Slashdot Mirror
Author Unknown
With unattributed text, says word sleuth Don Foster -- an e-mail from a Hotmail address, the rantings of the Unabomber, or an anonymous letter to the editor -- it is increasingly possible to connect the voice with the creator of the document.
Most anonymous texts, from Elizabethan playscripts to libel on the Net, offer stylistic evidence that reveal a lot more than many scholars and detectives have previously realized. Words are our own intellectual DNA, writes Foster, a professor of English Lit at Vassar and perhaps the world's foremost word sleuth. Analyzed in the right way, they invariably will give us away.
Foster, the author of Author Unknown, has some solid credentials in this field. He solved a puzzle involving one of Shakespeare's sonnets that had stumped sleuths for centuries, identified the anonymous author of presidential tell-all Primary Colors (unmasking the journalist Joe Klein as the author), was enlisted by federal authorities in the hunt for the Unabomber, and was also asked to uncover the author of the celebrated Monica Lewinsky-Linda Tripp "talking points."
This is highly readable book, an intellectual who-dunnit of particular interest to people online who are continuously confronted with anonymous texts. And if you're hiding behind electronic anonymity, you don't want Don Foster on your case. The book tells juicy stories about some of the most sensational crimes and scandals of our time, but it's really about the human mind: the language, identity and the clues that individuals leave behind when they create text, digital or otherwise.
Criminals can ride or hide, Foster writes, but they can't disguise their words and language patterns. We are prisoners of our own language, writing from within a repertoire of certain thoughts and words and spellings.
Some words, Foster writes, are content specific. "Two documents about making salad from "dandelion greens" may have been written by the same person (in this example, Ted Kaczynski) or one writer may have borrowed from another; but if two documents about gardening mention the words "dandelion," "hoe", and "trellis," that may indicate not common authorship or indebtedness but only a shared topic."
Some of the most riveting parts of this book are the insights into Kaczysnki's character and intellect that Foster was able to piece together from a meticulous study of his letters and manifestos. Foster makes Kaczynski more comprehensible than anyone has managed to do, mostly by tracking down the influences, stories, books and writers that pop up again and again in his writings, beliefs and help explain an enduring mystery about his awful work: his choice of victims. We see how Kaczynski, holed up in various libraries, came across biographic references, academic writings or writings that triggered passions and interests (like the story of the Titanic, which obsessed him as a metaphor for failed technology and technological hubris, an obsession that cost lives).
Foster writes about how advanced Geographic Profiling used by police to track down serial criminals didn't work in the Unabomber case, since the offender's primary residence (Montana) and place of employment (none) couldn't have been pinpointed even with the help of advanced computer tracking software. But by locating Kaczynski's words and ideas, pseudonyms, even mailing addresses; by locating his books and magazines, reference works and principal intellectual influences, something could have been learned about the Unabomber's physical whereabouts, even down to the particular buildings in Utah and Northern California where Kaczynski was conducting his primary research, (including the probable dates of his most recent visits).
Foster writes that it was the famous Unabomber Manifesto, "Industrial Society and Its Future," completed and mailed in June 1995 that ultimately led the Unabomber Task Force to Kaczynski's Montana cabin. "I believe," he writes, "that the same wonderfully verbose document, partly written in California libraries, could have led agents to Ted Kaczynski even without David Kaczynski's invaluable assistance."
Foster was repeatedly led to archived stories from "Saturday Review" which Kaczynski read. The Unabomber borrowed heavily, says Foster, especially from the writings of Jacques Ellul. It was in February 1965 that American readers were first introduced to Ellul's "Technological Society" in that magazine. Kaczynski wrote his brother David that he was deeply impressed by Ellul's writings which closely mirrored his own beliefs. These ideas pop up all over Kaczynski's writings, letters and manifesto.
Author Unknown is a great read, timely and riveting, and with special relevance for cyberspace. "In a culture that encourages anonymous communication and the right to speak without responsibility for the content of the utterance, the spoken message and, eventually, language itself are depleted."
There may be a flip-side to Foster's new science. In the wrong hands, it could mean that the kind of free speech and safety sometimes associated with anonymity -- Anonymous Cowards on this site can be obnoxious, but they also pass on valuable information -- could be lost.
But thanks in part to Foster's ground-breaking research, speaking without responsibility may one day be tougher to do.
You can purchase this book at ThinkGeek.
true, but generating an account on /. doesn't necessarily reveal your identity. Ie, how many Bruce Perens clones are there?
The USPS can't track where mail originates?!? UPS and FedEx both can do that. Yes, they use omputer tracking software. No, it isn't advanced software (it was at one time). No wonder the government keeps competitors from delivering regular 1st class mail: the post office would go out of business in a week.
Don't the strings on the viola get in the way of erasing anything?
For someone who is so clever about forensic linguistics, it is a bit surprising for Foster to rather blatently misuse words. It almost makes me doubt just how meritorious is his braggadocio.
In point of fact, if Ted Kaczynski could have been found even without his brother's assistance, David Kaczynski's help may have been valuable... but it was not invaluable.
Buy Text Processing in Python
For more details, see Farringdon, Morton, Farringdon & Baker's book Analysing For Authorship, University of Wales Press 1996; ISBN: 0708313248.
ToiletDuk
Protector of the Wastes
I am sorry I have to disagree. The cream is increasingly being washed away by all the sour milk. The trolls down to just deplete the message they pi** in it as well.
You say this is not so but look at most newsgroups.
Common end lusers do not bother using newsgroups not because a pretty interface could not be drawn around it through a slick GUI client or all the porn but because moderated www boards served their purposes of talking with other women going through a breast cancer experience without some complete b*st*rd jumping on and taunting them
with jokes about breasts cancer or both.
The whole idea of anonymous communication is a geek thing and the whole idea of the good stuff getting the exposure is not necessarily true. Search engines and portal sites are the only way to get around all the cruft and crud on the web and it becomes more and more annoying to list through the same 50 people saying exactly the same thing on Slashdot and getting modded up the exact same way instead of being listed as redundant.
ACK
Neo/Zargon/Multiple Bogeys/Numerous Boogers on a certain rec.games newsgroup was like this. mutate, mutate, mutate.
Sounds double-plus-good.
Yes, the review said that, but it said that he believed the Unabomber would have been located even without such assistance.
Donte Alistair Anderson Roberts - hi son!
Karma: Chameleon
I didn't think it was a voltaire quote. I've seen it attributed to him in the past, but I've also seen variations attributed to a million others. Mark Twain most of all.
I didn't think you were wrong or right, I just thought that it was a funny parallel issue.
--
There are no trails. There are no trees out here.
I suppose one of the best uses of this technique would be to properly determine the originator of your sig quote.
--
There are no trails. There are no trees out here.
