Haha wow - sorry, but if you think you can create a "better Google" by limiting the scope of searches to specific sites that you've selected ahead of time, you're missing the point of the this entire decade.
So, first I thought "Oh no, not another DuckduckGo!" Then the article's first line had Google misspelt as 'Gogle' and then there were a few paragraphs with too many biased assumptions and with more of marketing slogans revolved around the importance of '<i>you</i>'..and so I thought it's a social network with a search engine integrated. But then, I thought "hey, isn't that what Google already does?"<p>And then I clicked the link, it's just a semi-polished front-end, with serif and sans-serif fonts mixed, that just builds-up queries for ONE particular search engine (Bing). Atleast, if there was an option to choose between any of the top 3 Search engine's, I would have been slightly impressed.<p>Honestly, this is a disappointment and the title is misleading.<p>It solves a very little problem. Anyway, I don't want this to be a 'hate-only' comment, so, I'll share what I know with the community, so far:<p>(Replace news.ycombinator.com with the site you want to search for)<p>1) If you want to search a specific site, then type in :<p><pre><code> site:news.ycombinator.com
</code></pre>
2) If you want to exclude keywords, then type in:<p><pre><code> site:news.ycombinator.com -mongodb
</code></pre>
This will exclude all results containing 'mongodb'<p>3) If you are searching for a specific string within the results, then type:<p><pre><code> site:news.ycombinator.com riak
</code></pre>
This will explicitly highlight the word riak in your results and also include only results that matches this string.<p>You could also search for a particular term in double quotes to be included with your result:<p><pre><code> site:news.ycombinator.com "paul graham"
</code></pre>
Hope this helps..
This post is everything wrong with valley culture distilled into a few ugly paragraphs, followed by an ugly site that redirects to Bing, which the paragraphs had just harangued.
They created a frontend to Bing, and a way to group site restrictions.<p>A more honest title would be "I built a better frontend to Google (by which I mean Bing)".
"Take a look at NeatSeek, and prepare to be amazed." You are right: I'm amazed by the amount of self deception and wishful thinking in this article. You missed the point so hard you hit yourself.
This site takes the problem of the search filter bubble[1] and pushes it to the extreme. It might be useful for people who want to avoid, at any cost, political opinions that they disagree with. However, if you're trying to solve a hard problem that you don't know the answer to, is it reasonable to assume that you <i>do</i> happen to know what tiny subset of the web contains your unknown answer?<p>[1] <a href="https://en.wikipedia.org/wiki/Filter_bubble" rel="nofollow">https://en.wikipedia.org/wiki/Filter_bubble</a>
A suggestion - As it stands right now, your title and blog is link-baity as it belittles the work of engineers at Google. Re-title your submission <i>contextual search</i> because you're searching within a silo or context of websites. The problem you're trying to solve is along the lines of millionshort.com in that you are trying to lessen the impact of over-optimization of results.<p>As for an explanation of the down-votes you're getting, I can only refer you to:
<a href="http://news.ycombinator.com/item?id=4962544" rel="nofollow">http://news.ycombinator.com/item?id=4962544</a>
blekko allows you to do this with their feature to create your own slashtags.<p>for unscatter.com I took the approach of using other sites apis but the caveat being I never intended it to compete with Google or bing or the like as it's results are based on recency rather than the relevancy.<p>in one incarnation of unscatter.com I did something similar to what you did except I used boss. one thing you should consider is there is a limit on how many characters are in a get request and you will hit it fast embedding all those site: declarations.<p>good luck.
Although i don't personally think that this project (nor the way it was presented here) is worth much, we have to agree that we are now in a place where limiting search results is relevant and filtering content will be more and more important as years go by and the information continues to increase at this rate.
Search programming:
c libraries<p><a href="http://www.bing.com/search?q=c%20libraries%20%28%20+site:php.net%20OR%20+site:java.sun.com%20OR%20+site:python.org%20OR%20+site:ruby-lang.org%20%29" rel="nofollow">http://www.bing.com/search?q=c%20libraries%20%28%20+site:php...</a><p>good job.<p>edit: All things considered. I think this blog might be a joke.