Apple Patenting the Obvious Again

I enjoy coding, and many would say I'm quite intelligent. But, by no means, am I some sort of super-being. With this in mind, I'd say if I was able to think up a technique then it's more than possible it was too obvious to patent.

So, given Apple have just been granted this patent, I felt it was worth highlighting that it's not new, it is obvious and has already been done!


Apples' Patent  "Techniques to pollute electronic profiling" essentially describes having a bot browse on your behalf in order to pollute the stream of anyone who may be profiling your browsing behaviour. In the real world, this is most likely to be advertisers, but there's no reason it couldn't also be law enforcement bodies and the like.

The bot can be programmed with the user's (or Principle's in patent speak) interests, but is mainly aimed at 'browsing' information that doesn't fit with the users actual interests. The idea being that the database of "x likes y" held by advertisers becomes far less valuable thanks to the innaccurate data collected from the bot browsing on your behalf.

Nice idea, sure, but not patent worthy. I wrote a script to do this in 2008 (and I don't claim to have invented the technique either) to amuse myself when it looked like BT were actually going to roll out Phorm's Webwise system. I even published it on my site (back then, it was!) for others to tweak and use. Apple's patent was filed in October 2011, so it's safe to say my script pre-dates that (17 June 2008!).


Colour me Cynical

There's a cynical part of me that suspects this is aimed at raising advertising revenue for Apple. At the moment, anyone capable of deploying Javascript across the net (ahem.. Google) can track a users browsing habits, and tailor ads to the detected preferences. Throw a bit of data pollution into the mix and it becomes far, far harder. Apple, of course, could still be able to aggregate this data (and I'm not saying they definitely will) which would become particularly valuable if the traditional methods of stalking tailoring ads were to become ineffective.

Regardless of the motivations, it's the Patent I object to. Apple's recent behaviour indicates that it's become a litigious shit with a stronger focus on lawsuits than actual innovation (I had a Mac SE that I loved, last bit of Apple kit I actually liked) so the granting of another patent is never good news. Given that I know, for fact, that there's prior art it seemed important to share it for the day when Apple decides to try and file suit against someone.

I mentioned earlier that I don't claim to have invented the technique of polluting an advertisers stream, which means that there's probably a shitload of prior art out there. Firefox's TrackMeNot achieves a similar end from within a browser, and I'd hazard a guess that it's also been done for Mobile already. The only real difference is that neither I, or TrackMeNot bothered to implement a 'bot chat' function for chatrooms/IRC etc. Bet someone out there did though.



I've attached the files I posted back in 2008 (note the date on the PDF), but for easy reading, here's how simple Apple's 'invention' actually is


# Urlrequest released under the GNU GPL v2 copyright B Tasker 2008
# A simple script to clog a Phorm client with lots of unrelated keywords
# thus reducing its commercial viability

#Before running this script you need to turn WebWise on,
#grab the cookie from your browsers cache, and then turn WebWise off

cd /home/$USER/.urlrequest
wget --save-cookies=$WORKDIR/wgetcookies \
--load-cookies=$WORKDIR/wgetcookies \
--keep-session-cookies -r --random-wait \
--ignore-length -nd \
-R .gif,.jpg,.txt,.pdf,.js,.js.1, \
--spider -i $WORKDIR/urlrequesturls
rm -rf /home/ben/.urlrequest/*

# Be aware that you will be using up your bandwidth, 
# not at any great rate, but bandwidth does cost money!!!!


I've a feeling that I also wrote a more advanced version that based the delay between requests based on the size of the HTML page being retrieved (bigger pages often contain more text, so it represented someone taking a while longer to read), but can't find whether or not I posted it. It also allowed you to set the user-agent (the default was something rude about Phorm). I probably have it somewhere, but it's not that important.

It's not a great implementation (I've no idea why I decided to require ~/urlrequest and ~/.urlrequest for a start!) but it works well when populated with a sensible list of URLs. Granted the list accompanying the script wasn't exactly sensible ( is always a classic!) but the later version that searched for a category and then browsed on it's own enhanced it a bit.



There's probably reams of prior art, but as the USPTO seem to have missed it (there's a surprise), I thought I'd contribute my little bit. If it helps prevent any of Apple's iTrolling then it's worthwhile.

I have to admit, I was pretty amazed to see the patent given how obvious such a solution really is, but then who holds a design patent on rounded corners? I was even more amazed when it occurred to me that I'd created something similar more than 3 years prior to Apple filing their patent! It's just a pity I can't link to my original posting on though I may check the archived version so that I can look it up in the Internet Archive (some of my stuff made it in, be great if this was there too!)


File: urlrequest

File: urlrequest.pdf