Showing posts with label Search Engines. Show all posts
Showing posts with label Search Engines. Show all posts

December 6, 2009

Google's bad idea of personalizing search results!

I admire and respect Google only for one reason - they made it very easy for me to find the information I want, when I want it, as quickly as possible. If what I search for is available on the World Wide Web, Google gets it for me. I am stuck to Google ever since I started using Computer and Internet only for this reason. I won't mind even if Google takes all the other services it offers away. But I can't think of internet without Google, simply because I have learned many things after Google came along.

Search is the core of what ever Google does and they seem to be very proud about that. Personalized Search Result  is one of the key Google Search Features the company flaunts. Day before yesterday, Google announced that they are taking personalized search a step further by converting it into Personalized Search for Everyone. This is what I read on the Company Blog:

Today we're helping people get better search results by extending Personalized Search to signed-out users worldwide, and in more than forty languages. Now when you search using Google, we will be able to better provide you with the most relevant results possible.Previously, we only offered Personalized Search for signed-in users, and only when they had Web History enabled on their Google Accounts. What we're doing today is expanding Personalized Search so that we can provide it to signed-out users as well. This addition enables us to customize search results for you based upon 180 days of search activity linked to an anonymous cookie in your browser. It's completely separate from your Google Account and Web History (which are only available to signed-in users).
I like the idea of Google trying everything they can to get me the best result possible. However from the way Google explains the new Personalization Feature with an example, I think it is such a Bad Idea and a total waste of Google's time and computer resources. Following is how Google explains their way of Personalization:

For example, since I always search for [recipes] and often click on results from epicurious.com, Google might rank epicurious.com higher on the results page the next time I look for recipes. Other times, when I'm looking for news about Cornell University's sports teams, I search for [big red]. Because I frequently click on www.cornellbigred.com, Google might show me this result first, instead of the Big Red soda company or others.
Let me give you another example. When I search for Scocial Networking using Google, let us say I come across Facebook. I visit Facebook and I like it. Now, I visit the site a couple of times using Google. My problem is, if Google is going to put that result on the top just because I visited Facebook  couple of times from Google, then what about the other Social Networking Services? What if some one comes up with something better than Facebook and just because Google has put my favorite Facebook first the new guy gets pushed to the second page where I never go? I say pushed back because  the first ten results on the page are going to be in the order of my clicking frequency.

Lets not talk about Facebook. Let us say I want to know more about Social Networking. Google gets me ten great articles on the first page. I like all of them and I keep going back to read them. What if a new guy comes up with something better on Social Networking? How will I know, if Google puts where I go often first?

Is not Google reducing their Search Engine into a Bookmarking Engine by adding this sort of Personalization where what you click often appears first? I thought Google's way of intelligently understanding what I look for should mean much more than that!

[I have no issues with them using information on what I search to help advertisers target me. Since they give me the search result for free, they might as well make some money out of the advertisers using information about what I do on their site.]

Update: I was just thinking - what if Google puts the list of sites I click often when I search something just below the space available in that corner where they put sponsored links? How useful it will be if they do that, if the sites I have already visited are separated from the sites I have not!

December 4, 2009

Google is out to demolish the outcome of Babel

"Then they said, 'Come, let us build ourselves a city, and a tower with its top in the heavens, and let us make a name for ourselves; otherwise we shall be scattered abroad upon the face of the whole earth.'" (Genesis 11:4).
God came down and confused the language of men - you know the Story of Tower of Babel. That is how Bible tries to explain the existence of different languages. Since the Tower of Babel, we have thousands of languages and as many cultures. We had no clue what was happening in different cultures and languages across the world. Today, Google integrated their Translation feature into their Search Engine and from today on we can look up what people are writing in other languages too.

Google says this is just to help people who travel out of their country. But I think this means a lot to even others. Now, I hope to look up and see what other languages has got to say about the doubt I want to clear using Google.



Now, next time you go to Google, click on "Show Option" and add as many languages as you would like to the Search result. Google will pick the translated results for you and you can know what people in other languages think about your topic. I hope God won't come down this time and confuse Google!

