So, craigslist experiences some blowback..

[TonyL]

Oh, and the reason for craiglook having to change it's name? "Craig" in the title. It's all about brand dilution, it's infringing on his brand, and contributing to confusion, people might think that Craig has something to do with it, and he doesn't. Hence lawsuity time.

[Tony_SS]

Oh, and the reason for craiglook having to change it's name? "Craig" in the title. It's all about brand dilution, it's infringing on his brand, and contributing to confusion, people might think that Craig has something to do with it, and he doesn't. Hence lawsuity time.

I could understand that.. but I think they should have just changed the name and left the search feature. One of their claims in their C&D letter was that Craiglook was putting and extra load on their servers - which is BS since it only searched cached RSS feed.. so it did not even touch their servers.

[wmhjr]

I could understand that.. but I think they should have just changed the name and left the search feature. One of their claims in their C&D letter was that Craiglook was putting and extra load on their servers - which is BS since it only searched cached RSS feed.. so it did not even touch their servers.

That is not true. The SEARCH may not have put extra load on. However the hits on the SEARCH RESULTS would. In other words, CL likely has their hosting design based on the assumption (and intent) that CL is "local only". However, the larger search indexing would result in people for example from NY starting to look at ads from TX. That's more volume, in an area that CL does not want. That would increase requirements for I/O, CPU, Memory and bandwidth. They DO NOT WANT people from NY searching all over the country. That is not the intent of their service, and they've been specific and clear about that.

I understand your position Tony, but it IS their company and their own proprietary property. It is their data. I know you're a private enterprise kind of guy and it surprises me a little that you have a different opinion here. No insult or criticism intended - just an observation.

Bottom line is that again - IT IS THEIR COMPANY AND THEIR PROPERTY.

[Tony_SS]

I am pro-private enterprise and believe in intellectual property, but in regards to information property, that's where things get into a gray area. There have been absurd allegations and lawsuits from a law firm that makes it's living off of suing online forums, like this one, for someone posting links and copy from other various news outlets. Read more here. And here.

There has been much debate about information intellectual property. Other than using the name 'Craiglook' I don't believe they infringed on craigslist, based on what I know. We don't have all the details about situation here. Obviously it appears they were singled out, as others still exists, so I have to question that. We can only speculate on a lot things - no sense in debating the unknown unless we have some facts to go by.

But I believe there is a distinct difference between stealing and sharing information. Fair use exists for a reason. It keeps the copyright & lawfirm trolls from shaking down everyone and everything for their own greed.

[wmhjr]

But Tony, your position is exactly opposite of one supporting private enterprise. "Craigslist" is a for profit business having a product with brand value, and the fundamental value of their product is with the data/information that people directly input into Craigslist. Without that data, their product is worthless. In order to differentiate them from the spam filled "global" or "national" players, they have instead consistently concentrated on making their product "local". They also have built an infrastructure designed to handle expected load based on the use cases and characteristics for which the product was designed. So, when somebody else comes along and attempts to take that data, which directly violates the CL temrs of use, it's a far different case than what you describe above. Trying to make CL out to be like Righthaven is ridiculous. CL isn't trying to gain revenue by use of patent enforcement. They are simply trying to protect their own product, which in case you forgot, is FREE to almost everyone.

If CL took data from multiple sources it would be a very different story. However, because CL has marketed itselt first and foremost as a "local" and "community" service, and expressly forbits any data entry that is not manual by the user, then the nature is very different. Again, in the terms of use, CL has for a long time expressly forbidden data mining of its data by external parties without the express permission of CL. To suggest that data mining of CL is "fair use" is more than a stretch. I can tell you that it does not legally hold water. It is THEIR data. Were they to choose, they could restrict viewing to paying customers. Fact is, CL has been extremely open about access, providing a service at almost no cost. Asking them to open themselves up to even more operating cost is unreasonable.

Bottom line is that participation of CL binds ALL users to the Terms of Use, which expressly forbid data mining, meaning that factually Craiglook vioated the terms of use and infringed on CL. It may not be convenient for you and I, but it is that simple. You have to ask yourself, what evil reasoning can you ascribe to this action? How exactly is CL acting like Righthaven? Where's the beef? What money did CL get for enforcing this? It's like Occums razor. Once you eliminate all the other reasons, the obvious comes to mind. And in this case trying to make CL look like the bad guy is going to be pretty tough to do.

[Tony_SS]

So, is google 'data mining'? I'm sure I could find a lawyer who thinks they are. I'm not trying to make craigslist the boogey man - I don't think they are. I'm saying that whether or not Craiglook (now Claz) infringed on craigslist claimed property is highly debatable based on the what search engines do. This is nothing new. I understand you see it differently.

