A recent study by content licensing and tracking company iCopyright claims to have found a link between article tools, the links that let users email, Digg, save or otherwise interact with a story, and content misuse.

The premise is that article tools, by encouraging users to copy the content on a site for limited purposes, may actually be creating confusion on the reader’s part and encouraging them to make use of the content beyond the intention of the tool.

While this makes sense to some degree and the study’s research does seem to indicate that there is reason to believe article tools may incite some level of content misuse, there are also nagging issues and other problems with the study that make its findings, and its proposed solutions, seem less certain.

Basic Findings

The iCopyright study consisted of two parts, a survey of 787 mailing list users as to their habits with reusing content and article tools. The second part was a usability study conducted by an independent lab that compared iCopyright’s article tools to competing services.

The study found that, 99% of visitors use articles on various sites in ways other than simple viewing. This includes printing, saving, emailing and sharing links. Of those who interact with content, nearly 65% said that they did so “Frequently” and another 31% said they did so occasionally.

However, nearly half of those who used articles used it for what the study considered a commercial use. This is because, according to iCopyright, the tools provide no clear system to explain the intended use of the tools nor any clear means to enforce the wishes of the publisher. These tools became de facto copy machines that, according to the study, were repeatedly abused out of ignorance.

According to the study, copyright notices and other self-fixes failed to work. Usability testing showed that such notices were overlooked or missed and users simply printed as they wished, even if it was in violation of the license.

To make matters worse, removing the tools did not help as 96% of respondents said that, if an article were not available for print via an article tool, they would do so anyway, most commonly via the browser’s “print” feature. Also, 95% said they would email an article even if it were unavailable via an article tool, including 37% who said they would copy and paste the entire text of the work into the email, as opposed to 40% who would simply email the link.

(Note: I it worth observing that linking to an article in an email nor printing out portions (perhaps even the entire work) is illegal. Deep linking is legal, fair use allows usage of small portions of the original and implied license may cover certain other behaviors. Simply because the study reports on a behavior does not mean that it is illegal or immoral.)

Also interesting is that 64% of respondents said they would rather use supplied article tools as opposed to other solutions and that the most important tools were email, print, save, share and comment, in that order.

The study then went on to present iCopyright’s article tools, which have strict controls on how the content is used, as a solution to the problem and as an opportunity to turn such uses of the content into a revenue stream.

The statistics in this study are very interesting and very enlightening for anyone who is curious how end users interact with their content. However, the study isn’t without its flaws and potential issues that need at least some attention.

Minor Problems

Though the data in the study is interesting and useful, it does come with a few caveats that are important to remember.

First, the sampling on the survey makes it more likely to attract those who routinely share articles. Not only were participants already subscribers to iCopyright’s Clip&Copy, a news service targeted at those who are interested in finding articles to reuse, but they had identified themselves as “information workers” more likely to have a job that requires them to share news.

Also, this sampling is not likely to be a good representative of visitors to an average blog or site. As every site has a different audience with unique skills and tendencies, so will change the types and means of interactions.

Second, the definition of “commercial use” is used a bit loosely. The answer options on the related question are whether users copy the article for “Personal Use”, “To Share with Friends and Family”, To Share with Coworkers”, “To Share With Clients” and “All of the Above”.

However, the study only considered the first two options non-commercial even though not all times an article is sent to coworkers or even clients is it commercial. Though many of them may be, it isn’t all of them for certain.

Another issue I see is that, in one section making the case for using iCopyright’s tools, the article said the following:

One thousand visitors to a site might generate $90 in advertising, assuming a $30 CPM and three page views per visitor. However, those same 1,000 visitors can produce $1,000 or more in commercial use licenses, assuming a 1% conversion rate at $100 per license. When built in to article tools, content licensing can provide a return ten times that of advertising.

These numbers seem incredibly optimistic on both sides. a $30 CPM is very high and many times the norm, but may be obtainable by some sites (very few). However, a 1% conversation rate on a $100 is almost impossible to imagine, especially since responses on far less expensive products is often much lower.

