Web Directories

Wushe Zhiyang

Wushe Zhiyang

Yandex, Russia’s leading search engine which is thought to be mulling over an IPO for up to $1.5 billion, is rolling out a new feature today that aims to make its search experience seem a lot more intelligent.

Dubbed “Spectrum” and claiming to be able to read users’ minds, it uses what sounds like a combination of semantic technology and machine learning to “infer implicit queries and return matching search results.” In other words, Spectrum is able to make better sense of the meaning of searches based on its own classification system.

It’s based on what Yandex describes as query statistics:

The system analyses users’ searches and identifies objects like personal names, films or cars. Each object is then classified into one or more categories, e.g. ‘film’, ‘car’, ‘medicine’. For each category there is a range of search intents. [For example] the ‘product’ category will have search intents such as buy something or read customer reviews.

So we have a degree of natural language processing, taxonomy, all tied into “intent”, which sounds like a very good recipe for highly efficient advertising.

But what if a search query has many potential meanings? Yandex says that Spectrum is able to choose the category and the range of potential user intents for each query to match a user’s expectations as close as possible. It does this by looking at historic search patterns. If the majority of users searching for “gone with the wind” expect to find a film, the majority of search results will be about the film, not the book.

“As users’ interests and intents tend to change, the system performs query analysis several times a week”, says Yandex. This amounts to Spectrum analysing about five billion search queries.

Earlier this month we reported on how Yandex was also getting smarter through partnering with VKontakte, which is the largest social network in Russia. Under the arrangement, the public-facing elements of VKontakte user profiles will show up in Yandex searches in realtime, essentially creating a people search engine since results, where publicly available, will link to and/or display a person’s date of birth, place of birth, university or place of work.


We’re frequently told Brits don’t care a fig-leaf for online privacy. But one London-based startup is about to test that theory — it’s just launched an anonymous search engine, called Oscobo, initially serving up search results specifically for the U.K. market. (Although the intention is to scale the model to other European markets in time too.)

The founders are starting with the U.K. because they reckon Brits do care about not being snooped on online — certainly once they are made aware of how much tracking is being done in the background by dominant search engines like Google. And if offered an easy to use alternative, which is where they’re hoping Oscobo will come in.

Think of it as a DuckDuckGo that serves up U.K.-specific results by default…


One of the two co-founders, Fred Cornell, used to work at Yahoo, so has seen the evolution of the search and online ad industry up close. “I worked for Yahoo for over 12 years and I really like Yahoo but I saw first hand how the industry tracks users, uses personal data, keeps pushing harder to make even more money from harvesting more of the data,” he tells TechCrunch.

“Search engines, ad exchanges, advertisers, publishers, data traders, everyone’s at it. And I became uncomfortable with the whole scene and decided I wanted to provide an alternative and a better deal for users who are concerned with online privacy.”

Go back more than a decade and Cornell argues there used to be a fair ‘social contract’ in place between the web user and the online advertisers and publishers. But in recent years he says that balance — i.e. of looking at an ad and getting to view some free content — has become hugely skewed, with far more personal data being harvested than can be justified. (And of course he’s not the only one saying as much.)

“Over the last — particularly the last five, six years, with the rise of ad exchanges and data harvesting — I think that that social contract is in breach… more personal data is being collected than is actually needed and the user has very little say in this,” he says, adding: “People are starting to become concerned about what happens with their personal data, how is it being used and so on.”

He points to huge growth in the privacy search segment, outstripping the overall rate of growth in search (a $62.5 billion global market last year), as another encouraging factor. Last summer, for instance, DuckDuckGo said it had grown 600 per cent over the past two years in the wake of the Snowden revelations about government mass surveillance programs.

Also on the rise in recent years: ad blocking — a technology increasingly associated with the privacy/anti-tracking movement, not just with pure-play ad-blocking. Last year Apple also threw its weight behind the online privacy cause very publicly. And where Cupertino walks others are bound to follow.

“We think that this is right on time to do something like this,” says Cornell.

“We’ve been following DuckDuckGo in the States and we’ve realized that via education they’ve managed to grow the traffic… They have really validated this marketplace,” adds co-founder Rob Perin, who used to work at BlackBerry. “The U.K. marketplace is a very ethical marketplace, I think people do believe very much in their rights.”

Oscobo is licensing its search index from Bing/Yahoo so does not have any semantic search tech of its own. Unlike European rival Hulbee, a Swiss tech company, which last year launched its own pro-privacy consumer search engine in Europe — and raised a bunch of money — another sign of growing interest in non-profiling consumer search.

Licensing its search index from companies that have already spent billions on competing with Google does at least mean Oscobo is sidestepping the problem of trying to compete head on with Google’s tech. On the advertiser side, they also have a deal in place with Yahoo’s ad marketplace — doubtless leveraging Cornell’s industry connections there.

So what’s the business model? How is Oscobo planning to make money if it’s not being evil tracking and data mining its users a la the Google goliath?

Its model is simple paid search, based on bare-bones search data (i.e. whatever string a user is searching for) and their location — given the product is serving the U.K. market this is assumed to be the U.K., but whatever search string they input may further flesh out a more specific location.

“We think it’s a bit of a myth that you need to track users, store IPs and profile them and cookie them to make money for paid search. What the advertiser is paying for is the intent behind someone typing in a keyword… So we still think that there’s a lot of money to be made in paid search without having to keep IPs and profile users and keep track of them wherever they go, offline or online or with mobile phones and so on,” says Perin. “We essentially throw the IP address away straight away, we don’t even log it. We don’t drop any cookies.”

