Friday, 19 December 2014

Is electronic voting a good idea?

Here's a not too geeky explanation of why it is impossible to create an electronic voting system which is trustworthy.


https://www.youtube.com/watch?v=w3_0x6oaDmI&feature=youtu.be

An excellent video. He does an exceptional job of explaining in relatively non-geek terms why it isn't possible to have a trustworthy electronic voting system.

That said, I think that there's yet another problem with electronic voting which is a different aspect of the trust problem. From my perspective, one of the key features of a conventional paper ballot system where humans do all the counting is that it is possible for pretty much any concerned citizen to look at the system carefully and convince themselves that it is trustworthy in the sense that nobody will be able to see how they voted or be able to fake the results or any number of possible things that a concerned citizen might be concerned about.

Now let's assume that you have an electronic voting system that actually is perfectly secure (remember, I said "let's assume") and is actually worthy of our trust. The only citizens who are in a position to actually convince themselves that the electronic voting system is trustworthy are a tiny handful of geeks who are in a position to actually inspect the system. The only thing that an arbitrary concerned citizen can do, assuming that they are not one of this tiny handful of geeks, is ask said tiny handful of geeks if the system is trustworthy and then try to trust the answer they get (obviously, if they actually do trust the answer that they get then they are fools since nobody in their right mind would accept an answer along the lines of "trust me, I'm a geek and I happen to know that the system is trustworthy").

The bottom line is that even if their electronic voting system actually is trustworthy (an admirable yet impossible goal as Computerphile explains in this video), only a concerned citizen who is a complete fool would ever trust that their electronic voting system is trustworthy.

As it turns out, this conclusion leads directly to yet another problem with electronic voting - since any concerned citizen with half a brain will realize that they cannot trust the electronic voting system, the fact that the electronic voting system is essentially being forced on them will breed distrust and cynicism in their democratic system as a whole. This is fatal to any democratic system because, quite frankly, even if it is possible for a concerned citizen to truly and properly convince themselves that the system is trustworthy, the vast majority of citizens actually have faith in the system only because they have faith in the system (i.e. their faith is circular and illogical and unfounded but it is all that they have to work with).

Are photographs works of art?

A photograph recently sold for the astonishing sum of $6.5 million dollars. Here's an article which discusses the sale and which essentially concludes that photographs are not art.


For what it is worth, I quite like the photo described in the article. There is also no question in my mind that this photograph and many other photographs are works of art. While easier to create than a painting or a sculpture of similar compositional complexity, there is no question in my mind that a lot of skill goes into creating a great photograph.

That said, I am uncomfortable with the notion of any photograph (other than a few very special cases that I will get to shortly) having a significant value. The reason for this discomfort is that when one buys a print of a photograph, it is not only impossible to know if there are other identical prints in existence, it is also usefully impossible to know if additional identical copies of the print can be made.

Another way of looking at the issue is that a work of art by a famous artist is valuable in significant part because it is exceptionally rare. In fact, the work of art is so rare that there is often exactly one instance of the work of are in existence in the entire world. Furthermore, even if the artist is still alive and still has literally everything that they need (the scene/model, the equipment, the paint, the lighting, etc., etc., etc.), it is generally if not essentially always impossible for the artist to create another identical copy of the work of art. Even when it is possible for the artist to create subsequent copies which are identical in all the ways that even the most expert expert could use to figure out which copy is the original (i.e. which copy was created first) then there will still only ever be a relative handful of such identical copies. Consequently, the person paying a large sum of money for a conventional work of art has some assurance that the work that they are buying will remain relatively if not truly unique / original and thus preserve much of its value.

In contrast, photographs, as a rule, simply do not have this same property.

There are a few exceptions. For example, if one could purchase the oldest photograph ever taken then one would have in their possession a physical artifact which could be demonstrated to be THE original oldest photograph ever taken because the technology used to create that particular photograph is such that making another copy of the photograph that a seriously expert expert could not identify as being a copy is impractical to the point of being impossible. There are a relative handful of other photographs in existence today with the same traits - that they are effectively impossible to duplicate in such a way that a seriously expert expert would be unable to identify which was the original and which was the copy.

For those interested, the oldest existent photograph in the world was a scene taken by Joseph Nicéphore Niépce in 1826 or 1827. It is a very dark and rather difficult to see image of the view out of an upper floor window of his home in Burgundy, France. See


for more information about Niépce's photo.