Though this is seems to be a simple example with off-the-cuff numbers to make a point, it doesn’t seem to mesh with current realities, especially with the economic downturn.

However, the biggest problem is the study’s proposed solution, to use iCopyright’s article tools with their protection, is not well supported. There is no research about how many people would be willing to pay for an interaction or what they would do if confronted with a fee for an action and they don’t wish to pay for it.

The study had already showed that, without any article tools, users will revert to other means to interact with content. If they are confronted with a fee, even a reasonable one, and don’t wish to pay it, those tools are still there.

Say, for example, a site charges for a print out of a work, if the user doesn’t wish to pay, they can still use their browser’s print button, copy/paste, etc. The only difference is that this time the user will know the wishes of the publisher and may be discourages. However, exactly how big of a deterrent that is, is not fully answered.

Still, none of these issues stop the study from providing some interesting insight into the way article tools are used and how the affect copyright. It will be interesting to see if these results change any major publisher’s approach to article tools as a result.

The Big Picture

Though the study has its flaws and is, understandably, a promotion for iCopyright’s own article tools services, which include its iCopyright for Creators service (previous coverage), a service targeted at bloggers, it does have some interesting statistics and some great insights.

For me, the takeaways are simple. Article tools see a wide variety of use and, if your goal is to limit what users do with your content, they are not the way to go. If you have less restrictive licensing, such as with this site, they probably make sense as the study shows that users strongly prefer such tools over the alternatives.

However, standard article tools as a means of protecting content or limiting use is a poor choice. But given that users who want to share your content will do so, it makes sense, in my opinion, it makes sense to apply them and channel that energy through a system you have some control over. Yes, there will be abuse of the tools, but at least it is your system and not something completely outside your ability to enforce, monitor and track.

Whether you use iCopyright or another set of article tools, it is important to understand the advantages and the drawbacks. As smaller publishers, most people reading this don’t have to worry too much about the bottom line and whether each infringement is a dollar lost, something the target group for this study does, but we still have to develop effective content strategies to make our content work for us, not against us.

Disclosure: iCopyright has donated to this site in the past and I have consulted for Attributor, a company that competes with the Creators service (though not the licensing service).

However, last week, Mike O’Donnell, the President and CEO of iCopyright, was kind enough to give me a guided tour of the backend. Though I wasn’t able to access anything hands on or experiment with the technology with my own content, that will have to wait until the service is available for iCopyright for Creators users, I was able to see what the service does, how it works and what it can do.

So here is a brief look at what the iCopyright Discovery system can do and how it will likely look when it is available for Creators users shortly. Please bear in mind that this is not a review, just a tour of the key features of the service.

The Basic Premise

The big idea of Discovery is this: Discovery parses your content as you put it up on the Web, accessing either a created XML file or your RSS feed, and then searches for copies of it on the Web.

The service then searches for matches of your content, highlighting ones that it determines to be the most important, and gives you options for remedying the situation. Among the actions it can perform are removal requests, which fundamentally DMCA notices, license requests, which goes through iCopyright’s existing licensing system, and forwarding to legal counsel.

This idea is fundamentally very similar to Attributor and Blogwerx, both of which are still in private testing. However, the execution of the system is going to be what is important. On that front, iCopyright has devised an interesting workflow system that seems to string the process together very well.

Setting Up Discovery

When a user first signs in to Discovery, the first page they’re likely going to head to is, oddly enough, the “Settings” page. The reason for this is that, without visiting the settings page, you have little control over the matches you see and you can’t use several of the remedy options.

From this page, you can set your enforcement agency, useful if you are part of a group that handles your copyright enforcement, and the email address to your legal counsel. This will let you enable addition redress steps down the road. However, the most important settings are the search sensitivity and risk assessment as they determine the matches you see down the road.

