Dapper Update

Greg says:

Well, I’m sorry you have resorted to shutting down the debate. I don’t see how I insulted you “for being an American” – I didn’t even realise that was an insult.

The only mention of the US or America in my entire comment was the phrase “the side-effect of the US exporting its laws through trade and other agreements”. No one could deny that this is a fair characterisation of the process of “harmonisation” that follows trade agreements with a more-powerful trading partner. It’s always been that way, and always will. Nothing to take offense at in there.

On re-reading, perhaps it was the reference to the right to be free of torture? I can understand that libertarian-minded US citizens might be uncomfortable about that this week, but “relaxations” on torture seem to be widespread among the democracies right now.

In any case, you have decided to give me the last word, for which I’m grateful.

1) Dapper use on balance. Yes, I’ve read all your comments. And, sure, copyright is fuzzy. But that doesn’t explain why you presume the bad uses outweigh the good. Normally, I’d let something like that slide, but it seems to be a big support for your view that you keep returning to. That’s why I pushed you to back up your claims.

2) Server Load. As shown by the hotlinking example, it’s not a copyright issue, it’s a server-load issue. Any damage done by Dapper can be handled by existing processes without regard to copyright law. It’s got nothing to do with copyright. This is why it’s not a “strike” against Dapper’s opt-out status.

3) Vigilance. You state that “vigilance doesn’t work”. I disagree. It’s not 100% effective, but that’s not the test. (Immunisation “works” even though it’s not 100% effective.) Vigilance is “effective enough” to make opt-out feasible, when backed by the state. It’s the same principle for criminal law and civil law, which are also compaint-driven (and hence relies on vigilance).

4) Oppportunity costs. From your response (“This is no cost to the organization”) I gather you haven’t spent much time in large businesses. Forging agreements is very costly because of the transaction costs (expensive lawyers, endless meetings, negotiations, wasted time, extra staff and risks of not concluding a deal despite huge efforts). We’re not talking about just emailing some guy with a blog, or someone with photos on Flickr.

The real costs are those borne by society on transactions that didn’t happen – all those mashups going unmade, all those orphan works locked up, all that content simply going to waste because the costs of finding the right person and asking to use it outweighed the benefits. Add it up and we’re looking at billions and billions of dollars unrealised. Sure, we don’t see those costs appearing on anyone’s balance sheet, but they’re real. Ask any economist: at the end of the day, all costs are opportunity costs. This, for me, is the real issue with opt-in.

When we started this thread, it was about you badmouthing Dapper and making unsupported claims about their lawfulness. Then we got onto opt-out versus opt-in as both a matter of existing law and a matter of public policy. We haven’t resolved that, but let’s leave it with a thought that we can both agree on, about how the world could be different:

I hope that more people come around to this view and we can see the re-emergence of cultures based on sharing and attribution, not property rights and passive consumption.

JB says:

Greg,

We’re getting to the point where you aren’t even reading my comments. I said this previously. This is not my view, but this is the law. Yes, I do usually agree with the courts decision, but it’s not my logic that is fuzzy, it’s the law itself. I’m going to repeat this.

opyright law was intentionally left ambiguous.

I am making a guess based upon past decisions. It is an educated guess, but a guess.

The decisions are made by men, if you want hard lines, take up trespass law.

The same issue could arise as a result of “hot linkingâ€? images and the like. Aren’t there already technological, social and legal mechanisms for dealing with this? Could not the same protocols be applied to Dapps?

Yes, and people have been held liable for damages to ones server due to hotlinking of images. What’s the point? That’s actually a strike against Dapper, not one for it.

(Correction 10:00 AM: I looked it up and though I found many cases where people voluntarily stopped hotlinking and some even paid out damages for it, no case law exists yet that I could find. All cases have been resolved out of court an in the favor of the original site. I spoke with a lawyer friend who reads this site though and he agrees that image hotlinking would likely fall under the same trespass of chattles and other laws mentioned previously. So the answer is still yes, this is still a strike for Dapper.)

Also, Dapper may not have enough Bandwidth to damage Myspace but A) What about after or if they gain funding and increase capacity or B) Sites that are financially important but much smaller?

A single T1 connection is enough to DOS a site much larger than it, it may not be able to bring it down fully, but it can slow it to a crawl.

Dapper works by scraping the page elements ie HTML, DOM and CSS.

As I said again in my previous comment, hidden areas of a site can be exposed in a browser through other means but might be useless without Dapper’s exploitation. It’s all in the previous comment. This is admittedly a weaker area, but it’s something Dapper has to worry about.

