3Taps Files Countersuit, Says Craigslist Is Anticompetitive | Hacker News Reader

3Taps Files Countersuit, Says Craigslist Is Anticompetitive

techcrunch.com

53 pointssbashyal61 comments

I read the core of 3taps' counterclaim — pages 25-30 — and I recommend it. It's a good read.

Specifics of 3taps' scraping aside, I'm starting to see that this is a particular case of a general problem that also applies to Twitter, with its recent API changes. They and Craigslist each control access to an extremely large and growing volume of user-generated content that people post there with the intention of making public. Users post that content to Craigslist/Twitter primarily because each is the largest website of its type. And both sites are, increasingly, trying to preserve their control by restricting third-party access to the UGC inside them, at the cost of reduced innovation.

It's worth dwelling on the fact that large, powerful corporations can be built off the backs of user-generated content in this way. What do you do when a service like this adds a term in their TOU that you don't like? When their web UI is lacking basic functionality? Runs really slow? Doesn't work in your favorite browser? In any of these cases, you hit a big roadblock, because you don't really have a choice. The content is on Craigslist/Twitter — even though these companies did not create it.

This problem seems to be one that will recur over and over on the internet. I can see three main avenues to fight it:

1. Technological — Create decentralized services that are competitive with, and somehow eventually overtake, the centralized ones. This is what StatusNet, Tent.io and several other efforts are trying to do in the social networking space. But this is hard. Decentralization makes it much more complicated to even build a working service at all, let alone one with a consistent, smooth and intuitive user experience. And success would result in less profit, compared to solving the much easier problem of building a successful centralized service. So developers who could pull this off are strongly incentivized not to.

2. Organizational — Start nonprofit organizations that are committed to interoperability and innovation in their bylaws, and have them run centralized services. This has been done successfully at least once: Wikipedia, operated by the Wikimedia Foundation. Their nonprofit status doesn't mean everyone is happy with their management, or that Wikipedia is perfect. But they provide full data dumps, and the content is under a free license, and their Statement of Purpose (https://wikimediafoundation.org/wiki/Wikimedia_Foundation_by...) would make it hard to justify changing that. So, no matter what, you can always innovate with content on Wikipedia.

Wikipedia was the first site of its kind, so it got big the same way major commercial sites did — by having the right idea and execution at the right time. It would be hard to replicate that success in 2012 by creating a nonprofit-owned Twitter clone. But now that Wikipedia is established, it seems to work.

3. Legal — Just as there are privacy protections in the law (moreso in the EU than the US), so could there be regulatory requirements for web services over a certain size to allow some degree of interoperability.

If we don't find some kind of solution, I imagine that in 5 or 10 years, large incumbent internet sites like Twitter, Craigslist and heck, let's throw Facebook in there as well, could be the new RIAA and MPAA, using the aggregate value of their content to bludgeon competitors into submission and restrict innovation, to the detriment of the public, except that in this case, it's that very public that created the content enabling these companies to bully others in the first place. Craigslist v. 3Taps is not the end of this.

I hope that 3Taps follows through. This countersuit may help set a clearer precedent for who owns the data posted on public forums and what third party developers should be allowed to do with it.

Earlier this year I was searching for an apartment in San Francisco. Craigslist was my only source for places to apply. The good apartments got snatched up pretty fast so I wanted a way to make sure I saw every listing that came through. The problem: there are easily 500 new listings added every day, and approximately 20% of these were re-posts of the same place. It's very difficult and draining, just trying to keep track of what you've already seen and where you left off since the last time you checked.

So I did what any self respecting iOS developer would do: I made an app. It worked pretty much like an email inbox. Each new listing was like a new message. Refreshing the feed was like checking your email. Reposts were marked. You could bookmark and store things. The app colored the message cells based on whether they were from the most recent import or not (i.e. the app keeps your place).

The best part? It worked pretty damn well. I gave it to a few other people via Test Flight and they were hooked as well.