The search sensitivity feature allows users to tell Discovery how many matches they want. They can set it so that only the worst matches appear in the system or so that they see almost everything. This is done by tweaking the minimum match ratio, meaning how much of the original work must appear in the copy, the minimum risk factor, discussed below, the minimum site activity and the minimum number of copied words that must appear in the match, useful for sites with short posts.

The Risk Assessment tool is easily one of the most interesting features in iCopyright Discovery. It lets users set the criteria for determining how much of a risk a match site is. You do that by setting sliders for Unique Visitors, which looks at the estimated traffic of the site, the number of inbound links, whether the site displays ads or how much of the content it copies.

These sliders are intended to be abstract in nature and are used to indicate which attributes are more important than others. For example, if you set all to 10, they would be weighed equally. However, if you put one at 5 and the others at 10, the first one would be weighed much less.

These attributes, when combined with the site’s actual use of the content, are used to determine the risk level of the site itself. This, in turn, plays a major role in determining the priority the site is given when analyzing suspect pages.

Sorting Matches

Once you are done telling Discovery what matches you want to see, the system then does a refresh, which takes about an hour according to O’Donnell, and you can then view your matches or “suspects”.

The match sort is organized by a combination of variables, focusing heavily on suspect pages with the highest risk. For each suspect, the system displays the URL of the work, whether it displays ads, whether it links back to your site, roughly how many visitors it gets, the number of inbound links to the site, the match percentage and the risk.

From this page, you can go through the matches and either archive the match, which functions similar to Gmail’s archive function and takes no action, move it to the Whitelist, either pending or approved, or send it to the redress list.

If a site is moved to the whitelist, that means that the use is licensed and future matches from the site will be ignored. You have the option of telling the system to either ignore matches on the URL, the subdomain or the entire domain.

If you move it to the redress list, you can then take further action on the match, including licensing the work or filing a removal demand.

Taking Action

The redress list, as you see below, looks very similar to the suspect list and contains much of the same information. However, the options for what one can do with a suspect are different on this page.

From this page, you can then either offer the site a license, which will send out an email encouraging the site admin to go through the existing iCopyright system, file a link request or send a removal notice.

Removal notices, fundamentally, are DMCA notices though they are written so that, at this stage, they can be sent to Webmasters directly. Link requests are more like informal license offers, but ones where the only stipulation is a link back.

All of the letter types are fully customizable and Discover offers a templating system that lets you build your own letter that automatically inserts the necessary information.

Once you file a redress, you can then track the status of it in the Redress Offers Status page. From there, it will let you know if the redress has been completed and, if it hasn’t, makes it available to be escalated.

If a suspect match is moved to the escalation list, then the user has a whole new series of options for how to deal with the site.

The options include the ability to, forward the situation to your legal counsel (if set up), notify the ISP, which sends a more traditional DMCA notice, notify the enforcement agency (if set up), send a notice to the ad network or demand removal from the search engines.

All in all, the initial Redress List can be looked at as the cease and desist/licensing phase where the Escalation List deals more with the DMCA/lawyer phase.

However, no matter what redress steps you take, Discovery offers a powerful means to track and monitor the progress of the steps that you took.

Tracking and Monitoring

Once you’ve taken a redress action against a suspect site, you can then track and monitor everything that has to do with that particular match.

It provides much more than just a brief history of what has taken place, giving a detailed history of every email sent, comments left in the system, both automatic ones and ones left by the user, as well as other information about the site.

The idea is to maintain a record of every action, including emails, phone calls and other steps, for the purpose of aiding in any potential legal case.

Once the matter is resolved, escalated outside of the system or the match is whitelisted, the case can be archived and thus removed from the suspect pool, allowing you to move on to other matches.

Some personal thoughts

It is very hard for me to offer any real review of the service. Without actually being hands on with the service and using it against my own content, there is not much that I can do.

