Captchas To Keep Idiots Out Of Comment Threads
dangerousminds.net
dangerousminds.net
Pair the following five comments with their corresponding definitions:
1. “Raganwald is a blowhard whose pathetic attempts to score karma reveal him as an insecure dweeb who can’t get over being kicked around in grammar school. He and his mumblings should be should be flushed down the toilet bowl like the turds that they are.”
2. “What, raganwald is talking about beauty in code? Have you seen some of his code? Ignore him.”
3. “Raganwald sounds a lot like a Ruby fanbody, and we all know how those people think."
4. “Anyone who has spent that much time on Java clearly has no taste in software and cannot be relied upon for sound reasoning. Ignore raganwald.”
5. "Of course raganwald would say that there’s something wrong with Waterfall, he’s a Certified Scrum Master, he’s just pimping his own credentials.”
And the definitions:
A. Fallacy: Ad Hominem Abuse
B. Fallacy: Circumstantial Ad Hominem
C. Fallacy: Tu Quoque (“You Also”)
D: Fallacy: Guilt by Association
E: Not a Fallacy: Insults
Example: 1-E, This is an insult, but not fallacious.
Poe’s Law strikes again: http://rationalwiki.org/wiki/Poes_Law
I don't think it'd be a 100% bad idea to implement some sort of knowledge-based testing, wire it up to varying levels of initial access/privs on a site, and see what happens, as an experiment. User privs could then evolve once the user had actually posted or commented or rated things.
If the person answers "Duh," you put them in some kind of post jail and make them earn their way out. I couldn't have named all those, but I recognized them as false criticisms immediately.
I found this puzzle both entertaining and educational. I guess that means I'm not qualified to post here ;-}
I have a really dumb spam filter on my blog which simply asks that the user "Type the word 'humour', but with American spelling". I've had a couple of complaints that this request is unfair to non-native speakers. My response is that you don't have to know it, you just have to know how to look it up.
(And yes, I saw that this stuff wasn't serious, but I thought it was a worthwhile point to make in any case. CAPTCHAs don't need to be things that users can pass without outside help.)
More importantly, after having made a bit of effort to be allowed in, the users might value the forum more.
There is little more annoying than someone who thinks they see logical fallacies where none exist and take the conversation from bad (real arguments intermixed with old fashioned insults) to worse (nothing but an argument from fallacy)
"I don't claim to know what an Ad Hominem is." would also be a passing response in my book...
Very measured writing about yourself.
Copacetic shield.
42 75 74 20 79 6f 75 20 73 68 6f 75 6c 64 20 68
61 76 65 20 73 65 65 6e 20 74 68 61 74 20 73 6f
6d 65 20 6f 66 20 74 68 65 20 63 6f 64 65 73 20
61 72 65 20 6e 6f 6e 2d 70 72 69 6e 74 61 62 6c
65 20 63 68 61 72 61 63 74 65 72 73 20 3b 2d 29>"He and his mumblings should be should be flushed down the toilet bowl //
So arguably you're making the case to ignore his [implied] argument based on, in the antecedent sentence, his character. Sounds ad hominem to me.
Just saying.
Since when?
The assumption here is that you can tell a comment is going to be stupid by testing whether or not the person understands homonyms. The only thing you check with that is language ability. I'd argue you'd be better off searching for common internet memes and cliches instead of original thinking -- since that's a sign of a lazy or uninteresting mind. The author is kind enough to inadvertently supply us with one "...If Fox Nation implemented something like this, they’d have zero commenters..." But you could test stuff like this based on any preconceived viewpoint, such as Obama-socialist, Obamacare, etc. The use of shortcuts and blindly-repeated jokes and phrases is a good sign that you're not going to be getting much from the comment.
Not sure how you'd code it, though, but I'm certain you could come up with some semantic magic given enough input text. I'd imagine you'd use n-grams and some Bayesian logic. You'd have to have a pre-existing corpus of the person's writing, though.
Anyway, it just reiterates that, while funny, this method of captcha is pretty much useless.
edit: Why can't I use an apostrophe in a URL? HN seems to be stripping it out, so couldn't link to the Wikipedia page without a crufty "%27" showing up...
