30 October 2014

Edith Widder: The weird and wonderful world of bioluminescence

Fact: most life on earth can produce light. Widder present her research of these creatures with a captivating, and often humorous, collection of videos.

Edith Widder: The weird and wonderful world of bioluminescence

Bonus: You can purchase your own collection of bagged dinos (like the ones in her bottle) to poke for you own blue-light special amusement from Sunnyside Sea Farms. I highly recommend these and love to poke mine dinos each night before I go to bed. If you watched Life of Pi then you may remember these scenes:


Those are dinos.

15 October 2014

"Rocks Breaks Scissors" by William Poundstone

Humans love to think we know what random is and that we can be random when we want to. As Poundstone demonstrates, we are far less random than we imagine and, interestingly enough, we have trouble identifying randomness when we see it. For example, consider the following string (taken from the book):
--||-|--|-|-|-|-|---|-|-|-|-|-|||-|-||-|-|-|-||-||
Mostly people would consider this to be fairly random sequence (with "-" and "|" having an equal chance of appearing). It is not. The "|" has a 75% chance of appearing. A truly random sequence might look like this:
|----|-|||-|||-||---||------|||--||-------|--||-|-
One reason for our poor randomness skills is that we imagine that things should be uniform in their randomness. The first sequence looks more random because the two symbols alternate more frequently. Gamblers often experience this problem: a run has held steady so they bet that a contrary pattern must appear. True randomness means that each outcome has an equal chance of occurring during each iteration of an event. That is, seven heads in a row does not change the 50/50 odds that the next coin flip will be heads too.

The first half of the book is filled with quick rules and explanations for examining randomness in a variety of situations.

Interesting tidbit: When vetting numbers, look at the second digits. They usually should be fairly well distributed if they are natural. Odd spikes of over usage are signs that the numbers may have been tampered with.

07 October 2014

Yasheng Huang: Does democracy stifle economic growth?

Huang argues that democracy does fuel economic growth but such growth can proceed much quicker when helped by spurts of authoritarianism.

Yasheng Huang: Does democracy stifle economic growth?

01 October 2014

Why you should not try to catch the IT guy

The tawny haired man stumbled through the door, just missing the aluminum frame of the glass door as it pulled open. He dragged a foot as he made his way to the reception desk, all the while clutching at his chest.

“Can I help you?” The receptionist asked in a cool tone. The man looked put generally well put together and in no need of hospital services.

“I need help,” the man said in a distant voice.

She looked him over. His suit was well pressed, lint free and his lavender shirt had a large stain running down it but was otherwise in good condition. There was a bit of a tear near the top of the stain. She froze. It was not a stain but blood running down his shirt.

“Wha… what happened to you?”

“I need help,” he repeated then collapsed on the floor.

The receptionist switched into auto-pilot mode and swooped to his side.

“Here,” she said grabbing his hand and pushing it to his wound. “Keep pushing here.”

“I need help,” the man uttered again, though he complied with the order.

The woman scanned the hallway for a moment before locking eyes with a doctor. “Probably the closest one,” he thought before calling out. “Doctor, here, now!”

The doctor grabbed his pockets to keep them stable then began jogging. He scanned the scene before him: the receptionist pressing the man’s hand to his chest, the man himself covered in blood and barely conscious of what was happening. A hand grabbed the doctor’s his arm and broke his focus. He skidded to a halt.

“Doctor,” an old, frail man said. “When can I expect my results back?”

The doctor stared into the old man’s eyes in disbelief.

“Um, I don’t know.” He muttered. “I have to help the man in the lobby.”

“But I’ve been waiting.”

“Are you dying?”

“Well, no, but I want my test results.”

“Mr. Jones,” the doctor said firmly while detaching himself from the man’s arm. “There is man in the lobby who will die if I don’t help him right now.”

“Now see here young man,” the old man started angrily. “I have given this hospital a small fortune in my time. They’ve even named a wing after me. I am entitled to better service than this.”

His angry words were wasted though, the doctor was already down the hallway working on the stab wound.