How much money? Oscobo says the privacy segment of the search market was worth about 0.1% in 2014 but reckons it will grow to between 0.5 to 0.7% this year (a projected growth rate of 200% to 300% year-on-year). Which may not sound like much but the overall search market is forecast to be worth $71.8 billion this year so you can see why they’re keen to cut themselves a very small slice of that.

“We’ve got a proven business model. This generates revenue — it’s a very simple model. It’s advertiser driven. So we’re not here to grow the community and milk it later. We should be financially viable from day one,” says Cornell.

“Google have other objectives [than search]. We are forfeiting [user profiling] data to prove a pure and open service where the social contract is you come to our site, the first two links you get will be sponsored ads. If you choose to click on them it’s fair enough there’s an agreement there. If you don’t we don’t look to see where you go afterwards and when you turn on your mobile phone.”

“In terms of targeting there is a very well defined marketplace for U.K. ads for Yahoo and Bing, and that’s for the U.K., the marketplace we tap into, and then we target the keyword,” adds Perin.

The startup is privately funded at this point, including by the co-founders. Depending on how quickly they intend to scale — by launching horizontal pro-privacy products for other European markets — they say they might seek to raise additional funding.

“This year we have a roadmap. We will be rolling out into other countries. We will be providing country-specific search in those countries. For the time being we’re focusing our attention on the U.K., and as it does expand of course we’ll be open to investors,” says Cornell. “Our challenge in Europe different to DuckDuckGo is they have one big market in the U.S. America’s always lucky to have that. We go cross culture. So we’d have to have this conversation in German and Italian and Spanish and whatever.”


China's powerful internet regulator has moved to rein in the country's search engines following the death of a young cancer patient who had used Baidu to find an untested 'cure' from poorly labeled sponsored results.

Internet search providers must now clearly label all paid-for search results and step up their oversight of advertisers on their sites, the country's Cyberspace Administration said in a new set of regulations.

They are no longer allowed to remove negative content about their advertising clients from search results, it said.

"If paid listings are in-distinguishable from normal search results, they could mislead users," the agency said.

The new rules follow widespread public anger over the April 12 death of Wei Zexi, 21, amid concerns that for-profit sponsored links on the search engine had led Wei to an ineffective treatment.

Wei searched Baidu for treatments for his synovial carcinoma—malignant tumors that grow in soft tissues, usually around joints—and found one offered by an outsourced oncology department in the Beijing No. 2 People's Armed Police Hospital.

He later complained online that he had trusted the hospital because it was at the top of Baidu’s search results and not clearly marked as a paid-for link, sparking complaints that the company's current pay-for-listing policy is ethically dubious.

The new rules come as the agency also moves to "clean up" comments sections on news websites, warning news sites not to lure the reading public with "clickbait" stories.

Ren Xianliang, deputy head of the Cyberspace Administration, said in a video statement that news websites should "proactively foster a healthy, positive internet culture, and let cultured comments, rational posts and well-intentioned responses become the order of the day online."

Websites have a duty to "allow the internet to better benefit the people," he said.

Everything's already censored

Hebei-based veteran journalist Zhu Xinxin said the clampdown might give rise to other problems, however.

"There are a huge number of people in our society, and all sorts of things go on," Zhu said. "People have all kinds of varied needs for information."

"By selectively controlling the internet, by trying to solve one problem, they risk creating a lot of other, unforeseen problems when people search for results," he said.

Last week, China's state media regulator further boosted controls over media content with new restrictions on foreign television shows, saying that only independently produced TV with "Chinese cultural genes" would make it to air or online in future.

Online activist Lai Rifu agreed, saying that the new rules are superfluous.

"Actually, most of the controls on search engine results are aimed at managing what ordinary people are about to see online, and they are already very effective," Lai said.

"Anything we might want to see online has already long since been deleted anyway, so these rules won't do anything," he said.

He said anyone seeking information critical of the ruling Chinese Communist Party wouldn't be using Chinese search engines anyway.

"Most activists or dissidents have long since stopped using Chinese search engines, as well as a good many websites," Lai said.

The move is the latest in a long string of controls on what Chinese internet users can see online, and comes amid an ideological campaign launched by President Xi Jinping earlier this year.

The party's internal disciplinary arm has warned its powerful propaganda department that it is failing to exert enough control over public opinion, particularly online and in universities.

Meanwhile, Xi has hit out at "western" ideas entering Chinese public debate, adding that he wants all public debate to be shaped by the Communist Party and not by "hostile foreign forces" peddling values like democracy, human rights and the rule of law.

Earlier this month, authorities in the central province of Henan set up an online task force comprised of volunteers from schools and universities who wage an ideological "struggle" on behalf of the ruling Chinese Communist Party.


As the number of reported data breaches continues to blitz U.S. companies — over 6 million records exposed already this year, according to the Identity Theft Resource Center — IT budgets are ballooning to combat what corporations see as their greatest threat: faceless, sophisticated hackers from an outside entity.

But in reality, a bigger danger to many companies and to customers' sensitive data comes from seemingly benign faces inside the same companies that are trying to keep hackers out: a loan officer tasked with handling customers' e-mail, an attendant at a nursing home, a unit coordinator for the main operating room at a well-regarded city hospital.

According to Verizon's 2015 Data Breach Investigations Report, about 50 percent of all security incidents — any event that compromises the confidentiality, integrity or availability of an information asset — are caused by people inside an organization. And while 30 percent of all cases are due to worker negligence like delivering sensitive information to the wrong recipient or the insecure disposal of personal and medical data, roughly 20 percent are considered insider misuse events, where employees could be stealing and/or profiting from company-owned or protected information.

