Porn recognition software?

The USPTO continues to amaze me. I mean, one would hope that any silly patent that can be awarded has already been granted. How could it get any worse? Regard US Patent 6,751,348 for protecting an "Automated detection of pornographic images ". The abstract states that the patent covers:

A method of detecting pornographic images, wherein a color reference database is prepared in LAB color space defining a plurality of colors representing relevant portions of a human body. A questionable image is selected, and sampled pixels are compared with the color reference database. Areas having a matching pixel are subjected to a texture analysis to determine if the pixel is an isolated color or if other comparable pixels surround it; a condition indicating possible skin. If an area of possible skin is found, the questionable image is classified as objectionable. A further embodiment includes preparation of a questionable image reference shape database defining objectionable shapes. An image with a detected area of possible skin is compared with the shape database, and depending on the results of the shape analysis, a predefined percentage of the images are classified for manual review.

I love that "If an area of possible skin is found, the questionable image is classified as objectionable". Good-bye to the well-loved "I know it when I see it" Casablanca test.

Yet another Google story

(Via IPKat) It seems like everywhere one looks these days the techie press and the blogosphere are all talking about Google. First Google and China, then the Google Search function. All interesting stories, particularly the strange fascination with "free porn" on the part of American lawmakers and Fox News (surprise, surprise).

There is another story that has caught my eye. The IPKat reported that Google has won a copyright case regarding Google's cache function. The Google cache is part of their indexing process that keeps copies of pages. If you have never clicked on one of the "Cached" buttons on each search, this is how one looks. There is no doubt that there is a copy made in some Google machine, but is that copy an infringement? A Nevada judge has ruled in Field v. Google that the cache constitutes copying, but that it falls under fair use provisions. In his decision, he took into consideration that the cache has considerable public functionality and that Google provided technical means that allowed to exclude being included in the cache. On the contrary, claimant Field's conduct was suspect:

Field'’s own conduct stands in marked contrast to Google'’s good faith. Field took a variety of affirmative steps to get his works included in Google'’s search results, where he knew they would be displayed with 'Cached'” links to Google'’s archival copy and he deliberately ignored the protocols that would have instructed Google not to present '“Cached'” links.

Interesting reasoning, but I wonder how would Google fare in a jurisdiction that does not follow the broad American fair use concept. Prof Lenz has three posts about the case, starting here.

Patenting emoticons in mobile use

I thought that I had read it all when it came to silly patents. Apparently not. The Register reports on U.S. patent application 20060015812 by mobile giants Cingular, attempting to patent the use of emoticons in wireless devices. The abstract reads:

A method and system for generating a displayable icon or emoticon form that indicates the mood or emotion of a user of the mobile station. A user of a device, such as a mobile phone, is provided with a dedicated key or shared dedicated key option that the user may select to insert an emoticon onto a display or other medium. The selection of the key or shared dedicated key may result in the insertion of the emoticon, or may also result in the display of a collection of emoticons that the user may then select from using, for example, a key mapping or navigation technique.

The patenting of a key that displays emoticons. Who would have thought of that? :-) There is no novelty here, and I would say that this application is thoroughly obvious, and therefore should not get a patent. The application goes on to explain what is understood as an emoticon:

:-) or :) Smile ;-) or ;) Wink :-D or :D Big smile :-)) or :)) Very happy :0) Big nose smiley |-) Cool! >:-) or >:) Evil grin >;-> or >;> Evil grin with a wink :-X or :X My lips are sealed }:-) or }:) Devilish :-{circumflex over ( )}) or :{circumflex over ( )}) Tongue in cheek :-P or :P Sticking out tongue :-& or :& Tongue tied :op Puppy face 0:-) or 0:) Saint :-)8 or :)8 Happy wearing a bow tie 8-) or 8) Happy with glasses #-) I partied all night %-) or %) Drunk :-###.. or :###.. Being sick %-( or %( Confused :-0 or :0 Shocked :-o or :o Surprised :-| or :| Indecision :'-( or :'( Crying :'-) or :') Crying of happiness :-( or :( Sad

:-&

NOTE: It seems like Andrew Orlowski from The Register continues to get his stories wrong. The patent claim is not to patent emoticons, it is to patent a key in a mobile phone that inserts them into the text. Big difference.

Trolls eat Blackberry

(via IPKat) A very unfortunate turn of events for innovation. The U.S. Supreme Court will not hear the case of Research in Motion Ltd (RIM), the inventors and makers of the famous (and perhaps infamous) Blackberry, versus patent trolls NTP Inc. RIM has been trying to attack an injunction against them resulting from the patent litigation initiated by NTP, where they claim that the Blackberry infringes five of their patents (including patent 5,436,960). What I find disturbing about this case is that NTP clearly is just a troll filing broad patents (read the above link and laugh and/or cry). They never built a thing, they sit on their patent and wait for the true innovators to start making a product, then attack them with an expensive infringement suit.