Image from Google Blog

August 16, 2009

webfinger: Soon you could be using Gmail ID for more than emailing!

What do we use our Gmail Ids for? To send and receive emails, to log in to Google related services and sometimes, use other web services that require email IDs and not usernames. We use any email Id by any service provider, for that matter, to do similar things.

Try searching an email id to find out more about the owner of the ID! You will get absolutely no information about the person as there is no way a person can attach information to his or her email id. Email IDs are not linked to personal information the way social networking IDs are linked to personal information. For example, if you search for a long lost friend, it is quite likely that his or her social network profile pages like Twitter, Face Book or Linkedin would show up in the search. But not if you use an email ID instead of the name. Because as mentioned in the Google Code Summary Page on webfinger: "If I give you my email address today, you can't do anything with it except email me. I can't attach public metadata to my email address to give you more information." Now, through the "webfinger Project" launched on April 28, 2009, Google is trying to make it possible for people to attach personal information to their Email Ids. In other words, Google is on a project to make email Ids searchable!

What are the implications of this innovation? I don't know if you know, this is not a new idea. This was something tried during the initial Internet years. A finger programme was written by Les Earnest to help those users who wanted to find out information about other users using their emails. However, by 1990, websites and companies stopped using this service as it was a threat to security and privacy of individuals. Hackers used this service to carry out social engineering attacks on companies and individuals.

Why is Google bringing back a service that was tested and dropped, years ago? Why do they, all of a sudden, want to make an email id both 'writable' and 'readable' like a URL? Experts say that Google is trying to come up with an alternative to open ids! I understand the concern about the failure of Open ID System, as people don't use Open IDs often. It is perfectly fine if Google is only trying to make people use their email ids instead of Open ID URLs, to access all the web services. But I don't understand that part where Google says it is also thinking of making emails 'readable'.

I'm already thinking of taking off my profiles from the popular social networking sites. After Twitter and Face Book allowed me to customize my URL the way I want to, I am increasingly worried about the idea that I'm now more prone to social engineering attacks. [Thank God, I have an option to keep all my information closed to me and the people I know on Face Book and Twitter. But not on Linkedin!] Because every time you search my name, I show up in the search and it is not always good to leave trails online.

Google already collects information about what I search. They use my Gmail ID to keep track of what I search! Now, they are trying to get me attach personal information to the very same Gmail ID I use. If that is done, Google knows who I am and what I do online. That is a lot of information, I can give away, don't you think?

And if Google makes my email searchable or 'readable' as they calim they are going to do, it means people can dig out my Google Profile [Or my Face Book Profile or my Orkut Profile or my Linkedin Profile or my Twitter Profile] using my email as the key word. Now, they can do that only using my the Name. Where your online life is concerend, your email id is more credible than you name, as your email id is directly linked to you unlike your name. A name or username is not considred to be credible or authentic online. Emails Ids are legal today and emails are accepted as evidence for this reason. Doesn't Google understand the political implications of this move?

August 12, 2009

Twibbon: Is it just about wearing a badge?

Now doubt Twitter is becoming a phenomenon on-line. And the number of sites that survive because of Twitter are increasing. Earlier I wrote about Twitterfeed, a service that allow you to post your feeds automatically to your Twitter Page. Another one I recently came across is Twibbon. Twibbon is about wearing a badge to show off your affiliation. But is Twibbon only about wearing a badge?

The habit of wearing an E-badge is becoming very popular these days. There are sites that force you to wear a badge to show your affiliation. There a badge is nothing more than a reciprocal link. You link me because I linked you is bad on-line attitude.



Twibbon takes E-badges to a new level. People wear a badge to support a cause. IE6 Must Die was a cause that many Twitter users supported and made popular that it trended on Twitter for a long time. When the King of Pop passed away, people who mourned the sad demise wore a black badge on their profile and Jackson trended on Twitter like never before. Along with the Badge, Twibbon sends out a tweet on the person wearing the badge. It is actually a link to the location of the badge so that other people can come and claim a badge for their profile if they care. It is viral, viral and viral all the way.