Often, that translates to employees on the front lines stealing patient medical data or client social security numbers, which can then be sold on the black market or used to commit fraud like collecting someone else's social security benefits, opening new credit card accounts in another's name, or applying for health insurance by assuming the identity of someone else.

"The Insider Misuse pattern shines a light on those in whom an organization has already placed trust," Verizon said in the report. "They are inside the perimeter defenses and given access to sensitive and valuable data, with the expectation that they will use it only for the intended purpose. Sadly, that's not always the way things work."

For the first time since 2011, Verizon found that it's not cashiers involved with most insider attacks, but many "insider" end users — essentially anyone at a company other than an executive, manager, finance worker, developer or system administrator — carrying out the majority of such acts. Most are motivated by greed.

"Criminals have a different motivating factor," said Eva Velasquez, CEO and president of Identity Theft Resource Center, a non-profit charity that supports victims of identity theft. "There are a number of jobs that pay minimum wage where individuals have access to this type of information, and so the incentive may be 'this isn't a job that is paying me enough to support myself.'"

Velasquez cites workers in an assisted living facility tasked with caring for patients, a job in close proximity to medical records that can be accessed by a few keyboard taps. According to the Bureau of Labor Statistics, such healthcare support occupations see mean annual wages hovering around $25,000, a salary that might make workers more vulnerable to stealing for self gain. Or, maybe worse, they fall prey to acting as a conduit for some type of organized crime ring looking to make big money by selling or manipulating stolen personal data.

"There are a number of jobs that pay minimum wage where individuals have access to this type of information, and so the incentive may be 'this isn't a job that is paying me enough to support myself."

According to the Verizon report, the public sector, health care and financial services — like credit card companies, banks, and mortgage and lending firms — were the industries hit hardest by insider incidents in 2015.

In one recent cases a Baltimore man is facing federal charges of identity theft and bank fraud after he used personal information of at least three nursing home residents to open multiple credit card accounts without their permission. A former employee of Tufts Health Plan pleaded guilty to stealing names, birth dates and social security numbers that were eventually used to collect social security benefits and fraudulent income tax refunds. A former assistant clerk at Montefiore Medical Center in New York who was indicted in June 2015 for printing thousands of patients' records daily and selling them. The information in the records was eventually used to open department store credit cards at places like Barneys New York and Bergdorf Goodman; the alleged actions are estimated to have caused more than $50,000 in fraud, according to the New York County District Attorney's Office.

While the number of breaches and hacks by outsiders has skyrocketed since 2007 in tandem with the surging digitization of information, the occurrence of insider jobs can be a read on the overall economy. It tends to peak during recessions and drop off when times are good, according to the Identity Theft Resource Center. In 2009, the percentage of insider attacks hit a high of roughly 17 percent; after a three-year slide, the amount today (about 10 percent) is slowly creeping back up.

"When the economy isn't doing well, you'll see people that are feeling stressed and taking advantage of opportunities they might not take advantage of otherwise," said attorney James Goodnow from the Lamber Goodnow team at law firm Fennemore Craig.

With the defining characteristic of an internal breach being privilege abuse — employees exploiting the access to data that they've been entrusted with — the best way to mitigate such attacks is to limit the amount of information allotted to workers.

"As business processes have started to rely more on information and IT, the temptation, the desire is to give people access to everything [because] we don't want to create any friction for users to do their jobs," said Robert Sadowski, director of marketing and technology solutions at security firm RSA.

Terry Kurzynski, senior partner at security firm Halock Security Labs, said that smart entities perform enterprise-wide risk assessments to find where their systems are most vulnerable and to spot aberrations in user behavior.

But sophisticated analytics does little to assuage situations where employees are using low-tech methods to capture information. "Most systems will not handle the single bank employee just writing down on paper all the bank numbers they see that day — that's difficult to track," said Guy Peer, a co-founder of security firm Dyadic Security.

Clay Calvert, director of cybersecurity at IT firm MetroStar Systems, said communication with employees in a position to turn rogue is key. "That's a big deterrent in identity theft cases; if an employee feels like the company cares for them, they're less likely to take advantage of the situation."

Hackers hiding in plain sight

Preventing the display of sensitive data in plain sight — say an employee seeing a confidential record as they walk by a colleague's computer — is the focus of Kate Borten, founder of Marblehead Group consultancy and a member of the Visual Privacy Advisory Council. She recommends companies institute a clean desk policy (ensuring that workers file away papers containing customer data before they leave their desk), implement inactivity time outs for any tech devices, and switch to an e-faxing system, which eliminates the exposure of sensitive patient data on paper that's piled up around traditional fax machines.

Experts also say that tougher penalties for and more prosecution of inside hackers would also be a disincentive for such crimes. "On a general level, there can be practical barriers to pursuit of a criminal case, such as the victim company's fear of embarrassment, reputational damage, or the perceived risk — real or not — that their trade secrets will be exposed in a court proceeding," said Brooke French, shareholder at law firm Carlton Fields.

But she added, "The DOJ and local authorities prosecute these cases all the time, despite what are seen as common barriers. The barriers are low when the actions are clearly wrong, such as a hospital employee stealing electronic medical records and selling them on the black market."

While the price tag for stolen information on the black market can translate to a lucrative sales career for some crooked employees, it's a costly phenomenon for organizations once they have realized it has occurred, which is often "during forensic examination of user devices after individuals left a company," said Verizon.

That's usually too late to enact damage control. According to the Ponemon Institute, the average cost of a breach is $217 per record.

"That's just the hard costs, what you have to pay for notifying customers or any type of remediation services," said Velasquez. "The bigger, broader cost is the reputational damage that shows itself not just to the entity that suffers the damage, but to the industry."

