How about the same for the Software Boutique (cursor in searchbox by default)? What are your plans for adding Startpage and Ixquick? Unfortunately the ISPs at my present location don’t allow me to use Google and DuckDuckGo. How else could I use the beautiful new startpage then? 
Learnt the trick for that. That’ll land in the next Welcome update. 
Currently undecided. 
From a technical point of view, these may make the searches more privacy invasive then visiting the page directly via HTTPS.
- Google/DuckDuckGo send queries using GET (exposed in the URL, ISPs/snoopers could read this)
- Startpage/lxquick use POST by default (nobody can see the query in the URL).
Startpage/lxquick does support GET queries too, as far as I can see, not POST. If privacy is the main concern, you may wish to re-consider whether you’d want us to add this.
If were to decide against adding it, there could be a modified version of the page that you could use locally in your browser. 
I don’t use start pages. They never really fit in my routine and my browser always opens to a blank page, so do my tabs. Can’t recall either the last time I hit the so-called Home button. Must have been a decade. It’s one of the things I remove from the toolbar. Best of luck and fun in this endeavor. But it’s not for me
I’m here for another reason, because of what @lah7 just hinted and the notion that is spreading on this thread that ixquick/startpage are better for privacy than duckduckgo. And why, despite not having plans to use this, I do feel that adding ixquick/startpage is probably not worthed unless it comes without pain.
ixquick offers no guarantee of privacy beyond that which they state. And you as a user can’t even inspect the url being sent to their servers without help of other tools. This is the reason I never used them, despite not being happy with DuckDuckGo privacy as of late. I get the idea behind POST. But a search engine should only be responsible for the security within its controlled space. Not downstream. GET with HTTPS is the only acceptable way to do web searches.
More, the company is removing ixquick in favor of startpage after a merger. startpage is to my knowledge not built with the same principles that forces it to fully respect the EU privacy act. What was true of ixquick, is not of startpage. ixquick is the only one of the two that operates fully in compliance with the EU privacy standards. But they are slowly discontinuing it and suggesting folks to move to startpage.
The sense in this? None. Suspicious? Very. Especially when we know virtually nothing of the dutch company that acquired ixquick some years ago. A “Holding” company, that by dutch laws seems to mean this is a family own company. And that no one knows nothing about.
DuckDuckGo is no saint either. They fully betrayed users when they partnered with Canonical and force Ubuntu users with a canonical referral that is actually very hard to remove. Despite nothing of value being sent to canonical, just the necessary for a profit split, this is completely contrary to what we were promised years ago when this project started. And that we are still being promised today on the home page. But they use GET. And that to me, means I don’t have to check every other week if DuckDuckGo is still a decent company.
3 Likes
Ultimately when you’re a big webpage you need to do… certain things to receive any financial backing, even if it means stabbing your users in the back.
Not saying that UM’s homepage will someday do this, I believe @Wimpy won’t stoop down that low, but if someone is hosting a search page and need space for either more servers or more financing to access other search servers (because most likely search server owners providing that service would charge website owners for being a front-end to access them) then often people in dire need will take the low way out if it keeps them able a bit longer.
We are not planning to try amd monetize the Ubuntu MATE start page. This is primarily an effort to complete the Ubuntu MATE branding and experience throughout the system.
4 Likes
Edit: I understood you wrong, sorry. Removed half the post.
For the referral, I don’t really see how the fact that my query is incrementing a counter of all the ubuntu users’s hits means they are tracking me.
Cheers
It doesn’t. I think I sad that explicitly.
However DuckDuckGo partnership set a precedent that was unexpected in the context of what this search engine advertised itself to be.
It’s really not an issue at this point. But I’ve been online since the only thing around was BBS. And I’ve seen over and over the slow but assured pattern of a decaying service. It starts with a blip, a little “nothing”, just like its nothing to increment a partner’s counter by implementing a url referral. And after that, by small increments of little nothings the process towards a bad service moves on.