Interesting. It looks like ASCII character 27 is supposed to be valid within URIs according to both RFC 3986 and RFC 1738 (via [0]). Maybe it's a simple component of a system for preempting SQLi on other sites via links from HN?
[0] http://stackoverflow.com/questions/1547899/which-characters-...
Say: George F. Will vs. Rush Limbaugh.
But you can also say, "It is poor form to loose (let free from restraint) your temper (agitated state of mind) in a discussion group."
The falcon cannot hear the falconer;
Things fall apart; the centre cannot hold;
Mere anarchy is loosed upon the world,
The blood-dimmed tide is loosed, and everywhere
The ceremony of innocence is drowned;
The best lack all conviction, while the worst
Are full of passionate intensity.
You make a good point that you were furthering discussion about common usage and not necessarily grammar.
The OED is useful here: it literally defines in as equal to on in the second definition of the prepositional form of 'in'. There's an extended history of the relation between 'in' and 'on', which includes the Latinate source of the Saxon term, its evolution in Old and Middle English, and its use today. Suffice it to say, the distinction is not a trivial one, and 'in' and 'on' are often interchangeable.
You can simply say: "He let loose his temper." so...
It is poor form to sleep in a meeting. It is poor form to scream in a meeting. It is poor form to loose your temper in a meeting.
---
"on" sounds much more natural to me, and I think that's enough for me. Idiom counts in a language, too. In and on might seem technically interchangeable, but (sometimes) if it sounds wrong, it's wrong.
You can say, "I lose my temper." So like you can say "It is poor form to sleep in a meeting.", you can also say, "It is poor form to lose your temper in a meeting."
As someone else commented, you definitely can say, "I let loose my temper." So, "It is poor form to let loose your temper in a meeting." is fine.
But, can you say, "I loose my temper."? I think you can, but it is very very unnatural.
Not that I'm calling you an idiot, you have a very fair point.
We'll forgive the less common words like your "grammer".
:)
Originally I wanted to write something along the line "whoever's grammar is loose lose" but I am almost sure this is not a correct form
Would you consider dyslexia an excuse, or English as a second language? Or even semi-illiteracy? I think it's definitely possible to be very clever and still not a good speller, and not bother to double-check on an Internet forum.
Einstein was notorious as a poor speller, which suggests he didn't double-check himself.
Ignorance and IQ are certainly different. But IQ is generally measuring "problem solving ability", and when to use a contraction is a problem. If one is very clever, one tends to have that particular problem solved. But, whether the misuse is from inability to solve the problem, or a choice to not solve the problem ("can't be bothered"), either of those is a reasonable predictor of job performance, health, and various other life outcomes.
On "Einstein was notorious as a poor speller" – after moving to the US, Einstein became completely bilingual but could never recall how to spell words correctly in both German and English. This is not the same issue as when to use a contraction, and isn't the same as "not bothering".
The ESL student learns, understands the theory of, and tends to be careful with contractions. ESL errors look and sound quite different than most of these errors encountered online.
The illiterate "I'm too cool for your TL;DR grammar-nazi wall-of-text" types have failed to prime they're Baysian neural networks by reading or doing homework, and generally have no idea their "doing it wrong".
I see what you did there.
As far as using common homonyms and misspellings as predictors of "job performance, health, and various other life outcomes", well, I'm not quite sure that's reasonable. For example, xd's comment below:
http://news.ycombinator.com/item?id=2868142
has an error in it "Too many people will spill there emotions" but I don't think it's reasonable to predict too much based on that. There's several reasons these errors are common, not just inability or indolence.
There's often a slip between mind and the keyboard - I know I often simply type the wrong thing even when I know full well what the correct thing is. Muscle memory or something.
Edit: 'their "doing it wrong"'. Ah yes, very good :)
Your argument reminds me of the argument that someone ought to wear a suit and tie to work: There is no excuse for not knowing that a suit and tie is what people wear to work, and anyone who doesn’t do so is not fit to work here.
EDIT: To clarify, it doesn't have to be that stupidity causes bad grammar, but if bad spellers are likely to be bad commenters, that's all you need.
Establishing correlation without causation is a minefield, IMO, although each of us may come to different conclusions. Filtering people on the basis of correlation and not the primary attribute is discrimination. I doubt filtering a forum is illegal discrimination, but I’m uneasy about where that might lead me personally if it were my forum.