I am always of the opinion that Internet is more about marketing than 'information' Of course it is a good way to gather and process information. But it is more about companies gathering information about what people are looking for when. No wonder why most of the services online, like search and social networking, come free of cost to the users.

Twibbon is another way of finding out what people are interested in, when. Twibbon lets you take a closer look at 'what people want', a degree higher than search engines, because people wear a badge usually because they are emotionally attached to something. And I am happy that they don't keep that information to themselves like most of the search engine companies do.

In short, Twibbon, like What the Trend is a great market research tool. I think it can go a long way by helping marketing wiz kids to formulate their strategies. Internet is mostly about Attention Economy now. And Twibbon is only about that.

Photo from Twibbon

August 8, 2009

PayScale: Get to know what you are worth!

Ever sat during a job interview wondering what exactly to say when the million dollar question about pay package came up? Ever wondered how much you are worth, with the skill sets you have? The other day I wandered Google to find out how a particular career I had in mind is placed in India, in terms of the pay scale. That was when Google gave me a link to PayScale.

PayScale is a web service that allows you to dig into the 'salary' side of different careers. It allows you to compare your salary with your peers. It allows you to see how a particular career pays. For example, see the following graph that allows you to see how much different jobs are worth in India right now:



Also see an example that shows you how a male or a female is paid in India:



Now, that is interesting. Even after so much of women's lib, a female is still paid lesser in India when compared to her male counterpart. Don't you worry. So is the case in United States, Canada, France, UK, you name it. Women are paid lesser than men in most of the countries that I tried. Women, you need to wake up!

PayScale was launched in 2002 by PayScale, Inc, an American Company.  It works gathering data from users, employers and employees who take PageScale Surveys. It has an option which lets you compare your pay scale with your peers. It is a tedious job, as PayScale asks you many questions over and over again. It asks you to answer same questions again and again. When you are answering those questions, PayScale is actually gathering data to update its database. It uses the data to generate reports and graphs that tell us who is worth what. They say they report only actual data to avoid erroneous assumptions.

Here is one thing I really liked about PayScale. When they give you a graph or a report, they tell you how many people contributed information to make that graph or report. Now, this helps me decide whether to take the report seriously or not. Especially because they are doing actual reporting and they are not telling me their assumptions.

I hunted down many graphs and reports on PayScale that gave me an idea about the career trends. As I was doing that, I found that there is lesser participation from people in India. I think, at least for the sake of building a usable database that helps people keep track of pay scale trends, Indians should participate in PayScale. The more number of people participating in PayScale, the more authentic it is going to be. I hope that in a few months or years from now PayScale will have more authentic information about pay scales so that we know where we are going when the million dollar question about pay package comes up during an interview.

August 7, 2009

Happy News Bloggers: Mr. Murdoch will charge for online content from next year!

Rupert Murdoch is an old man now. 78 years old. As an old man, may be he has lost that spark which saw him rise to the position of the media emperor he is today. Perhaps, that is the reason why he has announced the most uncharming business decision of his career the day before yesterday- a decision to make all his news sites available only to users who pay, from next year.

He said, :"Quality journalism is not cheap,..The digital revolution has opened many new and inexpensive distribution channels but it has not made content free. We intend to charge for all our news websites."

You should read people's reaction to that. Many reacted as though, it is not a big deal and they consider News Corps news sites as crap. Most of them said they don't care as they are not going to pay to read his sites anyway. News Analysers sounded very disappointed that a person like Murdoch took such a naive decision. Stories ran on blogs and technology review sites on how this decision is all about News Corps digging its own grave. Journalists are wondering how and why this is too risky for nothing. One reviewer even went to the extent of saying only Apple can save newspapers.

Is Murdoch serious about what he said or was he just testing the waters? True, his business ventures reported a net loss of $ 3.4 Billion this fiscal year. True, recession is at its best and on-line advertisement revenue is dipping. True, Wall Street Journal, a paid news site from Murdoch is doing relatively well and has users paying to read news. But, is he serious? Or was he, as I said, just testing the waters? Or, was it just a sudden outburst of frustration? May be, it was all the three put together.