Source:  http://www.cnbc.com/2016/05/13/a-surprising-source-of-hackers-and-costly-data-breaches.html

Tuesday, 14 June 2016 05:20



Use to: widen your search and ensure that you don't miss relevant records

Most databases are not intelligent - they just search for exactly what you type in. Truncation and wild card symbols enable you to overcome this limitation. These symbols can be substituted for letters to retrieve variant spellings and word endings.

a wild card symbol replaces a single letter - useful to retrieve alternative spellings and simple plurals

eg wom?n will find woman or women
a truncation symbol retrieves any number of letters - useful to find different word endings based on the root of a word

eg africa* will find africa, african, africans, africaans
eg agricultur* will find agriculture, agricultural, agriculturalist
Important hint! Check the online help screens for details of the symbols recognised by the database you are searching - not all databases use the ? and * symbols.


Use to: combine your search words and include synonym

Also known as Boolean operators, search operators allow you to include multiple words and concepts in your searches. The shaded areas on the diagrams below indicate the records retrieved using each operator.

AND retrieves records containing both words. Boolean AND search operator
In this example the shaded area contains records with both women and africa in the text.
It narrows your search.

Some databases automatically connect keywords with and.

OR retrieves records containing either word. Boolean OR search operator
In this example the shaded area contains records with women, or gender, or both words in the text.
It broadens your search.
You can use this to include synonyms in your search.

NOT retrieves your first word but excludes the second. Boolean NOT search operator
In this example the shaded area indicates that only records containing just Africa will be retrieved (not those with both Africa and Asia)
Beware! By using this operator you might exclude relevant results because you will lose those records which include both words. 


Use to: combine multiple search words

On most databases you can type in a search statement, which involves combining your search words using search operators. When creating a search statement you must use brackets to ensure correct processing of the search.

Words representing the same concept should be bracketed and linked with OR
eg (women or gender)
Groups of bracketed terms can then be linked with AND or NOT
This is an example search statement bringing together all the techniques described above:

(wom?n or gender) and agricultur* and africa*

Searches enclosed within brackets will be performed first and their results combined with the other searches.

This is how the search would look when entered into the CAB Abstracts database 

Example search in the CAB databasePHRASE AND PROXIMITY SEARCHING

Phrase searching

Use to: make your search more specific

Phrase searching is a useful technique which can increase the relevance of your results. Sometimes your search may comprise common words which, when combined in an AND search, retrieve too many irrelevant records. Databases use different techniques to specify phrase searching - check the online help.

Some web search engines and databases allow you to specify a phrase using inverted commas.
eg "agricultural development"
eg "foot and mouth"

Hint! Some databases automatically perform a phrase search if you do not use any search operators eg agriculture africa is not a phrase used in English so you may not find any items on the subject. Use AND in between your search words to avoid this.

Proximity searching
Use to: make a search more specific and exclude irrelevant records

Some databases use 'proximity operators'. These enable you to specify how near one word must be to another and, in some cases, in what order. This makes a search more specific and excludes irrelevant records. For instance, if you were searching for references about women in Africa, you might retrieve irrelevant records for items about women published in Africa. Performing a proximity search will only retrieve the two words in the same sentence, and so exclude those irrelevant records.

Databases which have this facility vary considerably in their methods
eg: Web of Science - women same africa - retrieves records where the two words appear in the same sentence.

Hint! Check the online help for details of proximity operators recognised by the database you are searching.


Many databases offer other more advanced features which you can use to refine your searches further. These techniques include:

Search sets

Your results are displayed as "sets", which can be combined with other searches or new words.

Field-specific searching

Most database records are made up of different fields (eg author, title etc.). Field-specific searching allows you to select a particular field in which to search, rather than performing a keyword search across all fields. Some databases allow you to type words into specific search boxes, whereas in others you will need to type in the field name or its code.

Hint! Check help screens for field names or codes, and other hints on searching specific fields.

Searching using indexes

It is possible to search some databases using indexes, which are usually alphabetical lists of authors or subjects. They allow you to refine your search using the correct form of names or terms as defined on that particular database.

Hint! Not all databases allow searching using indexes. Check the online help on a particular database for more information.

 Example of the limits available in the CAB Abstracts database

 Many databases allow you to limit your search in various ways. Limits are usually available on advanced search screens, or you can apply them after doing your keyword search. An example of the search limits from the CAB Abstracts database is shown on the left.

Check the help pages on the database you are using for detailed instructions on applying these limits.Examples of the types of limits you can apply include:

by date

by language

by publication type (eg journal articles, chapters in books, review articles that provide detailed summaries of research, book reviews) 

Source:  https://www.reading.ac.uk/library/finding-info/guides/databases/lib-searching-databases-search-techniques.aspx

For the moment, at least, cyberterrorists have not harnessed the technology they would need to destroy Western civilization from a basement lab in some remote corner of the world.


Although Director of National Intelligence James Clapper has said a “cyber-Armageddon” scenario is unlikely any time soon, new technological developments have the potential to allow terrorists to move from low-tech killings aimed at gaining attention and creating fear to high-tech sabotage aimed at disrupting the sinews and social tissue of society.

While defense budgets are declining in much of the developed world, the threat of terrorism has elevated homeland security concerns. Terrorists make no distinction between front lines and home fronts, between combatants and civilians.

Fear of terrorism, sometimes exaggerated, has put governments under pressure to prevent terrorist attacks before they occur, which means intervening before intentions become actions. One way to know what evil lurks in the heart of potential terrorists is to monitor what people say and write. Police states do that all the time, but democracies have strict rules about when and under what conditions that may be permitted. 