Giving a ridiculous example, at a certain time in history, correlation might have established that 90% of the people who used AOL to get onto the Internet posted idiotic things. Should all AOL users have been banned from forums? I suspect that many forums would have had a net positive benefit from such a policy, however the 10% false negatives disturb me greatly.
I would morally prefer other mechanisms for filtering the 90% out, preferably mechanisms that directly test user’s propensity to the “post idiotic things.” This is a personal view, and I can accept that others may not agree.
p.s. And then there’s the question of gathering actual data and not anecdotes. The worst way to proceed would be to claim that since we seem to remember that 90% of all idiotic comments contained grammar mistakes, there’s a 90% chance that a comment containing a grammar mistake is idiotic.
Those who miss nuances of thought and logic are often the same folks who miss nuances of grammar. It's a skill one can learn. Using that as a membership filter would be no different than forming a programming group and restricting it to programmers.
I'm trying to write as correct as I can, but the GP just proved that these mistakes happen to everyone, even the people that think that it might make sense.
What about non-native speakers? Their point might be valuable, the participation in a discussion really help- and meaningful, but still - these now/know, there/their, you're/your mistakes are common. And not an indication of intelligence.
Edit: Just to be sure here and kind of answering the posts below: Guys, I'm a non-native speaker as well (in fact, my accent is terrible. I try to write as decent as possible, but that's a different thing). So - my data point, without any backing but my past experience, is that this is indeed an error that non-native speakers do just as well. I didn't want to imply that non-native speakers are idiots per se according to the criteria of this blog.
In my head, "then" and "than" sound totally different. Same with "they're", "their", and "there", "loose" and "lose", etc.
Probably this just proves that my accent is crap, but for me 'then' and 'than' sound about ~the same~, ditto for 'their'/'there' and 'loose'/'lose'. Again - probably I'm just missing something here. Improving my spoken English is definitely on the list of things I want to do..
When I speak, of course, they sound identical, but my spoken accent is not the accent I think in.
You just triggered an image of my first english teacher. 'To be plus adjective, adjective plus noun, verb plus adverb' was her favorite. ;)
"I'm doing good" instead of "I'm doing well" "He eats real fast" instead of "He eats really quickly"
If enough people speak that way, I don't think you can call it "wrong" so much as a dialect in its own right.
Fast and good are de facto adverbs. Maybe in some American dialects it's common to use other adjectives as adverbs, but not in mine.
Agree that language changes.
Did you mean "He is probably an idiot." or "They are probably idiots."?
The Internet has democratized widely published writing. There is no quality control over comments, except through communities. At the same time, there are large populations who place are undereducated and place little value on education, while they do prize the ability to get a rise out of others.
An unfortunate consequence of this: reading more used to make me feel smarter and I'd sound more literate. Nowadays, through reading reading I tend to absorb things like "loose" for "lose" and other atrocious sounding quirks.
There is a clear market for quality text, and there is power in being the provider of it.
I wholeheartedly disagree. I am pretty much a total idiot (based on many years of evidence) and yet I can easily use the words listed correctly. The fact that I espouse bizarre and unworkable methods of government (see my username) certainly doesn't prevent me from spelling correctly. In fact, unworkable and impractical ideas may be a significant indicator of post-secondary education.
I'm surprised to see such a thinly veiled insult directed at one set of political beliefs sit at #1 on HN.
I'm surprised to see such a thinly veiled insult directed at one
set of political beliefs sit at #1 on HN.*
That being said, I think the not-so-thinly veiled insult directed at the average Tea Party clientele is backed up by casual observation. And that's saying something, coming from a foreigner who has no stake in the US party system (=me).
I am pretty much a total idiot (based on many years of evidence)
and yet I can easily use the words listed correctly.
What is the output of "date -u +%W$(uname)|sha256sum|sed 's/\W//g'"?
EDIT: whoops, I interpreted it as 's/\S//g', which I assumed would discard everything.
I, however, wish my Web had more input from illiterate people. There are a lot of them -- the majority of the world, an overwhelming majority when you restrict yourself to the English language. They have lives, thoughts, and stories too. And if it weren't for recorded music, oral historians, and the occasional documentary they'd be completely invisible in our media.