And it’s always about good intentions. We are assured. But you see, “we are giving this for free you know and it does cost money. donations have been slow… etc” starts to show up in a way of apology once some people start to realize something is changed. Apologists will come rushing defending, others will rush attacking. But it should be obvious to everyone the process has started and there is no turning back. Like they say, Hell is paved with good intentions. But what users don’t realize and what these service providers don’t want to accept is that really, if you set to offer something for free and it comes to a point you can’t afford it, then please stop. Don’t do it anymore. It’s ok, really.
Anyways, that came out in the way of an anecdote. But make no mistake, when things like this happen, experience has taught me that it’s time to stay alert.
I didn’t know about these differences. Could you explain the difference between POST and GET and how can I manually check what method a search engine is using?
Why would POST be worse from a privacy standpoint? I think the confusion for me here is that I associate “exposed in the URL” to possible privacy violations by ISPs?
In summary:
- GET exposes the search query in the URL; can be easily integrated, plus links can be shared.
- POST is like bits of data embedded into the request, trickier to integrate, which is why these 2 “privacy” engines use it by default.
See a comparison here of both types of requests: http://www.w3schools.com/tags/ref_httpmethods.asp
I’m no security expert, so I’m not 100% sure how this plays when accessing over a HTTPS connection, but there might be a “hand over” period between Ubuntu MATE’s server passing the request to the search engine’s server, which may expose something. So if privacy is the primary concern, I suggest using the search engines directly since you’re guaranteed the “privacy” they’re offering.
I don’t imagine these engines will be added, @marfig raises good points about them and I personally wouldn’t use either.
1 Like
Thanks for pointing this out! How would you evaluate Oscobo and Qwant then? I don’t know how to find out whether they use GET or POST.
Today is the first time I heard of them, @maro. Wouldn’t know what to say.
Of course that the POST/GET discussion is just a detail of the whole privacy issue (and not a very interesting one, despite me being the one who started it). There’s a lot more to it, and to understand or gain enough suspicion that a certain service may be hiding an agenda from you, you have to go deeper, looking even into the company(ies) behind it, their history and the individuals responsible. For instance, on the matter of ixquick, even ignoring the whole strange behavior of late, how can one trust their claims if we cannot find any information about the company behind it?
I would have to look into both these services. But right now I do have a couple of things to say about both:
Oscobo is a UK business. With the Brexit thing happening, soon it will have to forego any EU protection conceded by the Privacy Act that was approved this year by the EU council, and instead submit entirely to the UK law; And we know from recent news, the government is putting some pressure on anti-privacy legislation and entertaining the idea of security backdoors and whatnot.
Also,
Oscobo About Us claims that it was
founded by two guys who left their corporate jobs to work towards a
belief they both shared – that personal data should remain just that,
personal.
Disillusioned with the corporate world and that the private individual
is paying for what seems to be “free” internet content, they felt it was
time to turn the tables the other way round.
I don’t know about you, but I’m not sensitive to romantic notions of people who leave their jobs to work for the common good. I’m hard to impress. Especially by vague claims.
Without knowing these “two guys” names, who they are, what were they doing and how are they feeding themselves and paying their medical bills, I don’t trust statements like the above. I actually find them boring and slightly insulting of my intelligence. And it is really not just about calling it a lie; We genuinely should know if these “two guys” actually have the level of expertize that allows them to make claims of privacy based on such a hard technology to control and secure as is the Internet.
As for Qwant… it has a log in thing that supposedly gives you more features or somesuch. Immediately disqualified. It’s a search engine and should not be another place to store a darn password/username/email. If those features are so important, why not give them without a login? And if that is because the login helps ensure some sustenance to the business, then don’t make claims of privacy because you are already violating mine with your users database that only you control and that you will never allow me to visit and inspect its contents.
2 Likes
That’s a very good point. It has become all too common to advertise products through story-telling following a supposedly romantic conversion experience.
I am glad you started it. Your insights are not only extremely helpful but of great educational value and I hope you will stay in this community to share your knowledge and experience. Thanks for looking into this search engine material and participating in this thread despite the fact that you don’t use start pages.
2 Likes