That is where developments in information technologies are redefining relationships between citizens and their governments and creating new tensions. Governments now possess unprecedented capabilities to collect, store and analyze vast amounts of information about our private communications and individual lives. Some would argue that the mere possession of such files in government hands represents a potential for control and intimidation that is alien to the American form of government.

As national security and war are being redefined for the digital age, Silicon Valley will need to be on the front line of counterterrorism. Its inventors and entrepreneurs are driving the information revolution, and they must figure out how to protect vital systems against malevolent intrusions. It lies at ground zero of the battle between government efforts to protect society and individual rights of privacy.

Terrorist tactics have been employed for centuries, but technological developments in the late 1960s created new vulnerabilities and capabilities. Modern jet air travel gave terrorists worldwide mobility and provided what amounted to nationally labeled airborne containers of hostages and victims. Local terrorist campaigns could easily go international. Small arms and explosives had become widely available commodities. Most importantly, communications technology — radio, television and communications satellites — gave terrorists access to a global stage.

Terrorism is theater, violence choreographed to create an atmosphere of fear and alarm that, in turn, causes people to exaggerate the importance and strength of the terrorists and the threat they pose. The actual victims of terrorism are irrelevant to the terrorists. Our terrorist adversaries understand that communications are half the struggle — it is not simply what they do, but how it is perceived and portrayed.

In the late 1970s, analysts like me tried to figure what new weapons terrorists might try to acquire and adapt to their struggles. We worried about precision-guided surface-to-air missiles and, of course, chemical, biological and even nuclear weapons, but we missed the most important development of all — the beginnings of the modern internet, which would become a critical weapon in the terrorist arsenal.


As a propaganda platform, the internet has enabled terrorists to communicate directly with vast audiences, without editorial or effective government interference. It also allowed terrorists to communicate more easily with each other, creating virtual communities of like-minded fanatics. And it provided information about targets and instruction in bomb-making and other techniques of violence.

Social media takes things further and gives today’s terrorists the ability to communicate directly in a mode embraced by millions of young people. The so-called Islamic State effectively exploited social media to advertise its exploits and attract recruits.

However, the internet also allows vicarious participation without outright radicalization. One does not have to join a group. Participants can have a virtual yet real-life experience: Psychological satisfaction can be obtained by merely pretending to be a terrorist online.

In the pre-internet 1970s, the United States was dealing with an average of 50-60 terrorist bombings a year — a number that in retrospect seems astounding. In the nearly 15 years since the 9/11 attacks, there have been about two to three terrorist bombings a year, with almost no fatalities, in the United States.

These attacks were carried out by a variety of groups motivated by extremist ideologies and quarrels related to ongoing conflicts abroad. Since 9/11, however, two-thirds of the approximately 60 jihadist terrorist plots in the United States have involved a single individual. There is no real membership in a group, no institutional learning. New plotters are almost always amateurs. 

The threat posed by today’s terrorists is still primitive, manual and low-tech, but contemporary terrorists are becoming savvier navigators of the internet, with the potential to become high-tech adversaries who can threaten economic sabotage. Instead of holding individuals hostage, they might hold systems hostage.

In the 1970s, “red teams” of terrorism analysts, trying to think about how this might be done, found that it required immense resources to significantly disrupt society. The growing network of the Internet of Things may change that.

The capacity to destroy, disrupt, alarm and force society to divert vast resources to security is descending into the hands of ever-smaller groups with grievances that will not always be possible to satisfy. How democracies deal with this trend, and remain democracies, is one of the major challenges of our technological time.

Silicon Valley is up for the challenge, if my recent experience at TiEcon, an annual conference in Santa Clarita of innovators and entrepreneurs, is any indication. Some of the participants already have created technologies currently being used to assist in security. Others have exciting new concepts or ideas already in development. These represent new approaches to physical security and information protection; the detection of weapons, explosives, radioactive material and other dangerous substances; analytics; and other countermeasures.

With these potential advancements, Silicon Valley may already be placing itself at the heart of the terrorism battle.


Source:  http://techcrunch.com/2016/06/12/cyberterrorism-and-the-role-of-silicon-valley/

A long, long time ago I was talking to Mike Grehan about search engine rankings. He used the term “the rich get richer”, to explain why sites that live at the top of Google are likely to stay there.

One of the reasons is the ease of findability.

A writer who is researching a subject on Google is highly likely to click the top result first. If that web page answers the right questions then it may be enough to earn a citation in an article, and that link will help fortify the search position.

The rich get richer.

I mention this because yesterday I read a brilliant, in-depth post from Glen Allsopp (aka @Viperchill), which illustrates that the rich do indeed get richer, at least in terms of search positions.

In this case, the rich are major publishing groups.

The way they are getting richer is by cross-linking to existing and new websites, from footers and body copy, which are “constantly changing”.

There’s nothing hideously wrong with this approach, but it’s a bit of risk to tweak the links quite so often. Especially when the anchor text is something other than the site’s brand name.

As Glen says:

“As anyone who has been involved in search engine optimisation for a period of time might wonder, surely getting so many sitewide links in a short timeframe should raise a bit of a red flag?”
It’s interesting to see that Google not only tolerates it, but actively rewards this kind of behaviour, at least in the examples highlighted in Glen’s post.

The short story is that Hearst was found to be linking to a newly launched site, BestProducts, from its portfolio of authority websites, which includes the likes of Cosmopolitan, Elle, Marie Claire and Bazaar.

This helped to put the new site on the map in a rather dramatic way.

Party hard in footerland

Here are a couple of screenshots. The first is from March, when the anchor text was ‘Style Reviews’

cosmomarch 2

The second appeared later, with the link text changing to ‘Beauty Reviews’. Note that the link placement changed too.

cosmopolitan 1

I’m going to assume that these links are dofollow, which is a potentially risky tactic, and one that has attracted the dreaded manual penalty for some site owners.

