Reminds me of how google moved maps.google.com to google.com/maps so that they can ask for location permission in your browser for the whole google domain.
Similarly Google (and Facebook) moved to a combined privacy policy - it effectively grants permission for all services to collect all types of data, including data you wouldn't expect each service to be collecting. All while using examples that mislead the user into thinking such data collection is limited.
For example, if one reads the Privacy clause regarding collection of financial/transactional information they might assume that this is due to Google Pay, what they'd be missing is that even services such as Gmail, Maps and Photos are also collecting financial data. As mentioned, where examples exist in the policy, they always paint a more obvious, narrower collection of data.
According to Google's own admissions on the App Store, their services such as Maps, Photos and Gmail each individually collect location, financial history, purchases, contacts, user content such as photos, videos, audio (and any others), search history, amongst other personal data. The majority of this data has no bearing on the apps functionality whatsoever and comparable services don't collect -any- of this information.
Maps and Photos let you enter your credit card info to buy stuff in the app (you can order food in Maps and prints in Photos).
This is not unusual. Every app that offers food delivery or prints lists "financial info" on their App Store privacy label. And if you drill down into the details, it's specifically payment information.
As far as I can tell, the Gmail app does not collect financial info (it's not in the App Store privacy label).
I believe this is why Amazon no longer includes what you bought in their order status (confirmation, shipped, delivered) emails (an annoying change, imo)
They also wanted to eliminate the receipt brokering businesses. A couple of companies were tracking purchases and allowing users to call for refunds when prices suddenly dropped after purchase.
This was a primary goal of Google Plus: empower cookie / fingerprint joining. Even if Plus were to fail they’d still be able to harvest gmail and youtube for everything else.
To me it didn't seem a co-incidence that Google Plus was canned once Apple implemented their enhanced privacy features in Safari.
Google can already track you website to website with Chrome (e.g. shared browser history, amongst other methods), but on Safari per-site tracking such as how the Like and +1 buttons worked was needed.
"Be profitable" is enough to explain everything, no need to go further than that. Every large company does similar things, because that is where "be profitable" takes you until regulations catches up.
Every large company does similar things, but essentially no company on Earth is capable of doing it at the scale that Google does. This makes Google being evil a much larger problem than other large companies like Nike.
Is that actually known as the reason for certain, or is that reason being assumed?
Because I've seen that presented as a hypothesis but never any actual evidence. I recall another hypothesis had something to do with better Maps integration on Search pages.
I'm sure there are lots of potential internal technical reasons for such a switch. Location permissions is just one possibility.
I dimly recall it being noticed at the time but I suspect it was really a convenient side effect ie a contributory factor and not the primary reason.
I think "branding" is far more likely. google.com is the brand and a single entry point landing on search which then points you at what you "need". Note how you search and can click on the buttons underneath the bar to move into images, maps etc. Maps is just another specialized form of search.
Of course this opinion is not based on reality in the slightest. HN looooves to come up with wild conspiracy theories like this and reiterate them as fact, especially when they prove a corporation is secretly doing something evil.
Things that are searches (like maps) moved onto the search domain (www), other stuff like docs and ads stayed on property specific subdomains. Anything not a core google service (experiments and projects built by outside vendors) moved to withgoogle.com.
They also did that for chat. When hangouts was replaced with "chat" chat moved to mail.google.com. Which means allowing notifications for email allows it for chat as well.
Huh? There has been chat in the Gmail web interface since before Google defederated from Jabber, although I believe it didn’t have notifications aside from changing the window title, for lack of browser APIs at the time.
Wait no. I'm thinking the other way around. It was chat could enable notifications for hangouts.google.com. But when it moved to mail.google.com now allowing chat notifications allows mail notifications, which I didn't want.
It makes a lot of sense to unify web and geographic search in a seamless way. Many users would prefer not to have to grant permissions twice when they do a search like "<product> near me".
But your browser tells you when your location is being used? It’s not like Google can secretly use your location without your browser alerting you to it?
So I guess they have gone a full 180° on that "Don't Be Evil" thing. For Google employees with a moral compass, that must be a little confusing/upsetting.
I'm not sure there are Google employees with a functioning moral compass. If there are, they must have learned to just ignore it.
Can you work directly for an evil company, knowing that it's doing evil, and still consider yourself moral? Especially when you're got the skills to easily get highly lucrative employment elsewhere?
Unironically this? It might violate gdpr to get consent for the purposes of maps but then use it in more contexts. I guess they might include all purposes when the user is asked, and at that point it boils down to whether the user is being asked consent for overly broad purposes or whether it is legitimate to bundle all the Google apps together.
You can't redirect sites like that with DNS. All of those domains resolve to the IP of a load balancer (probably the same one minus some anycast routing), which then decides whether to show the requested service based on the HTTP Host header, not the DNS record. You can quickly verify this by looking up mail.google.com via DNS and putting that IP into your browser bar, which will redirect to google.com instead of opening Gmail.
A CNAME record would just mean they use the same load balancer.
$ curl -H "Host: mail.google.com" 142.251.16.17
...gmail-specific html
$ curl -H "Host: maps.google.com" 142.251.16.17
...gmaps-specific html
$ curl -H "Host: www.google.com" 142.251.16.17
...google search-specific html
For example, if one reads the Privacy clause regarding collection of financial/transactional information they might assume that this is due to Google Pay, what they'd be missing is that even services such as Gmail, Maps and Photos are also collecting financial data. As mentioned, where examples exist in the policy, they always paint a more obvious, narrower collection of data.
According to Google's own admissions on the App Store, their services such as Maps, Photos and Gmail each individually collect location, financial history, purchases, contacts, user content such as photos, videos, audio (and any others), search history, amongst other personal data. The majority of this data has no bearing on the apps functionality whatsoever and comparable services don't collect -any- of this information.