DuckDuckGo Cooks Google's Goose
thestreet.com
thestreet.com
- Somebody is trying to improve the search engine space which is too long dominated by just one party (which admittedly makes a damn good job)
What I don't like:
- I don't see any improvements
- Don't like their name
- Crappy logo
- Too long domain name (with some skills and little money, a great 4-letter dot com is a no brainer)
- The product is awkward to use: the entire UI is the #1 reason to leave and aesthetically not pleasing, too many colors (red, blue, green), no CI, favicons are distracting, screen real estate isn't used effectively, mouse over effects are distracting and feel like clumsy Web 2.0 sites in 2005 => even Bing's new interface is very much nicer
- Search results got better but seem to lack any relevance algorithm such as Google's Pagerank
- Shortcuts are nice but more some kind of an gimmick than real revolution and they are not helpful at all because people who do not use DDG regularly forget them: either they have to be displayed all the time somewhere next to the results or directly on the landing page or they have to be consistent and expressive as Google's syntax which is "site:<site_to_be_searched> query"—if this shortcuts are one of DDG's main features then they have to tell it.
- No autocomplete, no instant search
I think DDG's and the team main problem is a heavy lack of product vision—they just think search is a great business model (yes it is) and they want to do something to just be better than the market leader. And this 'something' is their biggest weakness, they just don't know what to improve, on what to focus. Google's search is so good, that it's gonna hard to improve but still there are some fields where Google can be beaten (anonymous search, irrelevant results for longer queries because of SEO gaming/blackhat). Instead they just brought up a really ugly design and the user wonders all the 15 seconds he's exploring DDG why he should use DDG.
DDG has to pick one of Google's weaknesses and focus just on this and they have also to position DDG accordingly and communicate this focus/USP heavily. Otherwise DDG is like G+ to FB, just a bad copy.
I can image how hard it must be to build a search engine from scratch, there's just too much tech involved and crawling all the web could take months even if you have hundreds of servers (imagine: Google crawls even smaller sites with more than 20,000 pages per day (!), how does anyone want to build such crawling power with little VC money and some hype posts from Fred Wilson??). But maybe that's the chance: a new player has to focus on something small—a feature, a content niche, whatever, just a MVP that is in its core much, much better than Google. Building and promising the entire Google search experience will lead to fail.
They do have ddg.gg as a short alias to the full domain, though that doesn't address the .com suggestion at all. For that, they also have dukgo.com which I thought was another alias, but it appears that's being used for a community platform now.
I think it would have made more sense to reverse those two domains and have dukgo.com point to the search engine itself, and leave the more esoteric ddg.gg domain to the community platform. People invested in the platform would have an easier time remembering the "weird" domain than the general public (i.e. the folks who they need to use the search engine) would.
You can also adjust the layout of your results: alignment, page width, font and font size, underlining of links, etc.
Clearly they've provided a multitude of options, you just need to try them out.
Also, they do allow site search using the same syntax as Google.
The length of the url isn't that big of a deal to me. I just type 'du' and the first url autosuggested is duckduckgo.
Yes, but just in their logo, that's ok because the rest of Google stands behind the expressive logo, is clear, has lots of white space, decent colors etc.
Now go to DDG: read header, blue search button on result page and green search button on landing, what the heck?? Strange colors in the text results and this ugly duck, sorry but getting a really nice duck on freelancer.com or 99design from a talented artist is a job of 500 USD.
> The length of the url isn't that big of a deal to me. I just type 'du' and the first url autosuggested is duckduckgo.
No, it's a big deal even if your browser provides autocomplete, The domain is everything: their brand, their name and will used anywhere and all the time. Just saying "check duckduckgo for searching this and that" or "i duckduckgoed my name" to your friend or colleague is too much communication overhead, too many syllables for something which is too important in our lifes nowadays.
And those both points bug me, because if they haven't done their basic homework on the frontend side who tells me that they done them on the backend side where it is more important (relevance algorithm)—why should I trust DDG's search results if they are not able to setup a proper brand communcation and interface?
I don't have any further disagreements.
somehow related, i like that they include links to google below their search results.
I definitely appreciate the effort that has been put into privacy and simplicity. The zero-click box has come in handy at times (though I often find myself clicking through the box anyway, in which case it's a glorified top result, and that's okay with me). My most used feature by far, though, has been the bang syntax, to reference the Java API or search HN or check Stack Overflow or pull up a Google map or explore Wikipedia. I think the bang syntax experience might be frustrating for people who navigate to DDG and then type in a bang query... but if you've got DDG set as your browser's default search, the bang syntax is a nice timesaver: open a new tab (autofocus into the URL bar), type in your site-targeted query, hit Enter, done. I save one page load every time I do a site-specific search, compared to what my experience would be on any other search engine, and considering the proportion of my browser time spent in searches, I think that's a decent amount of time saved. [I mean, there are people on HN who quibble over how many keystrokes they can save in their terminal aliases, after all!]
That doesn't mean I think DDG is perfect; my major annoyance is results from content farms (same as any major search engine), but I'm hard pressed to find a search engine that weeds out content farms without sacrificing the features I like from DDG. The name doesn't roll as trippingly off the tongue as some other sites, but is it really that much worse than any of the other search engines (or names of YC companies, for that matter) and does the name affect my ability to use the site?
As you hyperbolically express your disgust towards DDG, please remember that you don't get any points in life for hating a product that you aren't forced to use and that other people find helpful.
