Blekko

#slider
#adsense

Blekko - Yahoo News Search Results

Blekko Shuts Down, Becomes Part Of IBM Watson

Back in 2010, an alternative search engine called Blekko emerged. It came from Rich Skrenta, co-founder & former CEO of Topix and NewHoo (which went on to become The Open Directory Project or DMOZ). It aimed to crowd source search …

IBM acquires technology from curated search engine Blekko to bolster Watson

Blekko, the curated search engine, has been quiet over the last two years. Until Friday, the company hadn’t tweeted since December 2013 and there hasn’t been much more activity on its Facebook page over that time either. On Friday, the … Continue reading →

The story behind the first computer viruses ever

When we think about computer viruses, one tends to think about Windows or perhaps cross-platform malware that comes from visiting questionable websites. But truth be told, computer viruses have a long and storied history, both on the PC and Apple side of the equation. To be fair, most of the earlier computer viruses weren’t terribly dangerous. If anything, they were more often than not proof-of ...

Delicious/tag/Blekko

recent bookmarks tagged Blekko

Buscador Blekko

Posted on 27 March 2015

Blekko Search Engine With Its Slashtag Feature Is Proving To Be A Hit With Users - StumbleUpon

Posted on 26 March 2015

Blekko provides a fair and hassle free search experience

Posted on 19 March 2015

Blekko Provides a Fair and Hassle Free Search Experience - StumbleUpon

Posted on 17 March 2015

Use the Blekko Toolbar for Enhanced Browsing Experience

Posted on 12 March 2015

Use the Blekko Toolbar for Enhanced Browsing Experience - StumbleUpon

Posted on 11 March 2015

How Effective Is Blekko - StumbleUpon

Posted on 25 February 2015

Blekko Search Engine Provides Comprehensive Protection from Spamdexing

Posted on 25 February 2015

Blekko Search Engine

Posted on 18 February 2015

Blekko Search Engine - StumbleUpon

Posted on 17 February 2015

Top Answers About Blekko on Quora

Top Answers About Blekko

How much would it cost to build a search engine?


Well it depends on the value, depth, and scope of that programming. If you look at Google, Yahoo, Bing, even lesser known ones such as Mamma; they all employ staffs and overhead that costs them hundreds of thousands of dollars per year to generate. You can start small by working on unique algorhthms and then trying to partner with larger companies to sponsor or add to that.

I agree with Greg Lindahl that a search engine doesn't need millions to be effective or competitive even, but it does need at least several full-timers on staff working around the clock to market the search feature, set it up as unique among competitors and even then how to get it noticed by consumers so that ou can generate sufficient ad/marketing revenue to be self-sufficient and hopefully, later, competitive.

I would work on a unique search function, and research the revenue models of (not just Google) but Mamma, SuperCrawler, MetaCrawler, Blekko, Ciul and study how they grew organically. Yahoo is trying to build itself as an entertainment portal, rather than search giant, (and this hasn't worked for them in the past if you look at Yahoo Pipes, their "network," Geocities and so on), but how they started could be valuable research as well.

See question on Quora

Posted on 2 February 2014

Can the same URL belong to multiple categories in Blekko?


Yes. The categories in our results correspond to blekko's slashtags, The same URL can be in multiple independent slashtags, and some slashtags are nested within others.

The "Top results" and "Latest" categories are special: Top results is just the top 20 overall results, similar to what google or bing would show, and latest is dated webpages.

If a good result is dated, it will appear in one or more categories in addition to being in "latest".

See question on Quora

Posted on 18 September 2013

Roughly what proportion of search queries on Google, etc include boolean operators?


From my experience most searches in retail catalogs are 3 or less terms, include no boolean operators (or complete sentences) and are focused on either exact titles or product lines.

Users don't enter boolean/SQL queries into search - they use it to either:
1. discover products based on a broad query, then filter by adding another term
2. look for exactly what they want

See question on Quora

Posted on 13 September 2013

Roughly what proportion of search queries on Google, etc include boolean operators?


In blekko's data:

  • 0.1% of queries use a minus sign
  • use of AND/OR/NOT seems to be a lot less than 0.1%

It's hard to measure the 2nd number because people seem to like typing AND and OR in all caps. Here are some example queries that don't look like they're really boolean queries:

  • Is Jesus God OR God's son?
  • fat 32 OR NTSF
  • xylitol good OR bad
  • You receive an Error 1606 error message when you try to install OR remove a Microsoft program

If you're wondering why search engines don't seem to care that much about boolean queries, this is why!

See question on Quora

Posted on 26 August 2013

What is it like to work on blekko's Core Relevance and Ranking team?


One of blekko's unique qualities as an upstart search engine is that we have our own multi-billion webpage crawl and index. This puts is in the league of big players like Google and Bing -- in fact the entire list of search engines that have their own big index is Blekko, Google, Bing, Yandex, and Baidu.

As a small company, our entire company is outnumbered several times over by Google's core relevance and ranking team. And we have to do things like maintain our own NoSQL database, crawler, infrastructure, etc. As you might imagine, this means that our relevance and ranking uses a lot of machine learning, and we also emphasize our human-generated curation of websites into topic categories. We handle about 5 million queries per day, which gives us a modest amount of click data to work with. At the end of the day, the main reason that we can be competitive at all is our use of human curation, which the other players don't want to do.

So, what's it like to work on the team? Compared to working on a much bigger team that's already done a lot more work, at blekko you would have a lot more opportunity to work on fundamental things that are already mostly solved elsewhere. And you have the opportunity to blaze a new trail with our human curation data, although we've already done a lot of work there.

See question on Quora

Posted on 6 August 2013

Does anyone use search engine other than google?


Why, yes. Not only does bing (and search engines based on bing like Yahoo) have significant market share, but there are startups like blekko, which gets 5 million searches per day. Our main strengths are dividing the results into categories, which often means you get your result without having to do follow-up searches, and a super-strong privacy and do-no-track policy.

