Jaani's Linklog

The Linklog is a data feed from various social bookmarking sites that I frequent. It contains short entries deemed interesting but — whether because their subject matter is not directly related to the themes of this website (internet law and technology regulation) or simply for reasons of convenience — not justifying a full weblog entry. The format is simple: a title with a link to the aggregator source, and a one-line description.

Note that neither the titles nor descriptions are my own — they remain, with any associated errors, the work of the people who submitted each item to the feed from which each was derived.

Image of the day: Amazon book availability by decade (from a...



Image of the day: Amazon book availability by decade (from a random sample of 2500).  If this is representative of the wider catalogue, this paints a very troubling picture of post-copyright reprinting.

Image of the day: Amazon book availability by decade (from a...

Image of the day: Amazon book availability by decade (from a random sample of 2500).  If this is representative of the wider catalogue, this paints a very troubling picture of post-copyright reprinting.

Image of the day: Amazon book availability by decade (from a...

Image of the day: Amazon book availability by decade (from a random sample of 2500).  If this is representative of the wider catalogue, this paints a very troubling picture of post-copyright reprinting.

Self-represented litigant #1403: the comic book guy.



Self-represented litigant #1403: the comic book guy.

Self-represented litigant #1403: the comic book guy.

Self-represented litigant #1403: the comic book guy.

Self-represented litigant #1403: the comic book guy.

Self-represented litigant #1403: the comic book guy.

Building spilling booksDavid Pescovitz,...


Building spilling books
David Pescovitz, boingboing.net
“Biografias,” an instal­la­tion by Ali­cia Mar­tin at Casa de Amer­i­ca, Madrid. “5,000 Books Pour Out of a Build­ing in Spain” (via Imag­i­nary Foun­da­tion)

This art installation in Madrid is simply fantastic.

Building spilling booksDavid Pescovitz,...




Building spilling books

David Pescovitz, boingboing.net

“Biografias,” an instal­la­tion by Ali­cia Mar­tin at Casa de Amer­i­ca, Madrid. “5,000 Books Pour Out of a Build­ing in Spain” (via Imag­i­nary Foun­da­tion)

This art installation in Madrid is simply fantastic.

Building spilling booksDavid Pescovitz,...


Building spilling books
David Pescovitz, boingboing.net
“Biografias,” an instal­la­tion by Ali­cia Mar­tin at Casa de Amer­i­ca, Madrid. “5,000 Books Pour Out of a Build­ing in Spain” (via Imag­i­nary Foun­da­tion)

This art installation in Madrid is simply fantastic.

"The ECJ will be asked, ‘Is ACTA in line with the existing treaties, the [shared EU common..."

“The ECJ will be asked, ‘Is ACTA in line with the existing treaties, the [shared EU common law], the European Charter of Fundamental Rights? … If the court says no to any of these things, ACTA, in my opinion, is dead.”

- The European Parliament is set to refer the Anti-Counterfeiting Trade Agreement (ACTA) to the Court of Justice, which will add to an existing reference from the European Commission.  Long live the new age of rights-balancing in intermediary liability.

Google Spain not liable to enforce 'right to be forgotten'

In Alfacs Vacances SL v Google Spain SL [2012] (Unreported, Court of First Instance of Amposta, 23 February 2012), a Spanish district court held that Google’s Spanish subsidiary is not a proper defendant (ES) in a claim for an injunction to remove non-tortious but unwanted search results.  From reading this summary of the decision, it appears that the Court did not address the merits of the claim.

The claimant was the operator of a camping ground in between Barcelona and Valencia, where a tragedy occurred in 1978 which resulted in the immolation of approximately 200 campers by an explosion in a passing petrol tanker.  A Google search for ‘los alfaques camping’ now shows images of the disaster and a link to the Wikipedia page ranked above the company’s official website:


The claimant sought damages and an injunction requiring Google Spain to remove the images and reports of the disaster from “its” search results.

