Archive for May, 2006

Time Travel

Friday, May 26th, 2006

In relation to a recent Beanblog:

When I was somewhere around 10 years old, my folks and I went on a two week vacation to out west to see Yellowstone, Mt. Rushmore, etc etc. Two weeks in the truck with a camper and whatnot. I think I’ve blogged about this before.

Anyway, I was really bored a lot since we were doing things that I didn’t enjoy (nature related activities) plus there was a lot of driving.

So, one day I wrote down all of my theories on time travel. They were very detailed and logically well put, for a ~10 year old. I probably wrote a good 5-6 pages on why certain things would or wouldn’t be possible.

I was very proud of my work.

When we got back from the vacation, I was going to type them up on the computer for posterity. Only, that didn’t happen, because some kind of water like thing spillled all over them and ruined my ability to read them. Also, being home again didn’t give me the desire or boredom to try and rethink/write them. So they were lost forever.

Coincidence? Or possibly, was it some kind of future time police coming back to ensure I didn’t get too far along in my time machine build process?

Voting

Friday, May 26th, 2006

When voting for governmental positions, there are two key things of importance:

1. The vote must be anonymous
2. The accuracy of the vote must be authenticable

With “electronic” voting, these are very two tricky items. In the past, this is something we’ve taken for granted. Voting was done by punching a hole in a card (remember hanging Chads?*). Before that it was done by writing a name on a ballot.

But a big problem with newer electronic voting is that you have no way of knowing that your vote is truly correct. That’s because you push a button and submit your vote, and it whisks away into nana land into the internals of a machine. And while it may tell you that you voted for candidate X, how do you know that when the results are downloaded that you didn’t indeed vote for candidate Y. It could happen accidentally due to a software error or maliciously due to someone tampering with the machine.

If you don’t believe me, do a quick search on “diebold”, makers of many electronic voting machines, and see what people have to say about them. www.blackboxvoting.org is a good place to start.

The fact that there are people who power their toasters with Linux leads me to believe that people are more than capable of figuring out how these voting machines work and manipulating them into doing bad things. And it would be easy to ensure that the audit trail never figured it out.

The easiest solution to this is to print out a small stub that the voter can use to verify who they indeed voted for. This stub goes into a bin and is put there by the voter. This will ensure #2. Granted, there are still ways to circumvent this, but it’s a whole lot more reliable than a black box electronic approach.

But how to ensure #1? The problem is that it’s simply a bad idea to have your vote be able to tie back to you. Think about Iraq 5 years ago when they had “elections” and 99.999999% of the population voted for Saddam. If people’s votes were truly anonymous that probably wouldn’t have happened. And with the sleazeball politicians there are in the world today, even in the US, this type of anonymousness this is still very important.

Here is my proposal for fixing the voting problems:

A voter goes to their precinct to cast their ballot. They register, and once they register their name is crossed off a list. This can be done completely electronically.

They go to a booth and cast their votes on a machine. When they are done, they print out a small piece of paper (think, lottery ticket size) that has a list of the names of the people they are voting for along with a bar code that has the same information encoded in it.

The person looks this over and makes sure they agree with it. If they do, they take it over to a bin and drop it in.

Later, the local auditor uses this when counting the votes. They simply scan all of the bar codes and tally them up. For a recount, the names are listed on the front of the ticket. During the scanning, they will randomly select 1 out of every 100 and make sure the scanned information matches what’s printed on the ticket.

This is accountable, and 100% anonymous. And it sure beats the alternatives.

* hanging chad 0x!= hanging chad rangoon

Ridin’ Dirty

Monday, May 22nd, 2006

On my way home from work last Thursday, I ended up behind an old pickup truck with two people ridin’ dirty.

(more…)

An interview with me

Monday, May 22nd, 2006

KDE.news has posted an interview with me about our company’s business decision to use Linux and KDE in our testing environment. The interview was actually conducted over a year ago, but stagnated due to lack of time to post it. It was revived a few weeks ago and I updated some of the information to say more about what we are doing now.

While it’s KDE centric, note that we make heavy use also of Gentoo Linux, Ruby, QtRuby and Ruby on Rails all for our test stands.

In particular, if anyone in the Ruby community is looking to get some specifics or metrics for our use of Ruby in the enterprise I’d be happy to share that information.

Freshman Year Follies

Thursday, May 18th, 2006

My freshman year of high school was the first year that 9th graders partook in high school in my country. Previously it was done at the middle school level. This meant that both would be 9th and 10th graders moved up at the same time.

(more…)