I think owner-vigilance is a good heuristic for determining the cut-off for reasonable effort. If someone can’t be bothered (or justify their own costs of) monitoring their rights, then they shouldn’t be able to burden the state with it either.

Two questions arise from that, one. Why even have copyright if the state won’t protect it? As I said above, vigilance doesn’t work. I mention specific examples where vigilance not working to protect my own copyright and you respond with a philisophical question about it not being an “absolute right” and a petty shot at me for being an American.

I appreciate neither the change of topic nor the insult. You neither asked nor received my opinions on that matter and this is not part of the conversation.

It belittles your case.

Let me answer your two questions directly.

1) In light of the Justices’ remarks, does the Grokster ruling support the view that Dapper being “opt-outâ€? is lawful?

No. Napster offered an opt out procedure and it was found to be liable. It was not safe until it switched to an opt in system. Opt out alone is simply not enough. It is just one, repeat one, indicator. Though Napster predated Grokster, most assume that it fell under the same ruling since it was essentially the same technology.

2) Can the opportunity costs introduced by enforcing “opt-inâ€? (ie from lost content creation/consumption) outweigh the failure of civil law to compensate copyright holders for abuses under “opt-outâ€??

Once again, ask Napster. Second, what opportunity costs? How much does it really cost to ask someone if you can use their material? Nothing. If they had forged partnerships with a few major sites, I’m sure that others would have lined up. This is an issue of lazy marketing as much as anything. Dapper could easily make an opt-in situation work if A) their software provides real benefit to copyright holders and B) They are willing to put in some extra effort.

This is no cost to the organization.

As far as the rest of your statement goes, we live in an opt in world, that much is certain. Opt in versus opt out is an interesting debate I would love to take up, as I see both sides of it, but the fact is that the world we, and Dapper, live in is opt in.

At this point I’m going to say that this conversation is over. Since it has already degenerated to insults. I’ve decided to leave comments open, but will not be responding to any more of your letters. If you want a reply from me, you are fee to ask via email.

Greg says:

Thanks again for responding.

I’m not sure that you’ve grasped my point about “magic ratios”. You keep relying on statements that describe Dapper as “technology that has a very high potential for abuse and only limited legitimate use” or that “most of the [other tech's] uses are legitimate and the benefits far outweigh the negatives. The vast majority of Dapper’s uses are infringing”. You are (implicitly) basing your opinion on your assessment of the amount of infringing or illegitmate uses, or some sort of opaque cost-benefit analysis. However, when I ask you to explicate this, you throw your hands up in the air and say (paraphrasing) “Oh, I can’t explain it – it’s all too fuzzy. All I know is that the courts somehow make a call and I agree with them”.

It’s very difficult to make progress on an issue when you throw an analysis-proof curtain around your decision-making. That’s why I re-iterate: please provide some basis for your views in terms of the underlying cost-benefit/ratio/balance-of-uses or whatever. Appeals to “common-sense”, intuition or established custom don’t cut the mustard. In many ways, it would be more honest to say “I don’t really know why I hold these views; I guess it’s just the law of the land.”

The technical points are really not related to copyright issues surrounding Dapper being a scraper, but concern the uses of other people’s (tangible) resources in a networked world. They apply equally to search engine crawlers and web browsers, albeit to a different degree. But let’s go through them:

Server load. What you’re saying would be true if (and only if) Dapper had superior bandwidth and server resources than the total of sites it scrapes AND Dapper and derived apps made no use of caching. I would suggest both are unlikely. However, if true, the issue of Dapper-induced load from running a Dapp would be equivalent to a developer setting up a page that opened the target site in a frame. (Ie one page hit = one target hit.)

The same issue could arise as a result of “hot linking” images and the like. Aren’t there already technological, social and legal mechanisms for dealing with this? Could not the same protocols be applied to Dapps?

Cracking. I’m still not buying this one. In my view, Dapper works by scraping the page elements ie HTML, DOM and CSS. I don’t believe it can access text or elements inside a Java or Flash or media object embedded in the page. In which case, Dapps do not open up any new avenues for cracking websites that are not already afforded by someone with a web browser.

Damages. I accept that punitive damages may be sought in some jurisdictions for copyright violations. This seems bizarre and a testament to the power exerted by copyright holders, as it moves beyond rational economic compensation and into the realm of using the law as a means of intimidation and oppression. I hope – but doubt – that this practice is stopped.