While the transformation was interesting, isn't the whole object of the exercise to get out, after translation, something that could be at least comprehended to mean what the original did? In other words, by demonstrating that multi-pass automatic translations mangle the pediddle out of what was said, you've not preserved the meaning. It'd be easier just to run things through a random letter generator - that'd mess up identifying you, sure, but then, why would you bother?
Lemon curry?
Seriously. I always start reading posts without looking at the author. I'll get about 1.5 sentences in and say, "Ah, shit, is this a JonKatz article?", then look at the author. I'm always right.
(Vassar '90 Alum)
___
Cognitive Overflow
more than yo
How reversible would that be? Given the source code to Fish, could you take an output and reconstruct the original text? Would you have to know what different translations were done to it or could you guess? I just replied to post #15 about creating a program that cleans your "fingerprints" off of a work. Using the Fish is very cool, but I think there needs to be some random element that can not be reconstructed.
-B
The CIA used to (still does?) make different copies of highly classified documents using a computer that would use a thesaurus and other rules to change the texts, but still retain the meaning. If a copy of the document was later found in the wrong hands, they could trace it's origins to the mole based on the wording.
Imagine something like a travesty generator that can decompose your writing, sentence by sentence. Then, armed with a built-in thesaurus, grammatical rules, etc, it could re-cast your words into someone else's mold.
I've noticed on Slashdot, and other WWW forums, that occasionally someone posts a bunch of text from one of these filters, or something like The Postmodern Essay Generator. I usually spot the computer-generated randomness in the text after the first paragraph, and then stop reading.
If you're hiding, no one will notice you, so why bother writing at all?
Learn a new language and make your irresponsible ravings in that. Language is acquired by age five, and those patterns that you have are identical with the patterns of your parents, older siblings, and the others who were around you speaking your native tongue.
If I were to learn French or Spanish, the patterns would be of the Berlitz language company, or the Barron's education company, or whatever, and would be bland from a stylistic point of view.
I'm not entirely sure b/c I read the Foster book a month ago but I believe he addresses this and would disagree with you. Foster would figure out, from your "bland" foreign language writings, how long you have studied the language, possibly where/how and for what reason (lots of commercial jargon in your Spanish? Or many references to Don Quioxte?), what regions/dialects of the language(s) you are familiar with, and to what extent, what language you grew up with, and many other factors.
Also, there is evidence in the recent world of literature that many authors (Conrad, Joyce, Nabokov, Beckett) working with foreign languages (Joyce appropriating them for English work, the others actually working in foreign languages) do maintain certain stylisitic and contentual similarities between their native and foreign language writings.
He has, and the book doesn't comment on his conclusions.
That's it - I'm stuffing my anonymous postings
through BabelFish a couple of times...that'll
sort out their analysers!
www.sjbaker.org
Well, I took a shakespeare class a couple years ago, and the prof swore by this guy.... apparently he played a big role in the dating/chronological ordering of the plays.
Anyway, apparently this guy submitted some rather controvercial opinion to some board (don't remember specifics) based upon his analysis of a Shakespearean text... and apparently the board (I think it had three members) replied with a long statement that basically amounted to "that's absurd." So this guy did a little research into the writing history of the three board members, analyzed the reply, and pinpointed which one of them wrote the response to his opinion.
In short, I pretty much see this guy as legit.
Jose M. Weeks
I wonder if Mr Foster did use the word fingerprint
in his original manuscript and then the
editor changed it to DNA, surely text analysis
is as good as fingerprints, very high probability
but possibility for dup's. Is it really as
good as DNA?
English ---> [pick one] ---> English
Although altavista.babelfish doesn't list it as an option, you should be able to use machine translation technology to translate English directly to English -- that is, English to internal representation to English -- which should give you some flattening effect with the fewest number of steps. This might be enough for "real" applications.
I wonder if this is how they tracked down Jeff K.
--
This is not my sandwich.
--
A. Bullard
Should invading one's peaceful neighbours be opposed, or rewarded with trade deals?
this is what your sig became:
GIS: I do not speak for my society, and this mast does not educate the professional advice.
Then I guess this would probably be the first recorded instance of M$ Word actually doing something beneficial. That stupid program is always trying to correct my writing style.
:>
Now I'll start taking the suggestions more seriously.
"Rocky Rococo, at your cervix!"
Actually, I don't accept the behavior from either party.
"Because the other guy did it" is not a relevant excuse for inappropriate behavior.
Just ask my kids. I don't let them use that lame excuse either.
"Rocky Rococo, at your cervix!"
Katz uncritically repeats what I imagine is Foster's own, or his publishers' claim. In fact, Foster has tried to prove using linguistic methods that a flaky 1612 poem entitled "A Funeral Elegy" (it's not a sonnet) is by William Shakespeare. Many Shakespeare scholars violently disagree. Foster has certainly not "solved a centuries old mystery" to many people's satisfaction.
This is not a sig
Bastardized in-jokes make for painful deciphering.
If you're dedicated enough, you can fracture and steal other people's words to get across a completely new message. If you don't believe me, just ask William S. Burroughs...
Easy does it!
This comment has been submitted already, 276865 hours , 59 minutes ago. No need to try again.
blah blah yadda yadda.. how many cops testified that they were involved in malfeasance at Clintons direct order?
Next?
Maeryk.
(yes its OT.. I know..)
Feminine Protection? What is that? A chartreuse flame thrower?
I think what the author is trying to say is that much of language is defined by man's inescapable personality. As soon as you remove a real person from an utterance, its true message is dilute. The most meaningful uses of language are those that the speaker knows they'll be held responsible for--and more importantly, the listener knows they'll be held responsible for.
Can anybody guess who wrote this /. front page summary:
"Author Unknown is relevant to life online, where responsibility for words and anonymous authorship is an everyday issue."
I like Katz, but I can spot his paragraphs a mile away. (-:
I'm not so sure. I've recently become reasonably proficient in Spanish, and I recognize my own patterns in how I construct sentences.
just run your text through babelfish a couple times, and *viola* fingerprints erased.
"I don't need a compass to tell me which way the wind shines." - Mr. Furious, Mystery Men
I wonder how much the "grammar checker" features of Ms Word dilute the style of the original author©
Undoubtedly this sort of software will grow more sophisticated in changing individual's grammatical quirks into whatever Gates decides is "proper english©"
Maybe after the next version of Office comes out all you'll have to do is run you documents through Word to make them appear to have come from Microsoft's press office©
Karma: Bored. (Thinking about resurrecting the "Anyone else is an imposter" joke.)
Actually, Foster did this (at least 3 times to different peer-reviewers, IIRC) in the book. He identified an unattributed poem as a Shakespeare for his dissertation, and got some rather nasty comments back saying it was impossible to ID an author based soley on internal evidence, without any external corroborating data. When he wrote back directly to these reviewers, they were Not Amused, for the most part.
"Bugger this, I want a better world." - Jenny Sparks
PM Monday -> P. M. Montag (P. Monsieur Montag) -> P. Mr. Montag.
