I am doing the same. Chrome was the hardest move but Safari got much better recently and 90% of the websites are working perfectly only few exceptions. Has anybody tried Amazon Workmail yet? I would be interested in the experience.
At least it's a different one, and at least Amazon typically offers a straightforward exchange of money-for-service/product, as opposed to directly monetizing surveillance.
I agree, I'd rather suggest FastMail or even ProtonMail which has paid offerings. At least they make a profit from you paying for their services (at least in regards to FastMail).
Remind me again when Amazon was tracking me on every single website what I am visiting. Amazon's business is not primarily driven by ads and massive data collection. Google is an ad company with some distraction products like Android and Gmail so that they can collect more data and track you better. Amazon's revenue comes from the web store by selling goods and from AWS providing IT cloud services. I am not sure how could you equate Google and Amazon.
I am not really sure if you are serious or not but Amazon would love to have the data Google has access to. They are suggesting you stuff based on everything you do, and suggestions only become more accurate if they have all your data.
If you host your email with Amazon WorkMail they have access to all your emails just like Gmail or G Suite does.
Amazon is in the ad-biz and they're already near the top of the pile [0]. They're in the silent data gathering phase. All the data they have from 1) eCommerce 2) device like Firestick/kindle/alexa 3) apps like Audible 4) retail (wholefoods) and all their other operations can be combined (if not already governed by a single privacy policy) to leapfrog any competitor.
As you say, they are not dependent on their ad business, but they are making a lot of money from it. Expanding operations to improve profits is a no-brainer and they can existing distribution networks like AWS to have an extra edge.
With AWS services you can encrypt everything with your own keys etc. how cool is that? You also get the option to select a region where to store your data.
Do you have some examples of websites that don't work with Firefox? I've never considered moving to Chrome, and I've never had a problem (at least not since the death of IE-only sites).
I switched to Fastmail from Gmail years ago. In the beginning I missed the tagging when I went back to a folder structure, but you get used to it.
Other then that, I'm super happy and can only recommend the service.
Labelling is definitely something that I miss from Gmail; I live without it at FastMail, but I and a few others of us in the company would really like it. Fortunately, we’ve established the technical foundation for supporting labelling with JMAP, which can work with both folders and labelling; I have no timeline (I don’t think it’s on a concrete scheduled roadmap yet), but we are planning on supporting labelling as an alternative to folders.
Labelling was put into gmail and other google services because directories were too hard to implement and labels were seen as an easier approach.
A single table in a database mapping labelName to documentId can implement labels with no extra constraints, and it scales really well.
Put an index on labelName and you can easily list all documents with a given label. Put an index on documentId and you can easily list all labels a document has.
I don't understand why other providers go to the significant extra effort of trying to implement directories.
The largest issue with labels is that it isn't standards compliant. Gmail does some absolutely weird crud at the boundaries of anything not Gmail, such as IMAP or your Takeout export, neither of which understands labels.
FastMail's choice to wait until a new standard that supports it is available is a strong choice towards interoperability and standards.
An issue with fastmail folders is that you can't easily create or move them around in the structure without entering a settings page. Same with creating rules for folders, that's on a whole different page. This part of the system could really use some attention from their developers.
Congratulations! Personally, I'm not that radical but I agree with the motives.
I don't use gmail or google docs for anything essential and have my own email address for the past 20 years anyway, but getting away from Youtube is harder. There is a lot of interesting content on Youtube, like Numberphile and 3Blue1Brown, and I wouldn't know where to find this elsewhere. I also use Youtube for its intended main purpose, listening to illegally pirated music content. I don't understand how Youtube's management have succeeded in staying outside prison so far, it seems that the laws in this area are applied extremely selectively. Anyway, you can find and listen to almost any record from any time period at any time on Youtube without paying a cent, and I haven't found a replacement for that yet either.
Too big to jail. Google spends million lobbying so politicians that usually held local representativs accountable look elsewhere - I mean who would want to cut off their supply of free cash. Google/Youtube also hires tens of thausands of employees, including hefty chunk in California. So those local representatives like govenors or mayors won’t mention raidin Youtube because that may mean many people out of jobs - people that voted them in the office/s in the first place.
But please don’t try it at home. Unless you are a Google size or Google connections, you can be sure they will come after you. I mean they still are on Dot Com case and most likely he will be soon extradited to usa.
I can imagine that in the USA interest google stays on top, because when that spot is open and suddenly a foreign company competes for it it will be a big loss.
Maybe even as far as considering it to be of national security interest.
Actually. Truly pirated content - whole movies - are removed as soon as they're found (or even before they're uploaded). Music videos - and music used by others than the publishers - get all their ad revenue to the publishers themselves, so for the music publishers it becomes a streaming platform like Spotify (plus there's links to Google Play and such in the description).
> I don't understand how Youtube's management have succeeded in staying outside prison so far,
They manage this by letting the big media/music corps claim ad revenue from any video they feel like they own, no questions asked. That's how they manage to stay out of jail. I suspect Youtube is a pretty decent revenue source for these corporations.
It's pretty amazing that content creators haven't left yet; of course, the problems with copyright claims are likely to chase them wherever they go. At least they've become less dependent on Youtube's revenue, thanks to Patreon donations.
I think the problem is that most people use YouTube to find content (I’ve heard it’s the most popular search engine after Google). If you’re not on YouTube you’ll likely not be found.
It's basically impossible to tell, as a human, on YouTube which videos are sanctioned by Copyright owners and being monetized, and which are evading YouTube's Content ID systems. Clearly some are - e.g. videos where the content is horizontally flipped, or in a small box, or the audio is heavily distorted - but that's not universally true.
You can usually click "Show more" in the video description and see if there's any copyrighted content identified by Content ID. If there is, most likely the copyright owners are getting the ad revenue (if it's not an official channel).
I wouldn't call "not using Google" radical. We have grown used to hearing that independence on GAFAs means being "radical", even though it should probably be the norm.
>I also use Youtube for its intended main purpose, listening to illegally pirated music content.
Isn't the quality shit though ? It's good for discovery I guess, but if you have regular stuff that you listen often, wouldn't you be better off procuring good quality files ?
I agree that the quality is shit. It's still my main source for music.
I can't be bothered to create my own playlists, and the one's in Spotify etc are always full of stuff that I really don't like.
On YouTube it's just search for genre and you're done.