Furthermore, this is clearly something that has been done with intent. Design, not accident.

Glen says:

“It’s now obvious that the people working for Woman’s Day, Marie Claire, Popular Mechanics and Esquire had some conversion that went along the lines of, ‘Don’t forget, today’s the day we have to put those links to Best Products in the footer.’”
But did it work?

The results

Glen estimates that BestProducts attracted at least 600,000 referrals from Google (organic) in April 2016, so yep, it has worked incredibly well.

Here are some of the positions that the site has bagged in little over half a year, from a standing start:

Screen Shot 2016-06-07 at 14.51.32

Pretty amazing, right? Some pretty big, broad terms there.

Glen reckons that the following 16 companies – and the brands they own – dominate Google results.


I suspect that if you look at other industries, such as car hire, where a few brands own hundreds of sub-brands, that you’ll see similar tactics and results.

We are family?

The standout question for me isn’t whether Hearst and its peers are systematically outsmarting Google with a straightforward sitewide link strategy, nor whether that strategy will hold up. It is more about whether Google truly understands related entities.

Does it know that these sites are linked to one another by having the same parent company? And does that discount the link tactics in play here?

Certainly one of the webspam team would be able to spot that one site was related to another, were it flagged for a manual action. So is Google turning a blind eye?

Here’s what Matt Cutts said about related sites, back in 2014:

“If you have 50 different sites, I wouldn’t link to all 50 sites down in the footer of your website, because that can start to look pretty spammy to users. Instead you might just link to no more than three or four or five down in the footer, that sort of thing, or have a link to a global page, and the global page can talk about all the different versions and country versions of your website.”

“If you’ve got stuff that is all on one area, like .com, and you’ve got 50 or 100 different websites, that is something where I’d be really a lot more careful about linking them together.”

“And that’s the sort of thing where I wouldn’t be surprised if we don’t want to necessarily treat the links between those different websites exactly the same as we would treat them as editorial votes from some other website.”

Note that Matt talks about links to other sites, as opposed to “links with descriptive and ever-changing anchor text”. Somewhat different.

Screw hub pages, launch hub sites

Internal linking works best when there is a clear strategy in place. That normally means figuring out a taxonomy and common vocabulary in advance. It also means understanding the paths you want to create for visitors, to help pull them towards other pages, or in this case, other sites. These should mirror key business goals.

With all that in mind, I think it’s pretty smart, I really do, but let’s see how it plays out. And obviously it takes a rich portfolio of authority websites to play this hand, so yeah… the rich get richer.

Assuming this strategy works out in the long run we can expect to see lots more niche sites being launched by the big publishing groups, underpinned by this kind of cross-site linking.

Ok, so this fluid footer linking approach certainly sails a bit close to the wind and we may not have heard the last of this story, but it once again proves the absolute power of links in putting a site on the map. Take any statements about links not being mattering so much in 2016 with a large bucket of salt.

Source:  https://searchenginewatch.com/2016/06/07/are-related-sitewide-footer-links-the-key-to-dominating-google/

Wednesday, 08 June 2016 01:49

Is the web full of nasty tricks?

Ever signed up for something online and then found it very difficult to unsubscribe? Then you've been to a roach motel - one of many web design tricks employed by some firms to drum up more trade.

Microsoft provoked outrage recently when it was revealed that closing a pop-up box reminding Windows users to upgrade to Windows 10 was treated as consent to the upgrade, if a common setting was in place.
It was described by one frustrated tech reporter as a "nasty trick", as many people would expect closing the box to both remove it and prevent any related activity.

After a media storm, the tech giant backtracked.
But was it any worse than other cunning sales tactics?

Dark patterns


London-based user experience consultant Harry Brignull coined the phrase "dark patterns", which he describes as "a user interface that has been carefully crafted to trick users into doing things".
His website contains a library of the sort of tricks employed by websites great and small, submitted by readers.

Some tricks are no longer an option - for example, adding items to a shopper's online basket without their knowledge, such as through pre-ticked boxes, has been closed in many European countries by an EU directive - but there are still plenty out there.

The "roach motel" for example will be familiar to anyone who has happily signed up for a newsletter or subscription and then found it extremely difficult to unsubscribe - "a roach motel makes it very easy for a user to get into a certain situation, but then makes it hard for them to get out of it when they realise it is undesirable," the Dark Patterns website notes.

Also well known are what Mr Brignull calls "trick questions", where the process of ticking boxes to opt in or out of receiving marketing material is deliberately inconsistent. Royal Mail is one of those that asks people registering to opt out for the first question and then opt in for the second.

Receiving marketing from Royal Mail

Royal Mail says its wording is designed to reduce confusion. "Royal Mail aims to make it as easy as possible for its customers to opt out of receiving information that is not relevant to them," it said.
"At the same time, we aim to make it easy for interested customers to receive highly targeted information that they may find useful. We look to make these choices as clear and unambiguous as possible by using bold text where appropriate, simple check box options and plain English."


When Tim Jones from the Electronic Frontier Foundation asked on social media for a phrase referring to the practice of "creating deliberately confusing jargon and user-interfaces which trick your users into sharing more info about themselves than they really want to", the answers that came back were variations on the theme of "Zucking", a dubious honour for Facebook founder Mark Zuckerberg.
Mr Jones wrote about it in a blog post, settling on the less personal phrase "evil interfaces".

"Sharing your data requires radically less work than protecting it," he concluded.
But this was a while ago - last year the firm's privacy policy was praised by the US Center for Plain Language.