So, again, I did what any self respecting iOS developer would do: I put the app on the App Store. I figured that since my app was exclusively a craigslist client - a tool to make using craigslist better - I might have a better chance of not incurring craigslist's wrath. I put it on sale for $3.99 (it's called Craigslist Commando, if you're curious) and hoped for the best.

The app did pretty well given that I did zero promotion for it. It was brining in around $30-40/day after Apple's cut. Unfortunately, my hopes for escaping craigslist's wrath did not pan out. I received a cease and desist letter from them two weeks ago. I'm still waiting on hearing back from them. They've gone silent on me after our initial contact in which they told me to expect a draft licensing agreement. I'm guessing they went silent because they have bigger things on their plate right now. I have opted to keep the app in the store during this time, but I'm not very optimistic that craigslist's licensing terms will be tenable.

All of this to say: I have a pony in this race. I respect craiglist's rights to do what they think is necessary to protect their business. However something about this just doesn't jive for me:

a) People go to craigslist and create content about some aspect of their lives. They, as the original authors of the content, post it to craiglist with the intention that the general public can see it.

b) Other people go to craigslist to view content posted to a public forum in hopes of finding something. This process is frustrating and difficult.

c) I create a tool that makes the process of finding something on craigslist less frustrating and difficult.

d) Craigslist threatens to sue me for creating the that tool. This despite the fact that I'm helping both: the original authors of the content who would like to have their posts seen by their target audience and, the viewers of the content by making the viewing process more efficient.

I suppose craigslist has these rights. But it just doesn't seem right.

> a) People go to craigslist and create content about some aspect of their lives. They, as the original authors of the content, post it to craiglist with the intention that the general public can see it.

Craigslist users only agreed to have their content displayed on the Craigslist site, not on any other random site on the Internet who wants to relist their posting without the users explicit permission.

Yes, but it's definitely true that they understand that to mean public, and it's almost certainly true that they would rather have more eyes than fewer. It sucks that that isn't a legally compelling argument and this guy will wind up being shut down though.

I see what you're saying and that may technically be true.

I'm not really arguing the legality of the issue though. I'm just talking about the spirit of what the content authors were trying to accomplish which I believe is: have some random person on the internet see their post and, subsequently, arrive at some sort of mutually beneficial arrangement with that person.

There's any number of places on the net where someone could post content and "have some random person on the internet see their post" ... but they continue to choose Craigslist because it's got the audience. If they want more-broad usage rights for what they want to post, they're free to chose an alternative platform on which to post.

This argument is what the monopolist uses when they are faced with regulation. It isn't correct, because the user _isn't_ really going to be able to post on another forum.

This is just like if telcos say 'go to some other provider if you don't like my terms and conditions'. except, they have a monopoly.

Just because CL has a large user base doesn't make it a monopoly. CL can go the way of MySpace if a better competitor comes along. There isn't any barrier to entry to create your own CL + Padmapper type listing site that requires some sort of monopoly break up.

The hard part for you is creating a compelling enough competitor 15 years too late and also to make it worth people's while to start using your site vs. CL.

> The hard part for you is creating a compelling enough competitor 15 years too late and also to make it worth people's while to start using your site vs. CL.

which is essentially why CL has a defacto "monopoly" - there is no way someone can come and make a different CL.

That was CL's big mistake. These sort of unreasonable licensing terms that users never read are exactly the kind courts refuse to enforce.

Maybe you just want CL to win? That's different from whether CL has a real chance of winning. This case is unlikely to go to trial. That's because CL has a lot to lose and 3Taps has very little to lose. Some believe the reason there are few lawsuits on these issues that are not settled out of court is because websites like CL are afraid of having a court decide on them. Why would they want to avoid having a court decide?

I think you're talking in two directions here.

'any other random site on the Internet' is hardly the same as a single-purpose (native) app for making CraigsList listings easier to parse. I'm sure that most of the random Internet sites might not even be attributing the data back to CraigsList.

When it comes to apps like this, all Craigslist has to do to compete is make their UI for browsing/tracking listing better. If their interface was better, then people wouldn't feel the need to pay ~$3 for a separate app to browser their listings.