> 3Blue1Brown, and I wouldn't know where to find this elsewhere
Check out LBRY (https://lbry.io). 3Blue1Brown's content is available there, along with lots of great Youtube channels. If your favorite channel isn't there yet, reach out to the creator and encourage them to sync by visiting https://lbry.io/youtube (its a one-click process and they can get paid for it). Or let me know who you'd like to see there and we'll do our best to convince them for you.
> [...] they have access to most of our web browsing via Google Chrome (62.5% market share – although given the amount of broken websites (some explicitly Chrome-only!) I’ve found since switching to Firefox, I believe this number may actually be higher)
Anecdotal, but I got in touch with a pretty popular newsletter hosting tool to tell them the charts on Firefox didn't render correctly, only to be told to use Chrome.
I’ve see the odd app / website like that, if it’s not OSS I give them the finger and move on (while politely suggesting they should offer non-google integrated options), if it is OSS I start by opening a feature / bug request (if there isn’t one already) and submit a MR if I’m able to (which often I’m not as I’m more experienced in the Ops and Sec space than I have development (understatement of the century)).
Google went this route as soon as Chrome gained non-negligible market share. The "Works better with chrome" native ads or outright blocking other user-agents.
Emails are like postcards, easily read by anyone who handles them. Just because your mailbox provider isn't google, doesn't mean it doesn't come in contact with google somewhere else.
If you really don't want them to be able to collect the content of your mails then you and all who write to you have to encrypt them. And then you would still have the problem of the metadata to be accessible.
Get yourself a domain with some webhosting attached and start switching mail-by-mail. Everytime you receive something "important", log in to that service and change the mail address.
Thunderbird with filters does a pretty good job with keeping my inbox clean, although I'm aiming for much less mails anyway.
Buy [yourname].com on Gandi. Comes with two 2GB mailboxes or 50GB for something like $1/mo. Forward your GMail to your new address.
Then stop entering your old email address anywhere. If you notice yourself typing your old address, migrate the account to the new one. Eventually all the accounts that you care about will be moved over.
I agree with that strategy. I was using @gmail.com since 2007 and just setting up new e-mail on new domain and using it was enough to reduce old @gmail.com traffic to max 1 new e-mail per week.
Second the choice to use Gandi.net. 3Gb included with your domain. Their spam filter isn't as good as Fastmail's, which isn't as good as Gmail's. I'm back to training my mail clients' Bayesian spam filters like it's 2006.
I also set up an autoresponder on Gmail with the following message:
Hi -
I'm moving off Google Mail to a privacy-respecting provider.
Please update your address book to change my address to <me@mydomain>.
Thank you!
regards
Ben
Mailing lists dump that response into /dev/null and my human correspondents quickly and quietly complied.
The suggestion in the article is pretty straightforward. Forward Gmail to FastMail, and migrate your accounts and personal communications to the new address.
I did roughly the same years ago and it’s not difficult. Every once in a while I get a email forwarded from my old gmail but it’s effectively a complete migration.
I second Fastmail. GMail is 6MB now according to devtools, Fastmail is about half a meg. The latency in Gmail using any non Google browser makes it unusable.
I like mailbox.org but they also do not operate in a legal void and can be forced to hand over your data to law enforcement. Recently there was a ruling by the federal constitutional court (https://www.bundesverfassungsgericht.de/SharedDocs/Pressemit...) which clarified that providers don't have the right to ensure anonymity/untraceability of their users by technical means and are required to implement surveillance measures when asked to by a court. It does not mean that they need to track users indiscriminately though.
That said I still prefer having my mail hosted by a "normal" company located in my country of residence, as it would probably be much easier for me to get hold of them legally if they e.g. decide to delete my e-mail account or revoke my access to it.
Those numbers don’t sound right at all (we serve more JavaScript than that at FastMail, though not terribly much more, so it can’t be that small), but the ratios are about right; I last looked at Gmail shortly after the latest redesign, and when I measured it in Firefox (which tends to use less than Chrome, not so sure about Safari or Edge these days), Gmail was using 160MB or more (and I think I estimated about half of that to be Hangouts), while FastMail sits steady at 10–12MB, or add on another five or six megabytes if you use calendars too.
Note that these sizes depend a little on how you measure; some measurements will increase these numbers by up to a factor of two.
I think we're measuring different things. My 6MB figure comes from bytes in DevTools from visiting gmail.com to a working UI.
PS. if you do work for Fastmail, you can make a quick win by streamlining your import tools. You probably know from my MY records I came from GSuite (otherwise I can select GSuite), that means you know the server names to import from and shouldn't have to ask me.
Ah, I was thinking memory usage, you were talking downloaded data. Got it. Yeah, the ratios are quite similar! And again in Gmail, Hangouts is responsible for what seems to me an irresponsibly large fraction of it.
The import tool and related things are something we’re working on overhauling right now. I haven’t been paying close attention to it (I’ve been working on work that is a precursor to service workers and offline support, which I’m stoked about), but the designs I have seen are looking good.
Thanks, it's nice to hear you're open about plans. You've got a great product and it really seems like it's your time to shine right now (not just GMail falling off, but also making JMAP an open spec etc).
I avoid Google when possible, but... can we stop saying shit in these articles that isn't proven true? Google doesn't release your email address if you delete it. Squatting isn't an issue there.
When we just start saying things that are proven incorrect it makes the entire discussion look like a bunch of spooks, it's not helpful.
Sorry! I should've been more clear. I'm keeping it so there's no chance of them releasing it without my knowledge down the line. Whether their policy right now is to release it or not, I feel safer by not deleting the account.
"Privacy friendly" is more than a binary "has end-to-end encryption or not" when it comes to e-mail, especially given the issues laid out in the quoted document (which the quora answer then dismisses, arguably not having understood the basis for them or at least not properly establishing why they think it still provides a benefit). In nearly all cases you have to trust your provider in the end, so the more important question is "which organization do I trust the most".
As an Aussie who vaguely follows Aussie politics. I know that at the very least the current ruling party has some pretty funny ideas about internet privacy [1][2]. So it kinda scares me that Fastmail is based there.
That said, I'm currently using Fastmail since switching to it from Gmail a few months ago. The service itself is excellent. As far as I can tell, they don't seem to actively scan the contents of your email to build a profile on you like Google does [3]. So, at least for now, there's that.
Really? That must be recent, because when I stopped using Gmail late-ish last year, it was still doing that kinda-cool-but-still-pretty-creepy thing of overlaying airline reservation information gleaned from emails onto Google Maps.
That's true at the moment, but there's also no way to know what Google will decide to do with it in the future. Presumably if the account is deleted, it might not be possible to veto that decision in the future. Better safe than sorry, right?