See question on Quora

Posted on 9 June 2013

What if Search was a paid feature? What if Search Engines asked you money for every Search you made?


This will not happen because search advertising pays far more than anyone would be willing to directly pay. For example, an American making a query is worth an average of US$0.05 (5 cents) per query of ad revenue. That's a lot of money.

For a similar questions, see:



See question on Quora

Posted on 2 June 2013

How much would it cost to build a search engine?


The largest cost of running a search engine of Google/Bing/Yahoo scale is in provisioning many clusters capable of handling (tens of) thousands queries-per-second (qps) loads and returning answers in few hundred milliseconds.

The index has to be replicated because a single copy (replica) is not capable of handling such loads. There are no hard rules but if you have a single cluster handling few dozen qps reliably, then you are  doing (very) well. This assumes the index is not entirely in RAM, which is the case for Google and some of the others.

You would need hundreds of such clusters to handle query loads in thousands qps.

Building a single cluster would not be that much. Obviously, the following are rough estimates but they should give you an idea.

Take a 100TB index, at 10KB/page, of about 10B pages. If you do a good job of indexing standard page IR signals such as anchor text, title, headings. tf/idf etc. , together with decent link analysis and spam removal, you would have a world-class system, that would fit in, say, 9 machines at 12TB (e.g. four 3TB drives each).

Such a system could easily have 9*32 = 288GB RAM which would be plenty. A standard 1 GigE network should be sufficient. The cost would be around $25K, or less if you pick up a bunch of older machines from eBay.

Next major cost would be crawling.  You may want to check out How much would it cost to crawl 1 billion sites using rented AWS servers/bandwidth? to get an idea, but expect tens of thousands of dollars for crawling 10 billion pages. AWS would be a faster and easier way, but several times more expensive than setting up and running crawlers yourself.

I think the bare minimum of pulling it off would be on the order of $100K, not counting the time of people involved. Examples of such efforts include Gigablast and Duckduckgo.

Greg Lindahl mentions Blekko.com and Cuil, which are (were, for Cuil) fine examples, at higher operational levels, but still costing way below the big guys. Cuil, for instance, was much more disk-based

IMHO, the lowest cost solution would be a distributed search engine, such as my previous search startup, Wowd, or Faroo, YacY or Majestic-12. In such systems most of the cost is eliminated, by leveraging CPU, disk, RAM and bandwidth of users.

Consider a system of, say million users, the aggregate resources would be mind-boggling - 100TB RAM, 1 PB disk, 1Tbps bandwidth. But that is another story, I digress :)

See question on Quora

Posted on 17 October 2012

How did Cuil and Blekko build such large indexes of the web, despite being small startups?


Cuill started with 300 servers in 2006, then added another 1000 in 2007.

The crawl generated 100B pages, which when stored amounted to about 1P of data.

Each server had 12 500Gb disks, so the 1000 server cluster had 6P of disk space.

How could a small start up do this? Sell some equity at a large valuation then borrow money from debt lenders to buy the hardware.

See question on Quora

Posted on 7 October 2012

When thinking up a name for a new search engine how do you make the name so good so that people can say 'I'll just GOOGLE that'?


I think the premise of your question is flawed: People don't say "google that" over "bing that" because Google is a "better word."

I wasn't likely to say "google that" because the world "google" sounds like "goggles". I say "google it" because google is the definitive search engine. Over time "To Google" has become a verb form of a Proprietary Eponym, like Kleenex, Q-tip and Band-Aid.

Generally, if you were starting a company and wanted to become "verbed", I'd say pick  a nonsense or rarely-used word that sounds like a verb.

See question on Quora

Posted on 31 August 2012

Why don't search engines provide an option to sort results by date?


There are two huge problems for date sorting of search results.

The first is that date sort -- and I really mean date sort, not relevance sort of a date range like "past hour" or "june 3" -- means you only get 1 bit of relevance, where something is included or excluded. This means you may get a lot of spam. blekko plays some tricks here that work reasonably well, but sometimes the result is pretty bad.

The second is what date should be assigned to a webpage. The first date it was crawled? The date on the page? If a page changes slightly, does it get a new date? If a website puts the current date and time on every page, what do you do?

For this second issue, blekko tries to use an explicit date on the page, and doesn't show pages without explicit dates in date-sorted or date-range results. This is pretty good for news, blogs, forums, and press releases. When a newspaper puts its old archives online, we'll usually pick up the correct dates.

See question on Quora

Posted on 18 June 2012

Can Blekko survive even if it doesn't attract mainstream users beyond geeks?


It turns out that we have quite a lot of "mainstream" users: About 29% of non-navigational queries automatically invoke a slashtag, which gets rid of spam and boosts relevance. This is usable by everyone, not just people who are interested in doing something extra.

Try it out for yourself!

  • Cash back credit card: https://blekko.com/ws/cash+back+... -- other search engines send you to sales websites, while blekko sends you to sites that aren't selling you something.
  • Cure for headaches: https://blekko.com/ws/cure+for+h... -- other search engines send you to various dubious places, while blekko sends you to doctor-written websites.
  • Industrial design colleges: https://blekko.com/ws/industrial... -- other search engines send you to "lead gen" websites trying to make money referring you to actual colleges, while blekko sends you to actual college websites.

(This question was asked around our launch in Nov 2010, but I just got asked to answer, so I thought I'd update everyone with what actually happened.)

See question on Quora

Posted on 17 May 2012

In blekko's SEO data, what does it mean by "zapsite"?


"zapsite" links come from pages which we suspect are low-quality, i.e. pages on forum sites, user profile pages, directories of links, and so forth. These links are discounted when we compute ranks.