Mark Zuckerberg


User Testing journalist Jennifer Winter listed some of her pet hates in a blog post last year.
She illustrated the practice of "misdirection" in a mobile game where the buttons to start play and select a level were both long green tabs.

"Players are trained to associate the green buttons with game play," she wrote.
"But if you lost the game a screen appeared inviting you to buy a move - and the "buy move" button was also a long green tab.

"Guess how many times I've tapped the green button. I actually have no idea because I've lost count at this point," Ms Winters said.

man shouting at laptop

However, she also pointed out that purchasing the moves required a visit to the app store, so there was still a boundary before purchase.

"Wherever there are organisations that value short-term gains over lasting customer relationships, there will be designers who implement dark patterns in an attempt to trick users into taking actions they wouldn't normally want to take," said Hannah Alvarez from User Testing.

"Dark patterns are well known among designers and UX practitioners who care about their users and want to avoid creating poor user experiences.

"I believe some marketers and executives who don't have a background in user-centred design will implement dark patterns without realising they're doing so - they're just trying to get quick results."
Mr Brignull said deliberate dark patterns come about as a nefarious use of combining web design and psychology.

"Psychological insights that would once have been used to stop people making mistakes are now being used to get them to buy stuff," he told the BBC.

A or B?

A/B testing

Many tech firms conduct what's known as A/B testing - trying out different designs on a live site simultaneously to see which generates the better results.
"People don't realise it's on every large site you use - you will be tested on something like that. It is so easy to run," said Mr Brignull.

This was confirmed recently by Google, which experimented with turning the blue links on its search results pages black to see whether it increased the number of clicks.
"We're always running many small-scale experiments with the design of the results page," the firm said.
During her time as a senior executive at the search giant, Marissa Meyer famously led experiments with 41 hues of blue to determine the most clickable link colour.


"This is about selling people things," said web science expert Prof Leslie Carr of Southampton University.
"The [advertising] industry has been doing this for 150 years."
However, he thinks the tech giants, such as Microsoft and Google, have a greater responsibility than most to stick to the rules.

"You have to trust the platform," he said.
Ms Alvarez says designers should aim to develop "user empathy".
"That means questioning your own assumptions, asking how your designs could be interpreted in a different way than you intended, and regularly getting user feedback throughout your design process," she said.
"No matter how experienced you are, never trust your gut."

Source:  http://www.bbc.com/news/technology-36403020

Two months ago, the internet blew up over Google AMP (Accelerated Mobile Pages). And, just when you thought you passed the Structured Data Testing tool and implemented your AMP HTML file extensions perfectly, you realize your AMP pages are not showing up in the search results.

Are you asking: Where are my AMP pages in the search results? Does AMP happen in real-time? Or, does it take a few hours for Google to index your AMP pages?

Well, take comfort in knowing you are not alone. Yep, the SEJ team is right there with you.

In true early adopter fashion, SEJ launched our AMP pages on March 29. Since the launch, we have seen traffic to our AMP pages spike from 50 to 700 sessions. High-fives all around!

SEJ Google AMP Traffic

As you can see below in our breakdown of AMP traffic, the majority of our traffic is coming from play.google.com/newsstand.

Google AMP Referrers for SEJ

However, when we search in the search results, our AMP pages are nowhere to be found. Sound familiar? Others are having the same issue.

So, it’s no wonder that only 25% of SEO professionals have taken steps to implement AMP. And, while there may be more legwork involved in creating AMP pages, Google has created a new problem: With countless hours spent, it’s hard to know when and where your AMP pages exist in the SERPs.

Implementing AMP

For implementing AMP, Vahan Petrosyan, our lead developer here at SEJ, used the following plugins:


Glue for Yoast SEO & AMP

But, you will likely need a little more than a plugin if you want to implement AMP. Here is what Vahan had to say about using plugins alone:

“Simple installation of plugins gives only standard and limited functionality of customizing posts. I did custom development using that plugins actions hooks and modified standard look for our AMP pages. We modified styles, added slide menu from right using pure CSS3 techniques, sticky share buttons,and modified Google Analytics default tracking code in order to include custom dimensions and event tracking such as share buttons, menu items clicks,etc.”

Where are My Google AMP Pages?

Restructuring your mobile website to cater to Google’s new features is just part of the job description when you launch a website, right? But as we’ve come to learn, the theory “build it and they come” isn’t so true. Based on our experience at SEJ, Google AMP pages are not real-time. Since we launched our AMP pages on March 29, 2016, we have not seen them in the search results even though we’re getting traffic to our site from AMP pages. We noticed an uptick in traffic 9 days after launched our AMP pages. So, essentially it took Google 9 days to index our AMP pages.

“In general the same applies on AMP results as on normal web results: we have to crawl and index the page in order to show a result for it. Depending on the site this can take from a few minutes to days, but most of the time it’s pretty fast.”

One theory: While Google’s AMP pages are being praised as the fairy godmother to all the small businesses in Internet land, you may not see an immediate boost if you’re not first to implement. I’m noticing a large amount of the bigger publishers (mainly brands that partnered with Google in their beta testing) for AMP seemingly get first dibs in the Google search results carousel.

Now, with Google News featuring up to 14 pieces of AMP content from publishers that have already launched their AMP pages, it’s going to continue to be a race to see who gets their first. This could potentially hurt smaller businesses.

Another theory: Google will only display news articles. These news articles will only show up in the AMP carousel if Google views content as recent news for the topic you’re searching for. The AMP carousel is a Top Stories carousel so perhaps your news articles are newsworthy enough for the carousel. Google will only display Article, NewsArticle, and BlogPosting schema types into the carousel at this time.