I don't know how Yahoo went about it but Microsoft for example gave plenty of notice and announced that accounts that haven't been touched for X years will be recycled. When I'll make this move (soon) I will definitely squat the account even if I just end up scheduling sending a mail to myself once every X days.
In parallel I would keep an eye out for policy changes. Like when Google will inevitably tell me that in order for the service to stay free I have to provide a minimum amount of private data per month :D.
I would say that there is a difference between actively sitting on an email address and passively doing so or deactivating it.
To actively sit on the email I would leave in place all security alerts, 2FA, strong password, etc.
I would want to be alerted of any change in circumstance that affected that email.
If you deactivate, if you passively ignore communications, if you drop security... then you've taken the security of something that represented your online identity for a period of time, and essentially are being irresponsible about it.
Far better to just dial usage down, whilst dialling security up, and then to let it tell you if circumstances change.
That was true for Yahoo too. But only up to some point. Then they not only released deleted emails, but even emails that have not been used for a while.
That made me lose my Yahoo email to somebody else and caused all kinds of headaches.
> I’ve debated deleting the @gmail.com e-mail address, but I think it’s wiser to keep it and essentially squat the username lest someone else take it over and cause me trouble down the line.
No need. The google does not allow to register email addresses that was already deleted.
> I started by moving all of my websites off of Google App Engine and onto a dedicated box that I had already owned. That was straightforward enough.
That surprises me, that it was straightforward, I haven't used GAE for some time, but when I did it was python but with a custom ORM etc. - you couldn't just use anything.
Cool... Congratulation.
This is a project for me as well, which started two months ago. Still moving the drive files and re-registering all accounts where I used @gmail.com and my domain email. It's a pain after 16years. Everyday something pops up which I didn't consider.
I'm doing it with own server (hetzner) and protonmail.
For me the hardest part is other accounts associated with an old email address. I’m also moving away from google, but it is tiring to change accounts associated with gmail address to a new one
Third party email services have an import feature and you can set up forwarding of all emails from Gmail to your new address.
That alone allows you to make a move today (from my experience, it takes less than an hour), and then you can move your accounts as you go one by one. You'll probably get rid of a bunch of accounts in the process as well.
Backup codes and then the protection is against illegal SMS porting which is a social engineering attack hard to otherwise defeat. Your phone is fulfilling a different function and yes it does collapse both factors onto one device but the primary risk wasn't loss of phone, it was weak password and no variant second factor and then porting attacks on SMS.
I didn't understand the point your are making. I'm referring to the overall attack surface area of apps like 1password (which I think also have browser extensions ?). TOTP is better than sms, but why put it in the same app as your password ?
You have to ask yourself what's the primary threat. Yes,the point in strong sense of a second factor is a fully independent test. But the actual threat it mostly protects against is credentials threats. Not loss of devices or compromise of a keystore. SMS as second factor is way way worse because of the porting problem. Otp inside 1password is a compromise but it protects against the primary threat.
If you crypt your disk and use a good passphrase or a long pin and passphrase on a phone you are not that badly exposed.
I de-Googled completely a few years ago and it's nowhere near as difficult as people think.
The initial switching of accounts can be tedious, but it's a one-time job, and there are alternatives all all Google's services out there (some better than Google's offerings).
Personally the dreaded 'convenience hit' was temporary for me.
If you want to just avoid some of the tracking and logging, you could use an app like NewPipe which will stream only the video from youtube. I wonder if there's anything on the same lines for the PC.
I'm the same I heavily use the Watch Later feature, slap things in the queue then watch them on my telly alongside mty other media via Kodi with the YouTube plugin.
I guess one alternative would be having some kind of place I can chuck video links into and have them downloaded by youtube-dl then a metadata provider for Kodi to get everything nicely tagged.
More and more people are using Google docs instead of Ms office. How do you get away from having a Google account to edit a shared doc that another company or consultant had shared with you? Keep a dummy account just for these instances?
Totally anecdotal. Should have been clearer on that. When we are working with a consultant or someone else in a sharing context, Google docs is practically assumed.
Personally I switched to a mix of LibreOffice and Pages/Keynote and have been very happy with them. Google docs offers some awesome feature but I don’t think the privacy and ethics of using Google products justified its use and as of MS Office - I personally think that the software quality is terrible (Yes I know excel is powerful but there’s probably a better product or system for most of a spreadsheets solutions (but not all I’m sure)).
If I did I guess I’d create a minimal google account with as little info as possible, ensure it’s security / privacy settings were set as restrictive as possible and then have to settle for using a Firefox container just for google sites (there’s a good pre-made on on the extensions site), you could go as far as also spoofing your user agent etc but that might just be more of a pain than it’s value.
Another option is using your non Google provided email for a Google account. Given Google's huge reach I can certainly see the need for having a Google account for situations like you mention.
With wildcard aliasing offered by most email providers you can have that email address be anything you like, like googlesucks@your.domain.
It might kind of go against the whole de-googling depending on your outlook though.
You might like Collabora Code. It is a online thin client of libreoffice hosted in a webpage with collaborative editing potential. It should be fine for small groups as the main limitation is memory consumption. OnlyOffice looks promising as the design goal is to work with msoffice formats natively. Onlyoffice is client side run so the server could have more Concurrent connections. OnlyOffice has a professional looking companion mobile app.
The problem at this time is the challenge of getting either solution to host but at this time there may be services that do that for you. Hosting yourself could give you peace of mind that vital files are strictly in your custody.
I have been working on a wizard installer for both with nextcloud but the release is still far off due to my quality concerns.
Some privacy-conscious Gmail alternatives from Europe (I use them all):
- Mailbox.org (DE), from 1€/month, basic custom domain, aliases possible
- Posteo.de (DE), from 1€/month
- Migadu.com (SUI), from 4€/month, run unlimited custom domains with very flexible settings (mailboxes, aliases, forwarding)
There's also mailcow that's popular in the self hosting crowd. Anyone use it and can share experience ? I'm a bit lazy to switch email since the other end of the mail is going to be gmail/outlook etc. anyway. So not much net gain.
Great addition, from a quick trial I can say the admin interface is easy to use and has all settings you need as a manager of multiple users and domains.
I'm running mailcow for about a year now. So far so good, very little to no maintenance and the update process through the dockerized version is very easy. Using it with a few friends and 10 custom domains with wildcard addresses, no issues with the spam-filter either.
I'm only slightly worried if they ever abandon the project I'm out of luck maintaining it by myself. But since it's all based on standard software it's rather easy so switch to another solution I guess.