--

While most reasons for an IT worker to visit onsite do not involve keeping humans alive, they do in many ways involve keeping the business alive. Few people would behave as Mr. Jones when human life is involved however, countless are even more indignant when the business life is on the line.

To fully understand what it means when an IT worker walks onsite to fix an issue let me share a bit about how most IT people like to work: by remote.

One of the most potent blessings of the modern era is remote control. Today’s technologies allow for an IT worker to remote almost anything. Computers, servers, phones, cellphones, printers, copiers, fax machine and just about everything else in a business. At first glance, it may seem that working via remote is the lazy route, and it is, to an extent.

Working by remote means that issues can be corrected from the comfort of a desk. More importantly, it means that IT persons can stay highly efficient. Many location are physically large spaces. If workers had to walk to each computer to run an update or make a change, something that would take 2 minutes by remote suddenly takes all day to complete. But the real kicker comes when there is a problem.

Contrary to popular belief, most IT people do not have endless stores of solutions to every conceivable technical problem stashed in their heads. Much like doctors, we store a collection of the most frequent and reasonable scenarios and leave the rest to our resources to remember. For doctors those resources include textbooks, encyclopedias and such. For IT workers most of these resources are online in forums, knowledgebases and web searches. Online resources required online access which requires a functional device, preferably a computer.

In the old days (before effective remotes) we would have to leave a misbehaving device to find a working device to do our research on then return to the original device. If our research provided us with the correct solution the first time around (a rare thing) then we were all set. Most often however, we would have to make several trips back and forth.

Today, using a remote control, we can simultaneously remote a misbehaving terminal and research a solution from the same computer. Usually, one configured and setup just the way we like it (no one really enjoys using a computer littered with pictures of someone else’s family—or worse, cat—their Facebook and whatever odd sites they deemed worthy to bookmark). This research method maintain efficiency and when business computers are down, efficiency is key. Few businesses have a bevy of spare devices; often every hour a device is down is an hour of work lost which will need to be made up later, frequently on overtime.

All of this is to say that remote control work is a very, very good thing for IT workers (who can work faster and more efficiently), businesses (who experience less downtime and resource waste) and end users (who experience less frustration, and quicker response and resolution to issues).

Sometimes, however, things go so wrong that remote controls do not work and correction requires onsite work. Personally, I cringe every time this happens. Most other IT workers do to. Partly because we lose our precious efficiency but mostly because of the business equivalent to Mr. Jones: users who swarm the poor onsite IT worker.

On occasion, these delays are good and the users provide new and valuable information. These occasion are very rare. Instead, most swarms include peppering the IT worker with information they already have, regaling them with stories of their system going down and all the things they tried to bring the system back and how sad they are that things are not working correctly and… Honestly, most of this information is useless and, perchance it is not, a good IT worker will be able to quickly gather the relevant information they need to understand what is happening.

Do not be surprised if the IT worker does not ask you any questions though. Why? Because he already knows that something is wrong. For one, someone already called to complain. For two, if he could have remoted in, he would have. For three, IT workers are part doctor and part animal trainer: they can often tell what is wrong by listening to the whining machines and by poking around a bit. Like a sick wildebeest, servers are better at explaining their woes than the user who cannot get them to work.

Be very wary of the worst offence of all: trying to get the IT worker to look at something else. Unless he is just strolling through looking for work, he is probably headed somewhere to do something that was important enough that he could not do it remotely. Your belaying him causes two problems. He is not getting to what he had already promised to do and there is probably a server somewhere collapsed on the floor, bleeding out while you are whining about a slow email connection. The second is that he is going to struggle to remember your problem. IT workers are humans too, we can only remember so many things at once and your problem is probably not important enough to remember in the face of some other dire situation.

So the next time you spot an IT worker or catch them on the phone: wave, say ‘hi’ and ‘thank you,’ offer them a drink and get them quickly on their way. Oh, and send them an email about that annoying little problem you only remember when you see them. They will be grateful and probably be much more cheery when it comes to helping you in the future.