3 ms·

It's not shenanigans, it's browser sniffing instead of feature detection, and Google does it on too many of its sites. What's annoying about this "controversy"

It's not shenanigans, it's browser sniffing instead of feature detection, and Google does it on too many of its sites. What's annoying about this "controversy" is that the minuscule marketshare slice that is windows phones merits this much coverage, when this is daily life for the significantly larger (in absolute number) set of Opera users, on desktop and mobile. There's also a very real reason that Mozilla has considered adding (and is going to add?) webkit prefixes to a few CSS properties. So, sorry WP users. Unless the claim is also that the Opera/Google war is heating up like crazy, I'm pretty sure you're just being neglected. Welcome to the world of built-in user agent string spoofing just to make sites that would work fine not redirect you first.

azakai 14y ago
> I'm pretty sure you're just being neglected. First of all, when "neglect" gives Google leverage to harm a competitor in a different market (using maps to harm a mobile OS vendor), it is reasonable to think it might not be just neglect, but also strategy. Second, neglect is bad enough by itself. Google has been phasing out support for Opera in properties like blogger.com on desktop, and now is officially on record as saying that google maps for mobile is WebKit-only. That goes against the very basic principle of an open web accessible to all, that Google used to champion.
- magicalist 14y ago
  1) any kind of anticompetitive argument like that is a dangerous road to walk down. Is Google leveraging the popularity of maps.google.com to also marginalize rockmelt? Is a "install chromeframe" interstitial on IE6 some strategy beyond "we don't want to support your browser"? there is always a line, and considering how widespread UA sniffing and webkit-only prefixes are, you would need a considerably higher bar of evidence to demonstrate any sort of strategy at work instead of just lazy front end developers. 2) This we agree on. There are very few reasons to UA string sniff to determine functionality, and redirecting based on it is about as low as you can go on that axis. Yes, there will be users that click through the "this won't work in your browser" button and still get mad at you that your site doesn't work in their browser (just see app reviews for plenty of evidence of this), but there should still be some way of trying the content for the many browsers out there that a developer just doesn't have the incentives to manually test in. ("that Google used to champion" is just dumb rhetoric, though. it ignores the very simple fact that any organization is made up of people with very different opinions and attempts to trivialize the efforts of all the people at google still working hard on web standards and their implementations. I don't agree with all of Mozilla's decisions with regard to the open web (in fact, working group mailing lists are 90% disagreeing with each other on how to advance it), but that doesn't mean that you guys now no longer champion it)
  - cooldeal 14y ago
    From http://www.google.com/about/company/ http://www.google.com/about/company/ "Google’s mission is to organize the world’s information and make it universally accessible and useful."
    - magicalist 14y ago
      Please engage with substance or just misquote "do no evil" and be done with it. If you're just providing a counterexample to "universally accessible": as I already said above, maps clearly doesn't work in Opera Mobile and it's barely useable in Firefox Mobile. It probably doesn't work on a teletype or a Speak & Spell either. "Universally accessible" doesn't seem at odds with the fact that there's always going to be a browser support cutoff point. Google just picked a poor cutoff point in this case. It's only in light of google's commitment to cross-browser standards that this disappoints me.
    - recoiledsnake 14y ago
      ... and then make it inaccessible on Windows Phone.
  - azakai 14y ago
    > Is Google leveraging the popularity of maps.google.com to also marginalize rockmelt? If Google did prevent a browser like rockmelt, which is not a strategic threat to Google, from accessing maps, I would say that likely it is a side effect of something else. But Microsoft IS in fact a strategic threat to Google, and Google constantly thinks about how to compete effectively against it. To do otherwise would be stupid, which Google is not. > it ignores the very simple fact that any organization is made up of people with very different opinions Of course Google is a very large company, and has many people inside it with many different opinions (and as you say, many are great people that truly care about the open web). But like any corporation, it is hierarchical and its leaders at the top do define an overall coherent strategy for the company as a whole. A company that does not do that is doomed to failure, and there are of course plenty of examples in history. Google's overall strategy with respect to the web has shifted in recent years, with less support for alternative browsers on desktop and especially mobile. This makes business sense for Google - supporting other browsers in some cases might help them compete against Google, which like any company Google wants to prevent. This is not unique to Google; what is, is that years ago Google was very different - it didn't have it's own browser and OS, and it supported openness on the web, that openness helped Google as a new company making money on the web. But that openness is no longer crucial to Google's strategy, and that is affecting Google's products more and more.
chewxy 14y ago
Maps on Opera, Ubuntu 12.10: http://i.imgur.com/KKVbj.png http://i.imgur.com/KKVbj.png Maps on Opera, Nexus 4: http://i.imgur.com/0DXVz.png http://i.imgur.com/0DXVz.png You were saying?
- magicalist 14y ago
  I was talking about a number of sites all over the web (including several at Google) that user agent string sniff and tell opera that it isn't supported, not maps specifically. But, speaking of maps specifically, maps.google.com in desktop Opera works fine, but clearly you didn't try using Opera Mobile, because as soon as you start dragging the map, it disappears. (at least on a galaxy nexus running jelly bean)
  - chewxy 14y ago
    Maps on Opera mobile works (Nexus4). It just lags like nobody's business.
  - rquirk 14y ago
    Another example that I encounter semi-regularly is Google Groups. It does user-agent sniffing to block Firefox with Javascript disabled. Change the ua to something random and it works fine without javascript.

