XKCD Inadvertently Causes Googlebomb

MrCopilot writes "As I noted yesterday (and was joined by many others)... in an offhand observation xkcd has singlehandedly changed a small section of the Internet. Changing the results from a Google search for "Died in a Blogging Accident" from 2 to (at this writing) over 7,170 in a little more than 24 hours." If you aren't reading xkcd, you're missing out.

Comments

And this is just adding to it

[ShadowMarth]

Not that I don't love XKCD, but is this really /.-worthy? Oh well. Still, awesome, and each post only serves to compound the results!

Re:And this is just adding to it

[Anonymous Coward]

Yes! What are you, some kind of heathen?

Re:And this is just adding to it

[teslar]

Heresy. Anything related to one of the sites which are lucky enough to earn their own link on the main page of /. is always /.-worthy.

Also, the concept that observing any property of the internet within the internet can affect that property is interesting. If the choice is between reflecting on that or finishing that bloody piece of code I'm writing, I'll take the former, even if it may ultimately be pointless ;)

Re:And this is just adding to it

[Anonymous Coward]

Isn't that the Uncertainty Principle? It certainly would be cool if it was proved this could be applied to the Web or the so-called "Blogosphere".

Re:And this is just adding to it

[Arimus]

"Also, the concept that observing any property of the internet within the internet can affect that property is interesting. If the choice is between reflecting on that or finishing that bloody piece of code I'm writing, I'll take the former, even if it may ultimately be pointless ;)"

Sir,
I must formally give you notice that you are to hand over immediately to the appropriate authorities your geek license and your /. account.
Nothing should ever come between a geek and his code.

(well other than pizza and coffee - but that tends to be more between the geek and the keyboard if they're messy eaters)

Re:And this is just adding to it

[Goaway]

Indeed it is not. This kind of shit happens constantly, and it's not newsworthy just because a site you liked did it. Hell, xkcd has done far cooler things than a pathetic little change in the number google hits for a funny phrase.

Re:And this is just adding to it

[El Yanqui]

Also, the concept that observing any property of the internet within the internet can affect that property is interesting.

No fair! You changed the outcome by measuring it!

Re:And this is just adding to it

[UbuntuDupe]

No, you're confusing the Uncertainty Principle with the Observer Effect.

"The Heisenberg uncertainty principle is also frequently confused with the "observer effect". The uncertainty principle actually describes how precisely we may measure the position and momentum of a particle at the same time -- if we increase the precision in measuring one quantity, we are forced to lose precision in measuring the other. Thus, the uncertainty principle deals with measurement, and not observation. The idea that the Uncertainty Principle is caused by disturbance (and hence by observation) is not considered to be valid by some, although it was extant in the early years of quantum mechanics, and is often repeated in popular treatments."

Re:And this is just adding to it

[SNR monkey]

There seem to be a lot of xkcd readers here which makes it all the more surprising that someone has not pointed out that if you replace "blog" with "blag" (as xkcd is often inclined to do - "News/Blag") in the search "Died in a blogging accident", you get exactly two results. Just like the comic depicts.

Of course, now that I've posted this, people will probably go crazy running up its ranking too.

Re:And this is just adding to it

[Anonymous Coward]

He's obviously talking about work stuff you idiot.

Re:And this is just adding to it

[kylben]

No, you're confusing the Uncertainty Principle with the Observer Effect Nevertheless, "heisenblogging" would still be the obvious and useful term, should anybody suggest coining it...

Re:And this is just adding to it

[eonlabs]

It's ok, there's only one result for "Died in a toothbrush accident"

Re:And this is just adding to it

[Anonymous Coward]

Results 1 - 10 of about 1,710 for heisenblog.

Re:And this is just adding to it

[msuarezalvarez]

You cannot prove the uncertainly principle (the quantum physics one)

The uncertainty principle cannot be applied to the Web or the `blogosphere': it can only be applied to particles in the quantum scale.

We already have waaay too many people `applying' Gödel's theorem or Heisenberg principle to things these do not apply. Please do not add yourself to the list.

Re:And this is just adding to it

[kf6auf]

While not relevant, I would like to point out that I can in fact prove the uncertainty principle. You too can find this proof at the wikipedia page for the uncertainty principle. That being said, it still is totally irrelevant.

Re:And this is just adding to it

[nacturation]

Nothing should ever come between a geek and his code.
(well other than pizza and coffee ...) Or your mom.

Re:And this is just adding to it

[msuarezalvarez]

You cannot prove that nature obeys the uncertainty principle, much as you cannot prove any physical statement: what you can prove is that in the model provided by quantum mechanics, it holds.

And by submitting this story

[saibot834]

You probably change Google's result for "Died in a Blogging Accident" more than xkcd did.

Re:And by submitting this story

[Anonymous Coward]

You probably change Google's result for "Died in a Blogging Accident" more than xkcd did.

And you sir might want to follow the chain of events just one tiny step backwards and notice that it was, in fact, xkcd which caused the submission of the story. No xkcd, no /. story ergo xkcd is still to blame :)

Re:And by submitting this story

[OSUJoe]

That is only true if you ignore the fact that it never would have made it on to Slashdot in the first place if it wasn't for xkcd.

Any results from slashdot belong to xkcd too.

Re:And by submitting this story

[miruku]

/. being second result by tomorrow afternoon?

Re:And by submitting this story

[erbuc]

And by submitting your comments you have also contributed to the mayhem and popularity. E.

Re:And by submitting this story

[ShadowMarth]

257,000 and counting! Keep it up.

Blog writers prosecutions

[arigram]

Considering that many people around the world have been prosecuted for their blogs, imprisoned, tortured and maybe even killed, it is not just humor, its a terrifying fact.

Re:Blog writers prosecutions

[Caiwyn]

Considering that many people around the world have been prosecuted for their blogs, imprisoned, tortured and maybe even killed, it is not just humor, its a terrifying fact.

I've never been a proponent of medical marijuana, but somebody needs to get this guy some weed, STAT.

Re:Blog writers prosecutions

[Anonymous Coward]

You might want to look into the meaning of this "accident" word.

Re:Blog writers prosecutions

[The13thSin]

I'm pretty sure those deaths don't count as accidents...