Although workaholics everywhere are concerned that their gadget will disappear, bookies are not counting that the Blackberry will go out of season (to my count, at least three articles already use that title), the odds are 2-1 against that happening.

Moving up in the ecosystem

I have become addicted to The Truth Laid Bear's rankings. Incredibly, I've been upgraded to Multicellular Organism (I was an Insignificant Microbe just last week). I have decided to place a link indicating what my status is.

I have also noticed that I have now received more than 10,000 hits in the blog since I started keeping statistics. Thanks to all the constant and not so constant readers!

Piracy calculator

(via Denis Magnusson). How much are your illegally downloaded copyright works worth? (well, they are not really yours, they belong to the copyright owners, but you know what I mean).

Use the piracy calculator and find out! I was surprised of how little my stuff is worth, but I have never downloaded that much. Add that to the fact that I download music that I don't get in my local HMV (when are they going to carry Maná and Juanes anyway?). Most interestingly, whatever figure you get, that assumes that you would have bought the items, which is always a flawed assumption to make.

The truth about the Wikipedia German story

The techie press has been filled with reports that Wikipedia Germany was shut down for three days for hosting an article that revealed the name of German hacker Boris Floricic (aka Tron). The story has also been taken up by the blogosphere. But is the story accurate? Not according to Wikipedia founder Jimmy Wales.

Jonathan Zittrain posted a message in the Cyberpof mailing list reproducing Wales' opinion about the story. In it, Wales claims that the site was never shut down, not even for an instant, as the servers are in the United States and not in Germany. More importantly, the Wikipedia Foundation has never received a court order.

I have been doing some research of my own. The domain is still working at http://de.wikipedia.org. I performed a traceroute of the domain and it is indeed hosted in the United States, seems to be somewhere in Tampa, Florida. I would then guess that the story is false, and that it was one of those journalistic memes that have never been checked by the reporter. Did they even click on the site to see if it was available?

How did the story start then? It seems like there may be a case against the site http://www.wikipedia.de, which hosts a group of German Wikipedia enthusiasts. I don't read German, so a translation of the text on the site would be welcome.

It seems like the press gets it wrong (again).

[UPDATE] I've been looking more on this and I still cannot find corroboration that the Wikipedia foundation has actually been sued. The injunction seems to exist against the wikipedia.de site, but as that site does not host any content, the injunction would seem to be completely useless. Not only useless, counterproductive. Now more people know about Tron than ever before.

Software Patent Directive round II

(Thanks to Hector MacQueen for the link) IP Watch is reporting that a group of pro-software patent lobbyists and big industry representatives are intent in proposing that there should be another attempt to pass a European directive allowing the patentability of software (pardon me, the patentability of computer implemented inventions). The group met in Prague for the Progress and Freedom Foundation (PFF) conference, and it had a large number of representatives stating that a directive is vital for Europe's competitivity. Most of those interviewed claimed that they had learnt from the mistakes of the past and that they would lobby the European Parliament more efficiently.

Wonderful thing democracy. If you don't like the result, try and try again.

Nobody reads my blog

This is a very funny website with the top ten reasons why nobody reads your blog. True, I will never be a good-looking naked female from Tonga, but in my opinion the list is obsessed with tech consultancy.

Most intriguing to me is the fact that it does uncover a fact of web publishing. Clearly, the Internet has given us the chance to publish our thoughts easily and to reach the world. But is there anybody out there? For most blogs, the answer is no. As with many other things on the Web, blogging follows power law distributions, a few blogs get all the hits, while the large and vast majority of blogs gets few visitors (if any at all). Blogging follows scale-free networking, which can be followed through the cross-linking in blogrolls.

How do blogs compare? Check out The Truth Laid Bear to measure blog popularity. I'm ranked 37,346 (in the "Insignificant Microbe" category). By comparison, Boing Boing is 6th, Legal Theory Blog is ranked 724, Lenz Blog is ranked 8450, and IPKat is ranked 28,103 (surprising, but it seems that the really popular blogs are American).

More on GPL v3

I have been looking in more detail at GPL v3 and also reading some comments in mailing lists, and I am starting to become deeply troubled by the new version of the world's most popular copyleft licence.