I fundamentally disagree.

The users know CL is sharing it with the world and that's what the users expect. As the article says, CL is happy for search engines to access these. Just obviously not search engines that they don't like.

You don't get to pick and choose who accesses your website unless you put a password up. And if the content's not copyrightable...

If I place an ad on craigslist, I would like to know if my ad will show up on...SanFranGayApts.net or something like that too. Where did I agree to have MY ad syndicated? Nowhere, that's where.

And why would ad copy not by copyrightable? I don't see a creative commons license at the bottom of craigslist ads.

I'm not sure where all this entitlement comes from that you think you have a right to another company's database simply because it's index is public.

Craigslist's copyright claim is likely indefensible. Read the article for the reason why.

If you wanted your personally identifiable data protected from public access (being cached by search engines), then presumably you would not submit it to a web site that is open to the public (not password protected).

If Craigslist changes its name or sells its SF apartment listings to some site with a name you do not like, what can you do? Answer: Nothing, because you gave them an exclusive nonrevocable license to your data. If you really wanted control over your personally identifiable data then it would probably be wise not to submit it to a site that gets scraped and cached by Google (from which anyone can make fair use) and certainly not under a license that gave you no control over the data after it's submitted. You would want a site that has some sort of protections like passwords, that allows some degree of pseudo anonymity so your name is not embedded in an evergreen search engine cache; and you would want the right to tell the site to remove/delete the data at your direction, if that became necessary (e.g. in your example scenario).

In short, Craigslist really isn't in the business of protecting your data for you. Its terms are what they are because it's seeking to protect its own business, which relies on your data.

As a regular Craigslist user (buyer, seller and message board poster), my personal preference is to have material I post only accessible directly from CL. In other words, I do not want my classified ads, contact information contained therein, pictures, social banter, business ads and other information available to any external information service or website.

This is my expectation not only of CL, but of many sites. I understand that others may disagree. But, I suspect that my expectation of non-propagation of material is the default expectation for many or most sites that accept user submitted material.

If CL is somehow forced to let others scrape their content, I would welcome an option to turn on a switch that will block/hide/protect the material I contribute from external entities. On second thought, I would rather have the do not share option be the default and let me opt in to sharing everything always, share by material classification or one post at a time.

I would expect that many users would feel this way.

I can guess why. But could you tell us specifically?

Assume for the sake of the question that whereever the data may appear on the web, it would always have a "Source: " line to indicate its original source, i.e. the site to which you submitted it. Assume that the integrity of the data could also be verified, e.g., through a digital signature.

A few reasons came to mind when I originally wrote it. I know that there are many tens of other reasons why it would make good sense to keep CL content CL-exclusive.

Of course, I realize that this must be tempered with the realization that there are situations where scraping CL and re-posting the information on another site may make sense for the original poster (e.g., a hotel ad).

Here are a few of the negative items that come to mind for just classified ads:

- When I complete a transaction and delete a CL ad, I do not want other interested parties calling, texting and e-mailing me hours, months or years afterwards

- I already dislike the 10-to-1-ish ratio of SPAM/scammer/nutjob banter to communications from legitimately interested parties. I suspect any proliferation of my postings beyond CL will only tend to increase this ratio.

- The text and images within my classified ads may be more easily accessible to scammers who just need some content to farm and post elsewhere (e.g., other classified forums, eBay, AirBnB, etc.)

- Sites which do the automated scraping of CL can directly be or can become a target and/or searchable repository for intelligence gathering and source for creating new or expanding existing personal identification databases

There are countless other scenarios ranging from good to bad regarding this subject. For now, I'm convinced that the potential for the bad, the incompetent, the annoying, the negligent, the criminal and other negative scenarios outweigh the good. This is really meant to apply to how I use CL. As I mentioned before, there are undoubtedly specific scenarios where many of the potentially negative items would be moot for specific CL users.

Obviously, there are other factors here too such as the commercial viability/legality/precedent of allowing non-CL entities to use CL resources with or without restriction to directly or indirectly profit. From my perspective as a user, it's not really that important other than an arguable matter of principle.