Vigilance. You argue that vigilance is inadequate to absolutely “protect copyright”. I think you need to remember that copyright is not a moral or absolute right. We’re not talking about freedom from torture (!?) or something. It’s a device – an instrument – to further a public policy goal: advancing progress in the “useful arts and sciences”. It is not an end in itself. As such, it is to be defended where the costs outweigh the benefits. I think owner-vigilance is a good heuristic for determining the cut-off for reasonable effort. If someone can’t be bothered (or justify their own costs of) monitoring their rights, then they shouldn’t be able to burden the state with it either.

There are still two fundamental points that I don’t think you’ve canvassed:

1) In light of the Justices’ remarks, does the Grokster ruling support the view that Dapper being “opt-out” is lawful?

2) Can the opportunity costs introduced by enforcing “opt-in” (ie from lost content creation/consumption) outweigh the failure of civil law to compensate copyright holders for abuses under “opt-out”?

In thinking about whether copyright should be “opt-in” or “opt-out”, I liken it to the question of whether organ donation should be “opt-in” or “opt-out”. Or telemarketing, for that matter. In some places these are “opt-in”, in others they’re “opt-out”. It’s easy to imagine a world where these decisions are reversed, and it wouldn’t result in the destruction of civilisation.

The public collectively comes to a view through the political process, in consideration of the distribution of costs around society. I argue it should be the same with copyright. It’s an unfortunate fact that all recent significant legislative changes to copyright have been about tipping the balance in favour of massive copyright holders. This is due to the intensification and politicisation of media ownership, the new necessity for vote-buying (political advertising) and the side-effect of the US exporting its laws through trade and other agreements.

Copyright should be an “opt-out” system, since it is results in the least cost/harm to society. It seems to me that the Grokster ruling at least leaves open that interpretation.

JB says:

Greg,

First off, I’m sorry about the late reply. For some reason I was not notified about this comment. The second major glitch I’ve had with WP this week. Perhaps I should be submitting bugs.

Anyway, I do agree that we are much more on the same page and, though we might have some disagreements, I think we can work out much of what is going on.

As far as the “magic ratio” goes, you’re making the mistake of looking for hard lines where there are none. As I’ve saiad before, copyright law was intentionally left ambiguous. I’ve studied where the line has been drawn in the past and I can make an educated guess about where it would be drawn but I can’t provide a hard answer.

What would happen is that someone would file suit against Dapper, the matter would go to court and the judge would make a decision basedup up a framework set up by previous judgments. There is no magic ratio and two different judges can have widely different views on the matter. That’s part of what makes copyright law so frustrating and scary.

However, to quote a Berkley law professor on her first day of teaching her copyright class she, a lawyer, told her class that copyright law is “undreadable” and “almost impossible to understand”.

Could this go your way, sure, but I wouldn’t bet on it myself.

I would suggest that Dapps, by their nature, operate periodically and would not contribute significantly to server load. If MySpace fell over and they could convince a judge/jury it was because Dapper hit them once every hour … well, that says something about he courts, doesn’t it?< /I>

The amount of damages would be the earnings (revenue net costs) lost during the period from when the re-publishing started to when the “opt-outâ€? message was enacted. This latency period is going to be a function of how vigilant the content-provider is in monitoring leakage. The more valuable their content, the smaller their latency.

That statement is flat out wrong. In the U.S. as long as a work has been registered with the copyright office (as many serials of value are, punitive damages can be awarded and those accumulate per use, which would mean view in this case. Those can be thousands of dollars per use in some caases.

If the site had opted out already, which copyright defaults as an “opt out” and had a TOS that forbid scraping, the violations would start on the first scrape.

Vigilance is a nice thought, but ponder this. I consider myself VERY vigilant regarding my work. I have Google Alerts for all of my literature, I do random searches on other search engines and am even looking into services to help me protect my work more. However, I consistently discover infringements that began years ago. Simple vigilance does not guarantee that you will be able to detect work in a timely manner, especially if the person doing the infringing sells the work in a locked part of their site.

Vigilance alone is inadequate to protect copyright. It’s a major part, to be sure, but we also need cooperation and part of that cooperation is not releasing technology that has a very high potential for abuse and only limited legitimate use.

Of course the saving grace of Dapper may well be its extremely limited use period. Most people that I have talked to about it say that the service is not as intelligent as previously thought and has severe usability issues.

But as they grow and improve their product, they will run into these issues, that is almost certain. Myspace has already shut down several scrapers and other sites have done the same.

It’s only a matter of time before Dapper steps on the wrong toes. Sad, but true.

Greg says:

OK – so we agree we’re not talking at all about the criminal side of copyright law. Sorry for any confusion about this.