Our zapsite detector over-fires, but the damage to legit websites should be small. If you see a large number of zapsite links coming in to an URL, and you think they're legit, that's a bug worth reporting to support@blekko.com

See question on Quora

Posted on 9 May 2012

What are some web search APIs, now that Bing's will become commercial, and Google's is nonexistent?


blekko's free API allows the whole Internet to be searched. Contact us at apiauth@blekko.com for details. Given that there are only 4 search engines crawling the whole web (Google, Bing/Yahoo, blekko, and Yandex), we're pretty much it for free access.

Our slashtags feature allows interesting subsets of the Internet to be searched. For example, golf app can easily provide golf search and golf news using our API. For details of that, check out

https://blekko.com/tag/show for the list of good slashtags (or you can make your own)

and

http://help.blekko.com/index.php... for examples of advanced features around date sorting and slashtag combinations.

See question on Quora

Posted on 18 April 2012

How did Cuil and Blekko build such large indexes of the web, despite being small startups?


blekko bought a whole bunch of servers, wrote our own NoSQL database, and crawled and indexed billions of pages. It's a lot of work, which is why most search engine startups use someone else's crawl and index. (Currently, the only search engines crawling/indexing billions of pages globally are Google, Bing/Yahoo, blekko, and Yandex.)

When we launched in November 2010, we had 700 servers. We currently have 1,500.

If I remember correctly, Cuil raised a similar amount of $$ as blekko, and had around the same number of servers.

See question on Quora

Posted on 16 April 2012

What are the best vertical search engines?


Zvents (startup)' http://zvents.com is the local search engine used by most newspapers to power their event calendars.  With that much data coming in for such a difficult data set to aggregate, none compares for local.   No, it's not the sentiment or mobile based direction that's popular today but frankly, those innovations in local are built on a house of cards as the fundamental "Google" solution of a local search engine that is comprehensive, accurate, timely, and relevant is still largely unsolved.  Zvents' platform is light years ahead of the alternatives.

See question on Quora

Posted on 10 April 2012

What are the best vertical search engines?


https://blekko.com/ allows users to create their own vertical search engines (slashtags) from a list of websites, url prefixes, or patterns.

Here are some examples of slashtags that already exist, used alone and in combination with our /date slashtag that sorts results by date:



See question on Quora

Posted on 10 April 2012

Is there a way to tell a search engine to give you only travel- related links?


https://blekko.com/ has several slashtags -- curated search engines -- related to travel. Try them out:


subtags of /travel:


You can also sort by date, to see only recent news and blogs about Belize travel: https://blekko.com/ws/belize+/tr...

And if you've signed up on blekko with Facebook Connect, you can see Belize travel websites which are liked by your friends: https://blekko.com/ws/belize+/tr...

See question on Quora

Posted on 17 March 2012

How much would it cost to build a search engine?


Its critical to note that page index is hardly the measure by which one can claim to have a search engine. Search is more about relevance than about quantity (at least today). You may have the biggest index but if you don't know what goes on 1st position and what goes on 10th then you have a useless search engine.

See question on Quora

Posted on 13 March 2012

How much would it cost to build a search engine?


There are two schools of thought about this.

One is that you'll need billions of dollars, with the proof that all of the competitors who've tried to spend a mere $100M have eventually failed in the marketplace. As an example of analysis which elevates this track record into a firm conclusion, see http://www.fastcompany.com/17237...

The other is that you might need less than $100M, because of two examples:
  • Cuil built a 100 billion page index (similar in size to google's full index) for less than that, although Cuil subsequently failed for product reasons
  • blekko has built a multi-billion page index (similar in size to google's primary index) for less than that, and, arguably, the resulting product is interesting enough to survive.

I suppose we'll know the answer in a year or two!

See question on Quora

Posted on 8 March 2012

How many keyword combinations and link clicks do people try before finding what they are looking for on Google?


For the last few years I have seen Google search results as a joke. I am researching every day and almost every time that I use Google I have to sift through at least 100-150 results before finding what I want.

And the biggest annoyance is when typing in specifically what I am looking only to have Google deliver something different asking if I really meant what I asked for! Is that because Google makes so much money that it can afford to employ too many people with nothing better to do than find ways of screwing things up?

Note to Google: please give up trying to control the internet, give up on screwing around with search and wasting our time, and concentrate on polluting the world with yet another fast food franchise.

See question on Quora

Posted on 8 March 2012

Have the guys at Blekko considered partnering with the guys at CommonCrawl foundation?


I don't think we've ever talked directly to them about anything, no, although we do trade data with Factual, so it's not like the teams are strangers to each other.

The thing is, the way search engines crawl is different from the way other people crawl. We have separate crawls for fresh and non-fresh content. For non-fresh content, our limited index size means that we try hard to only crawl what we think is good content, using our search engine abilities to form opinions about the quality of pages we haven't crawled yet. To us, fetching the actual page is a trivial part of the process. Partnering with Common Crawl would neither increase our index size nor reduce our costs. That doesn't mean we think Common Crawl is doing anything wrong; we just aren't in the group of projects that finds working with them very interesting.

Update: On December 17, 2012, blekko donates search data to Common Crawl

See question on Quora

Posted on 18 January 2012

How often does Blekko update the index/data?


blekko's index updates at different rates for different parts -- the bulk of the data updates on a timescale of months, while "dated" material (e.g. blogs, news, press releases, forums) updates on a timescale of seconds.

The inbound link data is a combination of bulk data and dated material. You can sort the inbound links by date:

https://blekko.com/ws/disney.com...

EDIT: In mid-2012, the bulk data is now crawled on a 1 week to 1 month timescale.

See question on Quora

Posted on 26 October 2011

Who are the most influential Search experts that are not working at Google?


Key people in this space:


See question on Quora

Posted on 26 September 2011

How much of an effect do new search engines such as DuckDuckGo have on Google's stock price?


