Technical-Legal Confluence News

There’s a lot of talk out there about schemes to allow users to block Internet advertising. And there is a variety of proposals.

One is on “Internet Outsider” (all the way back to April 2007) is Henry Blodget’s “One Way for Microsoft to Kill Google”, link here is to put more emphasis on a “cost per impression” (CPM) basis as well as “cost per click” (CPC). The writer suggests that Microsoft create CPM network to compete with Google’s Adsense.

But another writer suggests something even more sinister and nefarious: simply build an adblocker into the next service pack for Windows Vista and make it apply to all ads and all browsers (especially IE and Firefox). Would that blow away the entire free content industry? Would newspapers stop offering their content free in online editions? Would not only Blogger disappear but a lot of shared web hosting go away? What happens to social networking sites? It’s a scary thought, because the plain truth is that the whole boom in self-promotion is predicated on a two-way street, the idea that ordinary users are willing to see and react with ads. I found this suggestion on a Wordpress blog called “Reluctant blogger”, also (apparently) posted “way back” in April 2007, here. It’s easy to imagine legal complications if Microsoft attempted this.

In fact IE7Pro a separate ad-on, offers an ad blocker.

There is also a product called “Adblock” as part of “Rick’s Easylist” here started by an entrepreneur in update New York who simply calls himself “Rick”.

What I don’t know is how much traction these products are getting. But when I started my own personal publishing back in 1996, I paid little or no attention to advertising. I started with a book and then used the websites to supplement the book to draw and interest readers to my political arguments.

It’s not clear whether these blockers affect ads that play before free videos, or even those, common on newspaper sites and imdb, that overlay the content and must be closed to see the content. Those are more disruptive to viewers, but may be necessary to support the free content.

Am I cutting my own throat by pointing this out? No, because readers will find out anyway. We might as well started a spirited debate about it.

Also, notice, on this particular blog, there are no ads. At least right now. That could change later.

The is a major brouhaha developing over fair use of wire service stories. This conflict has apparently started with some take-down notices from the Associated Press to the “Drudge Retort.”

The heart of the matter seems to be that the site reproduces links, titles, and leading sentences of AP news stories, without other comment enhancing the content. However, the stories usually result in visitor comments which, as a matter of natural course, add value to the original stories. There was also a report that AP wants to be paid $2.50 per word for any quote that exceeds four words, and there seems to be an objection to bloggers’ reproducing the titles of the stories, although that would be the accepted practice in the footnotes or a bibliography in a book or term paper as taught in high school and college English.

There are stories of negotiations going on, and the outcome is unclear. There will be more details added later here or on related blogs.

A typical news story about the incident comes from Workbench, here.

Here is a discussion from “buzzmachine” on the “ethics of the link layer” in journalism, including blogger journalism, here.

The account from the Media Bloggers Association (MBA) is here.

AP maintains that this controversy represents an intellectual property control and “ownership issue.” It could argue that the practices of some blog sites simply reproduce content and may deny the content originators legitimate revenue. But there are “philosophical” questions beneath, as to how visitors acquire news, and the legal observation that “facts” themselves cannot be owned (even though reporters work very hard to get original scoops).

There is an ongoing controversy involving trademark law and online advertising. Companies sometimes complain that search engine companies (like Google) or various other software products display ads of competitors (of trademark holders) as a result of searches on their wordmarks. There are various circumstances in which these situations may occur. Some of the complaint seems to deal with the natural result that search results for competitors may be displayed, or that criticism of the original mark holder may appear.

An important case, decided by the Second Circuit, was WhenU.com v. Vision Direct, Inc., in 2003. The Opinion is here.

The Electronic Frontier Foundation amicus brief is here.

Other cases include
American Blind and Wallpaper Factory v. Google, complaint (2004)

Letter to Google from American Blind’s counsel regarding Adwords (2003)

Account of oral arguments in Rescuecom v. Google (March 2003) as logged by “The Laboratorium,” is here.

It’s unclear as of yet as to whether the “prospective dilution” component of the 2006 law could matter in these cases.

The House of Representatives, in 2007, introduced a Bill, numbered H.R. 275, called the “Global Online Internet Freedom Act.” It was introduced by Christopher Smith (R-NJ).

The bill would establish the fact that it is the policy of the United States not to support restrict of political dissent and legitimate speech within the borders of other countries. However, the law would require, for the most part, American ISPs and search engine companies to comply with the laws of other countries within the borders of those countries so that countries that did not would not be put at an unfair competitive disadvantage.

The govtrack reference is this.

The Congressional Research Service Summary is here.

The text of the bill is here.