I’m glad we can also agree on an end-goal for what the world should look like. It seems we disagree on how to best get there ie “opt in” versus “opt out” uses of Dapper, and whether or not this is actionable under civil law.

Now, in making a distinction between the knife and Dapper, you state:

The difference between the knife and Dapper is straightforward, the most of the knife’s uses are legitimate and the benefits far outweigh the negatives. The vast majority of Dapper’s uses are infringing or at least bring on copyright questions.

I don’t know on what basis you make these claims. Have you seen statistics about how Dapps are being used? Or do you mean “potential uses”? Does this statement apply to guns and crack pipes too? How do you assess the benefits and negatives of knives and Dapps, respectively? Last time, you said you couldn’t determine this “magic ratio” – yet here you are, once again citing this (hypothetical) ratio as a reason for adopting your stance. Can I suggest you please either a) provide some empirical basis for this “magic ratio” or b) stop relying on it in your arguments. Ditto for your assessment of the proportion of uses that are infringing – your claim that it’s “the vast majority” is not backed up by evidence.

You’ve brought up a reasonable objection to the “opt out” protocol in compensation for damage already done prior to notification. Let me flesh out and rebut that argument.

Server Load. The marginal cost on a webserver of being scraped is equivalent to a browser hit. In fact, most scrapers (including search engine crawlers) are based on browsers and many are deliberately designed to appear as browsers. The issue of scraper-induced load should be handled in the same way as browser-induced load. That is, if I persistently hit ctrl-F5 on a website, it would cause the same damage as repeatedly scraping the page. That website has the right to refuse the connection request, block traffic, maybe lodge a complaint with their ISP, take court action for harassment etc. It’s got nothing to do with scraping and copyright – it’s a web traffic “denial of service” type issue.

I would suggest that Dapps, by their nature, operate periodically and would not contribute significantly to server load. If MySpace fell over and they could convince a judge/jury it was because Dapper hit them once every hour … well, that says something about he courts, doesn’t it?

Cracking. If someone used a Dapp to break crypto or access information that was not public, then that is a crime. It also doesn’t make sense – Dapps (at the moment) can only visit a subset of pages that a browser can, owing to technological limitations. I cannot fathom why someone would rather use a Dapp than their own browser (perhaps via an anonymous proxy) since there’s nothing to be gained by automating the process. I think this one might be a red herring.

Leaking Content. I interpret this one as being of the “you can’t put the toothpaste back in the tube” tpe argument. That is, once a juicy piece of content has been scraped, it can’t be undone because it’s been released “into the wild” and the value is lost. The problem with this argument is that if the leak can be effected by a Dapp, it can be effected by anyone with a browser that implements the dreaded cut’n'paste operation. So, if a website posted an image, text or sound file that was really valuable, that same object could be just as easily (and readily) scraped manually as by a Dapp. Could we expect this “scoop” to be Dapped, but not copy/pasted? Not really.

Showing that an automated scraper made infringement possible (where it was impossible earlier) would be required. Whatever exclusivity value the object had could be dilluted just as easily with a browser as a Dapp. Where’s the incremental damage attributable to Dapper?

A more reasonable scenario for Dapper damage would be as follows: a website regularly publishes new valuable information. It earns revenue through some combination of advertising, pay-per-view, sponsorship, subscriptions, sales referrals etc. A Dapp is used to scrape some of the content and re-publish it elsewhere. The derived work maybe more valuable (perhaps combined with other information or enhanced in quality) or has a lower price (eg fewer ads). Either way, the Dapped service has a reduced audience and hence revenue.

The amount of damages would be the earnings (revenue net costs) lost during the period from when the re-publishing started to when the “opt-out” message was enacted. This latency period is going to be a function of how vigilant the content-provider is in monitoring leakage. The more valuable their content, the smaller their latency.

These costs are going to be a small percentage of total earnings, since they only happen once. (Ie if you use Dapper’s opt-out form, then subsequent Dapps won’t work either.) Is it fair for these (modest) costs to be carried by content providers?

Yes, I believe it is. The reason is that the alternative – a default “no” like with opt-in – is that millions of innovative and legitimate uses simply won’t get to happen because of the administrative burden in enacting agreements (ie transaction costs). These wasted opportunities represent dead-weight losses to the economy, whereas splitting Google ad revenue with another site for a few hours is merely re-distributive (same revenue, just spread amongst different people).

The present arrangements are simply less efficient, but since they are locally optimal for the top 1% of content providers, that’s the law for everyone. This is because they have excessive influence over the political (and hence judicial) process.