That's a problem with using the Fish to hide your writing style. It not only mangles your voice, but also your content.
I lived in Barcelona for 4 years, working for a US company. There were many non-Spanish people there, Americans, English, Scottish and French mostly. Emails went around in English, Spanish and Catalan, but most people pretty much wrote in their foreign languages in the same style they would write in their native language. Some would be funny, word-for-word translations of how it would be correctly said in the native language. The fact that the person acquired the language well after age 5 doomed him/her. People had learned from many sources, but one's mental structure for language quickly subsumes any proffered style from a language school. In any case, most people learn mechanics from the schools, few stay to learn style.
My 8 pesetas..
Matt
"And we're happy here, but we live in fear, we've seen a lot of temples crumble..." - Concrete Blonde
Since I'm pretty cash-strapped and can't afford the book, could someone fill me in on what the story is behind the Sonnets? thx much
managers...why god invented purgatory
I tracked it down and played with it a bit -- its purpose is to help determine how much similarity a given anonymous message contains to a series of posts by a known author. If the resulting number is high, it indicates a likelihood that the anonymous post was, in fact, written by that known author. This is to keep people from generating 'tentacles of medusa,' and engaging in pseudoanonymity.
To quote from the included doc:
It's alpha code. Back when I originally found it (circa 1996), I emailed the authors to see if they had any updates, but they denied any intent to continue working on it. But -- get this -- it comes with source , so someone might improve upon it. (If anyone wants to continue the work on this, I'm not bad with interface design and I wouldn't mind furthering the development of this code. Contact me through the above email address.)
Disclaimer: as nifty as I personally think this code is, I am the most amateur and self-taught of linguists; its results might be pure bunk. Additionally, I don't remember the relationship of the code authors to the cypherpunk group, so no endorsements other than my own should be assumed.
Also Disclaimer: obtaining this code is left as an exercise to the reader. It's out there, and if you're savvy enough to operate the thing, you're savvy enough to locate it, as well. It took me about ten lazy minutes to discover it yesterday, when this topic reminded me of its existence. This disclaimer shall be negated for all inquiries including offers of Corn Chex cereal, Sam Adams beer, and Battlestar Galactica posters.
Word.
/></
Personal me, collaborative you
Sounds like a wetware implementation of Metatag analysis.
;) )
(There, who am I?
Mark Twain used Babelfish? I didn't think it'd been around that long.
Not A Sig
To see a very rudimentary version of this, go to http://www.spammimic.com.
Or try the following. First, write a message. Then, go to an online translator that supports multiple languages. Next,translate text from native language, in this case English, to another language, in this case German. Then take this translation, now in German and translate it to another language, French for instance. Finally, translate it back to your original language, which is our original English in this example. You now have a text that transmits the general meaning of your intended message with a totally obscured style.
A demonstration follows that uses the steps and text above:
to see a very rudimentary version of this one, you will try yourselves http://www.spammimic.com too much. Oder it what follows. You write initially a communication. Then, you go to a translator in line which supports multiple languages. Initially, you translate the text of the mother tongue, in this English case, with another language, in this case of Germans. Then, you now take this translation to them, on German and you translate with another language, French for example. Lastly, you translate it return to your original language which is our original English in this example. They have a text now which transfer the general importance from your communication under consideration with to totals a type makes vaguely.
Can I volunteer?
Original English text:
Result after translations: Now let's try it on a real-world message, say a ransom note:Original text:
Translated (this one's even worse than usual); I think that'll keep pretty much everybody guessing.Bugrit! Millenium hand and shrimp!
I recently have found that translating from different languages can have a profound effect on what the original message was. For example of a generic phrase: %I like to fish in the pond% was translated from english-> french-> german-> english and it came up with %self like on the angling inside the pond% So the entire meaning seems almost lost. The translation site is: %http://tranexp.com:2000/InterTran% any comments on hiding messages by converting them to different languages and back?
I was thinking along the same lines. How long do you think it will be before 'writing style' is admissable evidence in court?
I was under the impression that the clue leading to Kaczynski's capture was his brother ratting him out. Is this not the case?
And then the ever popular, "ALL YOUR BASE ARE BELONG TO US!"
Mas vale cholo, que mal acompañado.
Yes, but picture a Slashdot where your real name appears on the comments, whether you want them to or not. (And, just in case you have a common name like John Smith, that your social security number or other unique identifier appears there as well.)
That *would* be a great loss. Just think: No more insider whistle-blowing. Ever.
You don't need to know who said something in order for it to be true. "A triangle has three sides" is true for any speaker.
Anonymous speech can protect the speaker who can cite sources as a means of verification. This helps a lot if you have a strong moral obligation to say something to say that could lead to harm to yourself if it were known that you were the speaker.
Simply trusting whatever you hear simply because you know the name of the speaker is dangerous. Ex.: I'm Walter Cronkite for CBS News
I rang, you rang, we all rang for orangutang!
You see? You see? Your stupid minds! Stupid! Stupid!
the type of identification talked about in the article is not just syntactical and diction oriented, but also, especially in the case of the unabomber, centered around access to particular information and environments (i.e. at which library he researched). A foriegn language also can not at all cover that up.
sell your certainty and buy bewilderment
Heh, Kennedy arguing about morals.
That is a good one.
No bombmaking lessons here. The Linux Pimp
--It's Pimptastic!--
actually, as a writer, i think it is completely wonderful to be able to do so. i understand that some people will be scared by it, but think of it this way:
;)
- it can assist in preventing plagiarism
- it can assist in tracking email stalkers
- it's an important anthropological tool
- it gives terrific insight into yourself, if you look at it properly
being able to track someone through writing style can be a time consuming process. i would doubt that anyone is going to use it as a "Big Brother" tool unless you're already breaking the law.
One of the statistics professors, Wayne Larsen, at the university I attended helped create a program that would compare two documents and give a probability that they were authored by the same individual based on the usage of common words such as 'of', 'the' and 'or. The program is called WordPrint. Since I went to Brigham Young University, of course I have a link to them using this technique to analyze the Book of Mormon. http://caic.org.au/lds/byu_comp.htm
D'uh!
My other sig is extremely clever...
Obviously you have no opinions that deviate significantly from the norm. Otherwise you wouldn't have said something so stupid. Try wearing a shirt that says "Atheist" while visiting the small southern town of your choice. You may change your mind.
One day I feel I'm ahead of the wheel / the next it's rolling over me / I can get back on / I can get back on
And what a boring world that would be!
...
C'mon, mod me down, I'm not on the proxy yet
-- Do gravity waves prove the existance of God?
"Making linux GPL was the best thing I ever did" - Torvalds. I'd hate to see the worst thing...
I'm trying to imagine Slashdot without people who choose handles like 'Metal Machine Music' and yet persist in the notion that they're not just as anonymous as anybody who posts as 'Anonymous Coward.'
And, frankly, I can't see anything but very slight differences.