Some visitors know that, besides this new site and Wordpress blog, I have a number of blogs on Google’s Blogger, and a major site called doaskdotell.com. I have a site based on my legal name, johnwboushka.com, which is based on my IT resume only (as I am “retired” now). I also have minimal profiles on Myspace and Facebook, which I use very little. I have been self-publishing on the Internet since 1996 and gone through several different strategies and domains and have experimented with a number of strategies for deploying my material. As a result, it might seem a bit helter-skelter, although it makes sense to me, and to someone who spends some time with it.

Recently, there have appeared on the Google account forums a number of problems reported by bloggers, and a number of bloggers have expressed frustration at what they perceive as slow response or customer service. There are three particular areas of concern. One is disabling or removal of blogs for suspected terms of service violations. Another is similar disabling or removal when marked as “spam”, apparently by automated “spam blog” detection. A third is removal from Adsense, apparently for generating invalid clicks. Bloggers claim, with some credibility, that these are “false positive” situations and in some cases they indeed get reinstated. But they are maintaining that the service is slow to respond to the formal appeals that the service makes available. In a few cases the bloggers might have been publishing on domains that they rent and control, but in many other cases they were bloggers using the “free” service.

Let me say here that I am not nor have I ever been a Google or Blogger employee. I have no proprietary knowledge. I am only trying to make sense of what I see going on and make constructive criticism, both of bloggers and of the Blogger service. I want to stay on Google’s good graces. I have a customer service background myself (toward the end of my IT career particularly), and it is quite plausible that I could really help them if I worked there. I just might want to try for a job there later (and move out of “retirement”). So I simply offer here what “problem solving” that I can.

These are somewhat distinct issues. First, take the “terms of service” violations. These may include (OCILLA) “safe harbor” takedowns of material that other parties have alleged infringe on existing copyrights under the Digital Millennium Copyright Act. According to the law, alleged infringers have a right to contest claims, and there are well-established procedures that ISP’s and services like Blogger must follow. There are other possible TOS situations, such as perceived libel, obscenity or hate speech, child pornography, and the like. Terms of service rules are influenced not only by United States law but by laws in other major western democratic countries (like in Canada and Europe) which sometimes are stricter on some topics. Indeed, in some cases ISP’s might remove material on receiving complaints and authorities or other complainants might not proceed with further action. Whether the speaker could successfully appeal would seem uncertain. But one of the most important things to remember about TOS problems is that they can exist whether or not the service is free (Blogger) or rented from a conventional ISP (like Verio, Yahoo!, Network Solutions, etc.)

One practical tip for bloggers might be to be careful about embedded hot links to images and videos. Publishers should make sure that they really have permission from the original sources. Embedded YouTube videos seem to be permissible, but not if the original video was pirated or is otherwise illegal.

The second area here contains getting paid by advertising services, whether Google’s Adsense, or any number of other variously structured opportunities from Internet marketing competitors. Sometimes people get canceled, particularly when the business model is partly related to clicks rather than just product purchase. Actually, a few years ago, I had Linkshare with another domain, and advertisers would “pre-approve” you with them before you could even sign up with them, because some vendors (like airlines) won’t advertise on “controversial” sites. That could be an effective system. With a click-based system, the publisher is told not to click on his or her own ads or directly encourage others (particularly household members or anyone in repeated contact) to do so. Google, for example, offers a somewhat complicated scheme (including updating the Windows registry) to “test drive” the ads without their being counted. I can make a suggestion here, that the scheme could be simplified. For example, if a publisher has a fixed broadband IP address, she could furnish that to the advertiser with the “good faith” understanding that clicks from that address are not tallied. Advertisers probably can analyze clicks by IP source and other origin routing parameters anyway. Or some logon procedure could be developed.

I’ve saved the “best” for the last here – the problem of spam blogs. First for the easy part. Spam comments are a problem too, but they’re easy to control just by using comment monitoring (and perhaps additionally requiring the commenter to sign with a captcha). The more recent problem is spam blogs, or “splogs”, about which a lot has been written in the past three years. Wikipedia gives a comprehensive and rather definitive analysis here.

Literature suggests that splogs are generated by sophisticated software and made to “look real” by scraping passages of text from legitimate sites, and then throwing in links to products generally thought of as less reputable. It seems to be driven by the fact that some advertisers and customers do bite, and the unlimited opportunity of a “free” source seems to make the idea “work”, much as with spam in email. Splogs are not illegal in themselves, but they threaten the business model upon which free publishing services are offered. There does not seem to be a big problem with them outside of these services yet.

Another related problem that has been reported is that sometimes spam postings get put on legitimate blogs by spammers who break security, perhaps just to spam, perhaps to inflict vandalism.

One can imagine that automated spam detection software would be difficult to implement reliably without a lot of false negatives and positives. It is difficult for the legitimate blogger to determine what to do to avoid false positive detection, as any recommendation could be imitated by sploggers, and so therefore the algorithms themselves must remain proprietary and secret.