Re:Blog writers prosecutions

[Dr. Cody]

Don't you get it; it's a neologism, like "to disappear [someone]."

e.g. Mr. Hammond was accidented in the shower just days after blogging about President Obama's top-secret plans to invade Canada.

Re:Blog writers prosecutions

[Viceroy Potatohead]

Just being kind of pedantic, but I think those could be called accidents. An accident can also mean an unfortunate incident that happens unexpectedly and unintentionally, so getting tortured for writing a blog could be an accident if you didn't expect to, and didn't intend to, be tortured for writing it.

Re:Blog writers prosecutions

[Anonymous Coward]

they have oil.

This might explain something

[Xiph]

Never understood why my friends spam me so much, they must find it funny.

It appears that humour is viral.

"Died in a frist psot accident"

[pikine]

So apparently he didn't make it, and I'm making this nth post on behalf of the would-be first poster.

Re:"Died in a frist psot accident"

[Mikkeles]

'Your search - "died in a coding accident" - did not match any documents.'
so I'm safe :)

Re:"Died in a frist psot accident"

[maxwell demon]

I guess you just killed your safety. Surely Google will soon pick up your comment, thus making coding accidents more dangerous.

Re:"Died in a frist psot accident"

[GaryOlson]

Your search - "died in a meeting accident" - did not match any documents.

Damn, and I was looking for a good excuse to avoid the weekly systems meeting.

Re:"Died in a frist psot accident"

[Zugok]

'Your search - "died in a coding accident" - did not match any documents.'

Your search - "died in a slashdotting accident" - did not match any documents.
Ah, it was a myth all along!!

Re:"Died in a frist psot accident"

[jc42]

At this moment in time, there's exactly one google match for "died in a barbecue accident". And also for "died in a haircut accident".

There's gotta be a lot more ...

Re:"Died in a frist psot accident"

[GreyWolf3000]

Well, people certainly don't die from coding, unless you count failing to reproduce...

Re:I just wonder where my xkcd T-shirts are.

[lekikui]

Well, this isn't quite a thinkgeek scale business, from what I understand.

Try posting in the site issues forum and asking, maybe.

Incidentally, good call on Machinae Supremacy. Discovered them a bit ago, wonderful stuff. Have you heard of Abney Park?

Practical idea

[mapkinase]

Jews write "G-d" instead of "God". It's their thing.

May be we should try to write in metaquotes about google searches, modifying quoted search phrases...

Re:Practical idea

[the_humeister]

Yeah, I don't get that one myself. Is God really a proper name? I thought his name was Jehovah (or some other translation). That's like saying my name is Human.

Re:Practical idea

[Valdrax]

It is and isn't a proper name. God has many names, all of which should be treated with respect, though the tetragrammaton is supposedly his "true" name. The prohibition on "taking the Lord's name in vain" is observed quite diligently by many Jews to the point that even a word of indirection (liked "God") is not invoked casually.

Re:Practical idea

[Loconut1389]

Isn't it Yhwh/Yahweh?

http://en.wikipedia.org/wiki/Yahweh

Re:Practical idea

[hazem]

Obligatory Monty Python quote:

"All I said was this piece of halibut was good enough for Jehovah!"

Re:Practical idea

[mabhatter654]

first don't know why the Jews (and other religous types) are so uptight.. God is not he/she/its name anyway... unless King James wrote the Bible.

I do think you have a valid idea about search terms needing tags to identify them as "not text". let the HTML5 guys know!

Re:I just wonder where my xkcd T-shirts are.

[Spalti]

So, I'm obviously not the only one... I ordered at about the same time, and still haven't got my shirts, too. Ok, I'm living in Germany, so was expecting longer delivery times, but it's nearly two months now.

That's no bomb

[JackHoffman]

A Googlebomb is when a page becomes associated with an unfitting search term which doesn't appear on the page itself. This effect is caused when many website authors place misnamed links to that page, usually in an intentional and coordinated manner.

Re:That's no bomb

[Rob the Bold]

True. More of a googlewhack, really.

Re:That's no bomb

[VShael]

Such as the recent Liberal googlebomb which gave the term Liberal Fascism the primary hit return of the Urban Definition of "Fuckwad"

Re:That's no bomb

[Doctor Crumb]

No, it *used* to be a googlewhack, but now they've wrecked that by filling the internet with irrelevant hits with that phrase.

teh intarnets

[blind biker]

I count 'em 250.000 at this moment in time. The Internet is stupid that way...

Re:teh intarnets

[csnydermvpsoft]

Nope, still around 6500. Try adding quotes to your search query.

Re:teh intarnets

[Corsix]

http://www.google.com/search?q=%22Died+in+a+blogging+accident%22 (with quotes) is 8,350 hits
http://www.google.com/search?q=Died+in+a+blogging+accident (without quotes) is 256,000 hits
The XKCD comic and related articles refer to these searches done with quotes rather than without.

Re:teh intarnets

[blind biker]

Right you are, eccept it's 8130. I stand corrected. For now :D

Re:teh intarnets

[blind biker]

Yes, thanks, I stand corrected.

I'm Confused

[smackenzie]

I've read every TFA link in the post, but I'm not sure I understand what is going on.

1. What is the true definition of a Google Bomb? Are we confusing this with Google Washing?

2. Why is this incident a Google Bomb?

3. What makes this particular incident Slashdot newsworthy?

I think this might be a funny scenario -- but I don't get it!? Thanks for the info.

Re:I'm Confused

[matt me]

Slashdot, we know xkcd. If we read it, then we've already heard your story about katanas, ninja students or googlemooting. If we don't read it, then we don't find your story funny. Either way, posting this article is neither amusing nor newsworthy. Keep slashdot to stuff that matters.

Re:I'm Confused

[Feanturi]

Stuff that matters to whom? It mattered to the submitter, and it matters to anyone who finds it amusing or interesting. You are not Slashdot, and neither am I.

Re:I'm Confused

[Danny Rathjens]

One of the most famous google bombs was against microsoft. :)

In November 1999 Google users typing in the search string 'more evil than Satan himself' were given Microsoft as the top result. http://www.vnunet.com/vnunet/news/2126159/msn-search-brands-google-evil-satan

Re:I'm Confused

[skippy_twin]

To paraphrase Colbert:

I am slashdot, and so can you.

I take exception

[Yurka]

to "inadvertently". You have no reason to assume that the author is not smart enough to have foreseen (and even counted on) this effect.

