Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Ask HN: Has anyone else noticed Stack Overflow clones in Google search results?

news.ycombinator.com

Post reply on HN
Print view View original

11–20 of 67 posts

Re: Ask HN: Has anyone else noticed Stack Overflow clones in Google search results?

#11

Post by ibejoeb »

rchrd2 wrote: post #7
plorkyeran wrote: post #5

They've been around for years. A while back a few were sometimes beating SO in google results, but google eventually fixed that. Everything on SO is cc-by-sa licensed, so as long as the source is attributed (and it is on some of the sites), it's legal. The motivation is simple: the clone spam sites have ads on them. They're not actively trying to attack SO; they're just leeching value.

I guess it's a win-win situation. Stack Overflow's ranking goes up, because of all the sites referencing it, and the spammers make money. But as a consequence, the users have to sift through spam results.

I don't think so. Those sites redisplay SO content, but they're not linking to SO. They're actually trying to out-rank them, up their own traffic, and drive up their ad revenue.

Re: Ask HN: Has anyone else noticed Stack Overflow clones in Google search results?

#13

Post by Animats »

Google doesn't have a good way to establish provenance, and has trouble distinguishing copies from originals. It's a common complaint of blog operators that some bigger blog copied their stuff and got a higher ranking on Google.

Google could check when it saw something, but that won't work against fast scrapers. For that, you need trusted timestamps.

One solution to this would be to have a few time-stamping services. You send in a string, probably a hash, and it adds a timestamp, signs it, and sends back a signed result. Then provide a WordPress plug-in to use this service, hashing and time-stamping each blog entry, and putting the result in the HTML in some standard way. (Perhaps blog entry ). A few mutually mistrustful services for that would help; blogs with serious forgery problems could use multiple time-stamping services.

Search engines then need to look at timestamps as a rating indicator. If two results are very similar, the earliest one wins.

Re: Ask HN: Has anyone else noticed Stack Overflow clones in Google search results?

#14

Post by msie »

JasonPunyon wrote: post #10
volaski wrote: post #6

I don't understand why StackOverflow allows this, yeah creative commons is cool and all but it IS NOT COOL for actual users. I get so annoyed every time i search for something on Google and it leads to an its clone site. It's not like StackOverflow has better search than Google (which is ridiculous). I still have to search on Google if I want quality search result instead of StackOverflow. With more power comes more…

We use that license because it protects the content from us. No matter who comes along to run Stack Overflow in the future, Stack Overflow can't do something like put up a paywall and lock it up. Someone else'll just be able to host a copy.

Yes, look at what happened to imdb.

Re: Ask HN: Has anyone else noticed Stack Overflow clones in Google search results?

#15

Post by rhizome »

fiatjaf wrote: post #12

I don't understand why Google can't figure it out and remove these clones. They can do much harder things. Why couldn't them outrank sites based on equal text content or -- much better -- huge presence of ads.

They have made inroads in the past, but lately copycats have been cropping up in results (iswwwup.com is one I've been seeing a lot) again. I imagine there's a ranking algo update in the future that might fine-tune this more. To be sure, it's going to be an arms race, since the only purpose of these sites is adsense siphoning.

Re: Ask HN: Has anyone else noticed Stack Overflow clones in Google search results?

#16

Post by rhizome »

ibejoeb wrote: post #11
rchrd2 wrote: post #7

Earlier quoted context omitted.

I guess it's a win-win situation. Stack Overflow's ranking goes up, because of all the sites referencing it, and the spammers make money. But as a consequence, the users have to sift through spam results.

I don't think so. Those sites redisplay SO content, but they're not linking to SO. They're actually trying to out-rank them, up their own traffic, and drive up their ad revenue.

Sometimes they link back, sometimes they don't.

Re: Ask HN: Has anyone else noticed Stack Overflow clones in Google search results?

#17

Post by Osmium »

Animats wrote: post #13

Google doesn't have a good way to establish provenance, and has trouble distinguishing copies from originals. It's a common complaint of blog operators that some bigger blog copied their stuff and got a higher ranking on Google. Google could check when it saw something, but that won't work against fast scrapers. For that, you need trusted timestamps. One solution to this would be to have a few time-stamping services.…

Since establishing provenance is such a big problem for Google, perhaps it might be a good idea for Google to offer a time-stamping service itself?

Re: Ask HN: Has anyone else noticed Stack Overflow clones in Google search results?

#18

Post by TeMPOraL »

fiatjaf wrote: post #12

I don't understand why Google can't figure it out and remove these clones. They can do much harder things. Why couldn't them outrank sites based on equal text content or -- much better -- huge presence of ads.

I suppose it's mostly a social/legal problem. Google could get rid of 90% of those scrappers by basically hardcoding a preference for StackOverflow if it has the same content than another site. Same with sites like Wikipedia. But then obviously people will call foul play (probably scrappers and SEO people will be the loudest to complain). So whatever solution Google makes has to be general enough people won't call it unfair, and that makes the problem much more difficult.

Re: Ask HN: Has anyone else noticed Stack Overflow clones in Google search results?

#19

Post by dsjoerg »

Osmium wrote: post #17
Animats wrote: post #13

Google doesn't have a good way to establish provenance, and has trouble distinguishing copies from originals. It's a common complaint of blog operators that some bigger blog copied their stuff and got a higher ranking on Google. Google could check when it saw something, but that won't work against fast scrapers. For that, you need trusted timestamps. One solution to this would be to have a few time-stamping services.…

Since establishing provenance is such a big problem for Google, perhaps it might be a good idea for Google to offer a time-stamping service itself?

It's not directly a problem for Google — it's primarily a problem for sites that create original content.

Re: Ask HN: Has anyone else noticed Stack Overflow clones in Google search results?

#20

Post by ezequiel-garzon »

Osmium wrote: post #17
Animats wrote: post #13

Google doesn't have a good way to establish provenance, and has trouble distinguishing copies from originals. It's a common complaint of blog operators that some bigger blog copied their stuff and got a higher ranking on Google. Google could check when it saw something, but that won't work against fast scrapers. For that, you need trusted timestamps. One solution to this would be to have a few time-stamping services.…

Since establishing provenance is such a big problem for Google, perhaps it might be a good idea for Google to offer a time-stamping service itself?

Exactly! It seems to me Google could easily provide a proof-of-authorship API, especially for text. (You could publish a hash in some feed in case you don't fully trust Google in turn.) I'm not a big fan of conspiracies and such, but by now I'm pretty sure Google has some conflict of interest given that it hasn't offered such a service already.
Post reply on HN