IMAP idle doesn't really work on mobile devices anymore, because no recent version of Android or iOS allows apps to keep the radio alive with persistent TCP connections - you need something that supports APNS and GCM, converting every notification into an HTTP Post request to Apple or Google or another notification provider.
So far, there seems to be no standard extension to IMAP to integrate with those, so you need some app which has a server side component to do it. (Which is exactly what Gmail, Outlook, Whatsapp, Facebook Messenger, etc. all do).
k-9 works fine for me without gcm. So at least on latest aosp, it appears that imap idle is all that it needs. Also, signal, whatsapp and telegram work without gcm as well.
Edit: I don't remember if I had to whitelist k-9 under battery optimisations. Maybe that's a requirement to use notifications without gcm, as it's a common theme in all apps that support notifications without gcm. On Apple devices, you will be generally out of luck, but that's a foregone conclusion.
I've watched the changes of their laws over many years and I can tell you this: Germany is not the country of privacy love (anymore).
Given a decade or two without any changes in this development they'll very likely become like China some day.
Privacy is not a virtue in itself but a tradeoff between saving individuals and their data and public security (i.e. not helping human traffickers by totally blocking law enforcement).
I continue to trust a country where privacy issues are discussed and ruled objectively at the highest court and where law enforcers can only obtain IP addresses for severe crimes - mentioned in the article: drug trading, illegal arms trading.
Germany is nowhere near China with a civil society so active that every privacy threatening action creates strong counter action (i.e. new Bavarian police law). No, civil society cannot prevent everything that the powerful state wants, but neither can state representatives do as they wish (unlike the CP in China).
These providers also don't sell my data as Google does, because I pay for the service. It still strikes a good balance for me.
Edit: replaced "they" with "state representatives"
FYI: The bavarian police law has passed though many people tried to stop it. That also happened with a lot of other laws throughout the last years.
On the other hand people also love stuff like AMZN echo and trust all information to their smartphones.
What's going to stop this development then? A miracle?
How do you know providers don't sell data? Did they tell you that or promise it somehow? :-D
The obvious conclusion of this development is that we need technical solutions which work without the consent of governments and without putting trust into private companies.
> I’ve made it forward and delete any new mail it gets to @defn.io
Deleting doesn't do much, by that time Google already gobbled up all they needed. You'll get the real benefit when the trickle of emails to that address dries up.
> (62.5% market share – although given the amount of broken websites (some explicitly Chrome-only!) I’ve found since switching to Firefox.
There is a reason why this percentage is high. Other browsers suck, firefox for me takes forever to start and don't get me started about how long it takes to load a website. It has been copying chrome with it's last few updates but still way too far. I would happily switch to an alternative to chrome which is equally good or atleast is near it but there aren't any I believe.
What are you running on? I'm not saying you're wrong at all, but I've not seen any startup or load time differences on my macbook or win 10 desktop. If anything firefox is a bit faster and consumes less memory on my macbook.
This is amazing. So many people are starting to de-Google their life. It somehow become a popular subject. Earlier this week Fast Company wrote [1] about it, DuckDuckGo reached 1 billion monthly searches [2] and it's a common thing to see in Hacker News' posts lately.
I'm personally very happy with the raise of awareness. I'm hoping it's will reach the regular folks.
I realized this a few months ago when I started Simple Analytics [3]. I see that advertising is almost done automatically by the press. When Facebook or Google has bad press, it's great for privacy first tools.
> So many people are starting to de-Google their life.
Is this actually true or just an anecdote? Last time I checked the market share for core Google products was growing everywhere except in China, especially in Europe where EU has been killing off the remaining competition through GDPR et al.
I'm all for seeing newcomers take on the incumbent behemoths, but I also like true information.
I'm seeing this trend (it is a personal statement), and if I look at Googles popularity in Google Trends [1] it's going down in the last 5 years. It's still huge compared to privacy first products like DuckDuckGo [2].
Again, it's my view around me. I didn't have my business yet so it could be that my eyes are more open to privacy first products.
That's closer to the point, but market share dropping only implies fewer users if the total number of users is stagnant. If the search market is still growing it could just mean Google is growing at a slower rate than other search engines.
It can be both. They can get new customers in markets while old customers are actively trying to move away from them in higher numbers than ever before.
I'm one of them. While I'm not completely de-Google-ing my life, I did set up Firefox at work and use DuckDuckGo for my searches there now.
I do not like DDG. Google almost always gave me what I wanted, whereas DDG is about 80% of the time. I often end up wishing I'd just used Google the first time instead. 80% sounds like a good number, but it means that I'm often frustrated with it when I wouldn't have been with Google.
DuckDuckGo is really Bing under the covers... People are literally moving from Google back to Microsoft, and feeling good about it because it has different branding.
That's not a privacy issue. But indeed, you will support Microsoft when using DuckDuckGo. I believe it's necessary to use a big search engine before building your own.
Their ads also run via Bing Ads. This is something they really should change a believe. I tried to run ads for Simple Analytics on DDG but it feels very wrong to use Bing Ads for that. You can't also select "Run on DuckDuckGo only". I think they should move to something of themselves which is likely to happen if they grow bigger.
How is that source credible? He works for neither DuckDuckGo nor Bing. For the record I did the thing he suggests: the same search on DDG and Bing–the results are not identical.
DDG themselves claim to use multiple sources, including Bing (but not including Google). I've seen DDG == Bing mentioned enough times that I'm inclined to believe they leverage it heavily, but it doesn't seem as unnuanced as just spitting out Bing API results 1:1.
That statement isn't fully supported by the link you provided. It does however seem reasonable that DDG results are based in large part on Bing and Oath.
"In fact, DuckDuckGo gets its results from over four hundred sources. These include hundreds of vertical sources delivering niche Instant Answers, DuckDuckBot (our crawler) and crowd-sourced sites (like Wikipedia, stored in our answer indexes). We also of course have more traditional links in the search results, which we also source from a variety of partners, including Oath (formerly Yahoo) and Bing."
> We also of course have more traditional links in the search results, which we also source from a variety of partners, including Oath (formerly Yahoo) and Bing.
All organic links are sourced from Oath and Bing. The other 400 sources (and their crawler) are only used for widget style stuff.
I'll note that both of those forum links are from 8 years ago and may not entirely reflect current practice. Their "Sources" page also notes their own crawler the DuckDuckBot, and references connections with more specialized searches like Yelp and Stack Overflow.
Just to give a bit of context, that forum discussion was 8 months before Snapchat was founded and 6 months before Google bought Motorola Mobility (which split off just before the thread in question).