If you double checked your schema, head over to the Webmaster Central Forums and/or the AMP Error Reports in Search Console. You can also look at the new AMP filter in the Search Console.

When the SEJ team asked Gary Illyes why we’re not seeing AMP pages expand faster outside of news sites, he responded,

“The format itself is open to anyone who’d like to speed up their sites, it’s not limited to only one kind of site. Currently, we’re testing AMP results with a limited set of publishers, but we’re exploring ways to show AMP results from more sites.”
Also, Google does not guarantee your articles will show in the AMP carousel. It depends on the searcher’s search query and Google’s algorithms of whether or not they display your content, even if you’ve passed the data structuring test and validated your AMP pages.

We also asked Gary Illyes why AMP users are able to get traffic to their site via AMP pages but still not see their AMP pages in the SERPs, Gary stated,

“AMP pages, just like normal webpages, can be accessed in many ways: from bookmarks, from search results, directly from the browser, an so on. If I know that a site is AMP enabled, I will very likely stick the “/amp” to the end of the URLs on the site, and so the publisher will likely see a direct visit in their tracking software. This is not necessarily what’s happening to SEJ, but that’s my best guess.”

What’s Next for AMP?

For publishers, we simply wait. AMP pages are newbies to the SEO game. They are still trying to figure out where and how they fit into this crowded space. As Google continues to adapt and evolve their algorithm, the time and money invested will be worth it.

Source:  https://www.searchenginejournal.com/long-amp-pages-take-show-search-results/163006/

Think about the last dance you attended. It may have been last week or thirty years ago, but in a crowded room, dancing is almost impossible because you keep running into other people, or furniture or walls get in your way. And in a dance, others may laugh at your dancing style or clothes.

Now think about your online genealogy as a dance that you are trying to navigate, but you keep running into other people, walls, others’ opinions, and other obstacles. Janet Hovorka addressed this in her presentation at the BYU Family History Conference on July 29, 2015, titled “Six Steps to Choreograph Your Research across the Internet.”

A common complaint about FamilySearch Family Tree is “Why do people come in and mess up my tree, change the information, set aside what I have done, leave me out of the family tree, and stop me from changing what I know to be correct?” Frustrations abound in these public family trees. Hovorka clarified this when she defined a public tree—making life and comprehension easier.

What is a public tree? She called it a “community” or a “public conclusion tree.” These trees are found in FamilySearch.org, MyHeritage.com, and Ancestry.com, among many others. Many people contribute to these online trees, giving hints, sharing sources, showing pictures, and telling stories. But there are problems using these trees if you don’t know how to interpret these databases. And, of course, when your conclusions are changed or questioned, you want to throw up your hands and say, “I’m not going to do this anymore!” Don’t stop! Hovorka said, “The six steps [below] will keep you moving and dancing around brick walls in your family history research.”

These “personal conclusion trees” are great for collaboration, but these six steps will help you to have fun with your “dance” in genealogy.

Step 1: Know what you are really looking at online. Distinguish between the quality conclusions of others and original documents about your ancestors’ lives.

The sources and good interpretation are the keys—dig deeply into those sources, compare them, find them, and interpret them. She said that “good researchers dig down deep into the sources to come to the original documents of a person’s life.” Make use of helps about sources, and study the many different places you can find them.

Step 2: Create a personal conclusion tree and know how to sync it and what you can attach to it from websites across the net.

This is the key to keeping the frustration level down: Make your own tree, offline, on your own computer by using your favorite program. Hovorka says that offline software solutions offer clear copyright application, more features and reports, no privacy issues for living people, and usually more speed. Some software packages synchronize with online tree and easily compare databases. Some of the most popular software applications include: FamilySearch – Legacy, Ancestral Quest, and RootsMagic; MyHeritage – Family Tree Builder, Family Historian and RootsMagic; Ancestry – Family Tree Maker. She suggests choosing one place to start a tree and look for information on the other websites.

Step 3: Know the best search strategies and where to go if you aren’t finding what you need.

Planning—the word sometimes is daunting, but creating a plan is absolutely necessary. She suggests clear research objectives, a plan to find the sources, learn about records, ask others in Wiki, use social networking, try different spellings of your names, and other research tips. BUT, keep track of everything you do so you and those who follow you don’t have to do the same work again and again.

Step 4: Use tabs, windows, and multiple screens to quickly work between sites.

Learn how to use the online features of your computer. Check on tutorials for your particular computer.

Step 5: Use timelines to constructs a bird’s eye view of your ancestor’s life and keep track of everything you find.

Make up a chart, using your program, Excel, or other sources, and write down all the material you have about, for example, the birth of your ancestor—maybe there are conflicts, but get everything, and put it down by date or event. You will find much information on your public conclusion tree. Put everything on your personal record that you can find. Then continue with Step 6.

Step 6: Record what you find with complete notations and analysis so that you don’t create any unnecessary brick walls.

As Hovorka said, “Good researchers dig down deep in the sources.” They document, collect all they can find, keep everything traceable by citing good records, and remember to use the all-important analysis in the notes section of their source citations.

Don’t create any unnecessary brick walls, and keep dancing!

Source:  https://familysearch.org/blog/en/6-steps-choreograph-research-internet/

Page 7 of 8


World's leading professional association of Internet Research Specialists - We deliver Knowledge, Education, Training, and Certification in the field of Professional Online Research. The AOFIRS is considered a major contributor in improving Web Search Skills and recognizes Online Research work as a full-time occupation for those that use the Internet as their primary source of information.

Get Exclusive Research Tips in Your Inbox

Receive Great tips via email, enter your email to Subscribe.