Main

May 14, 2008

Fire near Office

fire.jpg
The building next to my office burnt down!

May 08, 2008

Cover story on Protothreads

0508esdCover.gif In my previous post on Protothreads, I gave an overview of Protothreads and talked about how working in Sweden was fun - thanks to Adam Dunkels and Thiemo Voigt. Protothreads got featured on the cover of this month's Embedded Systems Magazine (May 2008 Vol. 21 No. 5). The cover story is by Michael Dorin and is titled "Building "instant-up" real-time operating systems". Check it out!

February 06, 2008

Koen Featured in ICT Mag

Our group leader Koen Langendoen, recently got featured on the cover of a dutch ICT magazine. So apparently the sensornets research works of our group (like the infamous potato deployment) are not going unnoticed. koen_ICT.jpg

January 27, 2008

Lectures at Abdus Salam Center

abdus.JPG I will be at the Abdus Salam Center for Theoretical Physics (ICTP), in Trieste, Italy next month talking about wireless meshes. Abdus Salam is the first Muslim and the first (and only) Pakistani Nobel Laureate. He unified two (electromagnetism and weak interaction) of the four fundamental forces of nature, which is still the latest step towards the unified description of all four physical forces. ICTP was founded by Salam in 1964.
Salam wanted to bridge the scientific and economic gap between the rich and poor populations of the world - something that we are trying to do through Dritte. ICTP, following on its founder's vision, is organizing a School on Wireless Networking for Development, where i will be giving lectures on Feb 24th and 25th. I am excited about this trip, as I think that Abdus Salam's vision will inspire me to focus on fundamental problems in scientific research and to try and bring the benefits of science to the poor.

July 09, 2007

The Original Google Storage

I am spending the summer at Stanford working with Phil Levis. On my first day at the Gates Hall phil pointed me to something interesting in the basement of the building - the original Google storage (picture on right). Its fascinating to think that the multi-billion dollar search giant came out of THIS machine. n511689122_119153_7454.jpg

June 14, 2007

WSNblog.com

wsnblog.JPG
In case you were wondering why I was quiet for a while, most of my entries related to sensor networks were posted on WSNblog.com instead. Our wsnblog.com recently got featured on M2M magazine (although they got some information wrong about us). Joe Polastre has a decent summary of this issue of M2M magazine on Moteiv's blog. Some comments about this issue of M2M magazine would be:
  • I am keeping an eye out for ArchRock's IPv6 solutions - lets see how the industry responds to the idea of running IPv6 on sensornets, from a research point of view I have my reservations with that approach.
  • I recently attended a talk by Bob Metcalf at Cambridge, MA and it seems to me that Bob and Ember think that they have found the "killer app" in energy management - wonder how Metcalf's law would scale in that market.

February 13, 2007

Eric Brewer and Inktomi

Computer History Museum has this talk by Eric Brewer (UC Berkeley) where he goes over the rise and fall of Inktomi. For anyone remotely interested in the history of the Internet, case studies of startup companies, insight into the Internet bubble, or technology research in general; I would highly recommend watching this talk.

Inktomi was founded in 1996, by Eric and a Berkeley grad student, and went onto the Nasdaq 100 before it was bought by Yahoo! in March 2003. Yahoo search and MSN search are still powered by the Inktomi engine. Eric actually got quite emotional at times talking about Intkomi and I can clearly relate to why.

At the end he talks about the time when his 10% shares in Inktomi were worth a billion USD and he got interested in doing something for the "third world countries". Eric is the program co-chair for the ACM SIGCOMM NSDR workshop, Japan, Aug 2007 that I am organizing (along with Umar Saif). NSDR'07 is specifically aimed at bringing the benefits of networking technologies to the third world.

Below is the GoogleVideo embedded video of the talk:

August 03, 2006

Erdős Number

Paul_Erdös.jpeg I found out about Erdos Numbers from my alma mater faculty Arif Zaman - the father of Random Number Generation (Rand() in programming languages). After seeing the list of famous people with finite Erdos numbers I got curious about my own Erdos number (if it was not infinite). Let me explain Erdos numbers a bit;
  • Paul Erdős is the only person with an Erdos number 0
  • Anyone who has published a paper with Paul Erdős has Erdos number 1
  • So, Arif Zaman has an Erdos number 4 because he published a paper with George Marsaglia who published a paper with George P. H. Styan who published a paper with Paul Erdős.
The Erdos number is basically a measure of research collaborations taking Paul Erdős as the center. The Erdos Number project found that all winners of the prestigious Mathematics Awards (e.g. the Fields Medal) have finite Erdos numbers and also most Nobel laureates have finite Erdos numbers as well. In other words this shows that the research circles are smaller than what we imagine them to be e.g.
  • Albert Einstein (Physics) has Erdos 2,
  • John Nash (Economics) has Erdos 4,
  • Stephen Hawking (Cosmology) has Erdos 4
The online collaborative distance search page was not so useful in calculating my Erdos number primarily because I am not a Mathematician or a Theoretical Computer Scientist. So I had to manually calculate my Erdos paths and find the limit on the Erdos number (I had to find manual paths to someone recognized by the Erdos project database and reduce the overall path length as much as possible). Here are some paths I found (listing only one example path for each Erdos number): So my Erdos number is 5, which means that from my co-author index on the DBLP Bibliography Server after 5 clicks you should be able to see Paul Erdös.