It would seem, however, that even here some common sense rules could apply. The main problems seems to be links. Maybe some of it is excessive links in relation to text. But links to major news sites or government sites or other sites to substantiate what is said in a blog normally should not be a problem, because it isn’t logical that any search engine algorithm would base the ranking of a major corporate site based on references in blogs. Another issue would be “mutual admiration societies.” Blogger profiles allow the enumeration of other sites, but that capacity could be expanded to include a number of affiliation sites, in order to inform robots and search engines with “good faith” that announced affiliation should be taken into consideration in determining page ranking; once properly informed, robots would have no reason to consider such links as possible symptoms of splogging.

Often blogging providers (like Blogger) offer captcha verification to prevent automated content generation. But it is reported that sploggers get around this, and it seems that blogs are being taken down as suspected splogs even though the captcha technique ought to be available and work.

I’ve wondered about other possible symptoms. Perhaps the appearance of apparently nonsense words – but maybe they are tags in a source of a script that is being documented. Perhaps the appearance of many incomplete sentences, but those are common in informal blogs. Perhaps a wide range of subject matter. But even that is likely legitimate. The “range” could come from the use of metaphor, or it could come from the fact that the writer is “connecting the dots” and relating two or more issues or subjects not normally discussed at the same time to make a novel but legitimate argument.

So we come back to the “free service” argument. Anything that offers infinite resources for free will attract bad actors. So speakers who take advantage of it cannot claim much ownership or real rights. Perhaps all amateur blogging is in some sense a kind of “self spam” and fits into a gray area. Even rented ISP service has become so inexpensive with such generous bandwidth and storage space as to require little capital. Can this remain so forever, or will the difficulty in sifting out what is legitimate eventually make these services unprofitable for companies? Could bloggers face the same fate as weekend air travelers? That is the “anti-amateur” argument, yet “professionalism” in a world of extreme capitalism hasn’t offered a lot either.

Charging for emails has been proposed seriously as an antidote for spam in emails (something I support, if the “postage” is a microcharge per message); maybe the same concept is needed for what is now “free blogging.” There could be a limit on the number of posts, the disk space used (there is for images already), or the number of years retention, after which some sort of rental plan similar to that offered by ISP’s would have to be paid for. It could still be reasonably and affordably priced, and offer more sophisticated permanent archiving. Other possible concept could be a tiny charge for each post, just like a tiny one for each email, in order to discourage automated content generation on blogs. Perhaps the charge would be less or waived when a captcha is filled out.

July 24 2008

There are also a few reports of parties with Blogger accounts closed while the blogs stayed up. The blogger is directed to a link that mentions a “perceived terms of service violation”. There is supposed to be a manual review available. But if the blogs themselves stayed up, this sounds like a problem with automatic violation detection.

The “terms of service” violations could deal with all the “usual” problems: spam, obscenity, chain letters, virus propagation, etc.

Aug. 4, 2008

Blogger admitted somewhat profusely over the first weekend of August (to is bloggers at login) that it has a problem with false positives, and had even accidentally locked some blogs not identified as spam. There is a story at Blog Herald by Thord Daniel Hedengren, link here.

Besides the federal Can-Spam Act, many states have anti-spam laws. In Virginia the law is here.

The law is controversial because Virginia is reportedly the only state in the nation to make it a felony to send unsolicited bulk email of a religious or political nature. However, it is hard to see how the First Amendment could have anything to do with prohibiting spam headers. The Virginia Supreme Court, according to the AP (in a story printed in the Washington Times on June 4, p A12) will take another look at this narrow question. The case is Jeremy Jaynes v. Commonwealth of Virginia. Jaynes was prosecuted in Virginia because AOL’s servers are located there in Loudon County.

The main act used to prosecute misrepresentation on the Internet is the Computer Fraud and Abuse Act, originally passed in 1986, last amended by the Patriot Act in 2001.

The law is USC 1030 and the Cornell Law School reference is here.

The law was intended to prevent computer hacking and unauthorized use of classified information. It has recently attracted attention because it is being used in the Megan Meier Myspace case in Missouri, where a federal prosecution was launched in Los Angeles with an indictment based on this statute, where a woman had impersonated a teenager.

The PRO IP act (HR 4297) has the full title “Prioritizing Resources and Organization for Intellectual Property Act of 2007,” introduced by John Conyers (D-MI), with 19 cosponsors.

The govtrack reference is this.

The Executive Summary from the CRS is this It would create and Intellectual Property Czar, or a Executive Office of the President the Office of the United States Intellectual Property Enforcement Representative. It also provides some safe harbor for copyright registration with inaccurate information, prevents inappropriate disclosure of records related to copyright infringement, and prohibits importing and exporting of infringing works.

The full text link is this.