[wmhjr]

First of all, Google does not focus specifically on one information source. Second, Google does not allow you to preset parameters which exactly match the user interface of the core technology like these specific search engines are. If you are familiar with how search engines such as google, yahoo, etc operate then you must admit that frankly they operate completely differently - frankly for one reason so that they can avoid this specific type of problem. Craiglook specifically and solely targes ONLY CL data. That is something completely different. It is in fact data mining and in no way, shape or form resembles how the search algorithms function with actual "search engines" such as google. And I mean in absolutely no way whatsoever does Craiglook (or whatever other ones exist) resemble true "search engines". None. Google, Yahoo, etc use very complicated (largely unpublished and VERY proprietary) algorithms to generate result sets with rankings based on a number of specified and unspecified variables. Craiglook simply uses the exact metadata, and passes the request to get results that have been generated by craigslist resources having completed indexing on a ton of data. Data that they solely own.

But the real problem you keep missing is that this is private data, owned by CL, and managed by CL, with specific, published terms of use that explicitly prohibit this behavior in order to protect both their brand, and their intellectual property. It costs them money to provide this service, and they have certain methods to generate revenue to offset that. Without that data and that user interface (UI) they have absolutely nothing. They believe (probably with some merit) that their brand is uniquely "local" and want to avoid "national" or "regional" activity. I have no issue whatsoever with that. Furthermore, many of these "hack" search tools do NOT simply parse RSS feeds. They attempt to use bots and spiders to gather index data. But even the ones like this one that may use RSS feeds also still add load to CL systems - in 2 ways. First of all, it is inevitable that if you're parsing RSS feeds, the content will be dated, and you'll be returning outdated results - meaning that entire tables are scanned potentially if a user clicks on one of those links. That takes processing power, and potentially memory because obviously the data will not be cached. Beyond that, as I already mentioned CL explicitly does NOT WANT multi-region searches. When Craiglook returns a result set, actually selecting one of those results does not get run against an RSS feed. It then gets run against the production system, meaning more resources are being used than designed, intended or budgeted by CL based on their INTENDED service offering. Therefore, such use generates increased cost because they have to invest in order to provide adequate performance to all, even though some of it is being consumed by what are effectively "non-customers".

That to me is not acceptable - that is, to force CL to spend money so that somebody else can use their data in a manner that their branding explicitly opposes.

[wmhjr]

BTW, just for the record I want to explain why I'm spending time on this. I'm kind of in this business. I'm retired from my first career (active military) and working on my second. I run a group which focuses on delivering what can best be described as IT enabled services. Part of that is designing and hosting commercial, shared service, single instance systems. Unfortunately, a good deal of my time and budget is spent trying to avoid where possible, and mitigate where not possible, unintended use of our systems which costs me time, effort and money. Because IT in a very real way has become so consumer friendly, something I often refer to as the "Best Buy Syndrome" has grown. That is, because people can walk into a local Best Buy and pick up a 1TB drive for a couple hundred bucks, and get residential broadband for $50/month, they think this stuff is all real cheap to deliver with high availability in large scale. It's a good news/bad news kind of thing. People are far more aware of the potential of the technology and it has become far easier to use, but that has resulted in far FAR higher cost that nobody recognizes. Here's an example. There has been more unstructured data created and saved to storage in the past 2 days than was created and stored from this history of man to 2008 - in total. Fact. Resource utilization is not a linear function. It is an exponential function. And companies that deliver online services have to pay their bills. Last year alone I spent (business wise) over $2.1 million dollars in ONLY network circuit costs. Or that even using enterprise discounted pricing, MS SQL Server enterprise for commercial use is licensed at around $16 thousand - per core. So, a highly available clustered database with an active/passive 2 node 4 core cluster plus a disaster recovery location costs $256K annually - just for DB licensing. Don't even ask about Oracle. You know what they say - after a while you're talking real money I'm just saying that it's easy to point the finger and say CL is being unreasonable - so long as you don't know what it costs them to provide that free service that you're not even paying for.

[Tony_SS]

Craiglook specifically and solely targes ONLY CL data.

Wrong. They searched Oodle, Backpage and Ebay free classifieds. It was a free classifieds search engine that searched cached files.

[wmhjr]

Wrong. They searched Oodle, Backpage and Ebay free classifieds. It was a free classifieds search engine that searched cached files.

Didn't know that. Doesnt' matter though. The point remains intact that it's not a typical "search engine" and instead parses the RSS feed to take advantage of filtering already performed by CL. The interesting thing is that Craiglook/claz themselves prove my point. There's a page showing the impact to traffic going to CL up to, and following the change. You can see a dramatic drop in traffic - which is apparently what CL wants for reasons I already mentioned. The people I see complaining about it are typically the same people who complain that MS is evil, all software should be open source, etc. Of course, they're not complaining about how much development has happened as a result of MS as an example.

We'll just have to agree to disagree.

[Tony_SS]

Didn't know that.

You're forming opinions based on assumptions. So what about the traffic? Does CL own their traffic too? lol You can't claim infringement based on traffic. And now you're resorting to strawman's with your Microsoft 'open source'.

Yours and CL's cries of infringement are a pretty weak case against a free search engine searching the cached files of classified sites.

Agree to disagree. Cheers.

[wmhjr]

You're forming opinions based on assumptions. So what about the traffic? Does CL own their traffic too? lol You can't claim infringement based on traffic. And now you're resorting to strawman's with your Microsoft 'open source'.

