Concerns Over Google Modifying SSL Behavior
Lauren Weinstein writes "Google is handling SSL search queries on https://www.google.com/ in a manner significantly different than the standard, expected SSL end-to-end behavior — specifically relating to referer query data. These changes give the potential appearance of favoring sites that buy ads from Google. Regardless of the actual intentions, I do not believe that this appearance is in the best interests of Google in the long run."
Regardless of what business sense this makes/doesn't make for Google - it is better for the users.
The more traffic is sent via HTTPS, the better. The days of concern over the CPU overhead of HTTPS are long past.
Google is an ad agency. What do you expect? Google has to pass the referrer to their advertisers or monetization won't work properly.
Expecting ad sites to run SSL is unreasonable. That would run up the cost of operating a content farm substantially. Made-for Adsense sites would have to have their own IP addresses; virtual hosting wouldn't work.
Google passes Referer info from https to http how?
The gist: Google actively hides referer data when linking from the new SSL site, even if the site that is linked to is also an SSL site, except when the link is an ad.
Well, tough titties. It's Google's site, they can link to you any way they want. If they want to redirect the visitor in a way that hides the query from the linked-to site, that's their prerogative. They could simply make their whole search engine POST the query and you'd never see the search terms, not even with plain HTTP. What are you gonna do about it? Oh right, whine on your blog and have Slashdot link to it.
I turn off the referer header in all browsers and proxies I set up. With the exception of a few shady third-rate direct download web sites whose hotlinking protection trips over this, nobody requires it. One information leak less to worry about. Eat shit, SEO scum.
Both TFA itself, and the summary could do with a summary.
https move in itself is not bad... but the way it is implemented messes up statistics (you know that stuff came from google but no search keywords) and operation of some sites (display a page with the queried keyword to boost relevance). They say it affects less than 1% of the queries only logged on users).. but I think that is a low number.... who is not logged into gmail? maybe not everybody but I suspect figure is higher than 1%
Among others, they could in theory fix that with a redirect to an http site they own, then redirect to the final site.. I am sure there are other ways if they sit around long enough.
definitely a lot of webmasters pissed, that is sure
Never antropomorphize computers, they do not like that
You know, I'd be a lot more concerned about this kind of thing if we weren't hearing Slashdot stories crying wolf practically every day. I'm just not impressed with people trying to call Google evil anymore; none of these so-called revelations have panned out so far, so how likely is this one to go any differently?
I have not reached the security/SSL stuff in my IT course yet so could someone explain this a bit?
I did RTFA, but I am still at a loss as to how and where the problem lies. I typically don't use the HTTPS portion of the Google searches because I don't really care what they know I am searching for. Other places that are slightly more important, like FaceBook, I do browse using HTTPS.
"That's right...I said it."
If anyone was wondering how they do it, they're using JavaScript when you click a link instead of allowing the browser to open the link "normally." e.g.
window.open("").location.href = "http://www.example.com";
This results in the page opening as if it was a "new page" rather than as if it came from any
And who the hell is Lauren Weinstein and why the hell would anyone give a shit what he thinks about Google in the long run?
If I am paying per click for certain search terms, then this data SHOULD be passed along. The other alternative is to just get a bill from google and trust that it is accurate?
As an advertiser I need this information. First to make sure I get the clicks google is charging for me, and more importantly to determine which words don't have a conversion rate worth paying for.
You're the product, not the customer.
This meme needs to die. It superficially seems to have a message which rings true with slashdotters, but really doesn't deliver.
Just because a company is ad funded, doesn't allow a free-pass to provide crap service, whether that be search, or a social network.
You seem to be forgetting that this isn't television, and power users have unprecedented control over how content is displayed, if at all.
The second mistake you people make, is to think yourself part of some geek elite, where actually every kid or gamer can download the tools to control their web experience.
"You're the product, not the customer." basically says that an ad funded company is expected to act as evilly as possible, just because of the way it's funded. The reality is that sometimes there are conflicts of interest, getting it wrong tends to cause a backlash among more technically minded, and generally loud users. Facebook will tend to get away with more than google in this case, because of the technical experience of their users.
Do your part. Add to the conversation, and don't be a sheep by modding this meme up.
This just sounds like an individual gripe that somehow got accepted here at /. You don't like it, Google does, move along there's nothing more to see.
You know, if people don't like how Google runs their business: 1) Don't use it. 2) Start your own competitor. Google wasn't the first search engine. You can go somewhere else, but don't tell them how they should run their own business. That's nebby.
"It's the height of ridiculousness to say for those 9 lines you get hundreds of millions."
I hate referer information when I come from google, mostly because of sites that either:
1) Highlight my search terms in the page. You don't need to highlight every instance of 'of' in the page, and even highlighting the keywords is distracting.
2) Put a big fat "Welcome Google User!" (often with horribly colored letters for Google) that beg you to subscribe to the RSS feed.
I wish there was a chrome extension to hide referrer data just so that I could avoid that.
BTW: If you want an example of useless highlighting, google for VirtualBox and click on the VirtualBox website. I can't believe someone thought that people who can comprehend what VirtualBox is don't know how Ctrl+F works.
The right to protest the State is more sacred than the State.
Google is no more Evil than any company out there trying to make a buck. Do they care about their users? Sure, but only up to the point where it hurts the bottom line to do so.
This new tactic moves along the same line as their view on SEO. Do they want to make it more difficult to obtain better ranking in their site? Yes, but only to the point where they make it easier to pay to get better position within listings.
