Purpose:
No active development here, just a jumping-off place for
finding CueCat-related info. Linux-related links will be emphasized,
though I'll gladly link to other resources as well, particularly those
whose source code is freely (speech) available. Please submit relevant
links if you don't see them here (other than the well-known corporate
sponsors / early adopters who, if you need the reminders, may sound like
but are not named DigitalDivergence.com, See-Our-Que.com, Ray D.
O'Shack, WhyRed, & Forb's).
And no, unless suitably threatened by TM-holders, I won't be using colons in the name, as listed in the mfr's adverts. They make it annoying to type, and even more annoying to search for. However, if pressed, I'll gladly follow the lead of one /. discussant and refer to this item as the colon cat (is that how it gets read via text-to-speech interface?)
Slashdot is where I first read about the cuecat device, in their
Aug 22 edition . A number of the comments are informative and/or
thought-provoking. Searches for cue & cat via the usual search engines
turned up little of interest, unless you count messages re: an old
vulnerability in HPUX. I recently submitted this link to Google; lessee
if this gets any more traffic going (as of Sept 12, they're still
mum on the subject, except for the commercial cuecat sites--but
look below).
Pay a visit to www.flyingbuttmonkeys.com to read about Michael Rothwell's "foocat barcode" project. Mr. Rothwell's page also links to a document that nearly answers the question, "Can a letter contain the words 'flyingbuttmonkey' and still be considered a 'persuasive communication' ?"
The nice folks at mp3car.com have put up a rather interesting CueCat Reverse Engineering page, complete with links to generic barcode info and color photos of a cue-feline dissection. They've even put a form for direct cuecat data entry on their page, that lets you scan an item in the (relative) privacy of your home and get the translated version of the data via your browser (after disabling your CRQ software, of course).
http://www.new-sharon.me.us/upc.html
http://docwhat.gerf.org/software/perl/catscan/
Mark Duell tells me that no longer has a cuecat.html page up; nowadays you can read (just a little) about his Reverse base-64 decoder at http://www.fleacircus.org/~mduell/decoder/index.html.
Get the insider's view of the cuecat at matrixpm.com/~haveblue/cuecat/.
Christopher Curzio's
CueCat resource page
See the above-referenced Slashdot announcement and ensuing brawls ^H^H^H^H^Hgenteel discussions following the original article and a second, "slashback" discussion on SlashDot for details.
Despite my gratitude for their wonderful gesture, however, I must admit feeling some uneasiness about this happy event. A small, nagging voice persistently suggests that someone out there--totally unrelated to any of our benefactors, of course--has another plan in mind. It would seem probable that software could be compiled and distributed, giving no clues to users, that could enable the wonderful and obvious features ascribed to the :Cue system, while covertly enabling other things.
For instance--since the cuecat sends its unit-id to the computer, the software could convey that information to the target web host easily, secretly--and unnecessarily. Just one registration-form later, and the owner of the web site may have the user's name, cuecat unit-id, and some typical cookies, all at once. This could provide enormous "marketing value" to the website owner, who is now in a position to sell the user's identity to targeted-marketing interests. The user unknowingly gives up yet another layer of privacy, while someone else profits by that very loss. The privacy implications rival those of Intel's ill-fated unique processor-id, which similarly enabled the long-term, definitive identification of specific computer hardware over the network, by covert means. In this case, however, the risks could be easily sidestepped by making the software source code public.
Those who developed and distribute(d) the currently closed-source software for the CueCat, would seem to have incentives that do not favor rigorous privacy protections for users. Keeping it closed-source further heightens the risks to users while retaining maximal benefits for the advertisers. This requirement, that users blindly place trust in this software as given, without means to verify its security and privacy protections, creates a stiff price of admission for the use of the CueCat device. Revealing the code to the software would be one clear way to demonstrate goodwill, and to prove that unit-id information is not being made available as a customer-tracking device for the paying corporate customers. And for the user, it wouldn't feel so much like they were being asked to let the fox guard the henhouse. Developers could verify that unit-id information is masked at the driver level, thus reducing the risk of deliberate or inadvertent privacy disclosures.