Zero. The marginal buyer doesn't expect Google to have serious competition for a long time. DDG will only start to have an effect on $GOOG if (when?) their market share goes from 1% to 10%. Since they're well below 1% for now, they aren't a factor for investors.

Even in technology, most investors in profitable companies are totally blind to cool stuff that could destroy them. This is a longstanding tradition, and it makes plenty of sense. The probability-adjusted outcome of DDG's existence is probably a .5% drop in Google's revenue (i.e. that's a sub-1% chance that DDG totally destroys their business). There is very little Google can do to adjust their company's direction such that the drop in this risk-adjusted loss is greater than the cost to Google. So their ideal decision is to round these risks down to zero, which will work until it doesn't.

And at worst, they can use DDG as a sort of laboratory for A/B tests that are too extreme for them to run. As DDG becomes more of a factor, Google will invest more in copying. And I'll concede that given the choice between DDG with Google's features, and DDG itself, I (and most people) would pick Google. That's not an indictment of DDG; it just means I'm using them because of their differentiating features.

See http://duckduckgo.com/blog/what-... for DDG's view of their competitive differentiation.

See question on Quora

Posted on 12 July 2011

How much does Blekko rely on public cloud (i.e. Amazon, Heroku) for providing service?


We host our own infrastructure. Our only use of external servers is a little bit of CDN usage.

The reasons we  host our own stuff:

  • we plan on getting big, and if you're big enough, Amazon isn't cost effective
  • Amazon won't rent us the ratio of cpu/memory/disk/ssd that we want. We would end up with a lot more cpu power than we need, and no SSDs, both of which would waste $$.


See question on Quora

Posted on 5 July 2011

Can I create a slashtag in Blekko for a lenguage?


At the moment, the websites we are crawling are mostly English in mostly English speaking countries. So while you can certainly search in any language, the results won't be that great unless it's English. We also have a long ways to go to understand synonyms and the treatment of accented characters in other languages.

Your /spanish slashtag is, of course, extremely welcome!

See question on Quora

Posted on 16 February 2011

Does Blekko.com have any chances of successfully competing with Google? If yes how?


I don't think Blekko is really competing against Google. Their approaches are different and their goals are different. They're in competition in the same way a hometown hamburger joint is in competition against McDonald's. They both sell hamburgers, but the burgers they sell don't taste the same and they aren't going after the same customers. Sure, you might eat at McDonald's one week and the hometown joint the next, but when you're on the run or strapped for cash McDonald's is your first choice When you want to enjoy a great burger with the family, the hometown burger joint is your first choice. They happily coexist.

Rich Skrenta, one of Blekko's founder, has stated that he is happy being the third search engine on the market. He isn't building Blekko to take down Google. It's more of a specialty search engine, for when you want to sit down and enjoy your... search. Over the years many businesses have been started with the goal of being the next Google or "The Google Killer" or the next Facebook. Very rarely do they succeed.

Some people might give up Google altogether for Blekko, and that's okay. I don't think that loss will impact Google substantially, if at all.

Here is what Rich Skrenta has to say, from http://www.webpronews.com/topnew... :

"The web's not getting smaller," he adds. "It's not stagnant."

"We think users can benefit from a third point of view on search," Skrenta continues. "Our goal is to be the third search engine," noting that search began with human curation, with the Yahoo Directory in the mid-nineties. Back then, it was basically, social search, he says. Back then it was more human-driven, but now, he says most new pages are not created by humans. They come from "spammers flooding the web with poor content."

"You don't want to search the world wide web for health," says Skrenta, implying that when you look to the web at large for results, the quality is going to be greatly diluted by spam and poor quality content

Read  what search expert Danny Sullivan has to say about Blekko at
http://searchengineland.com/blek...

In the end, I think Blekko will succeed and might even become profitable.

See question on Quora

Posted on 5 February 2011

Can a specialized search engine like Blekko compete with Google and be successful?


I think they will always struggle to get the uptake they are looking for, which is a shame because as you say the slashtags make for a different/slightly more controlled search experience.

To achieve success and corner off a legitimate chunk of the market (let's say 10%) I think they will need to do a couple of things:
1. Improve the functionality of things like the /spam tag as well as some of the other data (SEO, etc.).
2. Crawl sites more regularly/frequently. This is not an easy task but it seems their results/cached data is often quite out of date compared to the other major engines).
3. Do more teaching and spread the word more about their product.

For me, as an SEO, I think Blekko has loads of potential. I've not seen it driving any major amounts of traffic (yet) but again this should change over time. If Blekko can improve some of the technical aspects over time and are willing to invest quite heavily in advertising I definitely could see them being a formidable opponent. I don't think they'll be replacing/surpassing Google any time soon in terms of market share but I could see them cutting into Google's portion of the market to the tune of ~10% of all seraches?

This is just my opinion, I just think that Blekko is presently a bit hard for a new user to understand (and the explanatory video is kind of hidden on the homeapge)... and until users start actively disliking Google and the results they return I don't think there will be the impetus to change the way people search.

I love the idea and I hope Blekko can stick around but I wouldn't expect Google to be falling below 50% of the search market share within the next 5 years.

Looking forward to hear what others have to say and hoping I'm wrong!

See question on Quora

Posted on 5 February 2011

What innovations are focusing on dealing with the information and knowledge discovery?


The rise of social technologies is having a hefty impact on knowledge discovery – and the easy answer to this question would be Quora! But equally innovation does not necessarily mean a new product, it can mean a new way of using an existing tool to bring benefit.

Knowledge share has changed the face of learning – it is now not simply about formal or informal learning, we have been introduced to networked learning. Twitter is an obvious example – the emergence of twitter chats are great for 'meeting' those with a common interest and learning from their experience. What I soon became more aware of when participating in a chat was by taking part and sharing my learning experience I was also marketing my knowledge to the wider audience of my general followers. Brands have also realised the potential of harnessing a community’s collective intelligence, creative crowdsourcing is common – www.mofilm.com one of my favourite examples - but those like Kodak are even drawing on their networks to impact product design.
 