Think of it this way. In exchange for a free, uncontested, effectively-permanent, state-backed monopoly (ie copyright), content-providers have to wear these small costs (monitoring plus brief leakage). If it’s not worth their while paying for these costs in defending their property rights, then it’s not worth the state granting the monopoly and it should default into the public domain.

In real property terms, if you can’t be bothered putting up a fence around your wilderness block and checking for/removing campers, it should become part of the adjacent state forest. This is because you are squandering the land and it should go to the highest-value use (in this case, camping).

So, to recap:

Compensation for damages due to server load, cracking and leaking one-off content is the same for Dapps as it would be for someone with a browser. The automated nature of Dapps makes precisely zero difference and no special laws or interpretations are required to handle the case of Dapp-initiated uses.

Losses to providers of ongoing (as opposed to one-off) information can be capped by their vigilance. It is preferable that these monitoring and (brief) leakage costs are carried by providers (and, ultimately, their customers) rather than having a crude blanket ban that sees dead-weight losses introduced across all content provision.

JB says:

Greg,

First off, I can explain why and, if you go over my previous comments, you get an idea. The courts weigh a variety of factors including the potential for damages, the potential for legitimate use, the promotion of the product, the stated intent and more. There is no hard line, but copyright law was written to be flexible.

If you want hard lines, you’re looking in the wrong legal field, just look at fair use. There are no hard lines there, just a framework for courts to make decisions by. The code was meant to be flexible and that’s to make room for new technologies, not block them.

Let me make this perfectly clear. I am not backing away from my statement that Dapper is criminal. I NEVER SAID DAPPER WAS CRIMINAL. You are the only one that has raised a criminal matter in this discussion. All of the cases I have cited are civil and all of the issues I have raised are civil.

I don’t even deal with criminal copyright matters on this site. You are twisting my words around there and I will not tolerate that.

Grokster is a CIVIL case and yes, that is the exact type of statement that would be considered a strike against it. The difference between the knife and Dapper is straightforward, the most of the knife’s uses are legitimate and the benefits far outweigh the negatives. The vast majority of Dapper’s uses are infringing or at least bring on copyright questions. Even their own examples raise issues.

Would the first step in most cases of unwanted scraping be to ask someone to stop? Sure. But what about damages already done?

Say, for example, Dapper scrapes a major site like Myspace and the dapp winds up lagging the server, causing a crash eating up significant resources. It’s happened before to Myspace. Sure, Myspace can ask them to stop, but the damage is already done, they’ve lost business, money and customers because of problems that Dapper caused.

Say also, for example, that a user finds a way to use Dapper to scrape a hidden area of a site. That information gets out and, from the streaming information that is being leaked. Depending on the information or content, that could be very damaging to a business model and asking them to stop once the scraping has begun does nothing. You can’t put the content back.

Since you can’t delete anything once it has been released to the Web at large. Asking someone to stop doing something that you’ve already stated is illegal after they’ve already done it is often times very silly.

Going with the garden party theme, if someone walks off with the utensils, and then gives them away to anyone who was previously interested in visiting my party, it does me no good to track him down and politely ask him to not do it again. Sure, I can get more utensils, but the damage is done.

All it would take is for Dapper to do permanent damage to one site, one host or one company to risk being sued. Dapper, if they carry on long enough, is almost certain to step on the wrong toes.

I hope that more people come around to this view and we can see the re-emergence of cultures based on sharing and attribution, not property rights and passive consumption.

I couldn’t have said it better myself. I completely agree with that statement. I just feel that it should be brought on through the free market system. The artists that choose the route will do better than those who don’t. I have chosen that route personally, I have not opted out of Dapper (then again, I already offer an API and a CC License, one doesn’t need to use Dapper).

I do NOT think however, that anyone should be forced into it, something that Dapper is trying to do. I believe in the right of the artist to choose how their work is used. The best ones will make the right decisions.

That I have faith in.

Greg says:

In terms of the classification of technology, I still don’t understand why photocopiers are in but Grokster is out. I suspect there’s no clear, justifiable reasoning as such; it’s more of a whim based on perceived damages to entrenched interests. By asking for a rational account of the determinations (which can’t really be provided), I was highlighting just how arbitrary these things are. As it happens, you seem to agree with where the courts in the US have drawn the line – but can you explain why?

Back to the question of criminal law – Dapper and copyright. It seems now you are backing away from your initial position that Dapper is engaged in criminality, undermined as it is by the wealth of legitimate uses and the US Supreme Court’s finding of a need to show intent. Where is their intent?

