• Does it work? UniWB to help determining exposure.

    01Nov
    Categories: Geeking, Photo processing, Science Comments: 0

    — in progress post — (hope to finish this week) — Here are my initial thoughts:

    I tried UniWB to improve determining exposure when shooting RAW using the RGB histograms.

    The idea behind uniWB is that it addresses the problem with the limitations of the histograms produced from shooting in “normal” mode (not using UniWB). The normal histograms show clipped channels, over-exposed areas, blown whites with no detail as right leaning histograms even if they do have detail in the actual RAW. This is because the histograms are determined using the in-camera generated JPG – which has a smaller dynamic range than the original real RAW.

    So far, I don’t think I like it, I prefer using spot metering, with the in-camera meter and then the Zone system. I am going to post more details about this approach and my results in my blog shortly.

    Tags:

  • The Problems with Voting Machines

    11Oct
    Categories: Science Comments: 0



     
    Barbara Simons on left [Photo by Peg Skorpinski]

    Diebold Touchscreen Voting Machine
     A Diebold Touchscreen Voting Machine.  

    I recently went to a talk on voting machines by Barbara Simons, a computer security expert from Stanford who is spearheading an investigation of the problems of these machines. I also had a one-on-one talk with her on this upcoming election. She has been involved since at least the 2004 election.

    Here are my thoughts resulting from our discussion and her talk: This is an important issue and it is of-course not a partisan problem as it affects both sides and our freedom to vote and impact the election. There are several problems:

    • Monopoly:  Only one company is in control of these machines  – no competition to drive up the quality of the hardware of the machines or the software that drives the machines.
    • Unverifiable Results: The software used is proprietary to the company that supply the machines: less control, less opportunities to verify the results or that the software works correctly. Worse the company refuses to have their software evaluated and verified by computer experts. The machines are also paperless which magnifies the problem, i.e., recounts are difficult if not impossible as there are no verifiable paper records. There are numerous examples of the problems caused with these unverifiable machines.
    • Security Holes: The software is plagued with errors and security holes. The company inadvertently released a beta version of their software in 2004 (I believe by an employee / software developer) on a bulletin board. Computer scientists found serious problems of this version and demonstrated that the software is easily hacked due to poor software practices.

    I believe that a solution is to publically release the software; releasing it allows it to be scrutinized by experts and consequently improves security and reliability. Promote competition in developing the technology of the machines: Allowing competition in both producing the machines (hardware) and software improves the quality of the software and machines. And I echo David Dill’s of Stanford sentiment when he said, “Our democracy depends on our having secure, reliable, and accurate elections.”

     

    Tags: , ,

  • The Universe, Understood.

    10Sep
    Categories: Kids, Science Comments: 1


    Photo: Copyright CERN
    public.web.cern.ch/public/

    Imagine, any day now we may make a leap in understanding the universe better.

    The CERN particle collider, the Large Hadron Collider (LHC), just had his first successful test in getting ready to smash particles.

    Particle colliders have been around for a while (at least since the eighties)- and we got our own colliders here in the US, e.g., Stanford’s SLAC (a linear collider) and there are some circular ones – not sure where they are located (but one I believe is Fermilab in Illinois).

    What is different about Hadron, in CERN, is that it can produce enough energy to potentially generate a Higgs boson (if it exist)– we don’t know whether it exists or not. It is a weightless particle, or a particle with no mass.

    The boson is necessary to explain a lot of theories in physics, for example you may know about gravity and you probably know about mass and energy (Einstein), this boson would unify the gap between gravity, mass and energy. Cool and exiting stuff!

    What about the danger?

    Some people think it is dangerous because it may create a miniature black hole, and some say it will decay while others say it will suck in other mass and grow and ultimately we all get sucked in. However, Stephen Hawkins, you may remember him from his special appearance on Star Trek, says that the particle collisions even greater than the one anticipated happens millions of times in the earth’s atmosphere so there is nothing to worry about.

    I also would like to add that Hawkins also placed a $100 bet that the experiment will not produce a boson.

    Tags: ,

Blogroll:

Meta:

Recent Posts