The danger of network learning is of course that we will never learn anything ‘new’ just expand on existing interests - and like minds are not great for innovation. The idea that we are constantly learning from our peer networks across social sites is being seized on by technologists and new collaborative platforms are evolving all the time - iTunes U is going to have a very interesting impact on Higher Education Qualifications. Which introduces a new question – how do we trust the source of information? Something we looked at in the media140 blog: http://insight.media140.com/en/2... .

See question on Quora

Posted on 6 January 2011

What are some good queries to test search engines with?


Comparing relevancy, coverage, span, freshness (real time and news search) and latency I use common terms (such as 'financial report', 'Obama'), uncommon terms (such as small company names, people names, complex queries), ambiguous terms(such as 'Bush', 'Magic'),terms in the news, and terms from different categories (Sports, Music, etc.).  .  Also test more structured queries such as 'Weather in...', 'Flight from ... to ...', 'Population of ...' and local searches such as 'Museums in...', 'Best restaurants in...'

See question on Quora

Posted on 15 December 2010

I use Yahoo Site Explorer but it's limited to 1000 results. Anyone know of a site that shows a complete backlink list?


There are several sources that show backlink data, and all have their strengths and weaknesses. Here are my personal opinions:
  • Yahoo! Site Explorer - http://siteexplorer.search.yahoo... - only shows up to 1,000 links to any given page or domain, but has a fairly comprehensive index that's most similar to Google's (right now) in breadth, depth and freshness.
  • Yahoo! Web Search - using www.yahoo.co.in or another regional Yahoo! not yet powered by Bing, you can run more detailed queries like linkdomain:mysite.com inurl:blog (which shows pages that link to your site that have "blog" in the URL). With this technique and diligence, you could conceivably extract all the links Yahoo! knows about.
  • Majestic SEO - http://majesticseo.com - a web crawl built to power a startup search engine in the UK, Majestic's index is exceptionally large, but not very fresh. One can find data from months or even years ago in the current version, and many of the linking URLs often no longer exist or no longer link to the target (we've run tests on this and see ~20-40% of the reported links still existing and being accurate - other SEOs have reported both better and worse numbers). Majestic has a good reputation, a strong core of users and a great API, though. They're also consistently improving, so these numbers may get better soon. Majestic also provides some basic metrics about the site's and linking pages (showing followed vs. nofollowed and number of links/linking root domains), which can be valuable for sorting.
  • Blekko - http://blekko.com - a new search engine from Rich Skrenta (previously of Topix), Blekko's index is still relatively smaller today than any of the others on this list, but they're growing and fresher than Linkscape & MajesticSEO (usually). You can see link data in chunks by using their /seo searches.
  • Linkscape (and Open Site Explorer) - http://www.opensiteexplorer.org - SEOmoz's web crawl (I'll try to be as unbiased as possible), which crawls 2-3 weeks out of every month and processes data over the following 1-1.5 weeks (producing a new index every 4ish weeks). Linkscape tends to be quite fresh when a new index is released and contains between 40-50 billion web pages (in our testing, showing about 40-60% of the pages in Google/Yahoo/Bing's indices). The key advantage we're trying to build, in addition to relative freshness and size, is metrics - showing how important a page/site is in the eyes of search engines using imitative and predictive metrics.
  • Google Webmaster Tools - http://google.com/webmasters - good list of links for your site, and many more than 1,000 but without any metrics and no competitive data.
  • Bing Webmaster Tools - http://bing.com/webmasters - as with Google's WM tools, they provide a good list of link data for your site (if you verify/register), but no competitive link information.
The only point I'd add is that while Google/Bing/Yahoo! are somewhat less innovative, Majestic, Blekko and Likscape are all working hard to grow index sizes/freshness/quality/usefulness. I know we're for example, investing 4 full time engineers over the next 3-4 months on making Linkscape considerably faster, fresher, more comprehensive and more predictive with the metrics. Majestic is likely doing similar, and Blekko is growing their engine (and thus data collection) quickly.

All of the other tools you'll find on the web are powered by one or a combination of these raw sources. Some of those tools have great features on top of these raw indices/APIs, but I'm not aware of any others that independently crawl and build a search-engine style index for link data.

See question on Quora

Posted on 12 November 2010

reddit.com: search results

Is DuckDucKGo The Future of Search or Why Unpersonalized Search Grows in Popularity

Highly personalized search is one of the key services that Google takes pride in offering to its users. Even though Yahoo and Bing are lagging behind, they try to follow Google’s lead in improving their custom search results too. The constant improvement of the personalized search has definitely facilitated countless people. However, it seems that a significant number of online users has decided against giving away their personal data and have opted for a less intrusive web search alternative.

What is personalized search?

Personalized search presents us with results that are tailored not only to our search terms but to our specific character as deduced by the search engine. Thus each person faces different information suggestions depending on whether he conducts a search on his own or on his friend’s PC. Basically the personalized search presents you with:

  • A list of recommended websites that save from answering to your query would also try to be consistent with your past searches and website selections.
  • A list of recommend local venues that would adhere to yours and your friends’ online reviews.
  • Local specific data based on your current location, etc.

What’s wrong with that?