Why does it matter that search engines use other search engines by proxy?
Using DDG for Bing, or Startpage.com/Ixquick for Google, or Searx for either of them, should not provide the original search engine with your data. Bing and Google should just see actually anonymous search queries coming in from DDG / Startpage / whatever you're using, instead of seeing the search requests coming from your browser session.
At least this is my understanding of it - please correct me if I understand this wrong. But if this is the right way to understand it, I don't get how using DDG would then be equivalent to using a Microsoft product directly.
I don't understand why people spread such fake news so readily. In this case you can TRIVIALLY prove this true or false. Did you decide to test this before stating it? No, obviously not.
I decided to search for "space engine exhaust". The reason I searched for this is because it has some hot keywords, but also some keywords that are probably fairly uncommon. The idea was to get a mix of softball hits along with some per engine unique hits. And it looks like it was a pretty good test query. Here they are:
And, lo and behold, they give very different results. DuckDuckGo may utilize Bing's results, but they are not "Bing under the covers" by any stretch of the imagination.
-----
Just for the sake of completion here is the google search for the same:
Interestingly enough I think is a good example of the increasingly large percent of queries where Google gives the clearly worst results. Their top 4/4 results all being for spaceengine.org which is a universe simulator, but very unlikely to be what somebody who was searching for 'space engine exhaust' was after.
I just tested this from various countries to ensure this was not biasing it. It wasn't. Bing gives localized results, such as German results when searching from Germany. DDG gives identical results from any country. So presumably everybody is getting these same results from DDG. Are you saying this is also exactly what you got from Bing?
That's one of the issues for me: those are your results, but there's no telling what anyone else might get when clicking on that link. "Feeds" of all kind, personalized by black boxes rather than explicit parameters we have access to, deprive us of a common (virtual) world to discuss.
Yeah, this is a very good point. I never use Google anymore and assumed that going incognito would have been enough to get roughly 'bare' results. Perhaps a reminder that just because you don't use Google doesn't mean that Google isn't tracking you extensively in any case due to analytics and countless other indirect forms of tracking and profiling.
I just kept reloading the same search through a bunch of different TOR identities + anti-finger printing, and it was like playing search roulette. And indeed there were, on occasion, actually some great results that don't show up elsewhere. It's such a shame that they're all masked behind some black box of tracking with another black box of ML and filtering. It's like two people going to the same library where librarian deciding to hide books from one person or the other because she, and her all seeing eye, thought they'd have less interest in them than other books.
> In fact, DuckDuckGo gets its results from over four hundred sources. These include hundreds of vertical sources delivering niche Instant Answers, DuckDuckBot (our crawler) and crowd-sourced sites (like Wikipedia, stored in our answer indexes). We also of course have more traditional links in the search results, which we also source from a variety of partners, including Oath (formerly Yahoo) and Bing.
In other words, all organic results are from Bing and Oath.
This is not a new thing. Historically speaking numerous "search engines" have actually used different engines under the hood. Yahoo hasn't been "Yahoo" in a long time, for instance, and I believe AltaVista moved from its own engine to somebody else's for a good long time before its demise.
(There's probably an interesting and nuanced discussion of the virtues and perils of a search engine provider depending on another company for something so critical. I hope DDG is investing significantly into their own engine, even if it's not ready-to-go.)
And it's a real change. I wouldn't even particularly care if DuckDuckGo did use Google, because the things about Google I find most objectionable would still be solved by proxying through DDG. Not 100%. The result might serve AMP pages, which is annoying, and not being a Silicon Valley liberal, I can clearly see how censorious Google is with its search results, or at least, how much they tilt the scales in their own political favor. But the primary issue, tracking me and the monetization thereto, would still be solved.
> So many people are starting to de-Google their life
I'm sorry to be that guy but... citation needed. What evidence is there that this is anything but a fringe movement within the tech community let alone mainstream in any way, shape or form? And "common thing to see in HN posts lately" isn't evidence of that, sorry.
> DuckDuckGo reached 1 billion monthly searches
Two questions:
1. How much time do you think it takes for Google to reach 1B searches? and
2. How many DDG users routinely use !g (or otherwise use Google results)?
As much as HNers like to bang the drum about "privacy" the term is ill-defined. Take search because its straightforward. Your search results are personalized by a bunch of factors including, but not limited to:
- Previous searches
- Location
- Language
- Inferred or actual demographics
The fringe privacy element is I guess most concerned with previous searches? Or is it all of the above? And if you say that location is not OK, take a simple search for "bakery". Isn't it better UX to show local bakeries?
Another question: the alternative to "free" (ie ad-supported) models is user pays. How exactly does this work for users in the developing world for whom $5/month might be a significant amount of money? Will they value their "privacy" the same way?
The assumption (by DDGers and their ilk) that using [Big company products] is some Faustian bargain is hyperbolic and unsubstantiated.
In places where $5 a month might be too much they are worth much less to google with fewer people targeting that group. Privacy isn't as big of a concern as access issues.
No one targets the 99% percentile directly. It is much cheaper to go non-targeted advertising and hit as many people as possible perhaps targeting at the country/region or language level only.
IMHO, the key is using your own domain. It's fine to outsource the actual running of a mail server to professionals if you can freely choose who that is, and switch if your opinion changes, self-hosting being one option.
I think deGooglying is easy but less important than stop using Facebook services yourself. As in case of Facebook you have to make your family stop using Facebook services before it can really make an effect
299 comments
[ 3.0 ms ] story [ 300 ms ] threadhttps://developer.amazon.com/mobile-ads
If you host your email with Amazon WorkMail they have access to all your emails just like Gmail or G Suite does.
As you say, they are not dependent on their ad business, but they are making a lot of money from it. Expanding operations to improve profits is a no-brainer and they can existing distribution networks like AWS to have an extra edge.
[0 ]https://digiday.com/marketing/amazon-ad-revenue-2-2b-132-per...
Your point is moot, sorry
Do you have some examples of websites that don't work with Firefox? I've never considered moving to Chrome, and I've never had a problem (at least not since the death of IE-only sites).
I don't understand why people upvote crap like this.
A single table in a database mapping labelName to documentId can implement labels with no extra constraints, and it scales really well.
Put an index on labelName and you can easily list all documents with a given label. Put an index on documentId and you can easily list all labels a document has.
I don't understand why other providers go to the significant extra effort of trying to implement directories.
FastMail's choice to wait until a new standard that supports it is available is a strong choice towards interoperability and standards.