I think most Craigslist sellers want as many people as possible to see their listings. Your criteria exclude Google, which indexes every publicly available user-generated content site and thereby benefits Craigslist sellers.

People should post their cease and desist letters from CL so others can see. I doubt you were the only person to receive one. Sometime when something doesn't seem right, it's because it isn't.

I also received a C&D two weeks ago via email and FedEx for my craigslist interfacing iOS app. I also got an offer for a license but haven't heard anything since Thursday.

Hey, could you email me? hi@adamawolf.com

Us little guys need to stick together ;).

> it's called Craigslist Commando

Issues over content ownership aside, calling it that is a pretty clear trademark violation.

You are not entitled to scrape content from other sites, even if they contain publicly submitted data. I do support this sort of thing in the hacker spirit (as long as you aren't being too agressive) but selling an application using these techniques is downright unethical.

You have zero control over what you are selling to people. They don't know that you aren't using an offical means to get the data and the application could stop working at any time. And if you were scraping via the client (as opposed to collating data on a server of your own) you also put your customers in a position where they might be violating the site's TOS and risk reprecussions (ip ban, etc).

> You are not entitled to scrape content from other sites, even if they contain publicly submitted data.

Isn't this Google's business model? They take snapshots of the public web and show organized and relevant snippets. Last time I checked, they never asked me for permission.

And they can be asked to remove listings from their search. Ask rupert Murdoch

They don't ask your permission, but they do respect robots.txt files as well as noindex, nofollow.

I am sure the orignal author does as well. This is more a question of "fair use", not robots.txt

Indeed - and i believe scraping data off a publically accessible webpage (e.g., one where you do not require registration and login/password) falls under fair use, provided you do not take up more bandwidth/resources than the average user of that site.

I recall there being some precedent for this sort of fair use - something like a phone directory - the information is not copyrighted, but the arrangement and layout is. So hence, you can't just iframe a site and present it, but obtaining the data, and deriving a new work from it should fall under fair use.

That court case was Feist v. Rural. The court ruled that copying the listings in a phone book was permitted because facts cannot be copyrighted.

I'm as much about free access to data as anyone, but holy shit do you have a warped view of the world.

Craigslist doesn't have to pay money to run their company and users don't have to use Craigslist.

But because Craigslist does and the users do, doesn't give anyone the right to scrape this data. The individual user owns their content, but Craigslist owns and runs the servers and access to this data for third parties is not a right -- it's a biz dev deal or partnership.

If the world worked how you want it to, no one would run any services because the load generated from scrapers would outweigh real user usage by a factor of 100x. And your competitors and parasitic freeloaders would just DoS you into monetary ruin.

You're not entitled to anything, and Craigslist doesn't owe you a business model.

TL:DR: This lawsuit is going straight into the trash.

The irony is that if Craigslist would just take the time and effort to add a bit more functional / elegant interface, these third parties would have no reason to exist, and all the legal matters would disappear

I understand that Craigslist wants to "Serve their community" and believes that changing their interface would interfere with that mission somehow (both, in terms of having to charge more in order to hire UI people, as well as engineer their Ops environment to handle additional load), but, I have to believe their is some way of simultaneously serving their community, providing a rapid response, as well as keeping their cost and overhead down while at the same time slapping a somewhat more elegant GUI and Search Interface.

It's just fear of screwing up a winning formula. Craigslist is making money hand over fist; they could hire some front end guys if they wanted to.

Well, they are also concerned about (A) Increasing their overhead, and (B) screwing up their uptime by introducing complexity (requiring even more investment in Ops) and finally (C) slowing down their responsiveness (once again, requiring investment in Ops, and Hardware, and engineering to get back that speed)

They really buy into the KISS principle over there, and it's worked well for them, but, the industry has moved forward, and it is possible to (affordably) add design and search flexibility without screwing up your uptime/responsive time.