Personalized search is often perceived as a true facilitator, however one should not forget that usually every coin has two sides – the same rule applies here. The challenges that personalized search poses before us include:

  • Being trapped in an artificial ‘filter bubble’ where the “intelligent” search engine of our choice decides what we need to see. Under the personalized search our choices are limited to the familiar sources that we have previously demonstrated interest in by recommending, sharing, or simply clicking and selecting them before the rest. Thus in the sea of information that is available on the web we remain blind to all the alternatives, we are stuck with the choices that the engine thinks relevant to our taste, interests, character. We live in our own filter bubble as Eli Pariser explains, but what is the most unsettling of it all – we don’t even realize it.
  • Being an easy target to advertisers Monitoring and keeping a record of our search history, the search engines make us a target of web ads that follow us everywhere on the web. Having compiled all the personal data like purchasing records, search queries, frequently visited websites, preferred news stories or even time spent on the site and pages skimmed through during your visit, the search engines offer a valuable resource to advertisers, who pay to be able to use it and snowball you with “targeted” ads.
  • Your search history is used for legal purposes Having a personal profile drafted at the given search engine your search history is kept on record and in case the government or a legal practitioner requires it, the search engine will immediately render the information.

How Is Unpersonalized Search Different?

  • You are not presented with a biased choice of articles that your friends have recommended.
  • Your IP is not being tracked, neither is your search history.
  • Your profile is not sold, because you simply do not have a profile when using unpersonalized search.
  • 3rd party advertisers and other parties are not able to build profiles about you in order to serve targeted ads, so your SERPs are not flooded with commercials.
  • Your Searches cannot be legally requested.

Why unpersonalized search engine like DuckDuckGo is quickly gaining popularity?

As Greg Kumparak explains DDG is still small to compare to a search engine giant like Google with its modest 1 billion searches for the year 2013. In order to make a simple comparison he draws our attention to Google’s searches for 2012 – they amount to “3.2 billion searches, or roughly 3X all of DuckDuckGo’s annual traffic, each day.” Nevertheless, DDG’s popularity marks an amazing rise for the past year and its users grow by the day. It attracts investments and partners that believe in the respect for privacy that the engine promotes by not using tracking cookies and not saving a record of its users’ IPs.

DuckDuckGo’s CEO and founder Gabriel Weinberg has openly stated that he has no interest in keeping or even collecting its visitors’ search history. Instead the search engine monetizes its services the same way Google does (showing ads triggered by the search term used) but without tracking its users.

Interesting result of this anti personalized search algorithm is the uniformity of results available to all of DDG’s users. Thus a specific search term triggers the same set of results for everyone regardless of search history and supposed interests. This provides unbiased information rendering that avoids the so-called filter bubble scenario.

Being more of a search engine aggregator DDG has a unique strategy for selecting its information sources. Unlike Google, DDG deletes the “search results for companies he believes are content mills”. Thus, while in Google’s top search results one often recognizes articles from eHow, DDG would never recommend them, as their thousands of daily posts submitted by paid freelancers are characterized as “…low-quality content designed specifically to rank highly in Google’s search index.”

The search mechanism of DDG actually aggregates the crawled results of several search engines like Bing, Yahoo, Yandex, Blekko, WolframAlpha, and many others. And while DDG does not invest time and efforts in web crawling itself, it focuses on its users’ privacy, positive experience, spam free results, and its own instant answers. It gathers the results from various sources, choosing the best vertical search engine on a given topic, while also reassembling and matching the results with the ones rendered by its private crawler –DuckDuckBot, and a number of crowd-sourced sites like Wikipedia.

DDG might just be the next step in the search industry and if it is early to be labeled as the future of search, it surely has a modern note to it that addresses the latest hot topic – about one’s right to privacy.

Nowadays the attempt to protect one’s personal information has lead to a lot of breakthroughs in technology. Take for instance how our currency has been changing throughout the recent years the modern virtual payment options like bitcoins promote lower transaction costs, immunity to inflation, and above all – anonymity. Privacy in web search is yet another rising need of the online users that alternative search engines like DuckDuckGo have proven to be able to address.

What do you guys think???

Source: http://optilocal.org/unpersonalized-search-grows-in-popularity-is-duckduckgo-the-future-of-search/

submitted by OptiLocal to marketing
[link] [3 comments]

Posted on 31 March 2014

Why Unpersonalized Search Grows in Popularity – Is DuckDucKGo The Future of Search

Highly personalized search is one of the key services that Google takes pride in offering to its users. Even though Yahoo and Bing are lagging behind, they try to follow Google’s lead in improving their custom search results too. The constant improvement of the personalized search has definitely facilitated countless people. However, it seems that a significant number of online users has decided against giving away their personal data and have opted for a less intrusive web search alternative.

What is personalized search?

Personalized search presents us with results that are tailored not only to our search terms but to our specific character as deduced by the search engine. Thus each person faces different information suggestions depending on whether he conducts a search on his own or on his friend’s PC. Basically the personalized search presents you with:

  • A list of recommended websites that save from answering to your query would also try to be consistent with your past searches and website selections.
  • A list of recommend local venues that would adhere to yours and your friends’ online reviews.
  • Local specific data based on your current location, etc.

What’s wrong with that?

Personalized search is often perceived as a true facilitator, however one should not forget that usually every coin has two sides – the same rule applies here. The challenges that personalized search poses before us include:

  • Being trapped in an artificial ‘filter bubble’ where the “intelligent” search engine of our choice decides what we need to see. Under the personalized search our choices are limited to the familiar sources that we have previously demonstrated interest in by recommending, sharing, or simply clicking and selecting them before the rest. Thus in the sea of information that is available on the web we remain blind to all the alternatives, we are stuck with the choices that the engine thinks relevant to our taste, interests, character. We live in our own filter bubble as Eli Pariser explains, but what is the most unsettling of it all – we don’t even realize it.
  • Being an easy target to advertisers Monitoring and keeping a record of our search history, the search engines make us a target of web ads that follow us everywhere on the web. Having compiled all the personal data like purchasing records, search queries, frequently visited websites, preferred news stories or even time spent on the site and pages skimmed through during your visit, the search engines offer a valuable resource to advertisers, who pay to be able to use it and snowball you with “targeted” ads.
  • Your search history is used for legal purposes Having a personal profile drafted at the given search engine your search history is kept on record and in case the government or a legal practitioner requires it, the search engine will immediately render the information.

