyberspace that it never had in the real world: zero losses. Things get stolen. They get lost. It happens all the time. The fact that people steal CD's hasn't stopped the RIAA from closing record stores until it finds an absolutely foolproof way to stop theft. And yet the RIAA seems insistent on having a bulletproof way to stop online piracy.

Jim Lai

Does anyone know if there were, or if not, why there weren't any arguments put forward on the basis of anti-circumvention measures in the Grokster cases? It would seem that the plaintiffs may have had a more convincing argument if they asserted that such technology was devised as a legal loophole which would then perhaps justify an analysis of % infringing vs. non-infringing uses as a determinate factor. If anyone has any ideas it would be a great help for my exam revision.

Rolex Air King watches made with swiss movement ! so cheap and high quality ! accept paypal ,and 14 days money back without reason !

If you want to buy more ragnarok online zeny, please come my company, we will give your best service.

Welcome to our silkroad gold store. We can provide all what you wanted.

Post a comment

By entering the words in the box, you are also helping to digitize texts that were written before the computer age. The words that you see were taken directly from old texts that are being scanned and stored by the Internet Archive. This CAPTCHA helps proofread the books. If the sample is too hard to read, click the recycle button to get another two. A space between each word is required. And thanks for the comment and help.