Why do I say that? I think Murdoch is missing a key point here. On-line content is just a tool. I don't believe that Internet is all about Information. Internet is about advertisements. Certain products are just tools to sell certain other products. Why do Google let me search for free? Why does Google or Wordpress permit me to upload content on my blog for free? Why do social communities allow me in for free? Why are there many sites that give me many of their services for free? Why are open source applications free? Why is the news free, after all? I think these things are free because they want me to stick around as long as possible. You stick around so that you buy something, sooner or later.

 This is not a new business model. In India we have temples and churches organizing festivals around their theme of worship. They spend a lot of money, organizing a fare, staging plays or other performances by professional artists etc. These religious festivals last for five to ten days. The plays or other performances go well into midnights. You don't pay anything to watch a play during the festival. Why do they do it for free? Because, they know that people who come to the festival will visit the temple or church to pray. But the few coins they offer when they pray is not the target! People come to these festivals to set up stalls that sell from glass bangles to kitchenware. They pay lovely loads of cash to people who run the temple or church as rent for these five to ten days. In fact, there are groups of people in India, who travel from one festival spot to another and make a decent living!

To me, Internet sounds like a temple or church festival. And people who make content available or provide spaces for people to meet are like the Temple or Church Committee Members. They run the show. Sites like Murdoch's or Google's or Facebook's or Twitter's are these plays or performances staged during this Big Bang Festival called Internet. People come to the festival, because these performances are for free. And as they come, some of them, sooner or later most of them, will buy from the people who set up stalls in Google's or Murdoch's or Facebook's or Twitter's premises.

What happens if Murdoch makes his sites 'pay-per-click' as he says? May be there are going to be people who are willing to pay like there are people to read Wall Street Journal. But imagine the number of people who would have read Wall Street Journal, if it was available for free?

And if Murdoch is going to hide his content behind the 'paywalls', what about those people who hit the search engines for anything and everything? Google, Bing and Yahoo! If they won't have traces of Murdoch's Sites, won't he disappear from the minds of people gradually? This is why I said, Murdoch's announcement is happy news to Bloggers.

I am sure he will not hide all his content behind the paywalls. He will float enough summary of news stories around to make sure that he is on search engines. May be he will enter into deals with Search Engines to make sure that he is there, when people search. But even then,, it is happy news for bloggers!

Experts say that about 90% of his traffic will die because of this decision. Let Murdoch take his ten percent! Where will the rest go? The rest will trun to sites that make content available for free. I think, if Murdoch will implement his decision, there is going to be an increase in traffic to blogs we have around. Is this not happy news bloggers?

August 1, 2009

When Yahoo gives in to Microsoft!

This was the talk of the town for the last six months - the Yahoo! Microsoft Deal! And finally it is done. Yahoo sells its user share in search to Microsoft for 88% of commission on advertisements sold on search results through Yahoo! Together they make 28% of market share in search, way lower than Google 65%.

In a few days from now, after the US Government approves the deal, you will see Bing on Yahoo! Both Steve Ballmer and Carol Bartz seem to be very happy about the deal and the notion of taking on the Search Giant, Google. I read the Bio of Carol and found that she was the lady who turned Autodesk around and made it into a profit making company. And I also understand that Steve lived Microsoft his entire career. Both are capable people. But I am wondering, what good can come out of Microsoft Yahoo Alliance!

28% in user share is not good enough to beat Google right now. But Yahoo and Microsoft can now focus more on their core competencies. Microsoft can focus on Improving Search. Yahoo can focus on their Social Networking Ideas. I think we are going to see a new era of Internet Search and Social Networking, as this alliance grows for the next ten years.

And I also think we are going to hear more from Google. In an attempt to strengthen their hold on the market, I hope they are going to come up with more search features.

July 30, 2009

Page Hunt: Microsoft tries to get Human Computation Service for Free!