Thursday, 11 December 2014

On the United States' failure to "do the right thing" regarding its commission and coverup of officially sanctioned acts of torture . . .

So the country of Jefferson and Franklin, of Washington and Lincoln, of Franklin Roosevelt and others now finds itself orchestrating and defending the coverup of what can only be described as crimes against humanity. Two successive administrations have essentially rationalized these crimes on the grounds that a U.S. President said that they were legal. By defending and covering up these clearly criminal and immoral acts, the U.S. has conceded the moral high ground on any number of issues for years and years to come.

The only good news is that both the infliction of torture and the coverup of torture are defined by international treaty to be crimes of universal jurisdiction. While the vast majority of the hundreds if not thousands of Americans involved in the infliction of torture or the coverup of torture may choose to avoid foreign travel for the rest of their lives, eventually someone will make a mistake and be arrested and put on trial by a country somewhere in the world which has the moral fortitude to call a spade a spade.

The sooner that day comes, the better . . .

http://www.nytimes.com/2014/12/12/opinion/after-report-on-cia-torture-no-more-disclosure.html

A custom set of Harry Potter books by Kincső Nagy

A very impressive custom version of the Harry Potter books.

https://www.behance.net/gallery/19617679/J-K-Rowling-Harry-Potter-

This is the work of Hungarian design student Kincső Nagy. It was apparently done as a degree project. No idea if it will ever become commercially available. With luck, we'll be seeing more of this artist's work!

How Marie Tharp proved Alfred Wegener's Theory of Continental Drift

Here's an interesting article that explains how Marie Tharp put together the evidence that proved German geologist Alfred Wegener's theory of continental drift was correct. Pretty cool!

http://mentalfloss.com/article/60481/how-one-womans-discovery-shook-foundations-geology

Thursday, 4 December 2014

MotionExposure.com

This guy does some really cool long exposure photography!

    http://www.motionexposure.com/Galleries

IBM System/370 "hello world" program

A "hello world" program written in IBM System/370 assembler language using the SIOF (start I/O fast release) instruction complete with associated channel program. Output will appear on the IBM 1403 line printer that you've (naturally) got connected to channel 0 at the usual address (i.e. 00E).
In order to avoid a bunch of extraneous details (some of which I don't remember anymore), we'll assume that the system is currently running in Basic Control (BC) mode.
EXAMPLE CSECT
*
* enter supervisor state
*
    %%% left as an exercise for the reader %%%
*
* establish a base register
*
        BALR  12,0
        USING *,12
*
* disable I/O interrupts (save the old mask)
*
        STNSM SYSMASK,0
*
* start the operation
*
        LA    2,CPROGRAM          Get the address of our first CCW
        ST    2,72                set the CAW to point at it
*                                 (EQU statements are for wimps!)
        SIOF  X'00E'              bang!
        BNZ   OOPS                Did it work?
*
* restore the system mask and continue
*
        SSM   SYSMASK
         .
         .
         .
*
* Something went wrong.  Just sit and stare at our navel.
* Someone will eventually notice that we need help!
*
OOPS    B     OOPS
*
* Our channel program
* (longer than it needs to be if memory serves)
*
* Sends "Hello World!" to a 1403 line printer

CPROGRAM DS 0D           Force double-word alignment

* first CCW - skip to channel 0 (i.e. the top of the next page)
* (does anyone remember if "Skip to channel 0" will take data as well?)

        DC    X'83'      Skip to channel 0 (assumes 1403 line printer)
        DC    AL3(0)     no data
        DC    X'60'      Command Chaining and Suppress Length Indication
        DC    X'00'      unused (always set to 0)
        DC    AL2(0)     no length

* second CCW - send the message text
        DC    X'09'      Write text and then advance to next line
        DC    AL3(HELLO) address of the data
        DC    X'20'      Suppress Length Indication
        DC    X'00'      unused (always set to 0)
        DC    AL2(LHELLO) length of message
*
* Our "Hello world!" message
*
HELLO   DC    C'Hello world!'
LHELLO  EQU   *-HELLO
*
* Somewhere to save the system mask
* (not re-entrant but you can't have everything)
*
SYSMASK DS    X
*
* Done
*
        END
Unfortunately, I wasn't able to convince anyone to lend me their mainframe so that I could test the program although I'm pretty sure that it would work.
Yes, I know - BC mode is also for wimps.