How Is Unpersonalized Search Different?

  • You are not presented with a biased choice of articles that your friends have recommended.
  • Your IP is not being tracked, neither is your search history.
  • Your profile is not sold, because you simply do not have a profile when using unpersonalized search.
  • 3rd party advertisers and other parties are not able to build profiles about you in order to serve targeted ads, so your SERPs are not flooded with commercials.
  • Your Searches cannot be legally requested.

Why unpersonalized search engine like DuckDuckGo is quickly gaining popularity?

As Greg Kumparak explains DDG is still small to compare to a search engine giant like Google with its modest 1 billion searches for the year 2013. In order to make a simple comparison he draws our attention to Google’s searches for 2012 – they amount to “3.2 billion searches, or roughly 3X all of DuckDuckGo’s annual traffic, each day.” Nevertheless, DDG’s popularity marks an amazing rise for the past year and its users grow by the day. It attracts investments and partners that believe in the respect for privacy that the engine promotes by not using tracking cookies and not saving a record of its users’ IPs.

DuckDuckGo’s CEO and founder Gabriel Weinberg has openly stated that he has no interest in keeping or even collecting its visitors’ search history. Instead the search engine monetizes its services the same way Google does (showing ads triggered by the search term used) but without tracking its users.

Interesting result of this anti personalized search algorithm is the uniformity of results available to all of DDG’s users. Thus a specific search term triggers the same set of results for everyone regardless of search history and supposed interests. This provides unbiased information rendering that avoids the so-called filter bubble scenario.

Being more of a search engine aggregator DDG has a unique strategy for selecting its information sources. Unlike Google, DDG deletes the “search results for companies he believes are content mills”. Thus, while in Google’s top search results one often recognizes articles from eHow, DDG would never recommend them, as their thousands of daily posts submitted by paid freelancers are characterized as “…low-quality content designed specifically to rank highly in Google’s search index.”

The search mechanism of DDG actually aggregates the crawled results of several search engines like Bing, Yahoo, Yandex, Blekko, WolframAlpha, and many others. And while DDG does not invest time and efforts in web crawling itself, it focuses on its users’ privacy, positive experience, spam free results, and its own instant answers. It gathers the results from various sources, choosing the best vertical search engine on a given topic, while also reassembling and matching the results with the ones rendered by its private crawler –DuckDuckBot, and a number of crowd-sourced sites like Wikipedia.

DDG might just be the next step in the search industry and if it is early to be labeled as the future of search, it surely has a modern note to it that addresses the latest hot topic – about one’s right to privacy.

Nowadays the attempt to protect one’s personal information has lead to a lot of breakthroughs in technology. Take for instance how our currency has been changing throughout the recent years the modern virtual payment options like bitcoins promote lower transaction costs, immunity to inflation, and above all – anonymity. Privacy in web search is yet another rising need of the online users that alternative search engines like DuckDuckGo have proven to be able to address.

What do you guys think???

Source: http://optilocal.org/unpersonalized-search-grows-in-popularity-is-duckduckgo-the-future-of-search/

submitted by OptiLocal to SEO
[link] [9 comments]

Posted on 31 March 2014

Trying to find a photo

I've been googling around and haven't found it. I've even tried searching on blekko and duckduckgo... nothing

There's a picture with two scenarios, side by side: one where a woman has a gun and another where she does not. On the picture with the gun, she's defending herself with it. On the picture with out the gun, she has her arms up, trying to protect herself from getting beat-up. Has anyone seen this?

submitted by happycrabeatsthefish to progun
[link] [3 comments]

Posted on 25 February 2014

Help understanding a significant data discrepancy between Web Server logs & Analytics

Hey there,

I hope some of you can help with understanding a significant data discrepancy (29%) in comparing our SEO visits obtained from varnish logs vs Analytics Organic visits.

We expect there to be a sizable difference in both sets of data, as we don't know - the inner workings of analytics - our differences in dealing with pre requests, - what Google consider to be search engines - how analytics manage bot traffic & traffic that has a high percentage chance of being bot traffic

The Varnish logs are huge (millions of visits per day). So we had to initially filter these to remove known & probable bot traffic. Detecting bots is pretty difficult (though we have a long list to work from), so this is most likely one of the significant sources of extra traffic. We can assume, it's unlikely make up the full 29%.

We factored in session id's too, so that only one visit is counted per session whilst filtering out those with just one page request as these are likely to be crawlers. The next step was to identify other verticals, such as paid, mail, social, referral & server requests (particularly from email).

We then categorized referral sources we considered to be SEO:

%bahdal.com%' %baidu.com%' %blekko.com%' %duckduckgo.com%' %bablyon.com%' %snapdo.com%' %avfirm.com%' %peekyou.com%' %yandex.ru%' %search.conduit.com%' %searchmobileonline.com%' %isearch.avg.com%' %static.flipora.com%' %search.mywebsearch.com%' %google%' %bing%' %yahoo%'

(there are a few more)

Then subtracted sources which are known to not be Organic traffic:

%accounts.google.com%' %mail %' %googleads%' %/uds/afs%' %/aclk%' %cse?%' %ogskillid%' %signedparameters=%' %/vetjob?jobid%' %picture=size90%' %ogaction%' %emaillanding_invitecontact%' %medium=facebook%' %ask.com%'

(I've removed a few for the purposes of this report)

submitted by Cocopoppyhead to SEO
[link] [9 comments]