Actually, I take a separate exception to "inadvertantly".

Re:I take exception

[langelgjm]

You have no reason to assume that the author is not smart enough to have foreseen (and even counted on) this effect.

Don't be rediculous.

Re:I take exception

[x_MeRLiN_x]

I don't follow XKDC, but it didn't take me long to find out this has happened before.

This comic spawned a whole different type of [art|softcore pornography]. If you accept the warning and scroll to the bottom you'll see proof of how wrong you are. If you're thinking that these events aren't the same because WetRiffs and XKCD are apparently operated by the same person, you should see that thousands did mention WetRiffs on their web log.

By the way, 33% of your post was misspelt.

Re:I take exception

[langelgjm]

By the way, 33% of your post was misspelt.

Clearly, you missed the joke. And, I do follow XKCD, and am well aware of the exploits it often spawns, many of which are covered here.

Of course

[Derosian]

Of course this is only going to increase the number of pages in the google bomb.

The original Google Bomb is a VERY bad thing

[MichaelCrawford]

It's a form of Black Hat search engine optimization, in which you destroy a competitor's website. The way it's done is to set up a link farm of your own, but with every page pointing at your competitor's site. Eventually Google and the other search engine operators discover the link farm, but assume that your competitor put it there, and remove it from the index.

Thus they tell me at webmasterworld.

Re:The original Google Bomb is a VERY bad thing

[Anonymous Coward]

That's not how it works. When Google recognizes a link farm, it discounts the effect of these links. The result is that the link farm no longer contributes positively to the page rank of the target page, but it does not penalize the target page beyond that. Google has punished sites for shady search engine optimization, but in those cases the sites had always used on-site techniques which could not have been performed by an outsider. Anyway, if that spamming technique could kick other sites out of the Google index, it would be called a "Joe job" (in analogy to the false flag email attack.)

A Google bomb is when many people link to a page and use the same unfitting link text, and then the target page moves UP in the rankings for that particular search term.

Re:The original Google Bomb is a VERY bad thing

[ben there...]

Google has punished sites for shady search engine optimization, but in those cases the sites had always used on-site techniques which could not have been performed by an outsider. Google does penalize for duplicate content. For example, if you setup your domain to have the same content on http://www.slashdot.org/ and http://slashdot.org/ a mirror rather than a redirect (notice the www. is a redirect here). It also penalizes content such as wiki-type content that gets mirrored in several sites around the web. Some webmasters have studied the effect of someone plagiarizing their content in this way and causing that effect. Though obviously their experiments couldn't have been very controlled.

A Google bomb is when many people link to a page and use the same unfitting link text, and then the target page moves UP in the rankings for that particular search term. I agree with you there. It's the only usage I've ever heard of the term. Such as "miserable failure". The first hit for that search used to be Bush's biography at whitehouse.gov, until the articles about the phenomenon itself pushed it down. Google likely fine-tuned their algorithm sometime along the way as well.

Re:The original Google Bomb is a VERY bad thing

[martin-boundary]

Why does everyone assume that this behaviour by SEOs is *bad*? It's Google's algorithm that's got problems with corner cases, and they have zero incentive to fix the flaws in it if everyone just blames the internet users.

Does a google bomb affect *every* search engine? No. It affects *one* search engine with a lot of clout.

Does a google bomb involve illegal hacking of google's servers? No. It involves creating links on people's own damn blogs and websites.

It's sad that people buy the moral victimization that Google's marketing has come up with. This idea that people on the web shouldn't be allowed full free speech, because it's "bad" to write anything they want in case it causes headaches for Google's engineers. At best, it's fanboyism gone wild.

A search engine should reflect what's out there, period. If a lot of blogs link to one site, a search engine should reflect that. If it causes trouble to their algorithm, they should fix their algorithm. But above all, it's not Google's job to tell people that what they're doing on the web is morally "bad".

Re:The original Google Bomb is a VERY bad thing

[Anonymous Coward]

It's a matter of perspective. As a free speech advocate, I agree with you: It's your site, create links as you wish. As a web user, I agree with Google: I'm not looking for the website which is operated by the most successful link farmer. I want relevant search results. I don't mind the occasional prank with an obscure search term, but the technique also works (worked) with search terms which are expected to deliver useful results, and then it makes searching more arduous.

You're right, it's not Google's job to tell people what is morally "bad", but that's not what they do. They tell people that Google disapproves of Googlebombing because it complicates the relevance-calculation. They also adjust the algorithm to reduce or eliminate the effectiveness of Googlebombing and link farming.

So far, common wisdom is that Google only punishes for on-site misbehaviour (behaviour which, in Google's opinion, undermines the calculation of useful results: selling links, keyword spamming, cloaking, linking to bad neighborhoods, etc.) Google does not punish a site for off-site factors (being at the receiving end of a Googlebomb, being linked to from bad neighborhoods,) but a site will usually not benefit from these factors either.

Duplicate content can have negative effects even when the duplication isn't caused by the original site, but that is not a punishment, it's just part of the relevance calculation: Google tries to find the most relevant copy and all others are discounted (as you would expect: What good are dozens of identical sites clogging up the search result pages?) When the algorithm gets it wrong, a copy site gets the good ranking and the original site drops. This can be used against a site, but to exploit that problem, there has to be a big over all relevance difference between the attacking site and the attacked site (in favor of the attacker,) which is usually not the case.

A search engine should reflect what's out there, period.

That topic got closed years ago. Back then, some argued that Google's rank algorithm was a form of censorship because it favored some pages over others. The users have spoken: They use Google because search engines which "blindly" reflect what's out there don't return useful search results. The sites are all listed, but order is important and the users like the order which is calculated by Google's algorithm.

Re:The original Google Bomb is a VERY bad thing

[martin-boundary]

As a web user, I agree with Google: I'm not looking for the website which is operated by the most successful link farmer. I want relevant search results.

Why is that? Surely, as a web user, you would pick a search engine that gives results, not stick with one which provides irrelevant answers? I know I try a few search terms on google, then switch to ask.com and do the same, altavista, msnsearch etc. if I don't find something relevant.

You're right, it's not Google's job to tell people what is morally "bad", but that's not what they do. They tell people that Google disapproves of Googlebombing because it complicates the relevance-calculation. They also adjust the algorithm to reduce or eliminate the effectiveness of Googlebombing and link farming.