Hay thar.
Does the idea of being able to track someone thru patterns in word choice/thought layout scare anyone else? It is not that I am worried about anyone tracking me cause i have nothing to hide, but the idea that it can happen is somewhat disturbing. Is Big Brother learning a new trick? Normally I dont give that much thought to the idea of being tracked, its a fact of life in today's society, but going this far seems somewhat ridiculous to me. But thats just my opinion.
-Life is a Journey, --Not a Guided Tour! ---Trust me, I've already looked for the guide book.
I'm not sure that this is correct. I'm a native English speaker from Ireland. I learned French in highschool and by living in France, and German by living in Germany. If anything, my style of speaking / writing in English is mirrored in both languages and my identity is even more obvious since it is not really how a native speaker would phrase things (although not necessarily incorrect). On the other hand, my grammar in both languages is a mess, so maybe that wouldn't work -- it's a form of obfuscation. Interesting article, though.
"Seriously. I always start reading posts without looking at the author. I'll get about 1.5 sentences in and say, "Ah, shit, is this a JonKatz article?", then look at the author. I'm always right"
Is this because it's a topic/movie review/whatever that only katz would touch?
The moment the movie review of the next Rugrats movie hits slashdot (at the way they are connecting movies to the slashdot audience don't be surprised to see one) I'll know immediatly it's katz without even reading the article.
Katz didn't tell me anything I couldn't get from the back of the book cover or the book jacket. Same with his movie reviews. Why can't he talk about various subjects without sounding like he's taking some 5-line review and expanding it, or copying a review off of amazon or imdb.
Now the FBI is going to take writing samples along with our fingerprints.
is from Voltaire isn't it ? interesting that i haven't read any comments on how "AC"s are responisble for almost all of the 'issues' we know about. several comments have talked about the 'legal system' protecting us ...
how many pentagon scams, cia operations, fbi witch-hunts, corporate dumpings does it take before one gets the idea that this government/society does not have a very 'open' policy.
-dan
----
Jon Katz's writing style is certainly distinctive, and he has little need to put his name on anything he writes. Take a look at such gems as:
..."
"This is highly readable book, an intellectual who-dunnit of particular
This is the first line of the fifth paragraph of the article. Such attention to detail, such ability to edit his pieces, such grammar and style... Yes, America, this is uniquely Jon Katz.
Well, the lack of concrete proof ("well, one guy says so.. though you of course have to buy his book to know anything concrete...") and lack of real content (and Katz's opinions alone do not qualify as content) made this inevitably the work of Jon Katz anyway.
I would imagine that If you looked at my
Each of us has a style of writing. I don't think this is a bad thing. Each of us look at the world our own way, and write to match.
The cure of the ills of Democracy is more Democracy.
Erlang Developer and podcaster
...many years ago in a college journalism class:
My instructor was also the head of the Journalism Department. And he liked to give little "news conferences" where he would pretend to be a news source (which he almost was) and we would pretend to be reporters (which we almost were). We would ask questions and then write stories based on the pseudo-event.
As I was usually bored to tears, I developed a little game: I would see how much I could change his quotes without getting caught. I never changed the meaning of anything. But I was able to change the actual words almost all the time. Needless to say, this would have been highly unethical in a real-world scenario unless you knew a great deal about how the person being misquoted really talked (his or her style as it is analyzed in this book). An offended quotee might easily say, "I would never say it that way!" But I never got caught because, as a captive audience, students DO get a good chance to learn how their professors speak. I made him sound more like himself than even he did.
I'm sure this story has some relevance to the field of study this book addresses. But I'm not sure what it is.
It would be interesting to know if the author would find it easier or harder to spot faked style examples than the person being faked.
It would also be interesting to know how much of the distinguishing characteristics of a person's prose are stylistic in nature and how much is semantic. (Both seem to be implied by the review.)
It seems like there may be some implications for NPC AIs in CRPGs as well: If one could codify a wide variety of speaking styles, one could give each NPC a distinctive voice with much less design overhead.
BTW, I am the guy whose posts can easily be spotted by the elipses (...) which end most of my subject lines and begin most of my comments.
Eternal vigilance only works if you look in every direction.
There are times when anonymity is useful and even necessary. Take the case of a political dissident in a totalitarian regime. This person cannot express their own beliefs, their own experience of oppression and use their real name. To do so would be suicide. There are legitimate uses of anonymity. They are admittedly rare.
Just be sure to wear the gold uniform when you beam down -- you know what happens when you wear the red one.
From now on I think I'll prefilter all of my anonymous online posts using:
Let's see how easy it is to track those posts down after that.
--
CUR ALLOC 20195.....5804M
I'm interested in learning more about how to do this analysis. Is there freely available software out there to do these sorts of things? Published papers/algorithms?
--
Tweet, tweet.
As my sig points out I browse at +1, meaning I don't see any anonymous posts, unless they are moderated up. I had to do this because the amount of trolls became to distracting for me at some point. I do regret that I miss the occasional interesting AC's post but I do enjoy a mostly troll free slashdot reading.
;-) and using the characteristics to create a Mao like text saying things Mao would certainly not agree to.
However, in our western society anonymity is not so much needed because we have a decent (not perfect) amount of legal protection. However, in countries like China where you can be locked up simply by attending a conference where certain things are said, anonymity is important. If the Chinese government will ever set up an echelon like network they will most certainly pay attention to this book and may be even use it to filter the online material against known offenders. Maybe they'll even use it as proof and take the occasional mistake for granted. Scary stuff indeed.
Luckily, this book also provides us with the means to anonimyze texts. By randomly inserting synonyms and replacing expressions with semantic equivalents you'd get untracable texts. You could even fake identity this way for instance by analyzing Mao's text (hint to chinese protesters
However this requires knowledge, so ignorance is not a bless here!
Jilles
Rather than making manifesto A appear to be written by author B, a first step would be a program that just wipes your "fingerprints" off of text. Many sentances can have different structures while keeping the same meaning. A program that randomly shuffles words around would disguise your preference for a certain style. The problem with anything like this is that you would have to proofread the text after the program finished with it to make sure your meaning is still there.
-B
I don't know why this was modded down.
I had the same idea about how the method descibed in the book could be cicumvented by machine translation.
-----
nuclear iraq bioweapon encryption cocaine korea terrorist
Babelfish is not a one-to-one translator, so there is information lost. Since its vocabulary is limited, multiple words may map onto one word in another language. This makes it impossible to reconstruct the original text, regardless of what you know about the program -in essence, Babelfish introduces 'noise' to the system.
However, I don't think this will inevitably fix things - the author here was talking about stylistic usages - such as quoting from sources, or common recurring words/themes - these won't be pulled out by Babelfish, though Babelfish may make it difficult to note certain patterns.