Of course, the average comment thread on the web is a terrible way to interact with the literate and the illiterate alike. ;)
If the goal is to keep out some idiots (since these tests will admit 33-50% of idiots who can at least complete the OCR test but answer the question randomly) but permit moderately sophisticated bots, mission accomplished.
Providing more context to the captcha solution in general strikes me as helping the bots, even if it confounds idiots.
"Programming today is a race between software engineers striving to build bigger and better idiot-proof programs, and the Universe trying to produce bigger and better idiots. So far, the Universe is winning." - Rich Cook
I always blamed Captchas as being to hard and not user friendly.
This title got me wondering - maybe it was by design, and I am an idiot.
> Whats the abbreviation commonly used for the Hypertext Transfer Protocol?
This weeds out bots (because they generally don't bother with site-specific stuff) and many trolls (because they are either to stupid or to lazy to go through this process).
I think overall, bots would be negatively impacted by a more diverse captcha ecosystem even if the individual implementations are easier to crack. Spammers don't invest 10 or 20 minutes in cracking a domain-specific captcha system just so they can spam one single blog - they'd rather move on to thousands of sites that are protected by the same standard word reading puzzle. Of course, if your site rises into the upper strata of the web, you're also quickly entering a zone where the standard, hard-to-decipher image strategy isn't working so great anymore either.
google "look over there." == 1.5M hits
google "look over their." == 16.2M hits
google "look over they're." == 1.0M hits
Google doesn't support periods, apparently. But, other examples from the article are a little more easy to game:
google "do you know were we are going?" = 8 hits
google "do you know we're we are going?" = 0 hits
google "do you know where we are going?" = 6.1M hits
"Being a native English-speaker does not guarantee that you have language skills good enough to function as a hacker. If your writing is semi-literate, ungrammatical, and riddled with misspellings, many hackers (including myself) will tend to ignore you. While sloppy writing does not invariably mean sloppy thinking, we've generally found the correlation to be strong and we have no use for sloppy thinkers. If you can't yet write competently, learn to."
I still try to read through poor writing, but the comment has to contain something fairly insightful for me to get over a "then/than" mistake. That said I think a Naive Bayes classifier for "sloppy" and "not sloppy" would work better than this captcha system. Not to mention it's super-easy to get around by a bot, and getting rid of bots is the point of a captcha in the first place.
Excellent advice.
Ayn Rand wrote: "If you don't know, the thing to do is not to get scared, but to learn."
Edit: I'm disturbed by the amount of grammar nazi's here that think the quality of someone's grammar reflects their personality...
If the goal is to keep the conversation civilized, why not do something along the lines of "Doing X is a more nice thing to do".
Too many people will spill there emotions into a discussion with sub standard spelling and grammar which makes it very difficult for others to understand them, which leads to inevitable misunderstandings.
I have real trouble imagining that a good point made with substitution erros is somehow less critical to the discussion than a poor point that doesn’t confuse it’s/its.
How would you expect someone taking on a programming challenge of say, an operating system, without the basic understanding of boolean logic to fair?
More importantly, I think it's fair to ask that people learn basic grammar, and if we don't take it upon ourselves to insist on seeing correct grammar, then that will never happen. That's why I think being a so-called grammar nazi is important. Now, being a dick about it is a different matter, but somehow people think you're a dick regardless if you try to correct their spelling/grammar, so there's only so much that can be done there.
That said, there are definitely some levels of pedantry that can be varied. For example, I tend not to be too put off by it's/its. There/their/they're, however, is annoying for me, because the words themselves have such monumentally different meanings. Same deal with fair/fare. Then/than is annoying to me because I actually pronounce them (subtly) differently, so as I'm internally reading the words, it throws me even more.
Overall, I'm really agreeing with you -- I don't think there is serious potential for misunderstandings. And indeed, a good point made with grammatical mistakes is more important than a correctly-made bad one. But a good point made in `proper' English is, I think, better than a good point made in `bad' English.
Another good Muphry's law example.
First link I could come up with: http://singularityblog.singularitysymposium.com/pester-power...
You can find presentation of the same concept here: http://www.slideshare.net/desaiguddu/drag-and-drop-captcha-a...
If anyone of you are researching or preparing Survey presentation or educational research on CAPTCHA feel free to use the content from the presentation.
[Pre-submit verification of 'there' homophones . . . . 100%].