I beg to differ.

Google's official webmaster guidelines use loaded language: "cloaking", "sneaky", "little or no original content", words designed to make people feel these things are evil, when they are merely technical hurdles for an algorithm.

Only recently, slashdot had an article entitled "gaming google a gateway to crime?", prompted by a Google engineer's blog entry.

So far, common wisdom is that Google only punishes for on-site misbehaviour

I don't think anybody can object if Google penalizes any websites in their index for any reason, but it's a far cry from there to call it misbehaviour and propagate that meme in the press.

In fact, some of the so called misbehaviour Google warns against is trivially bogus. Take the idea that content shouldn't be duplicated, because it's hard for their robots to deal with properly.

If this was taken seriously, we'd have a brittle internet where information gets lost (forever!) each time a server gets shut down or is lost for some reason like a flood, say. We'd lose wikipedia completely if they ran out of money for their servers, because nobody would host "ethically" another copy to avoid tampering with pagerank. What a daft prospect!

Far from being "bad" behaviour, redundancy of information on the internet is an important benefit, certainly more important than making life easy for robots.

The users have spoken: They use Google because search engines which "blindly" reflect what's out there don't return useful search results. The sites are all listed, but order is important and the users like the order which is calculated by Google's algorithm.

People thought search engines couldn't be improved ten years ago when Google's engine got started. It was ridiculous then and it is so now. Certainly, there's no reason why a useful engine must impose any external structure on the web itself, by publically propagating technical myths about good moral behaviour, or floating silly ideas like the nofollow tag.

I'm the first to admit that relevance ranking technology is both hard and fiddly. That's what makes it scientifically interesting. But I would have a lot more respect for google's engineering team if they just got on with it, rather than complaining publically about "unethical" SEOs breaking their old algorithms.

Re:The original Google Bomb is a VERY bad thing

[arkhan_jg]

SEO IS bad for users of search engines. It means that when I search for something, I don't get relevant links to what I asked for, but instead some spammers latest scam.
This applies to all search engines gamed by SEO users, not just google.

You also forget, commercial speech is not protected speech under free speech laws. SEO Advertising is NOT free speech, it's an attempt to subvert the normal function of the web for commercial advantage of a particular user. I'm not required to read it, and neither is google required to index it.

What's your next suggestion, mail server admins shouldn't perform any spam filtering as its infringing on the "free speech" rights of the spammers? Free speech means the government won't censor you, not that it won't punish you for illegal speech, and it has nothing to do with commercial speech or the interactions between non-government people and companies.

Re:The original Google Bomb is a VERY bad thing

[TheoMurpse]

It's a form of Black Hat search engine optimization, in which you destroy a competitor's website. The way it's done is to set up a link farm of your own, but with every page pointing at your competitor's site. Eventually Google and the other search engine operators discover the link farm, but assume that your competitor put it there, and remove it from the index.

Thus they tell me at webmasterworld.

I've never seen it used that way until now. There really isn't an authoritative source on what it means, but I've always used it in the "miserable failure"=>Bush sense. Wikipedia, Search Engine Land, Urban Dictionary, and even Google Bomb Watch agree with me there.

Now to define what I mean: A Google Bomb is where an entity/group decides to have text X point to site Y, often for humor purposes (e.g., "french military victories" used to link to a fake-Google site saying "no 'french military victories' found, did you mean 'french military defeats'?" so when you hit "I feel lucky," "Google" would allegedly report that). They make a whole lot of pages contain(ing) the text <a href="Y">X</a>, and Google does the rest with its indexing.

Re:The original Google Bomb is a VERY bad thing

[martin-boundary]

SEO IS bad for users of search engines. It means that when I search for something, I don't get relevant links to what I asked for, but instead some spammers latest scam. This applies to all search engines gamed by SEO users, not just google.

Your argument implicitly assumes that a search engine must not evolve. If an engine doesn't evolve, then a spam vulnerability is a problem for users forever. If an engine adapts in response to changing web patterns, then users get the benefit of improved algorithms over time, and competition between search engines.

You also forget, commercial speech is not protected speech under free speech laws. SEO Advertising is NOT free speech, it's an attempt to subvert the normal function of the web for commercial advantage of a particular user. I'm not required to read it, and neither is google required to index it.

I'm not arguing that at all. Google can put whatever they like in their own index. I object to Google telling people (and convincing a lot of them) that some web behaviour like duplicating server content, putting in keywords or whatever is UNETHICAL. It's not a moral problem for the web population, it's a technical problem for their engineers.

What's your next suggestion, mail server admins shouldn't perform any spam filtering as its infringing on the "free speech" rights of the spammers? Free speech means the government won't censor you, not that it won't punish you for illegal speech, and it has nothing to do with commercial speech or the interactions between non-government people and companies.

See previous answer.

Re:The original Google Bomb is a VERY bad thing

[martin-boundary]

I don't give a damn about spammers. I have designed IR systems in the past, I'm well aware of the difficulties, and I object to whiners in general who complain that the "bad" people are breaking their algorithms.

Google has no business putting out the meme that some types of web designs are unethical or even criminal, just to cover up the inadequacies of their system.

Re:The original Google Bomb is a VERY bad thing

[martin-boundary]

So do I, but what does that matter? They are all fooled by mostly the same SEO tactics and unless Google was specifically targeted and one of the others was not, Google usually comes out ahead.

It matters in that condemning SEO tactics on grounds of convenience is unproductive and slows innovation. We all lose when a lot of people think some web designs are "evil". They'll continue to believe that when it's no longer relevant.

Those words describe intention, because the actual technical details are too long to put in a clearly laid out list. It's up to your interpretation whether "cloaking" and "sneaky" are good or bad, but they accurately describe what's going on: The attempt to present a different view to Google compared to what regular clients see. Is that good or bad?

But surely, intention cannot be judged so easily. If a thief uses a hammer to break into a window, does that make the hammer a "sneaky" tool? It's obvious that Google feels under attack from spammers, but that doesn't make it the whole world's problem. There can be all sorts of reasons for giving Googlebot a different picture than regular clients. It may cause a penalty in Google's index, but it certainly doesn't make the technique itself evil. Yet people are told so, as a shorthand!