I agree with the method - I can usually recognize people online simply by word usage and punctuation usage. Punctuation is usually a total giveaway - for instance, I tend to use dashes a whole lot, and rarely (if ever) miss capitalization or comma usage (In real time chat, it becomes even easier, as you can tell roughly the speed at which someone types). You kindof "pick up" these habits after dealing with other people online quite a bit. This just takes that learned habit to a new level.
Language is acquired by age five, and those patterns that you have are identical with the patterns of your parents, older siblings, and the others who were around you speaking your native tongue.
I'm afraid I don't agree. I use words like floppy, mp3, and lemon grass a lot more than either of my parents or brothers. I use geek in a non-pejorative way. etc... I also construct my sentences slightly differently, because I'm a child of the 80's not the 50's. So, my vocabulary and grammar is a mixture that results from many different influences.
I don't think learning a foreign language would help much, because in order to get past the 'Monkey is in the tree' stage, you will be reading newspapers and talking to other people. You will learn some of their ideosyncrasies and start using them. Mixed-up in your own, unique way, of course.
In Bill Teahan's PhD thesis "Modelling English Text" (http://www.cs.waikato.ac.nz/~wjt/), another nice approach for the identification of an author is mentioned (don't remember if Bill came up with it or someone else).
It involves a text compression system that must be trained with text (the more, the better). If you train the system with the works of N different authors and compress the text to be identified with each of the models, it usually gets compressed best with the model of the author that really wrote the text.
This of course requires a sample of the author's work to be available.
Forget the usual sky-is-falling conservative rant ("the right blah blah without responsibility for blah blah the content of blah blah are depleted"): the point is really that the expressive tools available to any individual's intellect and experience are limited, and this can be used to triangulate the speaker, if enough speech is available to do so. Of course, this isn't really a scientific process, and without a lot of sample text - as was available in the Unabomber case - it's not much more reliable than a police psychic. And of course, in the Unabomber case, the source really wanted his views and intellect known, so it would have been difficult for him to resist leaking out key details such as literary sources.
In the case of AC's degradation of message, this is more representative of the phenomenon that communities become more diffuse in their cultural focus as they get larger. Yeah, I like "underground" too, but at some point the reactionary dispair over the lack of Quality and Values (whether or not it is masquerading in hipster fashion) becomes a call to continue living in caves. Go ahead, caveman.
Boss of nothin. Big deal.
Son, go get daddy's hard plastic eyes.
Expanding a vast wasteland since 1996.
If it's trivial, then you ought to be able to tell me the IP address of the machine I used to do that language mangling, right?
After all, that's what trivial means--that anyone can do it.
That's also ignoring the fact that Babelfish-like programs are available for free download. How do you propose to trace down exactly which program was used to do the language mangling when there are a few million laptops around with that sort of software on it?
Really, I would like to see if it's as trivial as you're making it out to be. Just post the IP address I filed the Fish request from (hint: it's not the same as the IP address I'm posting this from) and I'll concede that it's a trivial exercise.
Voices From the Hellmouth
Rethinking Virtual Community
The New Geography
Embracing Insanity
Cyberdemocracy and the Public Sphere
Dark Hearts and the Net
Spirit of the Web
Is The Virtual Community A Myth?
--
I always had an idea to make a program that perfectly "mimics" other peoples writing styles just from the spelling of the words:
1. Get as much writing samples from your victims as possible.
2. have the program find all the words spelled incorrectly and remember them.
3. You write whatever you want the victim to say.
4. Program parses it and then spells all the words you spell correctly the victim spells incorrectly, and vise versa.
Evil? Perhaps.
There was a guy like this making his way through a sizeable portion of Usenet a couple of years ago. At first I was surprised when he showed up in several completely unrelated groups that I read. The likelihood that someone would not only share my interests, but not be far from my viewpoint was surprising. On one topic, it made sense. By the time he hit the third it was shocking.
However, what he posted was actually on topic, or close to it for most of the newsgroups. He was posting, in pieces, books that had lapsed into the public domain that were relevant to the newsgroup. If I had to guess, he was testing some autoposting software. In fact, I wouldn't be surprised if it was a testing of spamming software.
He was not hard to killfile though because he would stick to a particular e-mail address and subject (title, author and incrementing part number) for the duration of each book.
From just my description of his "work", I would bet that several people here can identify him, or at least his pseudonym if it was one, and a list of newsgroups he was seen in. Recognizing trolls when they change addresses has been a popular sport on Usenet for years. I'm looking forward to refining my techniques.
The net will not be what we demand, but what we make it. Build it well.
I must disagree. This disallows for the possibility of linguistic innovation. Furthermore, or the course of many generations, it would make the linguistic styles of members of stable populations increasingly indistiguishable from each other.
I know that I speak my own idiolect. Large parts of it come from the people around me. Parts of it don't. Particular bits of wordplay have become in jokes for my friends and family. Each of us originated some of them. Each of us use a subset of them.
I have watched my own "voice" evolve over the years. I can attribute some of the changes to specific influences in my life. I have consciously borrowed from languages that I have learned and people I have met.
Some of my choices in wording also reflect outside influences: profession, location, hobbies, etc. But some are my own choice. This in no way invalidates the theory that stylistic data can identify an author. It would be difficult for me to completely enumerate every identifying characteristic of my writing and hide it effectively. However, some of it I consciously created or chose for myself.
As for learning another language, to a certain extent you are right. One common misconception about multilingual people is that there is a measurable degree of fluency in each language. The granularity is finer than that. I have friends from college who are quite fluent bilingually, but cannot discuss their professions in anything but English. It is the language in which they studied and now work. They don't have a technical vocabulary in their other languages.
The net will not be what we demand, but what we make it. Build it well.
**I suppose one of the best uses of this technique would be to properly determine the originator of your sig quote.**
I, as I assume you, used to think this was a Voltaire quote. Someone here pointed me to the fact that it was *not* Voltaire, but was Tallentyre who said it. I went and looked it up on the one quote database, and it said "usually attributed to Voltaire, but not appearing anywhere in his repertoire, actually said by Tallentyre".
So nyahh.
*grin*
Maeryk
Feminine Protection? What is that? A chartreuse flame thrower?
**Referring to the sheriff and the police chief, Burger told the Review, "Both men stated at different times during the meeting that - based on their conversations with Governor Ashcroft - the governor said he would 'look the other way' specifically regarding the [Missouri] Supreme Court's ruling and asset seizures going to education.**
No, I want to see some proof. Not one guy saying "he said she said". Sorry.. I dont buy that. When the "he said she said" was coming out over the last batch of elected crooks, we got "OH! that's anecdotal! No proof!" we got "Ohh.. you are just mean.. there is no *proof*.. those guys are just lying to make them look bad".
I see no reason not to apply the same judgement to this sudden revelation.
I am not a "Zealot", but I also dont believe that *ANYONE* in washington.. from bush down to the lowliest aide, doenst have *something* in their closet. What strikes me as humorous is that the guy threatening to filibuster against him DROWNED someone, and is now bitching about morals.