June 19, 2006

MIT TechReview TR 10

techreview.gif The MIT Technology Review "10 Emerging Technologies" list (TR 10) for 2006 has been released. This year the list is something like:
  • Comparative interactomics
  • Nanomedicine
  • Epigenetics
  • Cognitive radio
  • Nuclear reprogramming
  • Diffusion tensor imaging
  • Universal authentication
  • Nanobiomechanics
  • Pervasive wireless
  • Stretchable silicon
David Culler appears again this year in TR 10 (about wireless sensor networks) under "Pervasive wireless". Some highlights from previous years would be Hari Balakrishnan (2004) for "Distributed storage" and David Culler (2003) for "Wireless sensor networks". 2003 is the year I was finishing my BS and was just moving into sensor networks research - needless to say that I was much excited on seeing sensor networks at No.1 of TR 10 of 2003. For archival purposes I am giving links to the TR 10 articles (as PDF) starting from year 2003. I cant seem to find a PDF for year 2005 so that year is a link to TechReview web article; if anyone has the original PDF for year 2005 you can feel free to email me :-)

The MIT TechReview TR 10 archive:

April 10, 2006

FOB: Fresh-off-the-boat OR Fulbright Scholarship

fulbright_logo.jpg
No offense to USEFP (United States Educational Foundation in Pakistan) or the Fulbright Program (they indeed are providing a great service to students with their scholarships) but, in my view, students should think twice before relying on a scholarship grant like USEFP Fulbright. So, here I am going to try and shed light on some points that you should consider before you get too excited about the Fulbright scholarship (these points are specific to PhD Fulbright grants via USEFP):


  • If you are applying to a PhD program in the USA most PhD programs have their own funding and in most cases if a university is giving you admission they would offer you funding as well. So the Fulbright scholarship makes sense for Masters study (because Masters programs are generally self-financed). For a PhD study, the Fulbright scholarship helps in cases where the university wants to admit you but they are short on funding.
  • You give a "list of preferred institutes" to USEFP and IIE and after that they tell you to sit back and relax. You are not allowed (or discouraged) to apply to US universities on your own. The way this works is that IIE doesn't notify you about universities they applied to (on your behalf) untill February and by February the admission deadlines of all good schools expire. So you just sit there thinking
    "Should I apply to UC Berkeley? Oh but what if IIE also sent my application there. The folks at UCB would think I am an idiot who sent two applications. So lets not apply!"

  • At the end of February there are two things that can happen:
    1. you get a notification from IIE that they applied to the school of your choice (well and good!)
    2. they did not apply to the school of your choice (lets say UC Berkeley) on your behalf because their "experts" don't think that you had a good shot. So they sent the application to University of Disney Land instead.

  • So if you are the "Disney Land" case they would try and convince you how U.of Disney Land is not that bad a university and you should accept the offer while you sit there thinking that:
    "Hey wait a minute! I didn't get rejection from UC Berkeley or CMU, infact my application never got there and CMU etc has a policy to fund all PhD applicants anyway. What were the benefits of the fulbright scholarship again?"

  • If you are not the "Disney Land" case and you are happy with your placement, pretty soon you would receive a notification that HEC is partially funding your PhD study and as the cost of your PhD. Program will be about $250,000 therefore,
    "everyone will be required to post at least a partial bond, that is, a bond for as much as their family owns up to a fixed amount, but not more. At this time you should be identifying the people who will post their property ..."
    Apparently, this bond will be fully redeemable without any cost to you or your family when you have returned to your home-country and served your home-country for five years after you complete your degree. Let me simplify this for you - the minute you accept the Fulbright offer and hand-in your passports to them you have sealed your fate for the next 10 years of your life. And incase you don't comply to the bond (lets say you feel like going to Europe for research) they would send your parents a bill that would have a lot of zeros on it.
  • Also, keep in mind that the "experts" at IIE deal with students from various different fields and deal with a lot of different universities and programs. They are going to evaluate your application (in a Microsoft Excel-style) GRE + GPA = high rank fashion. They would not know what SIGCOMM conference is and how and why is it different from WMSCI conference (computer networks researchers and faculty at places like MIT etc. consider having ONE publication in SIGCOMM conference a career-goal whereas WMSCI is a conference where MIT PDOS students pulled a prank and their junk paper generated from a random computer program was accepted. Most probably IIE people would be unable to differenciate between the north and south pole of computer systems research and publications and you cant really blame them for that)

Hope this information clarifies some issues for future PhD Fulbright applicants.

October 30, 2005

Acknowledgments

I owe eternal gratitude to Dr. Zartash Uzmi and Dr. Umar Saif for their constant support. I got interested in computer systems research because of Umar Saif and over the years he has been a mentor and helped shape my career to whatever it is and whatever it is shaping out to be. Zartash Uzmi dedicated countless time and energy to advise and guide me not only in work-related matters but in life-in-general as well.

This by-far is not the end of my acknolwedgments list but its amazing to see how people dont even say "Thank You" to those who help change their lives.