.... so my project for the weekend is to make a firefox extension that resets your search default every X hours/days.
Being able to search on almost any website by adding a simple keyword is a real plus and I will go even further and say that sometimes it accelerates the search process (say you know you want to search on wikipedia for instance).
However, I have to admit that most of the times I end up searching on google :)
http://help.duckduckgo.com/customer/portal/articles/216441-f...
https://encrypted.google.com/search?hl=en&q=%s
DDG definitely doesn't have the same quality results as Google, but it's close enough in most circumstances for me that the times I have to search twice aren't a huge factor.
You're suggesting people that already stated a distrust in Google do a thorough network traffic analysis after every update, so that they know their privacy is violated after the fact? That's a bit ridiculous, really.
Unless you turn off some 10 or more options, it literally phones back your entire online activity, in or out of incognito mode, in or out of any google service.
It's slightly different, but you can do almost the same thing with Keywords in the Firefox location bar, and they presumably could support the bang syntax (or whatever) as well.
Sysadmin: !apache !awesome !bsd !chef !codeweavers !cookbook !cve !datasheet !distro !distrowatch !eventid !everymac !exex !exploitdb !filext !fsf !fsfe !funtoo !gdiag !gnomebugs !gnu !howtoforge (!htf) !ip !irp !iso !linode !linuxfr !linuxhcl !linuxmint !lxr !mysql !nginxwiki !openbsd !openbsdmisc !opensuse !oracle !osvdb !parabola !peppermintos !playterm !postgresql !psql !puppet !qth !salix !salixforums !sectube !senderbase !spice !splunk !sqlite !susebug !susepkg !technetlib !timestamp !tldp !trisquel !use !vmkb !vol !wog !zahe
Sysadmin (debian): !debbug (!dbugs) !debian !debianforums !debianfr !debml !debmla !debwiki !dfiles !dman !dpackages (!dpkg !dpts) !dtag !fanch
Sysadmin (FreeBSD): !bsdman !fports !freebsd !freshports !fxr
Sysadmin (Gentoo): !gbugs !gentoo !gentoowiki !gpackages !gwiki
Sysadmin (RedHat): !rbugs
Sysadmin (Solaris): !sunpatch
Sysadmin (Ubuntu): !askubuntu !docubu !docubufr !omgubuntu (!omgu) !ppa !puc !ubottu !ubuntuforums (!ubuntuf) !ubuntuusers !uman !upackages !upkg !uu !uude
"If a file isn’t from a known source, Chrome sends the URL and IP of the host and other meta data, such as the file’s hash and binary size, to Google."
[1] http://blog.chromium.org/2012/01/all-about-safe-browsing.htm...
[2] http://superuser.com/questions/387724/how-to-disable-downloa...
Examples taken from some stackoverflow posts:
http://blekko.com/#?q=Exception%20caught%3A%20java.io.FileNo...
http://blekko.com/#?q=No%20value%20for%20parameter
http://blekko.com/#?q=logger.error('Opps%20we%20got%20an%20e...
By all means, try and replicate this complete list in your Chrome browser: https://duckduckgo.com/bang.html. I'll wait...
What makes the DDG !bang system awesome is its enormity. You don't look up what !bang to use to search something. You assume it exists, do it, and 99% of the time, you're right. That's a game changer.
I don't write Perl, but I know !cpan exists. I don't even have to look. I don't even know exactly what site it's going to search. But I know if I !cpan twitter, I'm going to be looking at a list of Perl libraries for tweeting.
That's the DDG killer feature. You want to do some kind of site-specific search or lookup, you just type in the most logical sounding !bang and assume it will work. You're simply never going to replicate that inside your browser.
Distributed search, similar to bittorrent, DHT-based designs and the like are notoriously difficult.
I've participated in such efforts, like the Seeks Project [1], Yacy [2], and related initiatives like Unhosted [3], and it takes a certain amount of dedication (and suffering ;) ).
However, I believe it is not entirely impossible that we see a true alternative sometimes. From what experience, what is needed is a slightly better set of distributed algorithms, a business model with the ability to sustain such a technical effort, and a range of features that no search engine can yet offer (because centralized).
[1] http://www.seeks-project.info/ [2] http://www.yacy.net/ [3] https://www.unhosted.org/
I've switched to duckduckgo about a month ago, and indeed I'm regularly making search on google because duckduckgo just can't provide the results I want as well (especially when it comes to non english search). But it's way less a pain that it seems : I just type "g<tab>", and I'm using google. And I still have the high privacy advantage of duckduckgo for most of my searches where the duck is enough.
The bonus of that is that I now frequently use "gi<tab>" to search on github or "y<tab>" to search on youtube.
More complex queries are typically done with Wolfram|Alpha. This goes a long way: https://duckduckgo.com/?q=graph+y+%3D+x%2Bcos^2%28sqrt%2842*...
So does it? Or is it safe only because NSA hasn't bothered to go to DDG yet, or do whatever they are doing to get Google's search data (cable splitters or whatever)?
