Web Hoaxes and related thoughts....

I read an article on yahoo titled "Top 25 Web Hoaxes and Pranks". It is an interesting read though repetitive at places because it is highly unlikely that being on internet you have not seen one of such emails already. These hoaxes range from 'last picture taken before WTC collapse' to 'someone being terminally ill' email. Infact, I thought some of these mails to be true (though I never forwarded them) before I looked at this article, specially the ones that showed extraordinary pictures (like the iceberg one or the Africa one). These are just 25 and there are heck more that lot of us would have received and fortunately or unfortunately which are not as popular so that they make it to PC World's top 25 list :).

I remember when I used to receive such mails in India, they always used to end with some superstition advice. That leads me to the question whether the content of a web hoax exhibit some geographic locality? And by that, I mean: are the kind of hoaxes you receive in India different from the ones in United States? Yeah I agree some of them are global and have nothing to do with a particular location. But, it makes sort of intuitive sense that since the main reason behind sending such mails is to reach as many inboxes as possible, so exploiting the local culture (or beliefs) is the first (and easy) step towards accomplishing this goal. For example: in India, given the enormous belief in superstitions, you most likely would have seen something like the following at the end of the mail:

"If you send this now to 40 people, good luck will come to you now,
if you send it to 20 people, good luck will come to you tomorrow"

or some such stuff. Now this might not be happening in India alone, but I am just writing a personal experience.

Another surprising bit to this saga is that generally these mails are not from unknown users i.e. someone in your friend-list forwards them to you. So, although the content is spam but because of the ignorance of the sender, it doesn’t count as one. Even more surprising is the fact that young and educated people who look at these messages believe them enough to send it to their other 'n' friends, mostly without giving a thought about magnanimous amount of 'useless' network traffic these messages can generate. Strange Enough!!

A simple practice that can help curb such messages: even if the email content looks oh so real, the first thing to do before forwarding it to other people is to confirm the veracity of the message. This can be done by simple googling or better still checking at the Urban legends website, and if the mail is indeed a hoax (and you have truck loads of time) possibly replying the sender back. Lets be responsible on internet.....

ITs Google all the way!!

I had a chance to attend a computer science technical conference recently. Although topic of the conference was data mining and it is not my research area but I thought I will get good exposure to the way the conferences work in US, so I chose to attend. My main goal was to attend the keynote talks which I thought are very high-level and easy to absorb for an outsider. I was indeed correct at the end of the conference and I liked almost all the keynote addresses. I will not venture here into technicalities of the topics of various talks, but I want to share some observations I made, specially about the talk given by one of the Google's researchers.

The topic of this talk was "Next generation Algorithmic challenges in Google". This was an interesting talk, not that I got all the stuff that was said but overall it was interesting and not completely OHT. Cutting short the context and coming to the portion of the talk, I really wanted to talk about, starts at the beginning of Q n A session. Some of the questions being related to machine learning were really out of my grasp in their entirety, the questions not related to the talk (but which were asked) is the topic of interest in this post.

These questions were related to the Google's mission statement (which centers around providing universal access to information) that was read out by the speaker during the beginning of the talk. Someone questioned something to the effect of - will you make information in my wallet universally accessible. To this, speaker replied that they provide tools and the rest is upto the user. Attached was another question, which inevitably pointed towards Google's policies in China, though it was asked in a very indirect way. To this speaker replied, china is a big country and some presence is better than none at all. (Of course, what is the point of asking such a question, after all Google is a business and they have to make profits to stay alive and it was their business decision. The kind of work they are doing might have different political implications in different countries governed by the local laws and you have to abide by them.)

Anyhow, now picture this: a person from a research lab comes to a technical conference to talk about algorithmic challenges and is asked questions about Google's policies. My question is, first of all, does asking such questions make sense? specially when speaker just talked about algorithms for the last one hour. Its really not a press conference where a company spokesperson is briefing about company policies etc., in which case these questions might make perfect sense. This happened not only at this talk but various other Google talks in the past that I attended. It surprises me why people bring the same topic again and again even though the talk was about core computer science. A better way to put such questions might relate to addressing privacy/security/copyright concerns when making the information universally accessible but then again they have to somewhat relate to speaker's talk and not just appear out of the blue.

Nonetheless, I do not know why but after coming out of Google's talks, I always get a feeling that everyone wants to get in there (and of course why not). I might be generalizing a lot but the kind of questions asked by people (who might be from other companies :)) makes me feel that. Good for Google though.