I wouldn't say labels are widely used, Gmail is still really the only provider doing them I know of. It's just that that one provider is a monopoly.
I don't use gmail or google docs for anything essential and have my own email address for the past 20 years anyway, but getting away from Youtube is harder. There is a lot of interesting content on Youtube, like Numberphile and 3Blue1Brown, and I wouldn't know where to find this elsewhere. I also use Youtube for its intended main purpose, listening to illegally pirated music content. I don't understand how Youtube's management have succeeded in staying outside prison so far, it seems that the laws in this area are applied extremely selectively. Anyway, you can find and listen to almost any record from any time period at any time on Youtube without paying a cent, and I haven't found a replacement for that yet either.
They won't sue as long as youtube pays them.
They manage this by letting the big media/music corps claim ad revenue from any video they feel like they own, no questions asked. That's how they manage to stay out of jail. I suspect Youtube is a pretty decent revenue source for these corporations.
Can you stand the ads that are played after every other song?
It's radical from the sheer volume of change you'd have to make to your day to day life alone.
Isn't the quality shit though ? It's good for discovery I guess, but if you have regular stuff that you listen often, wouldn't you be better off procuring good quality files ?
On YouTube it's just search for genre and you're done.
Check out LBRY (https://lbry.io). 3Blue1Brown's content is available there, along with lots of great Youtube channels. If your favorite channel isn't there yet, reach out to the creator and encourage them to sync by visiting https://lbry.io/youtube (its a one-click process and they can get paid for it). Or let me know who you'd like to see there and we'll do our best to convince them for you.
Anecdotal, but I got in touch with a pretty popular newsletter hosting tool to tell them the charts on Firefox didn't render correctly, only to be told to use Chrome.
Emails are like postcards, easily read by anyone who handles them. Just because your mailbox provider isn't google, doesn't mean it doesn't come in contact with google somewhere else.
If you really don't want them to be able to collect the content of your mails then you and all who write to you have to encrypt them. And then you would still have the problem of the metadata to be accessible.
Thunderbird with filters does a pretty good job with keeping my inbox clean, although I'm aiming for much less mails anyway.
Then stop entering your old email address anywhere. If you notice yourself typing your old address, migrate the account to the new one. Eventually all the accounts that you care about will be moved over.
I also set up an autoresponder on Gmail with the following message:
Hi -
I'm moving off Google Mail to a privacy-respecting provider.
Please update your address book to change my address to <me@mydomain>.
Thank you!
regards Ben
Mailing lists dump that response into /dev/null and my human correspondents quickly and quietly complied.
I did roughly the same years ago and it’s not difficult. Every once in a while I get a email forwarded from my old gmail but it’s effectively a complete migration.
https://protonmail.com/security-details
https://www.eclipso.eu/mail-app/
That said I still prefer having my mail hosted by a "normal" company located in my country of residence, as it would probably be much easier for me to get hold of them legally if they e.g. decide to delete my e-mail account or revoke my access to it.
Note that these sizes depend a little on how you measure; some measurements will increase these numbers by up to a factor of two.
I think we're measuring different things. My 6MB figure comes from bytes in DevTools from visiting gmail.com to a working UI.
PS. if you do work for Fastmail, you can make a quick win by streamlining your import tools. You probably know from my MY records I came from GSuite (otherwise I can select GSuite), that means you know the server names to import from and shouldn't have to ask me.
The import tool and related things are something we’re working on overhauling right now. I haven’t been paying close attention to it (I’ve been working on work that is a precursor to service workers and offline support, which I’m stoked about), but the designs I have seen are looking good.
When we just start saying things that are proven incorrect it makes the entire discussion look like a bunch of spooks, it's not helpful.
[1] https://www.quora.com/Fastmail-or-Protonmail-Which-one-is-tr...
As an Aussie who vaguely follows Aussie politics. I know that at the very least the current ruling party has some pretty funny ideas about internet privacy [1][2]. So it kinda scares me that Fastmail is based there.
That said, I'm currently using Fastmail since switching to it from Gmail a few months ago. The service itself is excellent. As far as I can tell, they don't seem to actively scan the contents of your email to build a profile on you like Google does [3]. So, at least for now, there's that.
[1]: https://www.cyberscoop.com/australia-encryption-backdoors-la... [2]: https://en.wikipedia.org/wiki/Internet_censorship_in_Austral... [3]: https://www.fastmail.com/about/privacy.html#sharing
I thought G stopped doing that for mail?
This does not say they will stop using the metadata for ad personalization.
I would love for someone to correct me on this one.
See https://support.google.com/accounts/answer/7673989, or go to https://myaccount.google.com/purchases, https://myaccount.google.com/reservations, https://myaccount.google.com/subscriptions.
That's not a guarantee. Companies can and do shut down accounts for inactivity, sometimes with no warning, e.g.:
https://news.ycombinator.com/item?id=18715866
In parallel I would keep an eye out for policy changes. Like when Google will inevitably tell me that in order for the service to stay free I have to provide a minimum amount of private data per month :D.
To actively sit on the email I would leave in place all security alerts, 2FA, strong password, etc.
I would want to be alerted of any change in circumstance that affected that email.
If you deactivate, if you passively ignore communications, if you drop security... then you've taken the security of something that represented your online identity for a period of time, and essentially are being irresponsible about it.
Far better to just dial usage down, whilst dialling security up, and then to let it tell you if circumstances change.
That made me lose my Yahoo email to somebody else and caused all kinds of headaches.
If you're in Europe you can apply the Right to be forgotten, implement that can be a real pain (data engineer here), link for the template[2].
[1]: https://support.google.com/accounts/answer/3024190?hl=en [2]: https://gdpr.eu/right-to-erasure-request-form/
No need. The google does not allow to register email addresses that was already deleted.
That surprises me, that it was straightforward, I haven't used GAE for some time, but when I did it was python but with a custom ORM etc. - you couldn't just use anything.
What kind of dedicated box are you referring to?
That alone allows you to make a move today (from my experience, it takes less than an hour), and then you can move your accounts as you go one by one. You'll probably get rid of a bunch of accounts in the process as well.
If you crypt your disk and use a good passphrase or a long pin and passphrase on a phone you are not that badly exposed.
I'm now using OTP Auth in iOS, which allows backing up the codes. For me the risks in that are worth being able to easily restore my codes.
The initial switching of accounts can be tedious, but it's a one-time job, and there are alternatives all all Google's services out there (some better than Google's offerings).
Personally the dreaded 'convenience hit' was temporary for me.