The original source is here
If you need to pass a test to post, sure you'll keep out a lot of bad comments. But you'll also limit the pool of acceptance. There could be plenty of people who fail the test having something positive to add. Standardized tests prove this. There are those who are lousy test takers but end up being very successful. Also, certain topics might be easy to answer with someone with sufficient experience in the subject without requiring even average intelligence.
So this limitation might increase the quality of posts, but also adds a kind of tunnel vision to the site.
Too bad that the current implementation of captchas ever saw the light, they are pushing a security issue to the end user. Seems like a good opportunity :).
The acronym is "Completely Automated Public Turing test to tell Computers and Humans Apart" but it isn't really that automated, so do captchas actually exist?
These words/phrases are wildly different in other languages, and thus well distinct in the mind of non-native folks.
For example, in French:
were -> étais; etions, etiez, étaient (2nd person, singular; 1st, 2nd and 3rd person, plural)
we're -> nous sommes
where -> où
Who is 'Wer' in German
Where is 'Wo' in German
Trivial, but I see lots of people (and sometimes myself) falling for this trap.
According to what they've taught me, it should be "Do you know _____ are we going?" instead of the proposed "Do you know _____ we are going?", shouldn't it (notice the "we are/are we" swapping)?
In 20% of my HN comments I need correct "there"/"their", "buy"/"by" [...] which I write down correctly when re-reading my comments, but subconsciously write incorrect the first time when I do not think about the words but the thoughts.
http://www.buzzfeed.com/chrismenning/captchas-for-keeping-id...
or blogspam count as a useful screen?
There's some potential for confusion in that the verb "to go" can be used for the future imperfect (I think, I was only taught my tenses in French and German lessons) as well as the present tense: note the difference between "We are going to London" or "We are going shopping" and "We are going to go to London" or "We are going to go shopping", or even "We are going to shop". The former is talking about an action we are currently undertaking (to go: implying motion), the latter about a future intent (to go: expressing an intent to do something in the future) to carry out the action in the former case. Using a different verb: "We are riding a bicycle", compared to "we are going to ride a bicycle"
"Where are you at?" is considered nonstandard, but it (and the form "Where you at?" are standard in AAVE.
Fill in the gap: a) it's b) its
"The hacker got access to the server and killed ... processes"
lolz
^You make nouns possessive by adding a saxon genitive to signify the distinct possessive-izing "s" sound. The word "it" is also made possessive with this same sound when spoken, but for the sake of a contraction (read: abomination ;)), we have decided to arbitrarily remove the saxon genitive and replace it with a simple "s". The excuse for this inconsistency is "well it is a pronoun, not a noun, so this is not an inconsistency" does not take into account that English is a language primarily spoken. Written English, where reasonably possible, should approximate the spoken constructs.
Demonstration:
He stepped on the cat*'*s tail.
He stepped on its tail.
At the very least, people who use saxon genitives with the word "it" do not deserve the ridicule people like to heap on them. "it's"/"its" is not a "there"/"their"/"they're" scenario.
First, we weren't alive back when "it's" was used this way. Nobody currently alive was. There's no argument of "they're doing it the old way", because it's not even in living memory. As of the current standards, and those of the past 200 years, they are simply doing it the wrong way.
Second, it can't even be argued that they are confused because of the old way. Nobody accidentally writes her's, their's or your's. The reason people write it's is nothing to do with the possesive apostrophe, it's purely because of confusion with the "it is" contraction.
I object to the assertion that people are accidentally inserting "it is" where they mean possessive "it". Certainly people such as myself who are strongly verbally oriented type the saxon genitive reflexively to reflect the spoken possessive-ization construct. In a similar way, I also very commonly "improperly" insert commas in my sentence to signify pauses instead of using them merely syntactically.
Furthermore, I maintain that "its"/"it's" is not equivalent to "they're"/"their"/"there". The difference between the first is purely convention, while the second set of words are only related to each other incidentally.
"Nobody accidentally writes her's, their's or your's."
Actually, I do quite often.
`The cat stepped on his foot.`
`The cat stepped on its foot.`
Still, I don't get annoyed at its/it's because it's such an easy mistake to make. I still do it occasionally without meaning to. I don't quite get how people consistently get their/there/they're or loose/lose wrong, but that's just me.