Coordinated with this bill is “H.R. 5889: To provide a limitation on judicial remedies in copyright infringement cases involving orphan works,” govtrack reference link here. The bill would permit the use of “orpaned” works without permission when the copyright holder could not be found, but allow compensation later should the owner step forward. The text of the bill is here.

The case Bindrim v. Mitchell, in California, in 1979, establishes the idea that “real people” who are depicted in a novel (in this case, “Touching“, by Gwen Davis) purporting to be “fiction” can bring legal action (such as libel or invasion of privacy) if a reasonable person would know that the character in the novel really is supposed to “be” that person. A UMKC law school link is here.

Time Magazine has an old article from Mar. 17, 1980 “Writers’ Rights and Wrongs: A publishing house throws the book at one of its authors,” link here. This refers to the fact that typically authors have to indemnify publishers against loss, and today Internet self-publishers have to indemnify ISPs (as part of “terms of service”), even though, in practice, such contract provisions have generally been only very rarely enforced for obvious business reasons. (ISP’s have pursued people in spam-related cases.)

It is not necessarily true that this principle would hold in other states in which it has not been litigated.

There is another case in Vermont, Garrido v. Krasnansky, this a divorce proceeding, in which a judge has ordered on party to stop talking about a case online. The case is interesting because the online talk was a blog that purported to be “fiction.” I did not find a court opinion on line, but here is a blog entry from Info Law on the case, here.

Another problem known to have occurred but not litigated (in Virginia) is where an individual used a character based on himself in a fictitious setting on-line in an unfavorable way to make a “political point,” and a principal at a public school where the person worked tried to have the person removed. Eventually the person left for other reasons. It is not clear as a matter of law whether the “Bindrim” principle would apply to a likeness of the self, especially in a state other than California or Vermont.

It’s also interesting to wonder when rulings can be used from other states to make arguments to establish the same precedent in a new state. “Full Faith and Credit” applies to contracts (usually marriage, until now), but not for other legal doctrines.

It’s possible that an objection to “real life resembling” fiction may become stronger if some of the content is actually based on factual events, even if names (and perhaps places or times) are changed.

The “fact or fiction” issue also obviously ties into the new controversies about “online reputation defense.”

The law that resides at the heart of controversy over digital copyright management is the Digital Millennium Copyright Act of 1996 (DMCA).

The text of the law is on Wikisource, here.

There is a supplement called “Limitations of Liability Relating to Material Online,” section 512, Cornell link here (also called the “Online Copyright Infringement Liability Limitation Act” (OCILLA), or the “safe harbor” provision.

The law is controversial for a number of reasons. One is that manufacturers are forced to use the copy protection technology of one vendor (Macrovision). A totally unrelated controversy concerns the takedown provision. Web hosts are immune from copyright infringement liability if they follow takedown requests, and then have little incentive to challenge the legitimacy of infringement claims. There have also been DMCA-related requests to take down links to infringing material, and this is a legally gray area.

Another concern is that the DMCA can hinder the efforts of “newbie” artists to manufacture and distribute their own works for free or low-cost consumption. Some people feel that DMCA and related laws are intended to protect “entrenched media” from low cost competition.

A website called Chillingeffects keeps a log of DMCA “cease and desist” letters.

The Sonny Bono Copyright Term Extension Act of 1998 is here.
extends copyright terms for 20 years.

There have been several attempts to modify the law.

109th Congress.

These bills are “dead” but it is likely that similar ones will be reintroduced.

H.R. 4569: Digital Transition Content Security Act of 2005, to close an “analog” loophole and require certain analog devices to have copy protection.
Govtrack link.

H.R. 1201: The Digital Millennium Consumer Rights Act of 2005 (DMCRA) would refine the DMCA provisions and restore some ordinary “fair use” for consumers as was common previously with CD’s, cassettes, and vinyl phonograph records.

Govtrack here.

Some court cases:

Chamberlain v. SkyLink (DC Circuit Court of Appeals, 2004, here)
(may require registration), a case involving automated garage door openers

Universal v. Reimerdes, distribution of DeCSS, reverse engineering allowing decryption of DVDs. Trial court opinion (2001), here, was upheld in Second Circuit (resolved in favor of copyright holder)

U.S. v. ElcomSof and Skylaorv (criminal prosecution), Electronic Frontier Foundation reference.

Lexmark v. Static Control Components. Intervention of a printer’s microcontroller to allow other printer cartridges does not violate the DMCA (6th Circuit). Link here.

MGM et al. v. Grokster et. al (2005), opinion here.

The Supreme Court allows downstream liability for software vendors or service vendors when their “business model” is based on encouraging copyright infringement.

I have running news occasionally on this issue on my main blog (intermixed with many other issues), link.

A major attempt in the 108th Congress was S. 2560, “Inducing Infringement of Copyrights Act of 2004,” govtrack here.