E M E R G I C . o r g: March 28, 2004 Archives

Sunday, March 28, 2004

On Elections

Considering that both India and the US have elections this year (India in April-May and the US in November), one thing surprises me. Why don't the challengers (the Congress Party in India, and John Kerry in the US) name their complete administrations - the people who would be in various positions of power if they won the elections. Governance is not just done by a single person. Releasing a manifesto is not good enough. I want to know the alternatives - person-for-person. What does this also do is to provide a wider platform for debates as there will be more people who will be discussing specific issues (what falls under them), rather than just one person talking about everything.

This is like the shadow cabinet that exists (?) in the UK. What we need is a shadow cabinet pre-elections.

General | PermaLink | Comments (2)

this will open up a can of worms. People who are named to specific positions will probably not campaign as hard, and the people who are not named will ask why they should bother!

Probably 5% of the people who join politics do it because they want to see change, and create that change, do good, etc. The rest are in it for the money and the power.

I remember reading something about how Incentives & bias in Robert Cialdini's book - Influence (I think or that Charlie Munger's article).

Posted by Prakash S

Elections in India depend on rather palpable things -- the price of onions for instance -- things that mean something to someone who is not literate, certainly not educated. Let us not delude ourselves that a stellar cabinet has anything to do with whether the party will win or not.

Posted by Atanu Dey

Search Engine Flaws

Writes Ramesh Jain:

I feel that these engines are still following a dead-end path. They are pursuing ‘information-centric’ approach. In information centric approach, the model used is that the user comes to the system with a precise question and the system has answer to that question so it provides the answer. This simple model was good for early databases and even for early search systems. Now, when I type a keyword and the system firehoses me with a list of 5 Million items listed 10 on a page, what am I to do except getting frustrated. Advanced queries are each a new query and as every ‘normal user’ knows, they are not much help. So what is fundamentally wrong with these search approaches. Here are somethings that I think are fundamental to search in the rapidly expanding cyberspace where current search engines are starting to choke and are ultimately likely to choke and become useless, opening up the space for a new fresh approach by a graduate student, or a start-up, working somewhere maybe even today.

1. The engines have the simple model of information based on keywords. Use of ontological filtering to reduce the list helps but does not really solve the real problem. The real problem is that a set of keywords are not semantically enough to express the context in which search is being done.

2. In most cases people have ‘multidimensional’ search in mind. The simplest and most common example is seen in LBS (Location Based Services) where a keyword must be combined with location to provide the context. This then gives more meaningful answer to a user. Now LBS is not useful only in the context of mobile phones. The main point is providing dimensionality to search. Dimensionality provides different constraints that further helps in expressing context. Current search engines primarily store a link to the sources containing the keyword. This model is too simplistic to allow and match relations leading to context. By developing more rigorous models and extracting and storing all essential information, search can be more contextual.

3. The search environment based on current approach of specifying keywords and getting a list of pointers in return does not scale-up. People are interested in solutions to their problem, not knowing that 5 Million pages on web contain their keywords. The environment should be more exploration based, rather than query based. In exploration based environment, I will have freedom to slice and dice my results in many different ways to explore and find what I am looking for. Also, this exploration must be multidimensional.

4. Current approach of showing a list is too primitive. We never like to look at a table of stock values over time – we look at charts. Two-dimensional data (stock value, and time) when reduced to a liner representation, theoretically contains the same information but practical is useless as the number of data item increases. After about 100 items, it is for all practical purposes completely useless. We need to develop powerful visualization mechanisms. This is not a simple point. Mathematically we know that when a two-dimensional information is projected onto one-dimensional space (extend it to higher dimension) there is loss of information. Same is true for human mind – in fact that is more true for human mind.

Search Engines | PermaLink | Comments (7)

check out guha's work on TAP DB

Posted by pakiya

The 4th point is pretty good. Search engines can group results under some logical categories (even multidimensional) aided the user to filter out irrelevant content.

So say, for a search on Infosys, they group it under stocks, company information, news articles, products/services, advertisements etc. Ofcourse Multi-dimensional categorisation will again be a big plus. So further they may be grouped into Indian, Internation sites... Commercial sites, weblogs/personal sites... and many such

The Graph theory is pretty inviting and a pictorial representation rather than a textual one can do wonders to the speed to decision making.

Posted by Kshitij Chandan

The search space is really complex. The TAP project and other projects based on semantic web technologies, are trying to make headway but there are some fundamental problems of the AI(Artifical Intelligence) world which still need to solved. An ideal solution is still far from reach. But something will make headway perhaps with the huge effort going on in this space.

I agree with the author that current approach of showing a list of results is too primitive. There have been a number of attempts to visualise information and search results in multiple interpretable ways. But these haven't been able to make much headway. I believe, a clear mix of both the basic and complex interfaces ("Simple yet Powerful") is needed to serve both the basic and advanced users.

Posted by sunil

I think that the problem is probably not where Prof Ramesh Jain puts it. From reading his complaints, it appears that he wants a search engine which would be a mind reader. Or better yet, you come to the search engine not knowing exactly what you want but expect that the search engine will formulate your question for you and then help you with the answer. It is a search engine, for crying out loud. The user has to put in some effort -- such as typing in a set of words that delimit the search. The machine cannot do that. Not just the machine, no human can delimit a search for you.

One cannot blame a dictionary, for instance, if one doesn't quite know what one is looking for in the first place. In that case, the bigger the dictionary, the more frustrating the experience of consulting it. But if one knows what one is looking for, the larger the dictionary, the more likely one is going to have a satisfactory result.

Posted by Atanu Dey

I think Mooter is a good example of visualization of search results.

It is not useful always but they are working in that direction.

Suhit

Posted by Suhit Anantula

First, Thanks Rajesh for posting my thoughts. These thoughts were the result of some popular articles that appeared around that time and suggested that all the major efforts are still along the same direction. In any case, I have written several articles, some of those can be found on my university home page (jain.faculty.gatech.edu) or the best source may be MediaVision Column in IEEE Multimedia. I have several research articles that I’d be happy to send to you if you have serious interest in this area – jain@ece.gatech.edu.

Now more close to the points made in these articles. Let’s think out of the box because if assume that Google is the best and we should follow that direction (incidentally I find Google the best) then we are not looking for solving a problem, but improving an approach. Ontology based approaches, including Guha’s work and all the work in Semantic Web is based on that kind of assumption. As we all know, and one may refresh his memory by reading Language in Thought and Action, that a word in any languge does not have just one well defined static meaning. Meanings change depending on when the dictionary is compiled and has multiple meanings. So Ontology based approaches start helping us by providing some context. But context itself is very ‘dynamic’ and person and situation dependent.

There is a difference between a report and a judgemental piece. A report language is not suppose to deviate from facts. On the other hand a political piece, be it from Gandhi or from Bush, is rarely going to be close to a report. And language has different tools that allow us to do that in very subtle or in not so subtle manner. That’s the reason that once we try to go away from textual search to multimedia search (which based on my knowledge of is going to be a must given our increasing dependence on sensors in all aspects) the current techniques will be totally inadequate.

So my ‘complaints’ (I like to think of my thoughts more as suggestions to improve) were not for the current generation system, but how can we really get to the next generation systems. In DOS age thinking about, how can we get to Windows, or to the next frontier in the battle of the desktops.

Posted by Ramesh Jain

Hi, nice work, if you have the necessary time, please vistit me, you'll find interesting stuff, articles about men health.

Posted by penis enlargement