Is 302 Hijacking a Major Issue?

Source Title:
Banning Dogpile
Story Text:

DaveN is threatening to 'out' Dogpile partners who 302, GG doesn't seem to think it is much of an issue.

Is the growing 302 noise just mass-panic or is this a real issue that Google (so far not any complaints about Yahoo!, Ask or MSN?) need to fix ASAP?



It's a major issue for me right now .... just yesterday I finally settled down to see if I could figure out why a certain page has sunk out of sight on a term it used to rule (#1 most of the time, for over three years).

Sure enough, I found an outside site showing up in the command, and sure enough, my missing page is showing up as Google's cache for the "evil" URL.

It's devilish hard to *prove* cause and effect, but c'mon, something is definitely screwed up here and I'd bet big bucks that what happened to my formerly well-ranked page is not just regular algo ups and downs.

I intend to write to Google about it, and see if I can get the evil URL removed, but if anyone wants to see the details send me a private message within the next few hours.

Google response

I do thinl Google have responded to this and claim that they do exactly what the protocol tells them to do. 302 is a temporary redirect and they treat it as such. However, that is just not good enough today! Yahoo have solved it by assigned different rules to 302s from outside domains. Google should do the same instead of playing this childish game.

as Mikkel says

... Yahoo has fixed it already - around June 2004 or so as far as i recall. You can find the details in Tim's post here.

MSN should fix it as well, but i guess the reason that all the talk is about Google is because they have the biggest market share, so a similar problem at MSN and Google affects most people through Google. I don't know anything about Dogpile (apart from using it a few times, latest probably one or two years ago).

As for ASAP, well... i'd rather not comment upon that given how long the problem has been known, but Yahoo were not particularly fast even though they were first.

As for "panic", i'm really tempted to write a classic quote that i will regret later, but i will use a few more words - the essence is that there is some amount of "overreaction", yes, and having in some sense contributed to this i also feel that it's justified. It's definitely not "crying wolf" as people have already been eaten (figuratively speaking of course - that was not the quote either, btw.)

However, fair warning had been given a really long time in advance, i must say that (and not only by me, also by others). There's been plenty of time to solve the problem, test the setup, correct errors, and implement it a few times over. This is a simple problem, and there are several simple solutions - it's not like string theory or anything.

it's not rocket science either

The dependence on quoting the rfc is really quite lame and tiresome.

The 302 may have told the crawler where to find the page, but the page still exists where it exists and ought to be credited for the content. It, the index, is after all an index of pages relevant to a search term. Not a rat's nest of links as lionised by the much beloved origins of pagerank. End of story.

What makes this even more tiresome than waiting on DMOZ submissions that are never going to be approved is the constant introduction of new 'features' and acquisitions while this simmers in the background. Was prefetch for firefox more important and useful than fixing this? Apparently to the inmates in charge of the asylum, the answer was yes.

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.