I guess one alternative would be having some kind of place I can chuck video links into and have them downloaded by youtube-dl then a metadata provider for Kodi to get everything nicely tagged.
I put stuff in my watch later list during the day and then watch videos in the evening on my PC.
Source? I haven't heard much about this space in a long time now, and due to that I get the feeling that it's kinda meandering.
It isn't even very good, or have many features, but still seems to be the best.
In my opinion, the area is ripe for disruption, but I suspect nobody wants to spend their life re-reinventing the word processor in javascript.
If I did I guess I’d create a minimal google account with as little info as possible, ensure it’s security / privacy settings were set as restrictive as possible and then have to settle for using a Firefox container just for google sites (there’s a good pre-made on on the extensions site), you could go as far as also spoofing your user agent etc but that might just be more of a pain than it’s value.
With wildcard aliasing offered by most email providers you can have that email address be anything you like, like googlesucks@your.domain.
It might kind of go against the whole de-googling depending on your outlook though.
The problem at this time is the challenge of getting either solution to host but at this time there may be services that do that for you. Hosting yourself could give you peace of mind that vital files are strictly in your custody.
I have been working on a wizard installer for both with nextcloud but the release is still far off due to my quality concerns.
I'm only slightly worried if they ever abandon the project I'm out of luck maintaining it by myself. But since it's all based on standard software it's rather easy so switch to another solution I guess.
So far, there seems to be no standard extension to IMAP to integrate with those, so you need some app which has a server side component to do it. (Which is exactly what Gmail, Outlook, Whatsapp, Facebook Messenger, etc. all do).
Edit: I don't remember if I had to whitelist k-9 under battery optimisations. Maybe that's a requirement to use notifications without gcm, as it's a common theme in all apps that support notifications without gcm. On Apple devices, you will be generally out of luck, but that's a foregone conclusion.
2. Compare the privacy policy of Google and these companies.
They are forced to gather data about you as well (https://www.heise.de/newsticker/meldung/Bundesverfassungsger...) and since the government is taking measures to further invade your privacy it's not that safe there anymore.
I've watched the changes of their laws over many years and I can tell you this: Germany is not the country of privacy love (anymore). Given a decade or two without any changes in this development they'll very likely become like China some day.
I continue to trust a country where privacy issues are discussed and ruled objectively at the highest court and where law enforcers can only obtain IP addresses for severe crimes - mentioned in the article: drug trading, illegal arms trading.
Germany is nowhere near China with a civil society so active that every privacy threatening action creates strong counter action (i.e. new Bavarian police law). No, civil society cannot prevent everything that the powerful state wants, but neither can state representatives do as they wish (unlike the CP in China).
These providers also don't sell my data as Google does, because I pay for the service. It still strikes a good balance for me.
Edit: replaced "they" with "state representatives"
On the other hand people also love stuff like AMZN echo and trust all information to their smartphones.
What's going to stop this development then? A miracle?
How do you know providers don't sell data? Did they tell you that or promise it somehow? :-D
The obvious conclusion of this development is that we need technical solutions which work without the consent of governments and without putting trust into private companies.
Deleting doesn't do much, by that time Google already gobbled up all they needed. You'll get the real benefit when the trickle of emails to that address dries up.
> (62.5% market share – although given the amount of broken websites (some explicitly Chrome-only!) I’ve found since switching to Firefox.
There is a reason why this percentage is high. Other browsers suck, firefox for me takes forever to start and don't get me started about how long it takes to load a website. It has been copying chrome with it's last few updates but still way too far. I would happily switch to an alternative to chrome which is equally good or atleast is near it but there aren't any I believe.
I'm personally very happy with the raise of awareness. I'm hoping it's will reach the regular folks.
I realized this a few months ago when I started Simple Analytics [3]. I see that advertising is almost done automatically by the press. When Facebook or Google has bad press, it's great for privacy first tools.
[1] https://www.fastcompany.com/90300072/its-time-to-ditch-googl...
[2] https://twitter.com/DuckDuckGo/status/1091709578444750849
[3] https://simpleanalytics.io/?ref=news.ycombinator.com
Is this actually true or just an anecdote? Last time I checked the market share for core Google products was growing everywhere except in China, especially in Europe where EU has been killing off the remaining competition through GDPR et al.
I'm all for seeing newcomers take on the incumbent behemoths, but I also like true information.
Again, it's my view around me. I didn't have my business yet so it could be that my eyes are more open to privacy first products.
[1] https://g.co/trends/363ud
[2] https://g.co/trends/a7kcZ
So it's anecdata..
[1] https://www.statista.com/statistics/267161/market-share-of-s...
I'm one of them. While I'm not completely de-Google-ing my life, I did set up Firefox at work and use DuckDuckGo for my searches there now.
I do not like DDG. Google almost always gave me what I wanted, whereas DDG is about 80% of the time. I often end up wishing I'd just used Google the first time instead. 80% sounds like a good number, but it means that I'm often frustrated with it when I wouldn't have been with Google.
Source: https://www.quora.com/How-is-the-Bing-API-used-by-DuckDuckGo
I’m not particularly concerned with the backing engine as long as they’re not tracking me and I’m getting acceptable results.
Their ads also run via Bing Ads. This is something they really should change a believe. I tried to run ads for Simple Analytics on DDG but it feels very wrong to use Bing Ads for that. You can't also select "Run on DuckDuckGo only". I think they should move to something of themselves which is likely to happen if they grow bigger.
You need a better source, IMHO.
DDG themselves claim to use multiple sources, including Bing (but not including Google). I've seen DDG == Bing mentioned enough times that I'm inclined to believe they leverage it heavily, but it doesn't seem as unnuanced as just spitting out Bing API results 1:1.
https://en.wikipedia.org/wiki/DuckDuckGo#Overview
It's all Bing and Oath.
That statement isn't fully supported by the link you provided. It does however seem reasonable that DDG results are based in large part on Bing and Oath.
"In fact, DuckDuckGo gets its results from over four hundred sources. These include hundreds of vertical sources delivering niche Instant Answers, DuckDuckBot (our crawler) and crowd-sourced sites (like Wikipedia, stored in our answer indexes). We also of course have more traditional links in the search results, which we also source from a variety of partners, including Oath (formerly Yahoo) and Bing."
Bing returned just one result - An issue on Github
DDG shows multiple results. The bing one is result number 4.
I think they include Bing results but augment it with their own spider.
[1] https://duck.co/help/results/sources
[2] https://duck.co/forum/comment/27893
[3] https://duck.co/forum/thread/4350/did-you-know-that-duckduck...