To understand why this action raises such substantial concerns, take a look at this insightful piece by Nate Anderson at Ars Technica.  Here was content (photographs of an old accident near the site of the claimant’s business) which was not defamatory, did not infringe copyright, was not proscribed by the doctrine of unfair competition, and did not fall afoul of any Spanish law.  Nevertheless, the claimant argued that Google should have a responsibility to remove the images and links, because they are adverse to the claimant’s commercial interests.  Once deindexed, the information would effectively disappear from the web since, although not deleted per se, users would have no feasible way to locate it.

In other words, the interest being protected is not a right to private and family life, but a right to revise history.  According to news reports, the Spanish data protection authorities have commenced over 130 proceedings against Google and other intermediaries for the deletion of historic data.

With any luck, we’ll have a reference to the Court of Justice before  long.  As if we needed another reason why a so-called ‘right to be forgotten’ is a laudable but extremely misguided idea.

Google Spain not liable to enforce 'right to be forgotten'

In Alfacs Vacances SL v Google Spain SL [2012] (Unreported, Court of First Instance of Amposta, 23 February 2012), a Spanish district court held that Google’s Spanish subsidiary is not a proper defendant (ES) in a claim for an injunction to remove non-tortious but unwanted search results.  From reading this summary of the decision, it appears that the Court did not address the merits of the claim.

The claimant was the operator of a camping ground in between Barcelona and Valencia, where a tragedy occurred in 1978 which resulted in the immolation of approximately 200 campers by an explosion in a passing petrol tanker.  A Google search for ‘los alfaques camping’ now shows images of the disaster and a link to the Wikipedia page ranked above the company’s official website:


The claimant sought damages and an injunction requiring Google Spain to remove the images and reports of the disaster from “its” search results.

To understand why this action raises such substantial concerns, take a look at this insightful piece by Nate Anderson at Ars Technica.  Here was content (photographs of an old accident near the site of the claimant’s business) which was not defamatory, did not infringe copyright, was not proscribed by the doctrine of unfair competition, and did not fall afoul of any Spanish law.  Nevertheless, the claimant argued that Google should have a responsibility to remove the images and links, because they are adverse to the claimant’s commercial interests.  Once deindexed, the information would effectively disappear from the web since, although not deleted per se, users would have no feasible way to locate it.

In other words, the interest being protected is not a right to private and family life, but a right to revise history.  According to news reports, the Spanish data protection authorities have commenced over 130 proceedings against Google and other intermediaries for the deletion of historic data.

With any luck, we’ll have a reference to the Court of Justice before  long.  As if we needed another reason why a so-called ‘right to be forgotten’ is a laudable but extremely misguided idea.

Smartphone design before and after Apple’s iPhone.  Image...



Smartphone design before and after Apple’s iPhone.  Image credit to @digerati.

The trouble with Wikipedia

The trouble with Wikipedia:

Scholarly expert on the Haymarket riots tries to edit the (top-ranked) Wikipedia entry for the same, is violently rebuffed by community editors.  In short, anyone can edit, but you must play by the rules:

So I waited two years, until my book on the trial was published. “Now, at last, I have a proper Wikipedia leg to stand on,” I thought as I opened the page and found at least a dozen statements that were factual errors, including some that contradicted their own cited sources. I found myself hesitant to write, eerily aware that the self-deputized protectors of the page were reading over my shoulder, itching to revert my edits and tutor me in Wiki-decorum. I made a small edit, testing the waters.

My improvement lasted five minutes before a Wiki-cop scolded me, “I hope you will familiarize yourself with some of Wikipedia’s policies, such as verifiability and undue weight. If all historians save one say that the sky was green in 1888, our policies require that we write ‘Most historians write that the sky was green, but one says the sky was blue.’ … As individual editors, we’re not in the business of weighing claims, just reporting what reliable sources write.”

High Court concludes The Pirate Bay infringes UK copyrights

High Court concludes The Pirate Bay infringes UK copyrights:

In unsurprising news, Dramatico Entertainment Ltd v British Sky Broadcasting Ltd [2012] EWHC 268 establishes that The Pirate Bay infringes the claimants’ UK copyrights by (i) authorising acts of copying and communication by its users; and (ii) engaging in a common design with those users to perform those infringing acts.  In a carefully-reasoned judgment, Arnold J sets out the background technology and analyses recent case law concerning the communication right.

