DuckDuckGo implements HTTPS Everywhere | Hacker News Reader

DuckDuckGo implements HTTPS Everywhere

gabrielweinberg.com

100 pointsecaron33 comments

"rubbish. I thought maybe it would "quack" whilst searching, but it does nothing (other than find what I'm looking for)."

My wife's reaction to duckduckgo.

Just wanted to get it out there for posterity :)

I think the name is the only thing that is wrong with DDG.

Another thing, I personally don't feel quite right about using a search engine that relies quite heavily on other search engines to give results. From what I understand DDG is not a search engine in the same sense Google or Bing is a search engine (pls correct me if I am wrong).

Certainly ranking of returned results is a hard problem as the extraction/indexing/spider of web data?

I don't have a problem with DDG being a sort of meta search engine. Just give me the most relevant results first.

The only rational reason to use DDG over Google is that for some reason you are on the extreme end of the paranoia towards Google or any other search engine. That itself is not much of a good argument as there are number of things you can do to make sure that Google doesn't track your search results.

From my point of view there are no compelling reason to use DDG over Bing or Google, the only reason DDG is a HN darling because we like to root for the small guy; lets see it for what it is.

Personally I would love to see a search engine that really gives Google and Bing a run for the money and disrupts the search engine market. Unfortunately DDG doesn't do that.

I'd love to know what you think might disrupt as I'm always looking for good ideas, but know that is what I'm trying to do. Our about page tries to explain why you should use it: http://duckduckgo.com/about.html

The most visible things to me are the 0-click info and the less spam. Then there are lots of little things that start to add up, many listed on the goodies page: http://duckduckgo.com/goodies.html

I have some ideas...

I put a note on my task list to make sure that I email you (from the feedback page) on some of the things I believe might put you apart from others and also better promote your existing features (some of which I just saw for the first time). I need some time to sit down and give it a go for few hours and put down my fragmented ideas in a proper way.

Hopefully tomorrow.

Oh wow, thx so much. Looking forward to it!

Just replying to my own post to mention that I didn't get around to it yet... No excuses really. Just being lazy. Hopefully soon :)

I came across these new search mashups today:

http://kickdash.com (instant, with autosuggest, in categories, but no general search)

http://www.liquidjelly.co.uk/supersearch/ (client side merging of bing+boss+google into 1 page)

http://instantlyweb.com/ (i'm feeling lucky frame + instant)

One thing that might be helpful is to have a video showing off some of the "cool" features. I would think that this would at least get more non-techies to try it out.

The only rational reason to use DDG over Google is that for some reason you are on the extreme end of the paranoia towards Google or any other search engine.

That's a pretty bold statement. I have gained concrete benefit from switching to DDG full-time, mostly from the 0-click info but also from generally improved search results.

But no, I guess you're the "rational" one, and my experience is not valid.

When did I say your experience is not valid? If you feel 0-click is must have feature and DDG search is better than Google search good for you.

The text you quoted was my opinion, was it not obvious enough?

Statements that begin with the phrase "The only rational reason to..." usually are not intended as expressions of opinion, but rather as statements about how there is only one rational rational reason to do such-and-such, and this is it.

I tend to only need to use google when i know what I'm looking for but forgot something about it or its something I want answered. Error codes seem rather hard to search for on DDG.

Has anyone compiled a good error code database yet? Perhaps I can integrate it.

Not really a DB, but for Microsoft I have used http://www.eventid.net/ quite a bit. I wouldn't say it is great, but I have found it helpful as a starting point.

Thx for the reference; that's a start!

I use DDG for the simple reason that it gives me results in English all the time. I travel A LOT and Google is always trying to guess what language they should serve me my results in. If I'm sitting at a customer's site in Germany I'm not interested in German results. I also might not be able to login to Google on this computer so it gets my preferences.

It might sound like a silly reason but it's a big deal for me.

It's a hybrid engine. I do my own crawling and indexing in addition to relying on external APIs. For the APIs I use, I've built a lot of relevancy and other logic on top of them so they often don't look the same as if you'd query them directly.

I started out exclusively with my own stuff, and then started using BOSS when it came out so I could focus on what I think are my value-adds. Now I use a bunch of stuff (including my own). My approach is to just be a better search engine. If an API does something awesome, e.g. WolframAlpha or embed.ly, I figure why not use it.

Do you think in the future you might exclusively use your own index and not be depended on others?

Let's say hypothetically tomorrow, all the search API you rely on, decides to cut you off, what then? How good is your own index by itself?

Just to be clear, I am not dissing on you, it takes a lot of gutso to take on search engine market alone, I applaud you for that, I don't see myself having that much guts to do it. I am trying to be realist about the meta-hybrid-search engine.