You offer the fact that Dapper advertises itself as letting you do “whatever you want” with the tool – hardly compelling. Advertising for knife sets frequently involves a “utility knife” being used to cut shoes, frozen spinach, stripping electrical wires etc. Are you saying that when those spruikers say “You can use this knife for whatever you like!” that people would assume this extends to murder or robbery or surgery?

Or course not. Such commercial statements are not taken as decrees or permission to break the law. Nor is it the responsibility of commercial operators to educate (potential) customers about their obigations under the law. The common law (spanning several centuries and continents) recognises these principles, and if they’ve been over-turned in one jurisdiction for a few years, then that is an anomaly and an unfortunate one that I hope is corrected.

So, is that it? Can we agree now that Dapper’s operation is not violating criminal law? (Subject to, of course, Dapper-specific legislation or a new legal ruling.)

OK, now let’s move on to the civil law issues. As I’m sure you understand, this is quite a different kettle of fish. This is the realm of contracts, torts and suits, not criminality.

The trespass and TOS violation issues would have to be decided in a civil court. Perhaps at the end of some other form of dispute resolution process. It would be very time-consuming, costly and risky.

If Dapper were scrapping someone who didn’t like it, their first step would be to ask Dapper to stop. Under their current model, Dapper would. End of story. (This assumes Dapper is acting in good faith. Of course, should Dapper ignore the request or if turns out their “opt-out” mechanism is just a lie, then this analysis wouldn’t apply.)

Now, what happens if someone doesn’t ask Dapper to stop, but instead launches a suit under some sort of trespass or TOS violation? Eventually, someone with common-sense – a mediator, a judge, a jury – is going to ask “Well, why didn’t you ask them to stop?”. What answer could they give that doesn’t sound like they’re litigious, unhelpful, motivated by spite, greed, insanity or other negativity? How would this affect their chances of a favourable finding, or indeed a settlement?

Let’s use the metaphor of trespass a little more, and suggest it is actual property (ie not “intellectual property”). Suppose you host a garden party on your land with an open invitation to the public (ie a website). Suppose your land also has pre-existing easements (ie fair-use rights under copyright law) on it. Suppose you wish to restrict where people can or can’t go. If someone wanders off from your public garden party, follows (lawfully) an easement and ends up walking somewhere you don’t want them to, you might call that trespass. (Maybe in the US this leaves you open to being lawfully shot at, yet alone sued – but I doubt it!)

As a matter of common-sense, custom and legislation – are you going to get very far in your case if you don’t first ask them to get back on the path? If the first they know of the issue is a summons in the mail some three weeks later, is your action likely to be successful? Even if you had a huge 5,000 word TOS posted on the front gate, is it going to be accepted by the courts? Maybe, in some cases. Judges can be bought, juries confused. But consistently?

I think it’s more likely that these civil remedies for scraping will only work where there has been persistent abuse after a “cease and desist” type warning. To pursue a matter like this after the scraper has stopped on the first request seems churlish.

For that reason, scraping implemented as “opt-out” is unlikely to be a significant headache under civil law. And I think this practice is reasonable and feasible, technically-speaking. It’s also fairer socially and more economically-efficient than an “opt-in” approach, since it stops people being needlessly denied opportunities to create derivative works.

I hope that more people come around to this view and we can see the re-emergence of cultures based on sharing and attribution, not property rights and passive consumption.

JB says:

Greg,

First off, I am sorry for the late reply, I was AFK for most of the day yesterday.

On the Dapper home page, it has the following statement:

The Dapp produces XML which you can use programatically in whatever way you like.

There is no copyright warning, save a very weak one buried in the FAQ, and that statement seems to indicate that they are encouraging users to use the content “whatever way” they like, including illegal ones. Many will take it that way and a court will most likely put more weight on that statement, and others on the home page, than on the FAQ.

As far as a specific formula goes, there isn’t one. The courts determine that on a case by case basis. They’ve done a pretty good job so far, though it hasn’t been neat. Photocopies and VCRs are not tools of infringement, Napster and Grokster were.

The problem with the law isn’t that they aren’t doing a decent job telling what is and is not infringing, but rather, that there is no way to know when going into it. Dapper, and anything else that uses copyrighted material, lives in a gray zone until specifically vetted or denied safe harbor.

It’s also important to note that there are OTHER legal issues at stake. If you read the other articles I have about Dapper, you’ll see that copyright is the least of the scraper’s worry. The biggest is most likely trespass of chattles.

To avoid a trespass of chattles case, or any trespass charge, you have to be invited in some manner, see also, opt in. There are ways to tacitly invite people, putting up a web page you invite people to view it, but judgment after judgment has found you don’t invite scraping.