*sigh*
As for the election being decided by the Supreme Court.. nah.. they specifically ruled *not* to rule on that, and let it be decided by the state.. a noble thing, AFAIC.
This was almost a shameless theft by the Gore camp.. but the good guys won on this one.
Maeryk
Feminine Protection? What is that? A chartreuse flame thrower?
**No, I want to see some proof. Hmm... Did you understand this little demonstrations? **
No, Frankly, I dont. It is one thing to accept uncoroborated evidence as fact, and yet another to doubt something that was duly testified to in court and investigated. You can be an idiot if you wish, and say "they are just as valid", but, well, the fool is you.
**As for the election being decided by the Supreme Court.. nah.. they specifically ruled *not*
to allow the State of Florida to finish do its count, thereby selecting Bush. The Supreme Court even said they made a mistake to stop the counts, but that because of their mistake it was too late so Bush would have to do. ***
can I distort your myopic world view for a moment?
1) the recount (the FIRST) recount was mandated by Florida state law. The second, third, ad nauseum were mandated only by Gores crack team of lawyers, to force the issue.
2) By the laws of the state of Florida, and the laws of the US, (electoral college) Bush won the election. All the civil rights lawsuits in the world wont help.
3) Gores claims (and clintons, as well) of "every vote a counted one" rang with hypocrisy even as they said them, as there were many other counties in Florida that were just as close, yet on Gore's side, which were *not* recounted, or even mentioned.
4) Gored DNC president RESIGNED because of this issue! Gore got *just* close enough to being a ridiculous whining little child over this entire thing, that Rendell was *removed* from his position when he stated it was time for Gore to give it up, it was Bush's game!
How can you, in good conscience, follow that and *still* claim the wrong man got in? you can have objections to bush, or be for Gore, but you cannot claim that the election was the cause..
I honestly believe that the biggest fear about Ashcroft is that he wont be a presidential ass-kisser like Reno was, and that he may actually have some sort of spine, ulike Reno. Course, with a republican regime, no-one will be able to do anything without it being under the microscope. I just have to wonder if it would have been any different if the Republican AG had intentionally gassed over a dozen children to death in a compound. Would he/her and his troops walk free? Somehow I dont think so.
Maeryk
Feminine Protection? What is that? A chartreuse flame thrower?
**"I believe," he writes, "that the same wonderfully verbose document, partly written in California libraries, could have led agents to Ted Kaczynski even without David Kaczynski's invaluable assistance." ***
Yeah.. read it again.. "could have" is not the same as "did" and the "even without David's assistance" means that they are acknowledging that that *is* what caught him.
Thats what I said.. he never said *HE* caught him.. he said that language analysis (for wont of a better term) did.. IE his brother recognized his writing style.. which was basically the basis of the review/book anyway.
Maeryk
Feminine Protection? What is that? A chartreuse flame thrower?
If you had bothered to read what was actually stated in the article, yes, it was his Brother that turned him in.. and yes, that is how they caught him.. but the author Katz is quoting points out in the article, (and, presumably, deeper in the book) how the man's words gave him away. (a good example being his brother being able to pick him out.)
Read the article again.. and tell me where this guy says he had anything to do with the capture?
He is merely pointing out an example.
Maeryk
Feminine Protection? What is that? A chartreuse flame thrower?
**Ashcroft is the topic. Will he be nominated for Attorney General in spite of breaking the law because of his personal views? Another hypocritical lier, just like the president-select.
**
Ashcroft is a man who believes in his convictions. APparently that is not acceptable to the people currently involved in thwarting him. There is *absolutely* no reason he should not be placed in that office.
How can you say "president select"? or are you one of the people who still hasnt realized that Bush was elected fully within the laws of the US and within the laws of the state of Florida? Wake up, smell the big oil.. the man was legally elected, and we will now come out of the 8 year nightmare that was clinton.
Speaking of hypocrisy, how can you hold Ashcroft on a moral pinion, when Clinton just A) pardoned everyone involved in whitewater and his fund-raising scandals, and b) pleaded guilty to perjury and took a plea bargain of 5 year loss of law liscence and 25K fine when did so?
Please.. dont be a hypocrite yourself.. I never said Ashcroft *didnt* do it, I said that compared to the quantity of scandal following Clinton, this is *nothing* and no reason to keep him from assuming the office.
Maeryk
Feminine Protection? What is that? A chartreuse flame thrower?
**You should be just as paranoid about taking what they say as facts **
Excellent point! And, on a related note, just because they post a link that backs up their belief, does not make it carved in stone.
The thing is, however, that *usually* you can track someone's posting history.. go look at the author.. then look up permutations of that name, and you can have a good idea of the validity of their claim.
Public scrutiny exists here just as anywhere.. and if someone with a nick *very* close to someone elses has posted two posts, whereas the other has hundreds, and karma of 40, it should be a clue as to what is going on there.
maeryk
Feminine Protection? What is that? A chartreuse flame thrower?
When publishing a article in a (respectable) research journal, academic peers always write reviews making recommendations and criticisms of the work. I'm always amazed at how my research advisor can read a review and pin-point who reviewed it within in minutes. Granted, the lists of suspects is fairly small but still. From the style of the writing (even how it's formatted: MS Word/LaTeX, etc), references, spelling (color or colour?), and criticisms (i.e., what the reviewer thinks is interesting and what s/he doesn't).
It's not useless either. Knowing the reviewer really helps in writing a rebutal that will get your paper accepted.
Nonsense! It may be slightly overstated here, but that's more likely to be Jon Katz's fault than the author's.
Writing style or voice is a pretty common and obvious thing, very much like musical style and voice. I heard two bars of a song in a store the other day, and immediately knew it was Mark Knopfler, although I'd never heard the album before. Read Vonnegut and then the "sunscreen" monologue; and figure out why people believed that it was his writing, or wasn't. (It wasn't, and didn't really sound like him, but rather someone entirely different who was influenced by him)
Whether or not you can disguise your writing voice depends on how well you understand _what_ you're trying to disguise. Is it vocabulary, experiences, sentence/paragraph structure, or something else?
It's not infallible, but it's definitely got some pretty rigorous (and effective) methodologies.
"People who do stupid things with hazardous materials often die." -- Jim Davidson on alt.folklore.urban
Based on years with USENET, IRC and MUDs, I'd say that sorting out "voices" from "mere" text is something that a lot of us have been doing on line for a long time.
As a MUD admin, I was often able to tell when someone "other" than the owner of a character was using it. Female admins were a lot better than me at it, actually. Once you start to put together a mental "image" of who this disembodied voice is, it is amazing how quickly someone else assuming the role "clashes" with the picture that formed in your mind.
The science here comes in being able to just yank other text and categorize it, and scale that power up enormously. Thus, people like our author here have stunningly impressive powers over those that they do not know personally.