Direct link to the counterclaim: http://3taps.com/papers/3taps-answer-counterclaim.pdf

As a developer who received a C&D from Craigslist for a pretty innocuous app (seriously, it was built in a weekend), I'm glad to see someone test Craigslist's legal position in court.

At the very least, I hope it will expose that Craigslist is not just a happy, hippy-infused neighborhood listserve like many seem to think; rather, like most other companies, it actually goes to great lengths to quash threats to its very profitable business.

If Craigslist can assert copyright protection over classified ads and can make people grant exclusive licenses to the ads they submit, then why didn't newspapers do this before the web? Shouldn't there be a nice line of precedent for Craigslist to cite in its Complaint?

I think those web developers who actually think they "own" UGC may be in for a rude awakening if something like this ever goes to trial. As a UGC site they are providing access, and that's all. There may soon come a day when users will not need third party websites in order to provide access to data they want to make public. Web developers should consider themselves lucky if they are running a UGC site and managing to make people believe they, not the users, own the content. Cease and desist letters alleging copyright infringement claims that have no legal basis (where is the case law that says anyone can copyright classified ads?) will only work for so long.

This is like scraping Amazons reviews since they are available via Google and claiming you can display that user submitted content how ever you like. I can't see 3Taps having much of any case here.

I don't see how 3Taps is any different a client than google is. Now, i dont know enough details, so i m not saying 3tap has a case. But, if they scrape data off craigslist, not via their api (to which they agreed to abide by terms and conditions), then they _do_ seem to have a case!

Is 3Taps different from Google simply because they show a map?

Maybe CL should have terms for their general website that say data can be scraped and cached but that it must appear in the cache either exactly as it does on CL, or with some things missing, but not with anything added (e.g. a map).

3taps isn't a client application for Craigslist, they essentially turn CL's data into a feed and then sell it to folks who then create services on top of that data.

Padmapper is a client of 3taps who's been the client app that most of this is focused on.

BTW 3taps doesn't scrape CL, they actually pull CL's data from Google's Cache so they can't be accused of violating CL's TOS.

Yeah, it would be futile for CL to go after Padmapper, so they are going after 3Taps. Why would they be more worried about 3Taps? Because anyone can do what 3Taps is doing. Anyone can pull CL data from a search engine cache like 3Taps does. There's no need to go to CL.

But one thing we know, if you look at the precedent, is that you can't sue Google and win. You either let them copy your site or opt out and lose the traffic. So what can you do?

Sue Google users! As far as I'm concerned, that's all 3Taps is. They are just aggregating what they pull from the cache.

http://en.wikipedia.org/wiki/Web_scraping#Legal_issues

Hard to say what will happen. They will sure as shit get sued, but what will happen is up in the air as far as I can tell. If I remember correctly, if you are somehow changing or making new content from the data, then it is protected under fair use somehow.

It's not at all like that. The interesting thing seems to be the claim that Craigslist posts lack creativity, which Amazon reviews certainly do have. 3Taps is claiming the posts can't be protected by copyright, because you can't copyright facts. I think their case has merit.

Are 3Taps verifying by hand that every scraped post are only facts? I've seen some pretty creative CL posts. Also it's been mentioned that assemblages of facts can in fact be copyrighted. See the Farmers Almanac.

You're right about that, and I think that's going to make their claim a bit unsteady. But I see an argument they might make:

1. Classified ads are facts, with negligible creativity put into their composition. 2. Facts are not copyrightable. 3. Therefore, 3Taps can scrape the ads. 4. If any ad actually does have creativity / an applicable copyright, the copyright holder can contact 3taps with a complaint.

Basically a "safe harbor" take on the whole thing. What do you think?

Feist v. Rural (http://en.wikipedia.org/wiki/Feist_v._Rural) seems to indicate the the arrangement of the facts is copyrightable, (if there is even a minimal degree of creativity) even though the facts themselves are not. The court held the alphabetic arrangement of names is not sufficient to warrant copyright protection, and it seems logical that mere temporal arrangement of classifieds would not merit protection either. However, it seems as though 3Taps would need to find a way to rearrange the facts presented within the advertizement, because that certainly is a creative process.