Several cases have been handled using this law and precedent is VERY strong. Myspace recently shut down two sites scraping it even though no copyrighted work was being infringed upon. It claimed that the sites placed an undue burden on the server and were a violation of Myspace TOS. Though those cases never went to court, it seems obvious that lawyers encouraged the services to shut down.

That brings me to another potential issue for Dapper: TOS violations. Most large sites forbid scraping in its TOS and, if the courts find those can be enforced, they can be held in violation of those as well.

After speaking with the folks at Dapper and talking about their business plan, the orange tip gun analogy is a fair one. Their end goal is to create opt-in content supply chains. They have no intentions of ever being just a scraper. Their stated intended legitimate use is not hindered in any way by an opt in service.

No, I can not clarify those percentages because they are for the courts to decide. Courts look at more than just a magic ratio, but also at the potential harm of infringements, the usefulness of the legitimate uses, the marketing material and much more. As I said before, they seem to have done a decent job up to now, though I wish there were more clarity.

I admit that’s unfair to Dapper, but that’s the legal climate that they live in.

It really doesn’t matter if you or anyone else takes my torn feelings on the issue seriously. I like Dapper’s end goal and would hate to see them shut down early because of a stupid mistake.

Right now they are wading, very sloppily, through what has been called by lawyers a “legal minefield”. As someone who has studied copyright law for years, I know that they are in a dangerous place and that danger is coming from all sides.

PS: Yes, Dapper is a U.S.-based company.

Greg says:

With due respect to the Justices, I cannot support the US Supreme Court’s ruling that the onus lies on the creators of new technology in the special case of copying devices. However, I gather you are not interested in arguing what the law should be, but rather what the law is.

Your argument about the onus lies relies on a quirk in your laws, historically and geographically speaking. This bold new ruling – and your interpretation of it – is not binding on me as I am in another (common law) jurisdiction. I am also not sure where Dapper resides and to what laws it is subject.

If what you’re saying is true and Dapper is a US company, then it seems that Dapper is in breech. However, unless the ruling specifically put some numbers around their judgment, I don’t see how photocopier manufacturers could avoid being swept up in the same ruling. Ie does the device have to allow you to violate someone else’s copyright at the rate of dozens of instances per hours? Or hundreds? Or thousands? Or only if you have an IP address? I would be interested to know how your Supreme Court limited the scope of their ruling.

But, bad law or not, it is the one that Americans must follow. Including the copyright holders, who would have to show intent on behalf of the Dapper developers. The Wikipedia article on the case quotes the Court:

“…in the absence of other evidence of intent, a court would be unable to find contributory infringement liability merely based on a failure to take affirmative steps to prevent infringement, if the device otherwise was capable of substantial noninfringing uses. Such a holding would tread too close to the Sony safe harbor.”

So, for Dapper (where noninfringing uses abound) refusal to install an “opt-in” mechanism cannot be taken as grounds for contributory infringment liability. Indeed, Dapper’s existing “opt-out” mechanism sounds far more like the filters the industry bodies urged on the file-sharing apps.

Does this not mean Dapper is lawsuit proof, under the existing ruling?

(Incidentally, the orange-tip toy gun thing is not a good example, since it doesn’t impinge on the intended legitimate use. Much closer to your proposal is if the law said each gun must weigh 15kg – thus severely handicapping it as a toy.)

As for the argument that the legality of technology depends on the ratio of legitimate-to-illegitimate uses … I’m not sure how you arrived at the determination of this ratio, nor the appropriate “cut-off” threshold at which technology should be banned. Presumably, under your scheme, guns, cars, knives, VCRs and hydroponic kits all make the cut. But not Dapper. Would you mind elaborating on how your arrived at these two numbers (ratio and threshold)?

Lastly, it doesn’t really matter whether you “want to” like Dapper or their technology. You are suggesting that they either are breaking the law or are a risky investment because they might be breaking the law. You also accuse the developers of being inconsiderate and lacking in forethought. And you are publicly urging them to shutdown their operation if they don’t adopt your protocols.

While you continue to badmouth and run down a company in this fashion, I don’t think anyone will take seriously your claim to wanting to like them or being a part of it.

JB says:

Greg,

There are two problems with what you say. First, the onus is on the new technology. Under MGM v. Grokster and the inducement test, if a new technology that is capable of mass copyright infringement does not take steps to prevent it from being abused, especially if it has the power to do so, can be held liable for said infringement.

