Google Serves Old Search Page To Old Browsers

Rambo Tribble writes: In an apparent move to push those using older browsers to update, Google is reported to be serving outdated search pages to said browsers. The older pages lack features available on the newer versions, and this policy compounds with the limits announced in 2011 on Gmail support for older web clients. As a Google engineer put it, "We're continually making improvements to Search, so we can only provide limited support for some outdated browsers." The BBC offers a fairly comprehensive analysis.

Comments

Yes

[fisted]

Yes! Where? I want it!

Re:Yes

[Anonymous Coward]

You could have Opera fake your User Agent header in order to masquerade as an older browser, but the Opera team dropped that and every other feature that made Opera remotely useful when they decided to turn Opera into a rewrapped version of Chrome.

Re:Yes

[gweihir]

That is why I am still on 12.17. Works well and does get security updates occasionally. For the very few pages that do not work, I can always use FF.

Re:Yes

[Anonymous Coward]

You can get the old Gmail back: https://support.google.com/mail/answer/15049?hl=en

I use it because it loads much faster than the standard page.

Re:Yes

[Anonymous Coward]

You could have Opera fake your User Agent header in order to masquerade as an older browser, but the Opera team dropped that and every other feature that made Opera remotely useful when they decided to turn Opera into a rewrapped version of Chrome.

I use PrefBar all the time to switch User-Agent on Mozilla (Palemoon).

I vastly prefer the old-and-busted Lynx-style Google Image Search. If I'm looking for an image macro of a cat, I want the sidebar with the URL that actually goes to thefuckinglolcat.jpg, not have to page through some link farm's 5000-line-long malware-ridden SEO-optimized clickbait link or a random gigantic forum thread to find the image I'm actually looking for.

With the current GIS (and "current" web browser, even with Javascript disabled), it sucks because it takes me to the page that hosted the image; a page that may not even hold the image at all anymore. With User-Agent set to Lynx, I get the two-frame view, and I can pop the image (whose plaintext URL appears in the right-hand frame) popped open in another tab long my browser has even started loading the "content" in the big left-hand frame.

Please, Google, do continue to not "improve" the experience for "older" browsers. I've had all the UX "improvement" I can take.

Re:Yes

[JMJimmy]

Yes! Where? I want it!

A thousand times this... I'd even take their old search algorithms over their latest "guessing" and "missing terms" crap. In all the time I've used Google I've never had to use anything beyond +/- and "" to find what I wanted... now it's a mess of buggy "search tools", constantly having to switch to "verbatim", a mass of exclusion terms, etc. It's become an absolute nightmare to use. Unfortunately there's no competition to turn to (I've tried the top 6 alternatives, they just can't compete except for Bing which is down right ugly and not much better on its assumptive behaviours)

Re:Yes

[Kazoo the Clown]

Actually, Proxomitron can handle https, you just have to know how to set it up: http://www.proxomitron.info/45...

Re:Yes

[Anonymous Coward]

> Why is it a terrible idea to do google searches in the clear? Who honestly gives a shit?

Even if no one gives a shit, the more encrypted traffic becomes the norm rather than the exception the more secure we all are. Kind of like how high gun ownership rates improve the safety of non gun owners too (well, at least the white ones).

Re:Yes

[ArsenneLupin]

I tried it (by setting the user-agent of my firefox to "Mozilla/5.0 (X11; U; Linux i686; rv:1.7.3) Gecko/20040914 Firefox/0.10"), and I'm delighted. Image search works again and it feels faster too.

I didn't notice the problem you're mentioning about link visited being broken. I searched twice for myself, the first time I clicked on one of my links. After the second search, that link was correctly colored purple, as it should. However, I did notice a small delay before it turned from blue to purple.

And there are no tracking cookies or similar inserted into the links, just the plain links. Overall a good experience :-)

I didn't check though whether the results were maybe outdated (newer pages not listed...), that would be nasty...

A solution to a problem

[king neckbeard]

So, I can get good old Google back by spoofing IE6?

Re:A solution to a problem

[ugen]

Yes, and that's precisely what I've been doing. Firefox + UAControl = score :)

I get to use normal-looking and convenient Google maps (instead of the recent monstrosity that shows up in Safari and other "current" browsers)
I can easily remove trackback links (because Google returns them in plain text to IE6 but goes to great lengths to obfuscate them for current browsers)
And otherwise Google looks clean and neat.

This is one feature of Google I happen to love :)

Re:A solution to a problem

[PopeRatzo]

I've got it about 60% serviceable.

That sounds about right. I tried netvibes for a few months after iGoogle went away and finally just gave up and use Awesome New Tab in Chrome. Unfortunately, you can't make Awesome New Tab the home page. You can make it come up automatically when you start Chrome, but not when you click the little house.

I've stopped missing iGoogle like I used to, but I'm still kind of pissed.

Seriously, it was a dick move that iGoogle went away. The problem with the igHomes and the netvibes is that they can't integrate all the google services as well as google. On the bright side, with all my futzing with iGoogle replacements and unsuccessfully trying to recreate iGoogle, I've come to understand more clearly how much of Google services' functionality comes from them invading my privacy. So now I use Epic browser and startpage.com and actively seek to thwart google wherever I can, at least in regard to my participation in its "business model". Sometimes, I use Tor browser just to be a dick to people who want to upskirt my private life. Not that there's anything interesting in my private life, but apparently, there's something there good enough that Google was willing to give me all sorts of free shit just to get a peek at my undies. Fuck them.