In the good old version 2, the viral aspects of the GPL only apply to derivative works that are distributed to the public. In other words, you take a work under GPL, change it and then distribute your own adaptation, then you have to re-distribute it under the GPL (I know that some people do not like the word viral, but there you go). This simple rule would generally provide clear-cut cases in which the GPL would apply and where it would not. For example, imagine that a hardware developer creates a driver module that interacts with the Linux kernel (distributed under the GPL). It is not part of the kernel, but it interacts with it. Under GPL v2, it is clear that this module is not a derivative, and therefore it does not need to be distributed under the GPL. In fact, I often receive messages in my Linux bootup process that the kernel is "tainted", this usually means that I am loading a hardware module that is not licensed under the GPL, or that is even proprietary (such as the wireless module provided by Linuxant).

What happens with GPL version 3? The situation with derivatives is less straightforward. The viral copyleft clause in the new GPL has been given a tremendous boost. Users and developers still have the right to install and use GPL software without restrictions. Developers also have the right to privately modify the software, unless they have initiated a patent suit "against anyone for making, using or distributing their own works based on the Program." The problem is that of modifications that are distributed. Consider section 5(b) (the old section 2.b):

b) You must license the entire modified work, as a whole, under this License to anyone who comes into possession of a copy. This License must apply, unmodified except as permitted by section 7 below, to the whole of the work. This License gives no permission to license the work in any other way, but it does not invalidate such permission if you have separately received it.

This would certainly apply to anyone who includes any sort of modification of code into their work. Imagine for example that you include modified modules from the Linux kernel in your work in order to provide compatibility with the kernel. It seems like you would definitely have to licence the entire program under the GPL! This would not apply if the work can be clearly identified as not being a derivative. The new GPL says:

These requirements apply to the modified work as a whole. If identifiable sections of that work, added by you, are not derived from the Program, and can be reasonably considered independent and separate works in themselves, then this License, and its terms, do not apply to those sections when you distribute them as separate works for use not in combination with the Program. But when you distribute the same sections for use in combination with covered works, no matter in what form such combination occurs, the whole of the combination must be licensed under this License, whose permissions for other licensees extend to the entire whole, and thus to every part of the whole.

This is extremely harsh in my opinion. Even if your work is not based on GPL software but it is distributed in combination with covered works, then it must be licensed under the GPL! What will happen with open source companies that do not choose to use the GPL, such as Mozilla and Apache? And what about interfaces? At first glance, I believe that this would also affect drivers and modules that interact with the kernel. However, the draft further confuses this by stating:

A compilation of a covered work with other separate and independent works, which are not by their nature extensions of the covered work, in or on a volume of a storage or distribution medium, is called an "aggregate" if the copyright resulting from the compilation is not used to limit the legal rights of the compilation's users beyond what the individual works permit. Mere inclusion of a covered work in an aggregate does not cause this License to apply to the other parts of the aggregate.

I said yesterday that the new GPL seemed clearer. Strike that. This paragraph further confuses the subject. This paragraph tries to rationalise specific cases in which the ridiculously revamped GPL copyleft clause will apply by inventing a new definition to what a compilation is. They try to determine that the software distributed in the same distribution medium has to be GPL if it is a "compilation", but not if it is an "aggregate". Why create the new terminology? For example, under GPL 3 my SuSE Linux distribution could not include any proprietary software at all, such as the PDF reader for Linux or even the Nvidia drivers for my video card! I know that some people do not like to "taint" their distributions, but I rather like that I am able to use the video card as intended and read PDF files right after installation.

I think that the drafters of the new GPL realised that the new copyleft clause would be controversial when they included an excuse within it:

Thus, it is not the intent of this section to claim rights or contest your rights to work written entirely by you; rather, the intent is to exercise the right to control the distribution of derivative or collective works based on the Program.

We have a saying in Costa Rica, "el que se excusa se acusa", he who excuses himself accuses himself. Why excuse the clause in the licence?

It is great that this is still in draft form, and I hope that some things get changed!

GPL v3 draft released

The draft of the version 3 of the GNU General Public License (GPL) has finally been released, and what a document it is! In general, the wording is clearer, tidier and tighter than version 2. The preamble is still as ideological as always (an in my opinion a tad superfluous). But the real surprise comes in the main text. Before its release, it was rumoured that the text would attempt to overhaul the protection of GPL software against software patents, but there was less talk about the amazing implication to digital rights management and technical protection measures. The new GPL could have a lot of implications for the future of open source usage in the entertainment industry. One of the most contentious sections is likely to be section 3 on DRM:

As a free software license, this License intrinsically disfavors technical attempts to restrict users' freedom to copy, modify, and share copyrighted works. Each of its provisions shall be interpreted in light of this specific declaration of the licensor's intent. Regardless of any other provision of this License, no permission is given to distribute covered works that illegally invade users' privacy, nor for modes of distribution that deny users that run covered works the full exercise of the legal rights granted by this License.