Posted on 28 November 2013

Stephen King's Pee Fetish and the Google Coverup

idiosyncopatic: original reddit link


I have read many (but not all) Stephen King books, and I have noticed a rather strange thing. It seems like in most all of his books at least one character pees. Now, most times it's an involuntary reaction out of fear, but there are also several normal peeing scenes. Now, this tends to stick in your head. I get it: losing conrol of your bladder is a vivid way to show terror; the fear must be intense to lose control of such a basic primal function. I do not have each reference listed, but I am thinking about rereading his books and making a refernce table. Please forgive my lack of sources for the moment. This sounds strange, but it's so frequent (pun intended) that when it DOESN'T happen, it's surprising. I have told a few people about this theory, only to get some crazy looks, but I since started reading the Shining and got very excited when on page pg 32 (kindle edition) Danny "smell[s] his own urine as he [voids] hemself in the extremity of his terror." I had never read this book before! But I was able to guess based on past experience. Now you explain to me how Google and Bing searches have turned up nothing. Rather strange, considering that (as Reddit tends to prove) no idea is original. How could it possibly be that NO ONE ELSE has noticed this? Simple: Stephen King is a rich man. He or his publishing company has paid off the search engines. How else could this happen. To test my theory I used an alternative search engine (blekko.com) THE FIRST RESULT is a forum entry on peesearch.com. here is the link to the discussion: http://www.peesearch.net/community/forums/archive/index.php/t-33725.html

SAY WHATT?????

You may cry "safesearch" results, but I am an adult woman and I don't need no stinking safesearch. That option is turned OFF in my settings.

What do you think? Mere coincidence by an overly-excited far? OR COVERUP?

If anyone is interested I will add some screenshots for proof.


Discourse level: 28%

Shills: 0%

submitted by funnymanisi to conspiro
[link] [7 comments]

Posted on 26 October 2013

Stephen King's Pee Fetish and the Google Coverup

I have read many (but not all) Stephen King books, and I have noticed a rather strange thing. It seems like in most all of his books at least one character pees. Now, most times it's an involuntary reaction out of fear, but there are also several normal peeing scenes. Now, this tends to stick in your head. I get it: losing conrol of your bladder is a vivid way to show terror; the fear must be intense to lose control of such a basic primal function. I do not have each reference listed, but I am thinking about rereading his books and making a refernce table. Please forgive my lack of sources for the moment. This sounds strange, but it's so frequent (pun intended) that when it DOESN'T happen, it's surprising. I have told a few people about this theory, only to get some crazy looks, but I since started reading the Shining and got very excited when on page pg 32 (kindle edition) Danny "smell[s] his own urine as he [voids] hemself in the extremity of his terror." I had never read this book before! But I was able to guess based on past experience. Now you explain to me how Google and Bing searches have turned up nothing. Rather strange, considering that (as Reddit tends to prove) no idea is original. How could it possibly be that NO ONE ELSE has noticed this? Simple: Stephen King is a rich man. He or his publishing company has paid off the search engines. How else could this happen. To test my theory I used an alternative search engine (blekko.com) THE FIRST RESULT is a forum entry on peesearch.com. here is the link to the discussion: http://www.peesearch.net/community/forums/archive/index.php/t-33725.html

SAY WHATT?????

You may cry "safesearch" results, but I am an adult woman and I don't need no stinking safesearch. That option is turned OFF in my settings.

What do you think? Mere coincidence by an overly-excited far? OR COVERUP?

If anyone is interested I will add some screenshots for proof.

submitted by idiosyncopatic to conspiracy
[link] [6 comments]

Posted on 26 October 2013

Ad-aware acting like adware- forced blekko search

Just got a new laptop for my father in law. So was getting the normal protection when noticed that Ad-aware changed my home page and keyword URL search goes through lavasoft.blekko.com . In firefox tried to change keyword URL but ever restart moves it back to the lavasoft default. Uninstalled Ad-aware and what ever fancy search bars, but it still forces me to use blekko. Why is this program acting like the programs it should be protecting us from?

submitted by chuloreddit to techsupport
[link] [1 comment]

Posted on 26 February 2013

Adaware free antiviris changed my default search engine even when I said "no" to toolbar installation.

Just a heads up... I received a prompt a few days ago to update the free version of adaware's antiviris software.

During the install, I made sure to untick the install extra toolbars and software buttons, yet when I tried to do a google search this morning, my browser's default search had been changed to blekko.

It's time for a new antiviris program.

submitted by anagrama to software
[link] [3 comments]

Posted on 29 November 2012

I think my PC might be infected

Hello,

There are several odd processes running on my PC. They have no description or username data. When I click "properties" or "open file location," nothing pops up. These are,

csrss.exe winlogon.exe atieclxx.exe 

screenshot

I realize that the former two are typical windows files and that the third is related to my ATI card, but I haven't seen these running before - or perhaps I have, but with a description and user data. I've heard that some types of malware pose as these processes.

Recently, I've noticed huge spikes in CPU usage (often reaching 100% and constant use of upwards of 50% of my computer's memory (although I have 3-4 Gb RAM, and my only memory intensive programs are Chrome / Skype / Steam). When I put my computer in sleep mode and turn it on again, sometimes the machine will turn on, the LEDs on my mouse/keyboard will turn on, the monitor remains unresponsive and the CPU light doesn't flicker. I forced shutdown once, and received a BSOD. When I rebooted, I didn't see a typical login screen - after the boot screen disappeared, the screen remained dark for 45 seconds, and suddenly showed my desktop as it was when I put my computer in sleep mode.

I've run scans in Malwarebytes, AVG and Microsoft Security Essentials in safe mode and found nothing in AVG/MSE. However, Malwarebytes found some spyware, which was deleted. However, I still notice these spikes in CPU / Memory usage, and the three processes are still listed and running. After a few searches, this seemed similar, but I haven't noticed any spam, except for Google Chrome changing my search engine from Google to Blekko -- however, this may have been a default process of one of the anti-malware softwares I've downloaded. But I am not entirely faithful that that's the case.

If you have any tips / suggestions / instructions / words of advice, I'm all ears. Thank you in advance for your help.

submitted by HideousInfant to techsupport
[link] [5 comments]

Posted on 12 February 2012