Google warns against duplicate content because it is not in the webmaster's interest to present a site in a way which causes it to rank badly in the Google search results.

Yes, yes, I know. And in the process it causes webmasters to produce less robust web server farms in the long term, because they believe it's best practice or they feel forced to do so, etc. Redundancy of information is one of the main strengths of the internet. It is surely not up to a mere search engine to bring out ethical redundancy guidelines.

BTW, it is telling that one of the most popular features of Google's UI is the cache. How can redundancy be evidence of bad intentions on the one hand, but a feature on the other? Why are websites discouraged to duplicate content, but encouraged to make it cache friendly?

I don't like the idea behind the "nofollow" attribute myself (I link or I don't link,) but I think Google is right to outline which optimization techniques are helpful and which techniques will cause the algorithm to demote a site. You're still free to ignore them and structure your website and link to other sites any way you wish, but you'll know what the consequences are.

Sure, but algorithms aren't set in stone. What was relevant to interoperability last year may no longer be today. Yet people believe that there's a morally right way to behave as webmasters, and a wrong way. Many bloggers believe that "nofollow" is a good thing (it doesn't actually affect their own PageRank, so they shouldn't have a strong opinion on using it. Although perhaps Google penalizes the mere presence or absence of it, who knows). Then there's a belief that reciprocal linking is "fair" in terms of ranking, while one way linking is not. The language people use is much too loaded with intentions, which remain even when the technical reasons no longer apply.

The meta-discussion, like when Matt Cutts speculates about career options for black-hat search engine optimizers, is not about techniques but about motivations, and yes, what is the motivation behind constructing complex and expensive schemes by which no user will ever be affected, except when he stumbles upon a useless website high in the results pages?

Does motivation matter? Surely, the issue is how to deal with the technical effect on the sorted index? All this speculation does is confuse intention and technique, and arbitrarily label websites as useless.

If the IR field evolves anything like the computer security field, I won't be surprised in a few ye

Re:The original Google Bomb is a VERY bad thing

[martin-boundary]

A hammer is hardly ever a sneaky tool. "Cloaking" describes the intention to give Google a different view of a page. The alternative would be to list all techniques which can be used to achieve that goal: user-agent detection, IP address lists, redirects, white-on-white text, etc.

Look, the word "cloaking" has negative connotations. It implies some degree of dishonesty, or failure to be open. As a placeholder for a list of techniques, it conveys too much.

The analogy with the hammer is a good example of this issue IMO, because the word "sneaky" refers to the thief's bad intentions, but I've misapplied it to describe the tool.

Listing the intention instead of the technique is both shorter and less rigid/limited. It does not involve a moral judgment.

Only provided the term used is neutral. For example, a neutral term for cloaking might be client adaptation. It describes what actually goes on much better: the server changes its output as a function of the observable characteristics of the client. This could be for deceiving Googlebot, or it could be for serving different content to graphical or text browsers, etc. It's obvious by using a neutral term that checking a standardized list of prohibited techniques is not in itself a robust algorithm.

It is not an "ethical redundancy guideline" to tell people not to create infinitely many URLs for the same page. It does not harm the web in any way to follow that guideline. If you want another Wikipedia dump, because you believe that the web needs more outdated copies of everybody's favorite all-encompassing article source, you're free to host one. Just don't expect it to rank well. Quotes, even occasional full page mirrors, are no problem.

That's debatable. The RFCs don't require a unique URL to represent a resource, and humans tend to use several different words to describe things anyway. The uniqueness issue with URLs is merely a technical limitation which arises because they are obvious candidates for unique identifiers.

The impact of this guideline is that webmasters implement redirects all over the place just to preserve the rule of one name for each resource. It's overcomplicated and brittle, yet people think it's best practice. Does it harm the web? Over the long term, I think it does.

Then there's a belief that reciprocal linking is "fair" in terms of ranking, while one way linking is not.

From a search engine's point of view, that's backwards.

It's neither. The effect of a single extra link in a measure such as pagerank depends on the topology of the graph, and cannot be fully predicted unless a fairly large neighbourhood is known. In other words, there's no simple rule of thumb. Other ranking measures have different behaviour altogether.

Does motivation matter?

From a technical point of view, no. But from a business perspective it does matter. A comparison of black hat SEO to petty crime or email spamming might be what is needed to steer non-technical people away from these "services." That is in Google's interest in more than one way: Less black hat SEO means better search results, and people who would attribute their misfortune to Google (after they get burnt by black hat SEO and its detection) might recognize that by using these services they're positioning themselves against Google, so it's not Google's fault that their sites got demoted.

Yes, I agree that's Google's interest. I don't believe Google's interest is a universal basis for adapting the web's content architecture, though. As I wrote before, a search engine should reflect (and interpret for users, if you like) what's out there, not try to change it directly.

I won't be surprised in a few years to see researchers and quality testers activel

The comic numbers were never right

[ergo98]

I happened to see the comic pretty much the minute it was posted (up late working), and curiosity had me searching on the various terms.

Not only did the search counts differ dramatically, even the relative ranking between the items differed.

It appears to have been a humor piece with intentionally bogus numbers, not a statistical summary of danger reporting online.

Re:OT: Comcast now shuts off people's connections

[Cafe Alpha]

Oh come on, schoolmarm, this is exactly the sort of thing Slashdotters are interested in!

Monkey see, monkey do

[0xdeadbeef]

Residual failure

Inadvertantly

[gmf]

Speaking of changing google results... This article is currently the third hit for "inadvertantly"...

Re:Inadvertantly

[Skuldo]

Not something to be proud of :)

Re:Wow.

[Anonymous Coward]

This is the most blatant Slashvertisment ever. There's already a permanent link to that terminally-humourless web comic on the front page already, does it really need an article -containing- an ad as well? Alright already, you already used that word once in the sentence already.

That and the rest of your comment already makes you look uneducated already.

Flawed stats

[tokul]

"Died in an accident" is over 208000. Do stats in pie chart and blogging will take less than 3%.

Re:Flawed stats

[coldcell]

News just in, 3% of all accidents are related to blogging!

Not just death by blogging

[Fear the Clam]

My wife showed me the "killed in a knitting accident" part, which was causing much mayhem in the ravelry knitting and crochet site.