In other words, the distribution of derivatives with works that contain DRM will be prohibited. The new draft also attacks the anti-circumvention measures in the WIPO Copyright Treaties by specifically stating that no work under the GPL shall constitute "an effective technological protection measure", and it would therefore not apply for such protections.

Surprisingly, the new version is not as adamant against software patents as it was expected, perhaps as a reflection that some of the biggest open source and free software players in the United States are acquiring patents as well. The new draft maintains the licensing of patents for those who use the GPL that already existed in the older version (section 11 of v3). The licence prohibits distribution of the software by people who have initiated or brought patent suits arising from the software. This is a clever attack on software patents, it does not forbid people to apply for them, but it frowns upon enforcement. Strangely, the licence allows developers to add a stronger patent retaliation clause that is not incompatible with the licence. This clause is spelled out, but it is not really a part of the licence, it is just not incompatible with it. This is terribly confusing in my opinion, and bad drafting practice. The compatible patent clause states:

e) They may impose software patent retaliation, which means permission for use of your added parts terminates or may be terminated, wholly or partially, under stated conditions, for users closely related to any party that has filed a software patent lawsuit (i.e., a lawsuit alleging that some software infringes a patent). The conditions must limit retaliation to a subset of these two cases: 1. Lawsuits that lack the justification of retaliating against other software patent lawsuits that lack such justification. 2. Lawsuits that target part of this work, or other code that was elsewhere released together with the parts you added, the whole being under the terms used here for those parts.

The final verdict will undoubtedly come soon, but it seems clear that the GPL is becoming more restrictive, and I would not be surprised if larger enterprises stay away from the new GPL and favour other licences.

Jacob against software patents

Court of Appeals judge Sir Robin Jacob has made a vocal attack against the American practice of patenting everything under the sun, and has stated that there is no evidence that software patents encourage innovation.

"The United States takes the view that anything made by man, under the sun, can be patented. And they have granted patents for business methods, mainly computer business methods. But as far as I can see, it would cover a new and improved method of stacking oranges on a barrel," Jacob said.

Interesting words from one of the UK's top IP judges.

Pledge against DRM

Here is a pledge at PledgeBank that proposes that "I will pledge to never purchase a CD containing any form of Digital Rights Management (DRM), but only if 500 people around the world will too." I have signed this as I generally dislike the practice of a company telling me how I can use my technology (I do have two Linux machines). I am also concerned about the privacy and security implications of technical protection measures as exemplified by the Sony case.

However, I wonder if threatening a boycott will really change things in the industry.

HD-DVD vs Blu-Ray: The fight continues

There is a very good article in The Guardian about the new format battle between high definition DVD replacements, Blu-ray and HD-DVD. Sony is banking on Blu-ray by including it in the Playstation 3, which will mean that they will have some captive audience and a large boost from the popularity of the PS brand.

Both systems will come with buffed DRM protection.

The high cost of patents

The EPO has just released a study conducted in 2004 by consulting firm Roland Berger Market Research on the cost of patenting in Europe. According to the study, an average EPO patent costs €30,530 EUR, where the costs are higher for Japanese and American companies validating in Europe. As further comparison, the average for a European patent applicant is €24,100 EUR, while an American firm has to pay an average €10,250 EUR for a patent by the USPTO, and a Japanese firm pays €5,460 EUR for a JPO patent.

The report goes on to state that the cost of a patent in Europe is higher because of higher translation costs. It should also be said that in Europe you get better quality and service than what you get at the USPTO. Nevertheless, the costs do seem to be prohibitively expensive for small firms, not to mention "individual" developers and inventors. The time of Thomas A. Edison is past.

IBM to improve software patent quality

IBM has joined other big players in the open source community and has expressed that it will use OSS in order to improve the quality of software patents in the United States. The idea is to have a searchable software code repository that will make it easier to detemine if the software has prior art or not. The other people involved are Open Source Development Labs, Novell, Red Hat and SourceForge.net.

This is yet another encouraging piece of news from Big Blue, but I wonder if the pro-patent lobby will agree to release source code.

Back from the tropics

I am now back from the tropics. Thanks to Arne Kolbe, who kindly accepted to fill in with some posts while I was away sipping piña coladas and Imperial.

Kill the Christmas Blues

Are you feeling the Christmas blues? Are you sick of eating fruticake, turkey or tamal? Do you think that if you see another picture of Santa you will leave the room screaming? The solution is finally here! Let Slingshot Santa cure all of your problems.
(Thanks to Rena for the link).

Database right review

As usual, James Boyle has made an excellent comment in his always informative FT column, this time about the recently announced review of the database right.

Could we be facing a turn in IP policy? As far as I know, there has never been a situation where an IP right has been eliminated.