However, don't believe that if your boss reads something you wrote, he might not add 1 and 1 to get two. When it is familiar with things, the human mind is a powerful associative tool. Do you really think you could fool your mother? Or brother? The Unabomber, as smart as he was, couldn't. He did get by the experts for a long time though.
If you proofread it, the point of the program would be worthless. As well, it would make which program you used immediately clear. Armed with the same program, investigators could work backwards to recreate your text. For example, Babelfish'd text (English->German->English, etc.) can usually be reconstructed with some skillful guesses.
--
On alt atheism, we have a guy who goes one step further by faking posts from group regulars. They are the kind of vile racist and misogynistic crap we normally don't get arround there, with inocuos titles and the name of someone you usually respect. (my newsreader doesn't give enough info on the thread heading to tell the difference.) Its really obnoxious. its like someone spreading dog shit around your favorite park so you're too busy looking out not to step in it to enjoy what you came there for.
The funniest thing is that, like above, he has the most easily recognizable posting style on the planet, yet continuously claims that it is this other guy doing it. Then he fakes emails from the other guy, still using the same posting style. Its pathetic, and he's not fooling anyone, but that doesn't stop me from stepping in his little troll piles when I'm trying to relax with some usenet reading.
This all started out (on alt.atheism, at least) with him sending a death threat to one of our regulars because of her posting in another group. (which, of course, he claims was really sent by this other guy). real loser.
-Kahuna Burger
...will work for Chick tracts...
(from the nice folks at the Onion, who always seem to say it best.)
Bush: 'Our long national nightmare of peace and prosperity is over.'
(from the article)
Bush had equally high praise for Attorney General nominee John Ashcroft, whom he praised as "a tireless champion in the battle to protect a woman's right to give birth."
"Soon, with John Ashcroft's help, we will move out of the Dark Ages and into a more enlightened time when a woman will be free to think long and hard before trying to fight her way past throngs of protesters blocking her entrance to an abortion clinic," Bush said. "We as a nation can look forward to lots and lots of babies."
Continued Bush: "John Ashcroft will be invaluable in healing the terrible wedge President Clinton drove between church and state."
--
share and enjoy
How can you, in good conscience, follow that and *still* claim the wrong man got in? you can have objections to bush, or be for Gore, but you cannot claim that the election was the cause..
The only thing that really matters, the only thing, is who got the most votes in FL. The votes should all have been properly counted. Spesific legal loopholes or whatnot don't matter.
Amber Yuan 2k A.D
"and dear god does this website suck now." -- CmdrTaco
Another good idea would be a code anonymizer. It would reformat and rename variables, Restructure logic to simplest form etc.
IMHO, he grinds his axes too loudly to be bullshitting. The first hundred pages or so have more iterations of "X said I was a fraud; here's why X was wrong; yeah, suck it, X" than I've ever heard from a supposed psychic.
One of the things not yet mentioned in the discussion is the quantity of writing necessary to do any real sort of comparison. For example, it would be difficult to identify me by my writings only, because I post so infrequently. If I posted more often, and wanted to remain pseudo-anonymous, I simply create new accounts. If each account has similarly few postings, it would be very difficult to sift through the hundreds of thousands of slashdot accounts to find the few accounts that are all most likely mine.
Even when people use account names, they're still pseudonyms that would be pretty difficult to link to real names by using writing styles, unless writings under both real name and pseudonym are prolific.
Dave
Woe is me, I'm doom'd
People will know my writing
From its form, syntax
for those who havn't seen it...
here is the manifesto:
http://www.wwfreepress.com/unaba.html
You hear that, all of you Anonymous Cowards? We'll hire Don Foster to track you down!!!
"Ancillary does not mean you get to rule the world." --U.S. Circuit Judge Harry Edwards, speaking to the FCC's lawyer
Now that we know the techniques, how about some countermeasures...
Um, no, we don't know the techniques. Not really. At least I don't, not from the Katz article -- though he's probably raised our awareness of what can/is being done. And yes, I'll remember to get this book and read it myself, but still I can't help but wonder:
Where can I learn more about the techniques described here? Are there software packages (open source, even) out there that do this kind of text analysis? Texts on how to do it yourself? Publications describing word-pattern detection algorithms?
(And where are all the karma whores who usually post these links, now that I need you?)
--
Libertarianism is rich wolves and poor sheep playing gambler's ruin for dinner.
The difficulty with this approach of masking one's identity is that you lose the ability to make a powerful, respectable argument. Words are weapons only when the author is able to hone his or her work to a razor-sharp edge; running a text through The Fish is the rough equivalent of bashing the weapon against a rock several times.
The first example was reverse-translated through Babelfish using English --> German --> English. The second is the original.
information wants to be expensive...nothing is so valuable as the right information at the right time.
Obliteracy: Words with explosions
Foster, a Vassar prof who solved a centuries old mystery involving...
Lisa: (depressed) At this rate I'm going to end up at Vasser...
Homer: (angrily) Young lady, I've had it about up to here with all this Vassar bashing!!
The surest way to make a monkey of a man is to quote him. --Robert Benchley
I have to agree on anonymity. I don't think being anonymous achieves much.
Just picture Slashdot with out the ACs. It would be no great loss would it?
Furthermore, I think people should, if they want to express an opinion say so in public - otherwise they can't really believe it.
Anonymous promulgation of opinions is cowardly and dangerous.
Down tracking bad creator of translations possible will be? Them grammar teaching so manual VCR not be not comprehending and similar sounding as Yoda?
(Do not sign anything.) -- Fell, Planescape: Torment
What will happen is that people who want to remain anonymous will simply wisen up to the emerging analytic techniques and learn to vary all of the relevant elements of their writing style.
I have already been doing this for years. If I post something to the net under an invented identity, I change the sentence structure, spelling, style, tone, various lexical elements---you name it. I'm also careful not to make any cultural or intellectual references similar to ones that I might otherwise make.
I can easily write something that could not possible be attributed to me. The reason for this is obvious: writing is, or at least can be, deliberate, subject to endless revision and clever disguise. It only acts as a fingerprint when it's poured out without deliberation.
Only in real-time performance activities is it difficult to mask one's traits: speaking, walking, playing music and so forth. The low level elements of these activities are beyond conscious control, unlike writing, whose every element can be controlled by the writer.
A most excellent site has several quotes that come from either the Unabomber's manifesto, or Al Gore's book "Earth in the Balance". It's suprisingly hard to tell the difference between the two. Give it a try:
s man/Gore.htm
http://ourworld.compuserve.com/homepages/ken_cros
This suggests two ways to evade the technology. One is to apply some filter that makes everybody's work look the same. A crude example would be if everyone used the Zippy filter - that style may be so strong as to mask the differences among the original authors, as well as the similarities between original works by the same author. It is somewhat analogous to writing in block print to hide one's handwriting.