The seller has the freedom to write text a description and include pictures. Photographs have a long history of copyright protection, as does written work. Simply copying the specification sheet for a particular model of TV into the ad, for example, might not be copyrightable, but how does 3Taps sort those from the rest?

I just stole this from a random Craigslist ad:

"Black futon in good condition. The mattress is a lot thicker than most futons. It is pretty easy to assemble. You must be able to pick it up though because I do not have a truck. I am moving in a week so I need to get rid of it, asking for 100 or better offer. Please use the link above to email me. "

That the object is a futon, that said futon is black, thicker than average, easy to assemble and in good condition are all facts. The seller could have presented those facts in any number of ways, but he or she chose this way (proper spelling, for example, and mostly written in complete sentences) because he or she thought that would generate a better response. The only way I can see 3Taps on solid ground is if they can take the ad text, use it to generate a set of facts, and rewrite the ad from that (analogous to the way the PC BIOS was reverse-engineered, but done mechanically (I can't imagine it would scale well to have humans rewrite the ads))

Excellent points, especially regarding photographs. I was thinking of that case already, and hadn't considered the arrangement of facts within a post; what came to mind first was the collection of posts as their own arrangement of facts.

My understanding was that, as the owner of a site, (even one that hosts mostly/completely content created by someone else), you get to dictate the terms on which any party accesses that site. It would be like if YouTube said that you may not access their site with anything but a web browser.

Strange that they went the copyright way, though. IANAL, but isn't accessing a service you've been explicitly forbidden to (via ToS or similar) a violation of the CFAA?

3Taps doesn't access Craigstlist so the TOS has no effect.

You are 100% wrong. The ad copy on craigslist IS copyright protected so it doesn't matter where you scraped it from. TOSs do govern the data whether or not you are viewing a cached copy or not. By your logic all TOSs are rendered useless if you browse the net via proxy servers.

You can scrape anything off the internet you like and do as you please...but you can't create a business out of it.

It is not black and white, despite your assertion. It should be tested in court.

"You can scrape anything you like off the internet and do as you please... but you can't create a business out of it."

I tried to tell this to the Google guys in the 1990's but they didn't listen! :)

Site TOS don't really matter too much or else Google could never exist. They're mostly to limit the liability of the publisher and not to prevent anything with the accessor. Googlebot is not a lawyer, it can't decide the legal implications of scraping or what exact activity can be done with the data it finds.

Google also overlays facts it finds by crawling the web in maps (Google Places and the One Box results) and uses the creative contribution of other people to provide things like reviews inline on SERP. Google is a bad target to sue though because they will punch you right back in the face.

I don't know of any precedent that gives sites that accept user-generated content grounds to sue when their users' copyrights are violated. I would like to hear of one.

Seconded.

Certainly this issue has come up before. Maybe the fact there is no precedent tells us something? Who would benefit from keeping the issue undecided?

If you're scraping user-submitted content by accessing (e.g.) CraigsList, then I would think that CraigsList could try and claim that it was unauthorized access of their systems.

Agreed, but my understanding is that 3Taps pulled the listings via Google cache.

People with the pitchforks seem to be forgetting something..

3Taps is not scraping Craigslist. Google is scraping Craigslist, and 3Taps is scraping Google. As far as common sense is concerned, Craigslist has no say concerning 3Taps data.

Why not? The primary question here is whether 3taps is engaging in copyright infringement as Craigslist claims.

Does everything scraped by Google becomes public domain? If Google News scrapes an article from the NYT, can I then scrape the article from Google and republish it?

If 3Taps wins, the gold rush is on to create CL interfaces using 3Taps' API. The major roadblock becomes images. Image URLs are provided in 3Taps' API but since no one can hotlink to CL, images are useless. How do you build a better CL interface if you can't use images?

Why do you need to hotlink? What is one more click?

    <img src="http://example.com/1.jpg">Photo</a>

No.

    <a href="http://example.com/1.jpg">Photo</a>

Yes.

Reply on news.ycombinator.com