If I am throwing the baby out with the bathwater, the U.S. Supreme Court beat me to it by several years.

Second, the legal uses for Dapper, when compared to the illegal ones are insignificant. Dapper is a service designed from the ground up solely for republishing. Does it have some legitimate uses? Sure. So did Kazaa and Grokster.

Governments already pass legislations to prevent products for being used for sinister purposes. Are most toy guns used in robberies? No. But they still have to paint their tips orange to prevent that from happening.

It was a simple step that did little harm to manufacturers and helped prevent many robberies.

My issue with Dapper, contrary to what you say, is not the technology itself, but the reckless application of it. Dapper was initially released without even an opt out feature, it was only after my urging they offered even that. Yes, opt-in would hamper them, but they will be hampered much more when and if a major company files suit against them for either secondary copyright infringement or trespass of chattles.

If the technology has a reasonably large number of legitimate uses, I’d bow out gracefully. If it were geared for personal use, I would do the same. Sadly, neither is true and my conversations with Dapper only proved that.

Finally, the Xerox machine analogy is flawed. A Xerox machine is far too cost prohibitive for massive publication for a work. Dapper makes it possible to republish content immediately to thousands, if not millions, of users for free. The threat is much higher.

In the end, I want to like Dapper. I really do. I think the technology is cool and would love to be a part of it. However, I can not endorse such reckless behavior for the sake of “cool”.

Dapper is neat and I think they’ve got some great long term ideas for the technology. Sadly, I think that the current execution is unnecessarily risky (considering their future plans are completely opt=in according to them).

In short, I don’t have the technology, I don’t even hate Dapper, I just hate the lack of consideration and forethought that went into it.

Greg says:

I disagree that the onus is on new technologies to prevent illegal uses, as long as it can be shown it is intended for lawful uses. This is the nub of it. Should it be a pre-condition of releasing knives that they can’t be used in robberies? Of course not. Should it be illegal to buy a bong because someone might smoke pot with it? Of course not.

Your wrote that “Dapper, with a simple change, could have stopped all bad uses of their application but didn’t.” That same measure would also stop a large number or lawful uses of Dapper too, throwing the baby out with the bath water.

There are at least two reasons why it’s lawful to have an opt-out approach: firstly, the scraping might be allowed under existing copyright rules, for example for research, review or other “fair uses”. The second is that content might be licensed in such a way as to encourage scraping, yet technical issues, lack of knowledge or simply administrative problems prevent the owners from opting in. I do not need the copyright holder’s explicit permission to make copy a few pages from a book or record television – it’s an “opt-out” sytem.

The opt-in suggestion you make would put Dapper at a considerable disadvantage to the next scraper service that didn’t artificially restrict itself in this way. Why should they hamper themselves unnecessarily? Would Xerox have released the photocopier had they throttled its rate to one page per hour? Or would they have been gazumped by the next people who removed such a pointless restriction?

You seem solely concerned with protecting the property rights of established businesses and entirely dismissive of the “opportunity costs” presented by a blanket ban of new technologies that might be used to violate those rights.

JB says:

Greg,

The difference between a video recorder and Dapper is simple. The video recorder made it simple and cost efficient to make ONE copy of something. To make hundreds, or thousands, would take too much time and cost too much money.

Dapper inherently makes it easy to republish content to thousands, if not millions. It is not designed solely or even mainly for personal use.

The question isn’t whether the technology is bad, but rather, why Dapper didn’t make the service opt-in. Given that Dapper has the potential for horrible abuses, it makes sense to take reasonable precautions BEFORE releasing it.

This isn’t about simply copying work, but republishing it. This isn’t about a “bad” technology, but rather, a careless application of it.

Dapper, with a simple change, could have stopped all bad uses of their application but didn’t. If past history of scraping is any indication, a majority of uses will be illegal in one way or another.

Even if you apply the “net benefit” test, Dapper will likely fail, especially once the sploggers take advantage of it. It could have easily been avoided though, if Dapper had been willing to take an extra step.

That’s the sad thing.

Greg says:

Your argument boils down to “Sometimes scraping is bad, so scraping should always be hard. These guys made it easy, which will lead to more bad things. Therefore, the whole thing is bad.”

Technology has changed the economics. We’ll need new customs and practices to deal with this. Laws will just have to catch up.

“There are also no answers as to why they released such an open and dangerous service before adding reasonable safeguards against abuse.”

You could be talking about the cut’n'paste operation, video recorders or the photocopier. Society has survived all of those things, and will survive Dapper too.

More generally, progess should be a held up to a “net benefit” test, not a “do no harm” test.