azakai 14y ago

> I'm pretty sure you're just being neglected. First of all, when "neglect" gives Google leverage to harm a competitor in a different market (using maps to harm a mobile OS vendor), it is reasonable to think it might not be just neglect, but also strategy. Second, neglect is bad enough by itself. Google has been phasing out support for Opera in properties like blogger.com on desktop, and now is officially on record as saying that google maps for mobile is WebKit-only. That goes against the very basic principle of an open web accessible to all, that Google used to champion.

magicalist 14y ago
1) any kind of anticompetitive argument like that is a dangerous road to walk down. Is Google leveraging the popularity of maps.google.com to also marginalize rockmelt? Is a "install chromeframe" interstitial on IE6 some strategy beyond "we don't want to support your browser"? there is always a line, and considering how widespread UA sniffing and webkit-only prefixes are, you would need a considerably higher bar of evidence to demonstrate any sort of strategy at work instead of just lazy front end developers. 2) This we agree on. There are very few reasons to UA string sniff to determine functionality, and redirecting based on it is about as low as you can go on that axis. Yes, there will be users that click through the "this won't work in your browser" button and still get mad at you that your site doesn't work in their browser (just see app reviews for plenty of evidence of this), but there should still be some way of trying the content for the many browsers out there that a developer just doesn't have the incentives to manually test in. ("that Google used to champion" is just dumb rhetoric, though. it ignores the very simple fact that any organization is made up of people with very different opinions and attempts to trivialize the efforts of all the people at google still working hard on web standards and their implementations. I don't agree with all of Mozilla's decisions with regard to the open web (in fact, working group mailing lists are 90% disagreeing with each other on how to advance it), but that doesn't mean that you guys now no longer champion it)
- cooldeal 14y ago
  From http://www.google.com/about/company/ http://www.google.com/about/company/ "Google’s mission is to organize the world’s information and make it universally accessible and useful."
  - magicalist 14y ago
    Please engage with substance or just misquote "do no evil" and be done with it. If you're just providing a counterexample to "universally accessible": as I already said above, maps clearly doesn't work in Opera Mobile and it's barely useable in Firefox Mobile. It probably doesn't work on a teletype or a Speak & Spell either. "Universally accessible" doesn't seem at odds with the fact that there's always going to be a browser support cutoff point. Google just picked a poor cutoff point in this case. It's only in light of google's commitment to cross-browser standards that this disappoints me.
  - recoiledsnake 14y ago
    ... and then make it inaccessible on Windows Phone.
- azakai 14y ago
  > Is Google leveraging the popularity of maps.google.com to also marginalize rockmelt? If Google did prevent a browser like rockmelt, which is not a strategic threat to Google, from accessing maps, I would say that likely it is a side effect of something else. But Microsoft IS in fact a strategic threat to Google, and Google constantly thinks about how to compete effectively against it. To do otherwise would be stupid, which Google is not. > it ignores the very simple fact that any organization is made up of people with very different opinions Of course Google is a very large company, and has many people inside it with many different opinions (and as you say, many are great people that truly care about the open web). But like any corporation, it is hierarchical and its leaders at the top do define an overall coherent strategy for the company as a whole. A company that does not do that is doomed to failure, and there are of course plenty of examples in history. Google's overall strategy with respect to the web has shifted in recent years, with less support for alternative browsers on desktop and especially mobile. This makes business sense for Google - supporting other browsers in some cases might help them compete against Google, which like any company Google wants to prevent. This is not unique to Google; what is, is that years ago Google was very different - it didn't have it's own browser and OS, and it supported openness on the web, that openness helped Google as a new company making money on the web. But that openness is no longer crucial to Google's strategy, and that is affecting Google's products more and more.

chewxy 14y ago

Maps on Opera, Ubuntu 12.10: http://i.imgur.com/KKVbj.png http://i.imgur.com/KKVbj.png Maps on Opera, Nexus 4: http://i.imgur.com/0DXVz.png http://i.imgur.com/0DXVz.png You were saying?

magicalist 14y ago
I was talking about a number of sites all over the web (including several at Google) that user agent string sniff and tell opera that it isn't supported, not maps specifically. But, speaking of maps specifically, maps.google.com in desktop Opera works fine, but clearly you didn't try using Opera Mobile, because as soon as you start dragging the map, it disappears. (at least on a galaxy nexus running jelly bean)
- chewxy 14y ago
  Maps on Opera mobile works (Nexus4). It just lags like nobody's business.
- rquirk 14y ago
  Another example that I encounter semi-regularly is Google Groups. It does user-agent sniffing to block Firefox with Javascript disabled. Change the ua to something random and it works fine without javascript.