I've had DDG as my default search engine. It's been a bit rocky. I've left it in place out of privacy preferences, and used it occasionally, but had actually started getting used to the google search method above.
I'm consciously reverting to using DDG more (it's getting better). Why? Because my privacy and rights do matter to me. I've still got Google literally a keystroke away (or !g from any DDG search, or !sp for StartPage's private proxied Google search).
And DDG's TTY mode is pretty awesome: https://duckduckgo.com/tty/
But if there's anything really untoward going on and people learn about it, it's going to look extremely bad for Google.
While I wouldn't say that's impossible, it's not a high-likelihood event in my book.
Of course, you're free to use alternative browsers.
Just ... keep in mind that if you're using a WebKit browser you're trusting Apple, and if it's WebCore, you're still trusting Google.
Mozilla's Gecko is still independent, however.
I'll switch between methods though.
Yacy is more advanced, uses a DHT, but does not really work for daily use, the accuracy being too low.
Unhosted is distributed architecture on top of the Web really. I've mentioned it because it may be useful and/or an inspiration to whoever interested in these matters.
Lets also add that Chrome!=Chromium. Chrome is based on Chromium. Not all of it is open and it is not compiled in the public eye.
No thanks.
So they will really need something more compelling to get the average Joe (who make up the vast majority mentioned earlier), who always get his results from Google's first page, to switch from Google to DDG.
Yet, in the context of search engines, I don't see anything that could be more compelling than always getting the most relevant search results.
> 3. Enormity has been in frequent and continuous use in the sense “immensity” since the 18th century: The enormity of the task was overwhelming. Some hold that enormousness is the correct word in that sense and that enormity can only mean “outrageousness” or “atrociousness”: The enormity of his offenses appalled the public.
Yes, until the NSA requests that DDG allow them to see all searches in real time. You could argue that it says "court ordered" but no one here has any clue what type of pressure the government used to get PRISM through at a major publicly traded company or what they are capable of with a small operation like DDG.
[added] as pointed out below, the NSA could and will just save the searches which would result DDG being just as insecure as google.
I am a little confused by your comment - has the word "enormous" come to mean "evil"? If so, maybe this new came about from people referring to the "enormity of a crime". It seems to have followed a similar etymological evolution to the word "gross" as in "gross depravity". But both words just mean "big".
Most of which can be addressed. There's also a UUID in each build. Privoxy can remove that AFAIU. I don't use Chrome myself, but Chromium.
As I said: you're arguing in the absence of evidence on a pretty widely available and analyzed application. Rather than simply throwing out hypotheticals, a more interesting exercise would be to do some research of posted information to see if there are known issues. Might even make for an interesting HN submission.
I might do that myself.
It's not a general synonym for "enormous" or "having vast size", except by confusion.
I filed a bug about poor integration of DDG with w3m (a console-mode browser) based on default placement of the search button in the tab order. I received a response within the day pointing me at the lite interface: https://duckduckgo.com/lite
... which works perfectly.
That said: these do tend to be somewhat hidden, and I didn't find them through the "Goodies" or "Settings" DDG pages just now. Hrm.
http://www.boomerangbooks.com.au/Enormous-Turnip/Katie-Dayne...
Obviously, the children's story uses "enormous" to imply that the turnip is very large, not very wicked. Somehow "the dastardly rutabaga" or "the heinous swede" makes a lot less sense in this context.
Edit: Admission - I used DDG (my default search engine), not google.
Making DDG my default search is the best thing I've done for improving my browser search capabilities. Coming from Opera, I'm already in the habit of using a great many custom search engines, which is nice if you're in Opera, but then Chrome feels very empty. Making DDG the default search gives me just about all of my custom search engines (and more), with pretty intuitive keywords, without having to configure and sync every browser on my system.
ever since ~2-3 years ago, Google is only very rarely my first stop in searching, I so much prefer to go directly where I want.
biggest part of the reason was their second-guessing my query and their results deviating more and more from a simple AND-query over my keywords. it felt like someone replaced my filet knife with a butter knife. (the other part was their instant search result updating and other "enhancements" were super distracting to me).
the only times when I go straight to Google (by which I mean prefix !g to my query), is for a few typical "local/Dutch" or "commercial/consumer" searches, you get a feel for what Google would answer better. Those types of queries are exactly what Google has been optimizing for the past few years, at the great cost of their general and precise web searching quality.
(PS, tip for the Dutch HNers: '!wnl' searches the Dutch Wikipedia. No you don't need it often but when you do, it's nice to have it quickly. I couldn't find it first because I always had it keyed to 'wn' myself, which sends DDG to a weather search site or something)
If it was just my desktop, and for some reason it only had one browser, okay. Customization is always cool. But I use too many different computers with too many different browsers in too many different locations that DDG's !bangs makes this really convenient and portable, getting the fast browsing workflow I'm used to in just a few clicks.