Microsoft is trying all that it can to make Bing popular. The latest from its arsenal is Page Hunt. Some say it is a 'clever' twist on 'human computation', meaning to say it is a twist on Google and their Quality Raters. It seems that Google employs people to rate the Search Results it generates and now Microsoft and their Raman Chandrashekhar have come up with Page Hunt [Is the name another twist on Google's Page Rank?] to help Bing rate Search Results.

For starters, Page Hunt is a game! I don't know if you can call it a game or getting your work done by someone else free of cost!

Microsoft says:
This Game is like search in reverse: you are shown a web page which you have to hunt down using queries sent to Live Search.
 Now, the idea is very simple. When you go to the Page Hunt page, you are shown a web page. All that you got to do is type the keyword you will use to reach the page shown to you and hit Bing! If you use one of the most popular keywords used by Bing Users to reach the page in question, you get points. If you don't, you don't get points. Which ever way Microsoft gets what it wants! It will know what key word you will use to reach the page in question.

I tried the Game and let me tell you it was quiet boring. To me it looked more like work than play. Now, helping Microsoft fine tune their Search Engine is serious work. I think they should consider paying people for doing that. After all none of their products are given for free to users and even if they give it free of cost, they do so because they have a competitor. I am sure if Google were ran by Microsoft, we would be paying to use it. Now that Google is there, Microsoft offers Bing for free. And everyone knows that is done with a hope of capturing the on-line advertisement market. The better the search results on Bing, Microsoft hopes that more people will start using it and that means more income from 'sponsored links.' Since Bing is about Revenue Generation, it is not fair that Microsoft seeks to fine tune their search results for free. They should be paying the Page Hunt Users like Google pays its Quality Raters.

Off the Topic 1: As I was reading Microsoft's write-up on Page Hunt on Raman's page, I found Page Hunt connected to Live Search and not Bing. [On Page Hunt Page I see Bing and not Live Search] Is Microsoft still confused if it is Live Search or Bing they go for?

Off the Topic 2: I am just wondering, is not 'Human Computation' equal to fooling people? All the while I was thinking Google Search is completely automated and I was always amazed at this. And now that I know it is human beings who are weeding out bad results, Google's Search Technology does not sound great any more? On the contrary, please read what signor john wrote in reply to a stub that doubted Google's Search Technology:

GoogleGuy (when he used to hang out here) once explained that quality evaluators are used for benchmarking purposes. In other words, the 10,000 evaluators aren't there to weed out sites per se; their job is to help train Google's "black box" by identifying examples of sites that need weeding.
As for why Google would have to keep creating new filters at this stage of the game, I'd say it's because:
1) Search technology doesn't stand still.
2) SEO doesn't stand still, either. 

July 24, 2009

OneLook: Meanings from everywhere!

I agree. There is nothing better than Google Definition Search when it comes to finding the meaning of a word. Type "Define" and the word Google will find the meaning out for you. Type "~" and the word, Google will find synonyms for you. This is one feature I use usually, as I am often short of words or find it difficult to understand some words. However, like I said in one of my posts earlier, Google has a structural limitation which sometimes makes it difficult to understand the word I want to understand. Google, as I understand, gleans meanings automatically and sometimes this makes it give you only half of the definition. What do I do with half of a definition? Obviously, I have to look elsewhere for the meaning. Now, how do I decide where I look from the hundred thousands of links Google gives me? That's how I stumbled upon OneLook.
OneLook is an Answer Engine. I think they link to Digital Dictionaries from every nook and corner of the world, gather meaning when you ask and present in a very structured manner. [So Wolfram|Alpha is not the first one to do that?] It gives you a categorized list of links so that you can look up the meaning based on what Category you have in mind. For example you search the word 'Home' The word 'Home' means different things in sports and computers. So when OneLook gives you the list and links to different dictionaries you can search based on what you have in mind at that point of time. If you have sports in your mind you can click on links under Sports Category. If you have computer in your mind, you can click on links under Computer Category.
The best thing I liked about OneLook is, whenever and where ever possible they give you the meanings. On one side of the search page they usually have Quick Definitions. These definitions come with sentences to show you how they are used in different contexts. Sometimes they give you a link to how the word is pronounced and sometimes how the word originated. They also have, usually, a list of similar words.