Right now there are many unknowns for me, including the following:

1. Match Detection: O’Donnell has said they are partnering with a major search provider to perform the detection but it remains to be seen how effective it is. Match detection is not easy, even with a big search partner, as Copyscape showed. The system will not be of much use if its match detection is not the best in its class.
2. Resolution Assistance: The hardest part about stopping a plagiarist is not composing the letter, but finding who to send it to. It is easily the biggest time sink in most of my cases and is the number one reason people approach me for help. It remains to be seen how effectively Discovery helps with this process.
3. Speed/Usability: Obviously, without actually using the system, I can’t tell how fast it moves and how much time it will save you. If the system is sluggish or error-prone, it could greatly hurt its usefulness.

This is not to say that these things are wrong with the current system, just that I don’t know right now and won’t until I can do a full review, likely later this year.

However, judging from what I can see, the system is very impressive. It looks very good, has a solid workflow built into it, though I somewhat disagree with having the ISP step be only available in the escalation section, and seems to be built with the user in mind.

What I like best about Discovery is how the user customizes the system to fit their needs, with their own definitions of what matches to worry about, their own letters and their own general strategy. Any such system should focus on automating what can be automated, but leaving the big decisions to the copyright holder.

What does worry me some is that the system is clearly geared toward larger clients. Discovery is designed to allow for multiple users to access an account and to work with attorneys as well as other rights enforcers. While those are great features for those that need them, it remains to be seen how the system will strip down for smaller copyright holders.

The other downside is that, according to O’Donnell, the version of Discovery for Creators will come with some kind of fee. Though pricing structure has not been discussed, he seemed confident that it would not be available for free.

Still, as these screenshots show, there is a lot to like in the Discovery system and the solution it promises.

It has a great deal of potential and Webmasters who are worried about tracking how their content is used should definitely take a serious look at what iCopyright has to offer.

Conclusions

There’s a lot of reason for me to be excited about the upcoming Discovery system. However, I have to restrain that excitement until I can use the system first hand and see both how effective it is and how smooth the process is.

No matter what though, I am happy to see that people are thinking about these issues and coming up with solutions. This has been a booming industry over the past few years and a lot of very smart companies are already involved and I am happy to be working in this field.

No matter what Discovery itself brings, it can only signal great things for copyright holders and Webmasters. Hopefully, this will help content creators not just enforce their rights, but understand how their work is being reused and encourage the kind of sharing that helps all involved.

Knowledge and tools can only help improve things, so long as those who use them do so wisely.

According to the press release, linked above, the service will search the Web for copies of the content, provide the creator with various means of redress and then monitor the site for compliance.

The service is not currently available with the iCopyright for Creators service, previously reported on here, but will be shortly. If you are not currently a user of the site, it may be a good time to sign up.

I have sent an email to iCopyright and I hope to gain access to a demo version so that I can test the service and provide a more thorough write-up shortly. Expect a review in the next couple of days.

Some Personal Thoughts

My first thoughts are that I am very excited about this service. iCopyright is a well-known and well-respected name in content licensing. Though their Creators service lacks functionality that prevents me from taking advantage of it, I highly recommend it to those that do not use CC licenses.

The pairing of licensing and detection services is a natural one, especially for larger customers that need to streamline the entire content control process. Whether bloggers and smaller users will benefit from this remains to be seen.

Though the press release makes it seem as if iCopyright functions similarly to other services such as Attributor, Blogwerx and other copy detection/resolution tools, its pairing with a strong license structure may give it an edge.

However, the real test of this service is going to be how accurate its detection is, how easy it makes it to handle unlicensed copies and how much value it is able to provide users. Until I’ve used the service, I won’t know.

Conclusions

If you haven’t signed up for iCopyright for Creators yet, now might be a good time. The licensing functionality is strong out of the box, especially for those who don’t use Creative Commons, and the promise of licensing, detection and cessation in one platform is compelling.

However, there are a lot of unanswered questions right now about the service including effectiveness, cost and features so I am not ready to get too excited. Still, it is worth noting that this will be available soon and could be a great boon to content creators.

I’m looking forward to giving this a more thorough review and updating this site with more information later.

Disclosure: iCopyright has contributed to this site. I am a consultant for Attributor.