I treat the deep Web search APIs as essentially commodities of the link graph. I was able to seamlessly switch between BOSS and Bing without anyone noticing. And there are several others I can use as well. So I don't feel much danger here. And now that BOSS is going to charge, I feel even less danger because it is turning into a real product.

My index is not usable in this fashion. I've re-purposed it mainly for 0-click and for spam removal.

Frankly though, I don't see why this comes up so much. Infospace is a public company that relies on other indexes pretty much exclusively. In other verticals, people routinely use more exclusive platforms, e.g. Twitter or Facebook or banking stuff.

In reply to the name comment I love it! It is different, fun, and super easy to remember. I can definitely see how it might not jive with some people though, to each their own.

I tried DDG for a week as my search engine. The search result quality was definitely on par with Google, the two biggest factors for switching back :

1) Google's "Did you mean?" appears to be a bit smarter and corrected more of my obscure searches than DDG. Wish I had recorded a couple of concrete examples to pass on, but they escape me at the moment.

2) Speed. Google returns results faster (at least on the days I was using DDG). Apparently my brain is use to sub-second search results because anything longer than that can feel like an eternity.

Thx. Recently I've made a lot of improvements to #1. Of course, specific examples are always super helpful to improve it further.

#2 varies a lot depending on location and network. Unfortunately I still don't have a lot of geographically distributed servers.

For me at least, #2 seems most often to have something to do with fetching the results themselves, rather than the network between me and DDG delaying the initial pageload. The results page will load very quickly (the header/sidebar, sometimes the zero-click results or disambiguation results), but then there's a very noticeable delay before the regular results get AJAX-loaded.

(It is true that the the better DDG is at getting me the zero-click kinds of results, the less of a problem this is.)

Have to agree on the first point, Google doesn't have all that datastore for nothing.

However, that didn't stop me from switching. DDG is my default search engine, and while it's a little slower than Google, that difference isn't significant. Do note however, that I'm using Chrome, and found the speed difference to be a little annoying sometimes on Firefox.

That said, I love the bang syntax. If your search fails in DDG, just add !g and you're immediately in Google. I've been using it for about a month now, and slowly gets a hang of what type of searches are better done in Google vs DDG (eg: shopping -> Google, documentation -> DDG, discussions (groups/forums or otherwise) -> Google, reviews (products,etc) -> DDG)

Google's "Did you mean?" has gotten increasingly annoying for me as they evolved it. It's now at the point where it will silently, automatically return results for a different word than what I searched for if it thinks I misspelled it (aka it isn't a common term).

This is infuriating 99% of the time, and I'm at the point where I just automatically enclose most of my single-word searches in quotes to begin with.

DDG is currently my main search engine. I've tried it for a few days a couple times in the past, but now I'm trying to give it more time - at least a couple weeks - to see if it grows enough on me to replace Google.

I'm finding myself using the "!g" command a lot to double-check results on Google, and I'm definitely noticing the speed difference with Google. But overall, I'd say that it's almost on par with Google for my searches.

Some things that would make me like it more:

-A bit more speed.

-A way in the settings to decide between using Yahoo News or Google News (that link in the sidebar on the right), same thing for images (google images or bing images) and maps.

-Some way to highlight the most likely or most popular result when the disambiguation page is very long (maybe just subtly point out which result is clicked on more often?).

-A way to personalize the DDG homepage with a couple of personal links (for those who use it as a browser homepage -- you could add a link to GMail or whatever, so that it even more seamlessly replaces Google).

I can see some pros and cons to all these suggestions (except more speed), but I figured I'd just throw them out there.

Thx for the ideas--some really good stuff in here. Some quick responses to your 1-4.

#1 always working on more speed--it's been a priority and will continue to be so.

#2 planning on ditching that sidebar. Check out http://duck.co/topic/new-ddg-header -- would love your thoughts.

#3 this is definitely needed and in process. It's half implemented now but is hit or miss for sure.

#4 many people have requested different variations of this, e.g. as far as iGoogle, but a few links would be cool.

#2 I might be missing something, but is there a way to do a quick image or news or map search with that new format without the sidebar? If there is, I couldn't find it quickly.

Yup, the new format (idea) would have a drop down right next to the search button where you would be able to select maps or images (just not in that prototype). You could still of course also precede or end your search with !m or !i for maps or images. What do you think?

DuckDuckGo is absolutely awesome. The only annoyance is that it completely sucks at non-english searches, but for programming related stuff it's really cool.

I continue to be impressed both by DDG's search results and by its dedication to privacy and security. Seems like a great way to celebrate their 2nd anniversary.

Reply on news.ycombinator.com