> We also of course have more traditional links in the search results, which we also source from a variety of partners, including Oath (formerly Yahoo) and Bing.
All organic links are sourced from Oath and Bing. The other 400 sources (and their crawler) are only used for widget style stuff.
They use Bing. That's something else than:
> DuckDuckGo is really Bing under the covers [...]
Just to give a bit of context, that forum discussion was 8 months before Snapchat was founded and 6 months before Google bought Motorola Mobility (which split off just before the thread in question).
Using DDG for Bing, or Startpage.com/Ixquick for Google, or Searx for either of them, should not provide the original search engine with your data. Bing and Google should just see actually anonymous search queries coming in from DDG / Startpage / whatever you're using, instead of seeing the search requests coming from your browser session.
At least this is my understanding of it - please correct me if I understand this wrong. But if this is the right way to understand it, I don't get how using DDG would then be equivalent to using a Microsoft product directly.
I decided to search for "space engine exhaust". The reason I searched for this is because it has some hot keywords, but also some keywords that are probably fairly uncommon. The idea was to get a mix of softball hits along with some per engine unique hits. And it looks like it was a pretty good test query. Here they are:
https://duckduckgo.com/?q=space+engine+exhaust
https://www.bing.com/search?q=space+engine+exhaust
And, lo and behold, they give very different results. DuckDuckGo may utilize Bing's results, but they are not "Bing under the covers" by any stretch of the imagination.
-----
Just for the sake of completion here is the google search for the same:
https://www.google.com/search?q=space+engine+exhaust
Interestingly enough I think is a good example of the increasingly large percent of queries where Google gives the clearly worst results. Their top 4/4 results all being for spaceengine.org which is a universe simulator, but very unlikely to be what somebody who was searching for 'space engine exhaust' was after.
https://en.wikipedia.org/wiki/Space_Shuttle_Main_Engine
https://www.nasa.gov/returntoflight/system/system_SSME.html
https://www.forbes.com/sites/startswithabang/2018/05/23/the-...
http://www.braeunig.us/space/propuls.htm
-------------------
The 6 nonshared results are:
http://spaceengine.org/manual/making-addons/creating-a-ship/
https://phys.org/news/2018-11-ion-space-aircraft.html
https://www.meineke.com/blog/dual-exhausts-just-for-looks/
https://en.wikipedia.org/wiki/Merlin_(rocket_engine_family)
https://www.space.com/34192-spacex-raptor-rocket-test-first-...
http://spaceengineers.wikia.com/wiki/Thruster_mechanics
-----
I just tested this from various countries to ensure this was not biasing it. It wasn't. Bing gives localized results, such as German results when searching from Germany. DDG gives identical results from any country. So presumably everybody is getting these same results from DDG. Are you saying this is also exactly what you got from Bing?
That's one of the issues for me: those are your results, but there's no telling what anyone else might get when clicking on that link. "Feeds" of all kind, personalized by black boxes rather than explicit parameters we have access to, deprive us of a common (virtual) world to discuss.
My top 4 results were http://spaceengine.org/support-old/ , http://forum.spaceengine.org/viewtopic.php?t=33 , https://www.forbes.com/sites/startswithabang/2018/05/23/the-... and https://www.nasa.gov/returntoflight/system/system_SSME.html
I just kept reloading the same search through a bunch of different TOR identities + anti-finger printing, and it was like playing search roulette. And indeed there were, on occasion, actually some great results that don't show up elsewhere. It's such a shame that they're all masked behind some black box of tracking with another black box of ML and filtering. It's like two people going to the same library where librarian deciding to hide books from one person or the other because she, and her all seeing eye, thought they'd have less interest in them than other books.
> In fact, DuckDuckGo gets its results from over four hundred sources. These include hundreds of vertical sources delivering niche Instant Answers, DuckDuckBot (our crawler) and crowd-sourced sites (like Wikipedia, stored in our answer indexes). We also of course have more traditional links in the search results, which we also source from a variety of partners, including Oath (formerly Yahoo) and Bing.
In other words, all organic results are from Bing and Oath.
This is not a new thing. Historically speaking numerous "search engines" have actually used different engines under the hood. Yahoo hasn't been "Yahoo" in a long time, for instance, and I believe AltaVista moved from its own engine to somebody else's for a good long time before its demise.
(There's probably an interesting and nuanced discussion of the virtues and perils of a search engine provider depending on another company for something so critical. I hope DDG is investing significantly into their own engine, even if it's not ready-to-go.)
And it's a real change. I wouldn't even particularly care if DuckDuckGo did use Google, because the things about Google I find most objectionable would still be solved by proxying through DDG. Not 100%. The result might serve AMP pages, which is annoying, and not being a Silicon Valley liberal, I can clearly see how censorious Google is with its search results, or at least, how much they tilt the scales in their own political favor. But the primary issue, tracking me and the monetization thereto, would still be solved.
I'm sorry to be that guy but... citation needed. What evidence is there that this is anything but a fringe movement within the tech community let alone mainstream in any way, shape or form? And "common thing to see in HN posts lately" isn't evidence of that, sorry.
> DuckDuckGo reached 1 billion monthly searches
Two questions:
1. How much time do you think it takes for Google to reach 1B searches? and
2. How many DDG users routinely use !g (or otherwise use Google results)?
As much as HNers like to bang the drum about "privacy" the term is ill-defined. Take search because its straightforward. Your search results are personalized by a bunch of factors including, but not limited to:
- Previous searches
- Location
- Language
- Inferred or actual demographics
The fringe privacy element is I guess most concerned with previous searches? Or is it all of the above? And if you say that location is not OK, take a simple search for "bakery". Isn't it better UX to show local bakeries?
Another question: the alternative to "free" (ie ad-supported) models is user pays. How exactly does this work for users in the developing world for whom $5/month might be a significant amount of money? Will they value their "privacy" the same way?
The assumption (by DDGers and their ilk) that using [Big company products] is some Faustian bargain is hyperbolic and unsubstantiated.
No one targets the 99% percentile directly. It is much cheaper to go non-targeted advertising and hit as many people as possible perhaps targeting at the country/region or language level only.
http://gs.statcounter.com/search-engine-market-share
Google will report numbers later today. But have been growing at 20%+ the last 10 quarters.
Obviously Google has nothing to worry about with those type of growth numbers.
Versus Apple reported last week declines in both top and bottom lines.
You also add a potential weak point in that not only your e-mail account, but also your registrar/DNS accounts could be compromised.
https://framablog.org/wp-content/uploads/2017/09/carte-roman...