July 13, 2009

Wolfram|Alpha: Out here to beat Google?

Ever since Google was launched, web thinks mostly about the best way to organize and retrieve information. Reading the mind of the one who engages in search and presenting him the information that he is looking for as quickly as possible was the goal so far. Now, I think, Search Engine Gurus are thinking about how to give information in a more systematic manner. So far, Google does it almost perfectly well. Sometimes, I feel, Google knows what I am searching for and gives me information as systematically as possible, within their structural limitations.

What do I mean by structural limitations? Google Search is all about giving you a list of sites relevant to your search. You got to click on the links on the search page and go to the site. Sometimes, it is very difficult to find the information on the page, even though Google told you it is there and you are sure it is there. However, these days Google gives you this information in a more structured and organized manner. For example, if you are looking for the meaning of a word, usually the first result on the Search Page would be the exact meaning of the word. You don't have to go and look up the information on another page. But, still Google is Google and a list of websites relevant to your search!!

I always used to wonder when Google will evolve into a site that will help me retrieve information the way Wolfram|Alpha is now promising to deliver. Enter any date! Enter any city! Enter any two stocks! Enter any two first names! Wolfram|Alpha tells you! And the guys claim that you are going to get the information in a very structured manner. And they deliver to some extend.


Excited about this whole idea of structured information, I straight away went and searched for 'Mangalore', the city where I live. Wolfram|Alpha gave me most of the things I would want to know about Mangalore. It tells me that the city is called Mangaluru. It tells me, the city is in Karnataka. It shows me a map and tells me where exactly the city is. It also gave me the local time, date and the weather. It told me that 417387 people live there. What more do you want to know about Mangalore? If you are not satisfied with the information you got, Wolfram|Alpha gives you a link to Wikipedia where it is written about Mangalore.

I also wanted to see what information it could give me on August 15, 1947. [Now, for those of you who have no clue why I bother about August 15, let me tell you - India got independence on that day!] Wolfram|Alpha tells me August 15, 1947 was a Friday. It was exactly 61 years, 10 months and 29 days from today. It also gives you an option to search it out on the web. I clicked the link that would take me to more information on August 15. Great! Do you know where it took me to? It took me straight to Google. Yes, Google!

Now, what does that tell us? Does it tell us that WolframAlpha is not competing with Google? Wikipedia calls Wolfram|Alpha an Answer Engine and not a Search Engine. "It is an on-line service that answers factual queries directly by computing the answer from structured data, rather than providing a list of documents or web pages that might contain the answer as a search engine might." Stephen Wolfram released it to public on May 15, 2009. Wolfram|Alpha calls themselves 'computational knowledge engine'. Knowledge Engine or Answer Engine, do they mean they are one step ahead of Google? I think they are not. First of all, I don't think they gather knowledge like Google does. They are giving you information from their knowledge data base mostly. And the fact that Wolfram|Alpha links Google to its site and uses it to gather additional information on a query tells me that it is more about on-line cooperation than competition.

I think Wolfram|Alpha has a long way to go! Currently, in a Wikipedia Style, they are encouraging people to become Volunteer Curators for the site and help them check available data and add data and enable an understanding of new ways in which people seek knowledge. There are many queries to which Wolfram|Alpha says: "Wolfram|Alpha is not sure what to do with your input." But this doesn't disappoint me and I perfectly understand the situation. Wolfram|Alpha is yet to perfect ways in which it gathers and presents information. So far, it does a good job with what it has. And I know, it is going to be better as they progress and people join hands with Wolfram|Alpha as Volunteer Curators.

How to prevent ad-type.google.com from ruining your reading experience?

Ad-type.google.com is a nuisance. It is the meanest possible click fraud one would ever encounter online. It affects your browsing experien...