Re:Not just death by blogging

[Qzukk]

And now there's a ton of hits for knitting fatalities too.

I wonder what the two hits for "Died in a Blogging Accident" were before all of this started?

Re:Not just death by blogging

[Zeseft]

http://profile.myspace.com/index.cfm?fuseaction=user.viewprofile&friendID=59755147
http://blog.myspace.com/thekevinwhite

Or search for site://myspace.com "died in a blogging accident" and ignore the first result. They're pretty disappointing if you're expecting to read some grisly tale of someone dying in a blogging accident.

"Died in a Blogging Accident"

[Midnight Thunder]

Do we really need to repeat "Died in a Blogging Accident" for nth time? I mean how many times do we need to state that someone "Died in a Blogging Accident"?

Anyhow, this was another xkcd comic that had its effect: http://xkcd.com/305/

I was curious too...

[rvtheace]

But I just took xkcd's word for it

Uhm

[Cafe Alpha]

yes.

You're sitting on a timebomb

[Nazlfrag]

"died in a computer accident" - 1 result.

Re:You're sitting on a timebomb

[Strange Quark Star]

No, I am afraid it is two now.

Re:You're sitting on a timebomb

[complete loony]

Sorry, wrong again. Now it's four.

This is what's wrong with Google.

[a_nonamiss]

What I find most interesting is now, after the "Googlebomb" try looking at some of the links that come up. More than half in the first few pages are the scum-sucking lowlife advertising sites. Clearly what they're doing is monitoring the "hot Google searches" and then googlepimping© their own sites to match those searches. Everybody knows this is going on, but the efficiency at which these people monitored Google searches, noticed that a particular search was popular, then got their own sites listed really surprises and frightens me. Google is fundamentally broken.

Re:This is what's wrong with Google.

[Yetihehe]

Yeah, just because it is not possible now to fight such overuses of google, it is fundamentally broken.

Re:This is what's wrong with Google.

[Animats]

More than half in the first few pages are the scum-sucking lowlife advertising sites. Clearly what they're doing is monitoring the "hot Google searches" and then googlepimping© their own sites to match those searches.

Searching for phrases in news stories sometimes brings up bottom-feeder ad sites. Take a headline from The Register, search it in Google, and see what comes back. I noticed this a few days ago when we got a writeup in The Register, and the bottom-feeder ad sites not only ranked above the real story, they pushed the real story to the second page. This only lasted for a few days; it looks like the bottom-feeders put up the latest stories.

Re:This is what's wrong with Google.

[onedotzero]

google.com/trends

It's up-to-date enough to scrape data and modify web applications that build keyword-based sites.

Re:This is what's wrong with Google.

[a_nonamiss]

I can't tell if you're being sarcastic or agreeing with my point. :)

Re:This is what's wrong with Google.

[msuarezalvarez]

Be more unique with your queries, then.

Re:This is what's wrong with Google.

[kingturkey]

If people are creating link farms for this search term then they're not getting their "hot searches" from Google itself, they too must have read it on a blog or xkcd. Google trends doesn't show any information on the search because there's not enough search volume. I tried because I thought the giant spike in searches would be interesting to look at.

Re:This is what's wrong with Google.

[Yetihehe]

Oops, I forgot sarcasm tag. I was sarcastic. The whole thing is a war-race, spammers try to outsmart google and google tries to fight back. For now spam is winning, but this doesn't mean that google is fundamentally broken.

Maybe not single-handed

[Beorytis]

I don't read XKCD daily or even often, but I found my way to the "Dangers" through Language Log a widely-read Linguistics-related blog that frequently includes Google-based frequency and attestation research. I'm sure Language Log readers were responsible for a chunk of the googlebomb.

FWIW, I was most interested in the "gardening accident". I didn't do the research yet, but I wondered how many of those references related to Spiñal Tap drummer John "Stumpy" Pepys.

Re:Maybe not single-handed

[blankinthefill]

While they might have helped, the origination of their reason for doing so was STILL the comic in question. I would say that still qualifies as single handed. Now, if both xkcd AND Language Log had boosted the search, but for different reasons, then you would be right, imo. :)

Died in a * accident

[kasperd]

Google will actually let you search for Died in a * accident. If you do so you can see what words people put in there. Right now the fourth result is actually "Died in a blogging accident" (right after three car accidents). I have used that to find out what might be the missing word in other sentences like Grab your * and double click or Either you are with us or you are with the *. Even more interesting if combined with the - operator to filter out the obvious possibilities.

Re:Died in a * accident

[Yetihehe]

Heh, actually for "Either you are with us..." fifth result is your post, talk about metasearches :D

Re:Died in a * accident

[Beorytis]

I, for one, welcome our new * overlords.

Re:Died in a * accident

[maxwell demon]

I just had to try that!

First two results are "insect" (that should be expected).
Next come two "BluRay", but from the same site, so they probably should be counted as one.
Then comes "robot" and "robotic".
And after that there comes "*" - but not your Slashdot post, but one in Language Log.
After that, there follows another "robot" and two "cybernetic".

I haven't looked at the following result pages, but the robots at least win the front page.

Re:Died in a * accident

[Anonymous Homo]

thats kinda neat

Re:Died in a * accident

[Jippy T Flounder]

what * were we talking about?

Re:Died in a * accident

[2bitcomputers]

I have noticed this effect before. My theory is that Google indexes all of the RSS feeds that people have on their customized homepages.

Re:Died in a * accident

[professorfalcon]

Now for the obligatory search.

Re:Died in a * accident

[complete loony]

So I google "welcome our new * overlords", and I get 403 "We're sorry..." "... but your query looks similar to automated requests from a computer virus or spyware application"

You've manually /.ed google!.

Death Penalty

[Fieryphoenix]

This is outrageous. Everyone involved with this obscene web comic should be prosecuted to the fullest extent for causing all of these horrific deaths. It must be put to a stop immediately or the death toll will rise past that of 9/11. And then who will we invade?

It's due to the death in China and Digg

[ckedge]

It's because of this:

      http://www.cnn.com/2008/WORLD/asiapcf/01/11/china.blogger/index.html

and also this:

      http://www.digg.com/world_news/Blogger_Beaten_to_Death_in_China_for_Filming_Argument ..where someone point out that "xkcd's coming wasn't quite so funny any more" but did not provide a direct link.