I came to understand that when somebody gives you something nice for free, and they're not related to you in some direct personal way, you should be suspicious as hell and look for the catch before putting out your hand. Of course, many of you have long understood this, but I had to get hit over the head with it because I'm a little slow.

Other solutions?

[Kludge]

But you will miss out on all the bloated javascript bullshit if you spoof an old browser.

If only getting rid of slashdot beta were that easy....

Re:Other solutions?

[J'raxis]

I have NoScript enabled on Slashdot, too. Only way this site is remotely usable, just like Google nowadays.

Re:Other solutions?

[AmiMoJo]

I use Classic with Javascript enabled and it's fine. What unbearable interference is it causing for you?

"Please don't throw me in the briar patch!"

[J'raxis]

This is supposed to motivate me to upgrade? Right now, on the rare occasion I use Google,* I have JavaScript completely disabled to make Google (search, image search, and news) actually work the way I want it to in my browser. If they're going to help with this by serving me their older---read "cleaner, simpler, faster"---search page, I say, thanks, Google!

* Google alternative. They use the Google index but don't track their users.

Slashdot Beta

[gargleblast]

The last remaining problem I had with beta.slashdot was its turning up in google results. I solved that with Firefox redirector and this rule:

Include pattern ... *-beta.slashdot.org*
Redirect to ....... $1.slashdot.org$2
Pattern type ...... Wildcard

Re:Slashdot Beta

[Anonymous Coward]

Awesome! I already had the redirector installed to make all NYTimes pages come up with pagewanted=all, it never occurred to me to use it on Buck Feta!

FWIW:

Include pattern ... *nytimes.com/*pagewanted=1
Redirect to ....... $1nytimes.com/$2pagewanted=all
Pattern type ...... Wildcard

Virtualize

[sjbe]

I dare not keep the Firefox browser current, and I'm using a plug-in that I depend on and is only available for Firefox. I don't keep the browser current because, even though I doubled the memory the laptop had when I got it (to the maximum that the old MB would support), and also replaced the minimal hard drive with a significantly larger hard drive

Why not virtualize this system instance? Then you don't have to worry about updating it or hardware failures. I have a piece of legacy software at my work we still need but that I've largely virtualized because for arcane reasons I cannot install it on new computers. Then I can give it as much RAM as I want. Works pretty well if your hardware is vaguely modern.

I have to say though that I've been using Firefox since before it was called Firefox and I've never had problems like what you describe. I'm on the latest version and it runs roughly as well as any other browser including Chrome and IE and Safari. I prefer Firefox mostly for personal workflow reasons but the others work fine too. I tend to avoid Safari on Windows an IE obviously isn't available outside of Windows so I tend to avoid it when possible.

There are several problems with this

[rudy_wayne]

Google says: "We encourage everyone to make the free upgrade to modern browsers -- they’re more secure and provide a better web experience overall."

Bullshit.

First, this simply is not true. Beginning with version 29 (which is now 3 or 4 versions out of date already), Firefox completely fucked up their browser and turned it into unusable garbage. Newer is not better. Newer is demonstrably worse. If I wanted a shitty browser with extremely limited configurability, I'd use Internet Explorer.

Second, you should be able to view any web page using any browser released in the last 5 years. If something doesn't work properly it means you are putting too much fucked up bullshit into your webpage.

Re:Old-Quality Results

[Neil Boekend]

I often like the "similar meaning" results but not always. Ergo it should be an option IMHO.
For example "Exact:" or something like it.
Or it could simply allow regex with a similar prefix. Then I would have a reason to learn regex.
They already use : as a special string. For example "site:http://www.slashdot.org " only gives results from /. .

Re:Anti-Trust

[lgw]

IE has been decent since 9, and good since 10. No reason not to use it these days. I like the UI better than the latest from FF or Chrome, though I hear good things about Pale Moon - need to try that soon.

Affected Browsers (non scientific)

[Saint Gerbil]

By changing the User Agent string on my brower I found that the following are affected by this:
IE9
FF6.0
Chrome 4
Safari 5
And all versions of Opera before they used the blink engine.

Possibly more when I have the time.

The timeline on when these came out is wavy as hell.

Re:It seems they really want to annoy their custom

[tepples]

I bought this Nexus 7 tablet on Google's store. Therefore I am the customer, no?

You do not need hundreds of open tabs

[sjbe]

When you have lots of tabs open (and I open hundreds of tabs)

Ummm, why? No disrespect intended at all but aside from stress testing the performance limits of the software I really can't imagine a reason why you would want to do that. The overhead of managing that many tabs would be far greater than any benefit. You certainly cannot actually use that many tabs for any genuinely productive purpose.

It should never take longer than 30 seconds and and 100% CPU to open a tab (like when you have 200 tabs open and 1800 tabs unloaded/hibernating).

Please explain to me any vaguely reasonable use case where you could possibly need that many tabs active. I promise to keep an open mind but I seriously cannot think of any reason I'd ever want to do that. I'm pretty sure I've never opened more than 20-30 tabs at once and I rarely open more than ten or so in normal usage.