Yours and CL's cries of infringement are a pretty weak case against a free search engine searching the cached files of classified sites.

Agree to disagree. Cheers.

No, I'm saying that the increased traffic COSTS CL MONEY! They have to pay for resources to handle that traffic. Bandwidth. Load balancer capacity. Web server capacity. Memory. Storage, as the indexing of unstructured data typically takes up between 50-150% additional storage. Firewall processing capacity. Those are facts and not assumptions. Unless there is some sort of magic pixie dust that handles the technology. It is your opinion that the argument is weak. Unfortunately, it isn't a legal opinion. I am not resorting to "strawman" arguments. I'm pointing out comparisons. I would recommend that if you're interested, you pick up a couple books on Information Technology Law, Privacy and Intellectual Property law. As I said, this is my business, so I may well have a different perspective. That being said, my perspective is based on doing this. There is not a week that goes by that I am not talking for at least a couple hours with in-house counsel AND external counsel for issues such as this. The internet is not a "free for all". It is a communications vector. Frankly, if you walk down the street and see an unlocked car with the keys in the ignition, there is nothing technical preventing you from driving it away. It is still theft. Allowing other vendors to access indexed data wholely owned by CL, indexed by CL and supported by CL, for which only the INDEX is cached and all hits are routed through CL once displayed, for the sole purpose of expanding the scope of their product and brand, when they explicitly want "their product" to be "local", which subsequently results in more cost to CL, seems to be a decision that CL should be able to make.

But, let's leave it at you have an opinion. I have a different opinion. I completely support CL in this regard and understand their position. I would likely do the exact same thing.

[Tony_SS]

And CL is profiting off the increased traffic. That is the cost of business with higher profits. Claz is a separate private enterprise. I believe its fair use. And that is why we have fair use laws. Otherwise we would have monopolies that would own everything they could to eliminate competition. Small business and entrepreneurs wouldn't flourish or exist. JMHO. A judge could see it either way. Information property is certainly a new issue with a lot of grey area.

[fantasygoat]

And CL is profiting off the increased traffic.

I'm curious how you think CL is profiting from increased traffic. They don't charge for most ads, they don't have any click count based revenue model. Increased traffic actually *reduces* profits because they have to pay for it.

I manage similar systems for a living so this discussion was interesting to watch unfold. You clearly don't understand that infrastructure costs money. Having some third party site scrape your site for data, without compensating you, is not good business. It is CL's prerogative to control the use of their data as they see fit, and in this regard I agree with their decision.

As a consumer I would like to have more options on CL such as area based search, but such changes should be through CL itself, not some third party they can't control.

[Tony_SS]

Fair enough.. that assumption was made assuming they price their services (that they do sell and profit on) based on traffic volume - the same way an ad is more expensive to purchase from a magazine with a high circulation rate.

So whats the infringement with searching cached data? Its not site scraping. That is what Search Tempest does.They and other 3rd party sites, that DO site scrape still exist - yet no one want to acknowledge that. They are not under the radar either.. Unless those are taken down, it just seems odd Claz was targeted only.

[fantasygoat]

That "cached data" has to come from CL at some point, to be cached.

I can't speak for the others, perhaps they are not big enough to appear on their network graphs.

[wmhjr]

Fair enough.. that assumption was made assuming they price their services (that they do sell and profit on) based on traffic volume - the same way an ad is more expensive to purchase from a magazine with a high circulation rate.

I understand. So just to be clear, CL does not operate under this business model. Increased traffic actually decreases their profit to some extent. I suppose you could try to make an argument that increased visibility would result in increased success of sales, which would then result in increased use. Problem is, only a very small subset of transactions are "fee based" so it would be an extremely difficult argument to substantiate. As opposed to magazines, who see revenue from not only the advertisement segment but also from circulation, CL gets no circulation revenue whatsoever. Circulation is actually a cost center and not a profit center for them. Completely different model.

So whats the infringement with searching cached data? Its not site scraping. That is what Search Tempest does.They and other 3rd party sites, that DO site scrape still exist - yet no one want to acknowledge that. They are not under the radar either.. Unless those are taken down, it just seems odd Claz was targeted only.

That's a more difficult question, and it's one that nobody on this site can answer definitively or with any authority. A lot has to do with how their intellectual property is actually protected, and we can only make some assumptions there. I'd say from just a general perspective, a great deal has to do with their published terms of use - which frankly govern both people posting there as well as people using those posts. They have clearly established the limits of their product, and how that private data is permitted to be used. Much like NFL football games which are broadcast over the public airwaves via OTA. You are perfectly legal in tuning to those stations (I'm not talking cable, dbs, etc - just OTA). You're also OK to record them for your own use. However, you are legally prevented from replaying those in a commercial establishment without the express permission of the NFL. That structured data which has been indexed at some cost (initially and recurring) by CL is wholely and completely owned by CL. There is zero confusion about that. The entire question here is how it is allowed to be used. CL has established limits for the use of THEIR data. They have given permission for it to be used in very specific terms. Violation of those terms could be considered infringement on a number of fronts, starting with but not limited to patent or copyright infringement.