Although this was a clear case, the Court was careful not to water down the applicable tests for secondary liability: authorisation, for example, remains “to grant or purport to grant to a third person the right to do the act complained of” — though it leads to a similar conclusion here to the broader “countenance, sanction of approve” test favoured in Australia (see, eg, Village Roadshow v iiNet).  Interestingly, Arnold J characterised the “means” of infringement as the torrent files:

The means used to infringe. The torrent files which are so conveniently indexed, arranged and presented by TPB constitute precisely the means necessary for users to infringe. It is the torrent files which provide the means by which users are able to download the “pieces” of the content files and/or to make them available to others.

This contrasts sharply with the approach taken at first instance in the Federal Court of Australia, which treated the “BitTorrent protocol as a whole” as the relevant means of infringement.  I suspect that not much turns on the distinction - except that the UK approach emphasises the importance of the application layer tracker service, rather than transport and session layer protocols.

A second interesting element concerns joint tortfeasorship.  Arnold J uses the language “induce, incite or persuade its users to commit [the primary tort]”.  This appears to be broader than the traditional approach to common design or procurement, and much closer to a common law version of “countenance, sanction or approve”.  It will be interesting to consider whether, on this approach, a marketplace such as eBay can be considered to “persuade” its sellers to advertise counterfeit products, or Google “incite[s]” confusing keyword advertisements by suggesting keyword targets.  This departure from the traditionally strict requirements of intent for joint tortfeasance may have unintended consequences.

A third high-profile takedown in as many days: this week looks...



A third high-profile takedown in as many days: this week looks like a losing one for pirates.  Notorious e-book platform Library.nu has been shut down, according to news reports.  The website, in combination with cyber-locker ifile.it, hosted around 400,000 titles — mostly academic textbooks and popular fiction.  Apparently, the operators received around €8m per year from donations, ads and premium accounts, though they claimed that most of this was spent on server costs.

Unsurprisingly, a consortium of publishers (including HarperCollins, Macmillan, Pearson, Oxford University Press and the International Publishers) led the take-down.  For academic publishers, it cannot come too soon: Library.nu was notorious for releasing ‘lossless’ PDF copies of latest edition academic texts weeks before they were even released — presumably leaked from offshore printers and compositors.  Unfortunately for publishers, numerous equivalent platforms continue to supply their copyright works with impunity.

Joining a growing list of takedowns this week, music...



Joining a growing list of takedowns this week, music blog RnBXclusive.com has had its domain name confiscated by the United Kingdom Serious Organised Crime Agency (SOCA). The move is unprecedented because it is the first (publicised) case in which a .com gTLD domain name has been seized by a UK government agency.  It is unclear what wrongs the website operator had committed, though a statement from SOCA suggests that they involve copyright infringement on a commercial scale:

The website in question specialised in RnB and enabled access to music obtained by hacking, including some which had not yet been released. IFPI estimates losses to legitimate businesses and artists caused by the site to be £15m a year. During the week running up to arrest phase the website had 70,000 users daily, mainly males aged 18 to 25 years. (emphasis added)

The concerning phrase here is ‘enabled access’.  If the music was not being hosted on the blog and was not actually procured or obtained by the website operator, it seems at least doubtful whether the website was committing an actionable wrong by linking or otherwise enabling access to the files in question.  At most, this might amount to the civil statutory tort of authorising copyright infringement.

Many have also pointed out that the warning message being displayed by SOCA in place of the site’s original content was highly misleading.  I look forward to hearing the matter determined fully in court.

In an astonishing display of executive power, popular business...



In an astonishing display of executive power, popular business form creation tool JotForm has had its domain name suspended by registrar GoDaddy at the ‘request’ of the US secret service.  It is unclear whether a court order was obtained or whether the request came from a state prosecutor.  Either way, this is a terrible denial of the right to a fair trial: JotForm was not notified of the action, has suffered irreparable damage to its business reputation, and it is unclear of what wrong they are accused.

