Wyszdom

As you may already be aware, I have a hard time falling asleep at night. My parents suggested that I get a TV for my bedroom and use that to put me to sleep. That would probably work, but I don’t feel like buying another TV right now. I already have a laptop stand and a laptop, so I’m going to be trying that tonight. Maybe I’ll find something on Hulu or YouTube or a cable channel’s site, or perhaps I’ll just watch whatever Leo is rerunning on TWiT Live. The only problem is that my computer is lacking a simple feature that is on most TVs: a sleep timer. Yes, I could go into my settings and set the sleep time to any period of time, but I don’t want to have to keep remembering to reset it back to its normal value the next morning. I’m sure there are freeware utilities which can accomplish this, but I was feeling a little geeky on WWDC Eve and decided to do some development myself.

It has been years since I’ve done anything with AppleScript, but a few minutes of Google searches and a couple of guesses had me create this:

tell application "Finder"
display dialog "Sleep time:" default answer "1"
delay (text returned of the result) * 3600
sleep
end tell

The script (or compiled application as I eventually saved it) asks the user how many hours to wait until putting the computer to sleep, and then sleeps the computer after that amount of time. The user input is multiplied by 3600 since it takes input in seconds. The script is very simple and doesn’t do any error checking, but it does what I want and it took longer to write this blog post than to write the script.

Google Squared is now available in Google Labs.

With Google Squared, you can choose a category of things you’d like to compare (like Colorado ski resorts), and then choose which attributes you want to look at for each item. It will automatically populate a few rows and columns for you, but it’s most useful if you really make it your own, and remove rows and columns you don’t want to see, and add rows and columns that you do want.

For example, right now for [Colorado ski resorts] I get columns labeled “Image,” “Description,” “Location,” “Snowmaking,” and “Telephone.” Out of these, I really just want location, so I can hit the “X” on all of the other columns. But I noticed that one of my favorite destinations, Telluride, isn’t in the list. So, in the lower-left of the Square where it says “Add items,” I type in “Telluride” and then click “Add.”

Now that I have a good list of resorts, I want to get an idea of how much skiing I can get done there. So, in the upper-right of the Square, I add columns for “Longest Run,” “Skiable Area,” “Base Elevation,” and “Summit Elevation.”

Here’s a video of me running through the above steps:

From here, I can edit any Square manually. This is automatically generated, so sometimes you need to clean things up a bit or just want to make your own notes. You can also just start with an empty Square and add items and attributes individually if you’re not getting the results you want with a search or want a more specific starting point.

Let me know in the comments of any particularly good (or hilariously bad) Squares you find. And when you’re done playing around with Google Squared, check out this great post about GoogleLookup by Beah. It has like a billion page views.

This will be about Google, but please remember this is my personal blog so nothing here is the official word of my employer. It’s all mine.

Today we released a fun video on the Webmaster Central YouTube channel that shows what goes into producing a video for the channel:

Going a bit beyond what’s covered in the video, here’s the basic process:

1. Matt is lured into the room using the scent of Pinkberry yogurt.
2. We record the video.
3. I transfer the video to a Macintosh Quadra 840AV and edit.
4. After editing is complete, I copy the final video file to a set of floppy disks. That’s right, disks. Plural. Since we shoot in HD, the files are so large that a single video often spans multiple 1.44 MB double-density disks.
5. The disks are uploaded to the YouTube office in San Bruno via data packets. One is pictured below:

Just kidding. 😉

The message I wanted to convey when Matt put me on video was that our setup isn’t really something that’s limited to a large company with a big media budget and staff. We’re using mostly “prosumer” level stuff, and you could get similar results by purchasing equipment today at even lower prices. Any standard video camera should be fine to get you started, and even a camera built in to your computer can work, depending on what the content of your video is. HD is a bonus, and well under the “thousand dollar range” I guessed in the video. But the most important thing isn’t which camera you buy, it’s how you use it. So rather than getting into the specifics of the equipment, I’m going to go over a few things you can do to set your video apart from the other 20 hours of video uploaded each minute to YouTube.

Get decent audio. I’m putting this first because it’s often overlooked by the producer, and is incredibly annoying to the audience if it’s bad. If you can get a microphone, that’s great. If you have to use the camera’s built-in mic, make sure the room is free of background noise (air conditioners, refrigerators, crowds, etc.), and have the talent speak loudly and clearly as if presenting to a large group. If your camera has a headphone jack, plug in some headphones and listen to the sound as you record.

Use a tripod, and don’t zoom. If you need to get a closeup of something, try cutting to a closeup instead of making the user watch a zoom. It’s much easier to get a nice image when the frame doesn’t move. If you’re unsure about how to compose your shots, start by looking up the “rule of thirds.”

Make sure the subject is well-lit. We actually don’t have any video lights in our rig at this point, and conference rooms tend to have pretty ugly lighting when it comes to video. So whenever possible, I try to pick a room that gets natural sunlight, or even go outside (just watch your audio!). Sunlight looks a lot better than fluorescent. If you’re especially lucky, you can find a room with windows on two walls. Then, you can have one window light your subject, and use the other as a background without your subject turning into a silhouette. Whatever you do, make sure your video isn’t too dark. Video cameras do not do well in low light, and after compression the image may be even harder to see. If you do need extra artificial light but don’t have a pro rig, you can get any bright light and then diffuse it by bouncing it off of a white wall or ceiling. I did this with a workshop light from Home Depot in the reconsideration video and also in my cooking shows.

Use manual focus, especially if your autofocus is fidgety. A quick way to handle this on cameras without a focus ring is to zoom in on the subject’s eyes, let it auto-focus, and then switch to manual mode so it’s set on that focal length. When you zoom out, your subject will remain perfectly focused.

Upload using the correct aspect ratio. Don’t stretch or squash your videos from their native dimensions. If it looks funky after you upload, try one of the tags listed in this YouTube blog post.

If you’re using Final Cut and are mixing in images, leave everything in its native format in the timeline, then export directly from the timeline to your final file. Your screenshots will be super-crisp if you don’t export them to a movie file (in a format like DV or HDV) first before compressing. If you do any kind of scaling of the images and your sequence is set to an interlaced video format, set the field dominance to “none” in the sequence settings.