Is this new process for handling SSL information biased towards their paying customers? Obviously, they are looking to differentiate their free and paid service. If this were to move into larger deployment, say all users (logged in or not), they would be able to offer, as a premium, to their paying customers rereferer data exclusively.
Follow the money! Does the fact that Google is out there to make a buck surprise anyone? I understand that's their goal and I don't consider them to be Evil because of it.
With that said it seems clear that they are not using standard SSL and therefore they should not be able to advertise that their site uses SSL or HTTPS, IMHO.
Encryption: I may not agree with what you say, but I will defend your right to encrypt it...
TFA implies that google is somehow causing my browser to send unencrypted data? I'm not an ssl expert, but i thought the expectation set by ssl is that communication between my browser and google would be encrypted. What google chooses to do with the data i sent them (my request headers, form inputs, etc) has nothing to do with ssl. As far as i know there is no SSL standard that says all data posted over ssl must only be transmitted via ssl from then on.
Google can take my referers and post them on the good year blimp at the superbowl. How is that significantly different from expected end to end behavior?
When a service is provided for free, you aren't the customer, you are the product.
Google handed out my referrer data before, to everybody, for free. Now they restrict it to clicks on ads. My overall privacy has increased. I imagine ad buyers would revolt if they didn't get the referrer data they have always gotten from Google. Google, quite properly, doesn't give a flying *bleep!* about webmasters collecting referrer data on clicks they are getting for free.
Outside of advertisement info, why is this "referrer" data important?
If this is somehow reducing my security, I can see a problem, but if it's just data to help websites know who their customers are, then why should I care?
Google provides a service. They give it free to the customer and if you want your website to have an advantage, then you pay a premium for access to Google's services.
To me this sounds more like a QQ, but I am interested to know if there's something I'm missing as I am not knowledgeable in this area.
This isn't Google somehow modifying the way SSL and referrers work in your browser -- after all, in the normal course of things, you browser is in charge of deciding whether to send a Referer header or not.
This is Google using a JavaScript method to intercept and handle clicks on their site. In some cases the JavaScript does a redirect through non-HTTPS Google so that the referer is sent. In other cases it goes directly to the result site, no referer (as expected).
They could (and probably do?) use a similar trick for non-HTTPS search users.
The days of concern over the CPU overhead of HTTPS are long past.
But the days of concern over the IP address overhead of HTTPS are still with us, and they will remain with us until Windows XP and Android 2.x go away. IE on XP and Android Browser on Android 2.x don't support Server Name Indication (SNI). And without SNI, a user agent can see only the first certificate on port 443 of a given IP address, not the certificates for any of the other dozens or hundreds of domains that may be hosted on that server.
Lately I'm finding Google is getting increasingly unreliable about finding references I want, specifically regarding politics, the economy, and Occupy.
Ask has been filling in the gap quite nicely, but I don't like what seems to be censorship by Google.
I do not fail; I succeed at finding out what does not work.
Referrer information is typically a function of the browser and is passed in your HTTP headers you're sending to the site you're going to. Normally referrer information doesn't persist when you click a HTTP link from a HTTPS page but do when you click a HTTPS link from a HTTPS page. According to the article what Google are doing here is ACTIVELY interfering with the normal functioning of this information. Adding javascript tricks to prevent it being passed to HTTPS pages when it's not a paid link and using similar tricks to ensure advertisers do get that information, regardless of HTTP(S) status. If google didn't give a beep about web masters collecting referrer data on non-paid links, they wouldn't be using the javascript tricks.
Excellent question -- I was very surprised to see absolutely no analysis of this in TFA!
Doing a very quick test googling my own blog from https://google.com/ the referer I end up seeing is like this:
"http://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&ved=0CBwQFjAA&url=http%3A%2F%2Fbrionv.com%2F&ei=fjynTpC4KoSqiQLFvezYDQ&usg=AFQjCNHi_Ia5lQINhrMRGTJyRLFc4ZOajw"
I don't have any Google ads on my site, so I guess this would be in the "Ordinary Site (http: = non-SSL)" category, which TFA claims gets no referer -- but I do get a referer, and it's an intermediary redirect that's on http, leading the browser to happily send that as referer info.
Following the same link from https://encrypted.google.com/ shows no referer, indicating that it either went through no intermediate redirect, or an https one (you can see by testing that there is one, also on https://encrypted.google.com/) that didn't pass on referer info from the browser.
SSL pages on my own site don't seem to be in index, but the intermediate redirects I see on other things like mailing list archives that are in there look the same -- http: redirects from https://google.com/ and https: redirects from https://encrypted.google.com/
I think it's just sending everything through an http redirect so everyone sees referer data, unless you search from encrypted.google.com.
Chu vi parolas Vikipedion?
I submitted a post that Google has stopped using the + symbol
to denote boolean AND, (ie specifically to require the word in
the results.)
It has been replaced with double quotes.
I for one find it EXTREMELY annoying after a decade and a half
of the 'correct way' to have to completely relearn the new way.
http://slashdot.org/firehose.pl?op=view&id=24913740
-AI
For me, it is far better to grasp the Universe as it really is than to persist in delusion
I thought referer was sent by the browser. How does google interfere with this process? The referer is sent as part of your HTTP requests. They come from you, not from the site you were previously on.
Does this have something to do with link redirectors?
lol dorkus you want to waste your mod points defending whatever passes for "honor" in clan Coward, here is another one for you! Cowards suck, get an account... if you can afford it! lololol