Crowdsourcing vs. Open Source

So, last night, I got bit on the foot by a mosquito, which itches like crazy, and was keeping me up.

I got started thinking about Wikipedia, like one does in the wee hours of the night, and I came to this conclusion (my thesis): Wikipedia suffers from credibility problems because contributors have vested interest and no vetting.

In distributing a pile of work among a great number of voluntary participants, there are basically two schools – crowdsourcing and open sourcing.

Crowdsourcing, abbreviated, is breaking a big task among a fluid group of anonymous, voluntary participants. One excellent example of crowdsourcing is reCAPTCHA, which utilizes humans’ ability to read distorted text to do character recognition on scanned text that computers have a hard time discerning.

Open Source, abbreviated, is providing information or skills you’ve learned to the collective body. It’s rarely anonymous. Linux, an operating system written collectively by thousands of people over the years, and provided for free to everyone else, is the most common example. Google “linux” and be amazed.

Wikipedia lies somewhere between the two. Wikipedia’s content comes from people who contribute what they’ve learned (or what they believe), with the idea that enough people validating (or correcting) eachothers’ statements will end up with something resembling the truth.

Here’s where the difficulty lies. With crowdsourcing, to use reCAPTCHA as the example, no one cares enough about whether the single word is read incorrectly to try to sabotage the system. There is no bias and nothing to be gained by coercing a group of people into convincing the system that a single word is wrong. With open source, the person doing the work has put in the time to learn the language, the time to learn the distro, and the time to learn whatever piece they’re working on. Essentially, the participants are self-filtering. It would take more than the casual user to become an active participant in the development cycle, write code, and get that code into the final build to cause any kind of harm.

Wikipedia, however, until recently, could be edited by anyone with little to no expectation that their statements would be traced back to them. This led to, obviously, all kinds of abuse, with people who have conflicts of interest making edits to suit their beliefs, employers, etc. The casual user, or a conspiracy of several users, can making their version of the truth the most popular belief.

I don’t know what the solution to this is, but I think that edit moderation would take us a long way towards it.

Whatever happens, I think that documenting the collective intelligence is a phenomenal idea that needs to be pursued vigorously. As long as we’re documenting the collective intelligence, and not the collective ignorance and mis-information.

That’s what Faux News is for.

DNS is a Pain

I went to MacWorld this year and attended a session called “Lucid Systems Administration“. It was a pretty good, though basic, overview of how to maintain sanity as a SysAdmin. A lot of it was pretty common sense stuff – use an organizer, don’t try to memorize everything, devote time to specific goals, prioritize, etc.

I say “common sense” knowing that there are a lot of people in my profession for whom “common sense” is elusive. They’re on “the spectrum,” the saying goes.

So I could see it being helpful for some of them…Lots of them, actually.

One thing that stuck with me was her final “tip”: Check DNS.

For any readers who don’t know what DNS is, it’s the service that translates human-readable domain names (like google.com) to machine readable (number) addresses. It’s easily one of the most critical pieces of the internet. When it fails, nothing works.

So, because it’s so critical, there are a few pieces of it that are designed to provide speed and reliability.

But when you’re troubleshooting DNS problems, those pieces become a huge headache. Primarily the one where changes don’t show up immediately. Sometimes they don’t show up for days. Literally. Fixing a problem is really frustrating when you don’t know if the change you just made has had any effect until the next day.

You know when you’re on an old computer or a slow connection and the text doesn’t show up until a second or two after you’ve typed it? Imagine that delay is 24 hours.

So instead of making a change, and immediately checking to see if it worked, you read and plan and re-read and come up with a plan B, and then consult some experts, and then make the change. Then you wait a day, see if it worked, and start reading again.

I’m not an expert on DNS, but I might be soon if this keeps up.

Coffee Service Update the first

The coffee is good.

You have to use one and one-half to two of the pre-measured packages to make a pot of good, strong Portland-style coffee, but it tastes good. Not Peet’s good, but certainly drinkable good. Somewhere between high-end gas station and low-end coffee bar. Like deli coffee.

Also, with our coffee service we apparently get to try an espresso machine free for a month. I suspect that this “dealer” has learned something from other “dealers” of “stimulants”. I didn’t think anyone in our office knew how to use it, but this morning I found out that our receptionist does.

The espresso machine is by Lavazza, and so is the espresso coffee, which comes in little pre-measured, pre-packed cup things. It’s better than you’d expect it to be, for pre-packed espresso, but not great. Good enough to serve in a deli.

We’ll see if the espresso machine sticks around. I’m sure that it’ll turn into a huge mess if it does. But until then, I’m going to see if I can get an irregular heart beat going.

Man I’m typing fast.