The comments are declining though, I admit that.
We all talk about scaling our companies, or scaling our technical systems, but when it comes to scaling a social system, are we really just going to throw in the towel and point fingers like we're in junior high?
We can do better than that. It's a hard problem, but isn't that what we're here for? Come up with a solution and quit whining.
It's "a bad name." It isn't evocative, it can't be (easily) made into a verb, and it lacks brevity.
I don't really like the name, but I'm oddly comforted that I can shorten it to ddg (and use http://ddg.gg). I can deal for now, but I do think it should change.
You know, there is a company called "Yahoo!" which did OK for a while.
I switched back to Google after trying to search for maps and local results. One of the really important features that Google offers me is Google Maps. Not having that kills DuckDuckGo for me, even if their normal search results were to work just as well.
If I'm going to do "map! place name" in order to get map results, I might as well just use Google; DDG isn't adding anything. Google is a place where I can type "thing I'm looking for", and get, most of the time, the answer I'm looking for, while with DDG that's just not true.
In areas where ddg isn't quite "there," they offer the !bang syntax.
!gm {query}or !gmaps {query} will redirect you to Google maps.
At first, I had a lot of reservations about switching and found myself using g! quite often. Then I realized that I wasn't necessary more impressed with Google results, I was just __used__ to it. For instance, I hated DDG's zero-click info sources, because I would mentally ignore it. It seemed like a waste of space. Now it's the first thing I look at when the search results show up -- and it's generally what I want.
There's a saying--- if you aren't buying a product, you are the product. So they have one paid link per page. Is that enough to make them profitable eventually?
I believe the idea of integrating and presenting results from various specialized sources has a bright future in search, but DDG doesn't handle it very well right now. It might be in the future Google will leverage the same logic to improve its search experience. Google's "Knowledge graph" seems to me to be a step in that direction.
And when Google does it, chances are that it will outperform years of DDG's effort in mere weeks.
Is that relevant to your evaluation of a search engine? I don't care how many hits the engine shows, just what it shows in the first 3 links or so. The rest is noise.
There, fixed that for you.
Not that there isn't enormous added value in presentation or that duckduckgo has a poor offering at all - but your search result quality will always be limited to that of your partner-competitors' (Bing index in this case). And the more you succeed, the more expensive your partnership grows: see Netflix.
Wait a minute. I just remembered the other added value: privacy. Now that I think about it, I know a few middlemen/middlewomen that I'd describe enthusiastically myself!
Taking the Bing API, building a frontend, a parser for few shortcuts and get some VCs to blog about me?
Sounds like a project from a 48h hackathon.
Keep in mind quite a few people have done this before, bitcircle, 4hoursearch etc... spring to mind but DDG has actually done some innovations in the space and is still around. The fact that people know it will still be around next year encourages you to at least play with it.
Really, why did anyone ever switch from Yahoo? It wasn't for the results, Google provided the search results for Yahoo for awhile. Many people, myself included, switched to Google for the clean interface.
Now I've switched to DDG not just for the results (which are 95% "just as good") but for the privacy, the variety (I like the idea of competing services), and some of the interesting things DDG is doing with providing information directly.
As I've said before, search is an intensely personal tool. Don't use a tool if you don't like it, but recognize that other people might like a tool even if you think it's terrible.
The ONLY thing DuckDuckGo can do that Google won't is "privacy" (maybe "less ads" is another thing). While that might be reason enough for some people to switch, it's barely going to make a dent in Google's dominance.
In the AOL case, while IP addresses were stripped, users were identified with unique keys so you could see individual users' search sequences. That is not the case here.
Use the tools that make you most productive. For me, that's DDG. For you, maybe not. But I think "...there is no reason to switch to them" just isn't accurate in general.
I finally admitted to myself that I was wasting too much time having to run so many searches twice and I switched back to Google.
I'm not particularly disapointed with google search, and if I try something I don't like, I won't try it again unless a friend tells me it's changed a lot.
That business is clearly one-shot, either v1 is good or you're going to have to fight uphill against a bad rep of your own making.
By quickly, do you mean in less time (search speed is more), or in fewer clicks (search quality is more / relevant links are above)?
I suspect we all tend to search for the same types of things over and over. Ruby developers search for Ruby docs and modules, stamp collectors search for auctions and information about stamps, etc.
So I fully understand that my experience won't be shared by everyone.
[1] indicates DDG uses third-party APIs for deep web searches. To me that sounds like how Inktomi was a "search engine API" and neither Inktomi nor Yahoo saw Google coming [2].
Even if DDG doesn't succeed, Google is now at the top of the hill and therefore is _the_ target for all competitors.
[1] http://help.duckduckgo.com/customer/portal/articles/216399-s...
[2] http://diegobasch.com/a-relevant-tale-how-google-killed-inkt... and HN discussion http://news.ycombinator.com/item?id=3924609
Ads can be a major pain, and any business that requires them is probably going to fail, intolerance to ads is building up as ads are more and more agressive.
Switching to DDG because of ads wouldn't make much sense, they'll get their own ads when they have to pay for servers, too.
Anyway, my point is that 1% of the search market is about half a billion dollars in revenue.
It'll be interesting to see if Gabriel keeps up with the affiliate model or moves toward an in-house ad solution.
Bing is 20% of search market yet last I checked they are nowhere near that amount in revenue
How so?
DDG offers privacy, something that Google and other big search engines do not. That is more than enough reason- especially with growing privacy concerns in the US.