The second way is to make ones own work look different every time. In other words, make your own work have very little internal correlation. Again, a crude example would be if sometimes one wrote in Hemingway's style, then Steinbeck, then Maya Angelou, etc. This is analogous to cutting and pasting words and letters from magazines, "ransom-note" style, to hide one's handwriting.
The fascinating thing to me is that someone has come up with a way (perhaps crude so far) to statistically recognize writing style, something that we people are relatively good at already - the Unabomber's brother fingered him upon reading the manifesto, for example. The thing to remember is that it's only as good as his algorithms, which we smart hackers can surely outwit.
I think what makes this difficult to do manually is that our writing style is probably a fairly low-level characteristic, like our gait, handwriting, or voice. Just as those can be overridden with training, surely one's writing style can be as well.
>>There may be a flip-side to Foster's new science. In the wrong hands, it could mean that the kind of free speech and safety sometimes associated with anonymity -- Anonymous Cowards on this site can be obnoxious, but they also pass on valuable information -- could be lost
AC's are usually posting anonymously so they can be obnoxious. In the *rare* (to my eyes) occasion that they pass on something of value, it is a two-sided coin. On one side, it is valuable info, on the other side, it is plausibly false. It is *easy* to spout off when you are "hidden" on the net.
Anyone who thinks that they are "anonymous" on the net is direly mistaken. You can be tracked, and if you are annoying enough to the powers that be, you *will* be tracked.
it continually amazes me, however, that they can track down a cyber-pedophile, or the Una-bomber, but they are powerless to stop mass mailings from ISP]s that specifically dont allow them.
Anonymity is all well and good, I guess, but typically, I dont pay much attention to what they say. I am willing to stand by my words, as long as the context they were taken from is moved intact as well. (IE: dont take my quotes on GWAR to stand as my quotes on how I feel about female rights in this country. The two are about as diametrically opposed as possible, yet with huge search engines and Deja News out there, it is easy enough for someone to do just that.)
Maeryk
Feminine Protection? What is that? A chartreuse flame thrower?
What Foster claims to do seems to me little more than the sort of nonsense that psychic investigators claim to do. I'd love to see any evidence or studies validating his techniques (a double blind study on a wide variety of text fragments would be a start).
Lincoln biographer Stephen Oates almost had his career ruined by some morons at NIH who ran some Lincoln biographies through a computer and claimed Oates had plagiarized based on some similarities with a classic 1952 bio of Lincoln.
Learn a new language and make your irresponsible ravings in that. Language is acquired by age five, and those patterns that you have are identical with the patterns of your parents, older siblings, and the others who were around you speaking your native tongue.
If I were to learn French or Spanish, the patterns would be of the Berlitz language company, or the Barron's education company, or whatever, and would be bland from a stylistic point of view.
Anonymous authorship was a formidable force in kicking the British out of the colonies circa 1776.
Bukvich
since the offender's primary residence (Montana) and place of employment (none) couldn't have been pinpointed even with the help of advanced computer tracking software
I, for one live in Montana and have no job, yet I think that I must be able to be tracked down by sophisticated computer methods, since I am always getting offers for pre-approved credit cards and offers of pr0n.
Hopefully I didn't put any [] around my words.
__________________
science is a religion
"In a culture that encourages anonymous communication and the right to speak without responsibility for the content of the utterance, the spoken message and, eventually, language itself are depleted."
I think Slashdot is a prime example of how this is not the case, certainly, ACs may "deplete" the message, but the cream rises to the top. The interesting, insightful stuff on the Internet will be linked to, will get traffic. Pages by the eponymous B1FF will be lost amid the dross.
Of course, there always will be exceptions to this. But the internet is in itself self-regulatory - the good stuff gets more exposure. Dilution? Perhaps. Lost messages? I think not.
Saying "Ted Kaczynski, the Unibomber", as Katz does (and not something like "Ted Kaczynski, the man convicted for the Unibomber bombings" is to show journalistic BIAS.
Let's not forget that the FBI called paper clips found in his home "bomb making equipment" (and not "letter writting equipment"). Let's not forget that while his court-appointed lawyers were saying, "Yes, poor Ted IS guilty, but it's OK 'cause he's insane!!!, as Ted was trying to fire the bastards and plead innocent, but NOO! the judge says... Ted isn't competent to speak for yourself in court.... You CAN'T plead innocent. Shut up and sit there and take what you got coming.
There is plenty of evidence to show that Ted was railroaded. The message in Ted's case is that any intellectual that abandons technology and moves to the mountians must be a killer. Anyone who writes anti-capatilist literature is a terrorist.
Just like in the OKC thing. If you oppose what happened in Waco, you are guilty of OKC bombings.
Where's the critical thinking? Where is the skeptical inquiry?
To stay a step ahead of the spooks: can one effectively FRAME someone by copying their writing style and influences?
Kill-filing this looney is impossible as his email address constantly mutates.
And Taco must use this to find the identities of all the trolls by comparing their troll posts with their regular logged in posts. Then Taco should find out the real world identity of the trolls by examing their email address and posting packets.
Once all the trolls have been identified, a crack team of Andover sponsored Army Rangers should break down their doors, grab the trolls, flog them with wet lasagna noodles, and send them packing to Belize, preferable Temptation Island, where they will be sentenced to death by oral sex.
Finally, Slashdot will be troll free!
Regarding using BabelFish, or other tools, I don't think that's a good countermeasure, clearly no one could tell who wrote the article, nor would they care, since the final presentation would lack style, and readability. In arguing against translating works of literature, Mark Twain demonstrated the folly of such an enterprise by translating one of his short stories to French and then back to English. The resulting text lacked Twain's style, and was unreadable.
My other sig is extremely clever...
(a) The Unabomber Manifesto
(b) An average Jon Katz article
Pencils down!
Cheers,
Who wouldn't recognize a JonKatz rant?
--
SCO employee? Check out the bounty
Online anonymity is easy. If you want to really screw up efforts to identify you by specific language details, just run your posts through the Fish a couple of times. What comes out is guaranteed to give Foster a first class migraine.
=====
On-line anonymity is simple. If you want to really upward screw efforts to identify itself by the specific language details just your masts by the fish execute a pair of the periods. Which comes out, is guaranteed, giving foster a first class klassenmigraene.
(English->French->German->English, for those who are interested.)
Now that we know the techniques, how about some countermeasures...
Travesty generators have been around for some time, taking random sequences of letters and filtering them into the style of Shakespeare. This makes for something between Shakespeare-sh gibberish and amusing reading.
Imagine something like a travesty generator that can decompose your writing, sentence by sentence. Then, armed with a built-in thesaurus, grammatical rules, etc, it could re-cast your words into someone else's mold.
In other words, scan manifesto A, scan writings of author B, building rulebases of both. Convert manifesto A into style of author B.
I don't believe we have this, but I don't believe it's much of a stretch, either. Kind of a computerized equivalent of cutting the words and letters out of magazines and newspapers, then pasting them back into your own message.
The living have better things to do than to continue hating the dead.