Canada Supreme Court holds ISPs exempt from local content rules

In Alliance of Canadian Cinema, Television and Radio Artists Bell Aliant Regional Communications LP [2012] SCC 4, the Supreme Court of Canada upheld a finding that Bell, Rogers and other Canadian ISPs are not broadcasting undertakings” within the meaning of the Broadcasting Act 1991 (CA), which means they are exempt from the obligation to contribute to local content.  The Court held:

[2] ISPs provide routers and other infrastructure that enable their subscribers to access content and services made available on the Internet.  … Content providers depend on the ISPs’ services for Internet delivery of their content to end-users.  The ISPs, acting solely in that capacity, do not select or originate programming or package programming services. …

[5] An ISP does not engage with [cultural enrichment and diversity] policy objectives when it is merely providing the mode of transmission.  ISPs provide Internet access to end-users.  When providing access to the Internet, which is the only function of ISPs placed in issue by the reference question, they take no part in the selection, origination, or packaging of content. …

[7]  … since ISPs merely act as a conduit for information provided by others, they could not themselves be held to communicate the information.

This seems like a sensible decision which confirms that ISPs will ordinarily be mere conduits for the purpose of regulatory obligations imposed upon broadcasters and other content providers.  This is consistent with the position taken in copyright cases, and should be welcomed.  However, it leaves unresolved the issue of whether a vertically-integrated ISP which provides both content and connectivity would be classified in this way.

The decision also substantially calls into question the draft proposal in Australia which would require retail ISPs to contribute to the local content fund.

Canada Supreme Court holds ISPs exempt from local content rules

In Alliance of Canadian Cinema, Television and Radio Artists Bell Aliant Regional Communications LP [2012] SCC 4, the Supreme Court of Canada upheld a finding that Bell, Rogers and other Canadian ISPs are not broadcasting undertakings” within the meaning of the Broadcasting Act 1991 (CA), which means they are exempt from the obligation to contribute to local content.  The Court held:

[2] ISPs provide routers and other infrastructure that enable their subscribers to access content and services made available on the Internet.  … Content providers depend on the ISPs’ services for Internet delivery of their content to end-users.  The ISPs, acting solely in that capacity, do not select or originate programming or package programming services. …

[5] An ISP does not engage with [cultural enrichment and diversity] policy objectives when it is merely providing the mode of transmission.  ISPs provide Internet access to end-users.  When providing access to the Internet, which is the only function of ISPs placed in issue by the reference question, they take no part in the selection, origination, or packaging of content. …

[7]  … since ISPs merely act as a conduit for information provided by others, they could not themselves be held to communicate the information.

This seems like a sensible decision which confirms that ISPs will ordinarily be mere conduits for the purpose of regulatory obligations imposed upon broadcasters and other content providers.  This is consistent with the position taken in copyright cases, and should be welcomed.  However, it leaves unresolved the issue of whether a vertically-integrated ISP which provides both content and connectivity would be classified in this way.

The decision also substantially calls into question the draft proposal in Australia which would require retail ISPs to contribute to the local content fund.

Eolas jury holds embedded hypertext patent invalid

Eolas, a company which many describe as a “patent troll”, has suffered a serious set-back in litigation against, Google, Amazon and others.  As has been widely reported, a Texas jury found that the patent was invalid for want of novelty.  So, what was the patent?  According to the specification:

A system allowing a user of a browser program on a computer connected to an open distributed hypermedia system to access and execute an embedded program object. The program object is embedded into a hypermedia document much like data objects.

Relevantly, claim 1 claimed:

A method for running an application program in a computer network environment, comprising:

  1. providing at least one client workstation and one network server …
  2. a browser application … to display, on said client workstation, at least a portion of a first hypermedia document received over said network from said server … and wherein said embed text format is parsed by said browser to automatically … display [it] and enable interactive processing …

In other words, the plaintiff claimed something akin to the <embed> tag, utilised in a web browser.  After expert evidence from Tim Berners-Lee and Dave Ragett, the jury found that the claimed invention was not new as at 17 October 1994.