I can't believe I'm the first one to point this out!

Re:It's due to the death in China and Digg

[Impeesa]

For what it's worth, I (like many others) searched it while there were still only the two matches. Both were references to the same Myspace post. I don't think anything relating to the incident you mention would describe it as an 'accident.'

Re:It's due to the death in China and Digg

[superash]

CNN news & Digg? and you are complaining that nobody at /. read a comment posted on the website digg.com?

You're new here, aren't you? :)

Re:It's due to the death in China and Digg

[rakslice]

Here's hoping China chills out a bit.

http://xkcd.com/340/

More than one GBomb...

[Venotar]

This is sort of a two-fer, since apparently "died in a knitting accident?" has garnered just as much attention :)

Humm what about /. ???

[laplace_man]

Now let's wait for Slashdot after effect. Here is even more curious people than on xkcd :)

Re:It seems text is quite important in google rank

[lintux]

The string "died in a blogging accident" actually has to be in the link to the comic. Most people probably used a different link text like "today's xkcd comic". As long as computers don't understand natural languages, it'll have to be like that.

Re:Oblig.

[blind biker]

By the way, how do you explain that "Died in a knitting accident" brings 460 hits and not 7? Was the XKCD cartoon lying?

Re:Oblig.

[Xinef Jyinaer]

Because

Results 1 - 10 of about 483 for "Died in a knitting accident". (0.16 seconds) Every time it gets mentioned the it gets more and more hits - especially on slashdot.

Inadvertently

[Xinef Jyinaer]

Inadvertently, you keep using that word., I do not think it means what you think it does

Not quite as many as thought...

[Dash Hash]

Not that this is really Slashdot-worthy, but... Who am I to decide what may or may not be worthy for this site.

Anyway, the actual number of results is far less.
Looking at Google right now, it shows "about 8,300" for the "died in a blogging accident" search.
However, actually going through and looking at the real number (skip to the end of the list, show all results, skip to the end again) and the results are much smaller.

Before enabling all references, there are a mere 243 results. Displaying all results, including those from the same site, yields only 849 hits.

While still interesting to see that it jumped from 2 to 849 in a day's time, it is still nowhere near 7000+ as advertised here.

Heisenberg

[Weaselmancer]

The amount of energy you spend studying a thing, changes a thing. Who knew this would apply to Google? Apparently Google must use quantum computers.

Re:Oblig.

[blind biker]

Fair enough, but I checked the other items on the list, and they didn't increase. This one did, by two orders of magnitude.

7170 or 3620?

[somegeekynick]

I see only 3620 hits as I type this.

'Missing out'

[chris411]

Each time someone mentions something to the effect that I'm 'missing out' by not reading XKCD, I give it a look. I don't laugh. I don't even chuckle. I just shrug and move on. I can't possibly be the only one, so how about you take your little 'missing out' preconception and shove it.

Original Results

[pgn674]

I think I found the original 2 results of the search, when the number of results was still down at 12. Both results pointed to this blog: http://blog.myspace.com/index.cfm?fuseaction=blog.view&friendID=59755147&blogID=106406778

Funnies

[kcwhitta]

So if we're missing out if we don't read it, why isn't XKCD in the Funnies Slashbox then?

Re:Funnies

[Lord Bitman]

It has MegaTokyo and User Friendly. Why do you think?

Re:Oblig.

[Xinef Jyinaer]

Results 1 - 10 of about 1,050 for "died in a skydiving accident". (0.23 seconds)
Results 1 - 10 of about 220 for "died in a elevator accident". (0.05 seconds)
Results 1 - 10 of about 505 for "died in a surfing accident". (0.11 seconds)
Results 1 - 10 of about 486 for "died in a skateboarding accident". (0.11 seconds)
Results 1 - 10 of about 168 for "died in a camping accident". (0.15 seconds)
Results 1 - 10 of about 160 for "died in a gardening accident". (0.13 seconds)
Results 1 - 10 of about 91 for "died in an ice skating accident". (0.07 seconds)
Results 1 - 10 of about 461 for "died in a knitting accident". (0.14 seconds)
Results 1 - 10 of about 8,360 for "died in a blogging accident". (0.12 seconds)
Curious indeed, seeing as some of them even decreased (elevator and ice skating)all the others managed to increase though most of them slightly, only knitting and blogging increased large amounts.
I'd attribute this to the fact that they were uncommon to begin with and then absolutely hilarious thus drawing more postings around the intertubes and on the blogosphere. Also thanks to the mod who made my original post redundant. You're my hero!!! A real value to society.

Died in a trolling accident

[4D6963]

In a off hand observation xkcd has single handedly changed a small section of the internet.

Oh my God, they changed the face of the Internet! (actually they mean the Web, not the Internet as a whole, sigh). Here, let me change a (smaller) "section of the Internet" :

Died in a trolling accident.

Right now, doesn't return any result. And now? OMG I did it! I has teh pawar ovar tah Intarwebs!

Re:Died in a trolling accident

[4D6963]

This is true because everyone knows the Web isn't part of the Internet.

Duh, saying this is like saying "everyone knows KFC restaurants are part of the North American continent". Apples and oranges, the Internet is a network, the Web is a service.

Therefore, and to further my bad analogy, if KFC changed its effigy from Colonel Sanders to General Custer, it would change the face of KFC restaurants, but it would be idiotic to say that it changes the face of America just because there are KFC restaurants in America.

Re:Died in a trolling accident

[Greymist]

There is a good reason for this, deaths related to trolling are never an accident.

Re:Died in a trolling accident

[mabhatter654]

They also created a shower pics website... wetriffs.com (NSFW) see http://xkcd.com/305/
and sort of were responsible for the Terminator TV series... http://xkcd.com/311/... or at least the show is close.

Re:Died in a trolling accident

[entgod]

Still no results. You just don't have the power in you.

Re:Died in a trolling accident

[4D6963]

Still no results.

Make that no more, I checked about one hour after I made the comment and also a couple of days later and it was there. Google is just not very consistent.

I think I'll let xkcd speak for me

[aztektum]

clicky

Re:I think I'll let xkcd speak for me

[Daimanta]

That's a great idea.

Step 1:Learn to draw
Step 2:Make popular webcomice
Step 3:Make a comic with something slashdot finds funny or insightful
Step 4:???
Step 5:Profit!