Tough luck for those who already settled with Eolas, which include Apple, Citigroup, eBay, and JP Morgan Chase among their number.  Restitutionary claim, anyone?

Eolas jury holds embedded hypertext patent invalid

Eolas, a company which many describe as a “patent troll”, has suffered a serious set-back in litigation against, Google, Amazon and others.  As has been widely reported, a Texas jury found that the patent was invalid for want of novelty.  So, what was the patent?  According to the specification:

A system allowing a user of a browser program on a computer connected to an open distributed hypermedia system to access and execute an embedded program object. The program object is embedded into a hypermedia document much like data objects.

Relevantly, claim 1 claimed:

A method for running an application program in a computer network environment, comprising:

  1. providing at least one client workstation and one network server …
  2. a browser application … to display, on said client workstation, at least a portion of a first hypermedia document received over said network from said server … and wherein said embed text format is parsed by said browser to automatically … display [it] and enable interactive processing …

In other words, the plaintiff claimed something akin to the tag, utilised in a web browser.  After expert evidence from Tim Berners-Lee and Dave Ragett, the jury found that the claimed invention was not new as at 17 October 1994.

Tough luck for those who already settled with Eolas, which include Apple, Citigroup, eBay, and JP Morgan Chase among their number.  Restitutionary claim, anyone?

"Either way, when international accords, like ACTA, are conceived and constructed under a cloak of..."

““Either way, when international accords, like ACTA, are conceived and constructed under a cloak of secrecy,” [Senator Ron] Wyden [D-OR] said, “it is hard to argue that they represent the broad interests of the general public. The controversy over ACTA should surprise no one.””

- ACTA ‘leaves to the discretion of each country the definition of what constitutes a “commercial” level of piracy, so some countries might choose to search travelers’ laptop computers and digital music players in search of illegal downloads. ACTA supporters reject the criticism and say action is essential when legitimate owners of intellectual property are losing tens of billions of dollars annually to counterfeiting and illegal sharing. They accuse some opponents of deliberately exaggerating ACTA’s provisions to fan fears.’

Liability for defamatory statements in Twitter hashtag feeds

Brett Debritz asks an interesting question:

what if somebody were to tweet an extremely defamatory or racially offensive remark using the Nova-nominated hashtag and it ended up on the company’s website for a sustained period of time? What if somebody sued? Who would be responsible: the author (if they could be identified) or the publisher?

Different jurisdictions will answer this question differently.  In England (and probably Australia), the question is likely to turn on two factors: (1) whether the website operator is regarded as the “publisher” of the relevant statement; and (2) whether the operator had actual knowledge of the statement.  By analogy with Bunt v Tilley and DesignTechnica v Metropolitan Schools & Google, there is a strong argument that the “publisher” is the person who authored the defamatory statement and caused it to be posted to Twitter.

Although Byrne v Deane suggests that someone who controls a facility (in that case, a bulletin board in a club house) to which defamatory material has been posted can be liable for its dissemination, there is little that a website operator can do to control the flow of comments in a Twitter hashtag feed: short of removing the feed widget entirely, tweets cannot be deleted and what shows up there is entirely within the purview of tweet authors and Twitter.  Whether Twitter faces liability is a more complicated question, and is likely to start implicating questions of safe-harbours under the E-Commerce Directive.

As to the second question, it is plausible that once the website operator has been notified they may become liable as a joint tortfeasor (or publisher) if they fail to remove the defamatory material within a reasonable period.  To the extent that the website is treated as a publisher of the material, actual or constructive notice would also preclude reliance on an innocent dissemination defence.  Upon receiving notice, the website operator would probably be required to remove or block access to the entire Twitter widget.  While this may seem disproportionate, there doesn’t appear to be any way to prevent the display of individual tweets - unless of course Twitter removes them directly.  The level of notice required, and the period of permissible delay, remain unclear, but existing case law (at least in England) suggest that some leniency is likely to be offered — particularly where the website operator acts with reasonable diligence and propriety.