Re:I think I'll let xkcd speak for me

[dkleinsc]

Of course, the true fun of this is that your link is now +5 (Insightful).

Re:I think I'll let xkcd speak for me

[TheVelvetFlamebait]

OK, while we're here and on-topic, who here hasn't experienced this effect on Slashdot? I know I have, I know friends who have, and I've read posts about the perception of Slashdot going downhill, but I haven't actually heard anyone speak out against this effect. So, if you're out there, this is a perfect time to speak up.

Re:I think I'll let xkcd speak for me

[Rynd]

I haven't seen it, but I must be new here.

Re:I think I'll let xkcd speak for me

[mfnickster]

You can skip step 1 (Randall did).

Re:I think I'll let xkcd speak for me

[jackpot777]

Ah, you must be new here...

Oh wait. You already said that.

Bugger.

Re:I think I'll let xkcd speak for me

[Nazlfrag]

As the alt text says, "Fun game: try to post a YouTube comment so stupid that people realize you must be joking. (Hint: this is impossible)"

Don't worry, it's a very big hill we're sliding down, and we're like a glacier compared to most sites.

They probably were, at the time

[serviett]

The author of XKCD has an script that uploads a new cartoon at midnight every monday, wednesday and friday. Hence, if he goes on a long vacation, his site will run as normal.. But this also means that he needs a backlog of comics that are drawn, but not posted - which he has said himself that he has.

I have no official numbers, but i heard somewhere that he's several weeks ahead of the posting schedule, maybe months - so you should have done your searching long ago to get the same results as he did when researching.

If you read the mouseover-text of the comic it seems like he's pretty honest about the numbers, I don't see why he should make them up.

Tag as "heisenblog"

[Anonymous Coward]

No, you're confusing the Uncertainty Principle with the Observer Effect Nevertheless, "heisenblogging" would still be the obvious and useful term, should anybody suggest coining it... Tagged "heisenblog". Everyone do your part.

Re:Oblig.

[blind biker]

I could buy your explanation, actually. It makes sense. I mean, died in a knitting accident? It's so hylarious you almost wish for it.

The only downside of what happened here is, those few people that _actually_ died in knitting accidents, are now swamped in a sea of useless links to the original comic, and same goes for the blogging accident. Assuming, again, that the XKCD comic didn't lie to us in the first place.

I noticed that mods usually turn their other braincell off, just before modding. I had one of my posts modded as "offtopic" a couple of hours ago, even though the post quoted text from the article submission itself. It was critical of Microsoft, though, so that would explain something.

Only 234 pages found

[paylett]

Google seems to over-report the number of results. If you click through the pages of results, there are only a few dozen pages. As at 8:27am+10

Page 23: Results 221 - 230 of about 46,900 for "died in a blogging accident"

Page 24: Results 231 - 234 of 234 for "died in a blogging accident".

In case anyone wants to influence the results.

[ttg512]

I put together a quick php hack which clones the comic in real time. Google of course thinks it's some sort of an automated attack but I suppose that's an ok price to pay. You can download the script at http://24.173.248.34:8191/public/xkcd369.txt (just rename to .php & serve). Have fun.

Re:In case anyone wants to influence the results.

[ttg512]

Since Google catches on so fast I didn't get a real screenshot when I put this together. Here's one http://24.173.248.34:8191/public/xkcd369_ex.jpg from about five minutes ago. Seems there's been a real spike in death by blogging since noon - Has /. killed anyone yet?

And this is a google bomb, because?

[Vexorian]

So, if I googled for "died in a blogging accident" I would find a lot of blogs commenting this one comic that is related to "dying in a blogging accident", hey' the system works.

And what about the universe?

[Darfeld]

Does observing xkcd shorten the life of the universe? I'm confused....

WHAT WERE THE TWO RESULTS?

[exhilaration]

Damn it, why hasn't anyone asked what the original two results were? Is it even possible to get that information anymore?

Re:WHAT WERE THE TWO RESULTS?

[Gromit#35]

They were 2 hits to the same blog
http://blog.myspace.com/thekevinwhite

Someone else (pgn674) has already supposed that these were the 2 hits (http://blog.myspace.com/index.cfm?fuseaction=blog.view&friendID=59755147&blogID=106406778), based on their search when there were just 12 hits. Having seen the comic early on and did the search myself, I visited the link when google was still reporting just the original 2 hits, I can confirm that pgn674 is correct that they were the 2 links, at least as google reported to me.

I was expecting more computer-related deaths

[Slashboo]

I'm suprised it's not much higher, what with all those exploding batteries out there.

Watch it peak at google trends

[aysa]

http://www.google.com/trends/hottrends?q=died+in+a+blogging+accident&date=2008-1-11&sa=X

Google hacking

[WK2]

This reminds me of google hacking, the method of using google to search for things that aren't supposed to be displayed on the web, such as passwords, credit card #s or social security numbers. First, you research what strings to look for. Then you type those strings into google. But all you get back are all of the "google hacking" sites that suggest you look for that string.

Funny, but strange

[crnet]

I guess this is the birth of a few other funny but strange searches too. Is that a * in your pants or are you just happy to see me?

meta-metaposting

[neminem]

And now, furthering the insane meta-ness of this whole thing, the number of hits for the phrase has increased even more dramatically since this was slashdotted, with many hits being blog posts about how it's amusing that this happened. So now, in addition to a bunch of posts about the comic itself, we have a bunch of posts discussing the comic's effect on google, thus continuing to further effect the results. Next, we need a bunch of posts pointing this out.

Other accidents?

[k3vlar]

Your search - "Died in a googlebombing accident" - did not match any documents. You'd think with google exploding and all...

Not a googlebomb

[dotancohen]

Yay incorrect use of terminology. This is not a GoogleBomb

Re:Not a googlebomb

[owlstead]

"yay the incorrect use of terminology"....

Yay what?

I think I'll let xkcd lick me

[12WTF$]

clicky
'c' licky

This would be funnier if slashdot url representation worked properly
"http://cu.nniling.us/301/" == "http://www.xkcd.com/301/"

Bombs away - I did my part.

[memojuez]

It's now at 264,000 and climbing.

This must be life after being /.'d

"George Whackjob Bush"

[Keith McClary]

Google for "George Whackjob Bush".

There is only one item.