Wyszdom

I do a lot of searching on Google. I literally perform tens of thousands of queries each year. If you want to know how much you search, check out Google Web History. It’s really fun to click on trends— I found out that on my personal, non-work account, the hour with the highest volume of queries from me is between 11 PM and midnight, and the most active day of the week is Thursday. Usually these queries are kept between myself and Google’s computers, but I’ll share a few of tonight’s searches with you.

A few hours ago, I came across this buzz from Leo Laporte. The image immediately made me think of a scene from a movie in which the shot begins with a man holding a magnifying glass up to his eye, then pulls it away to reveal that his one eye is actually larger than the other. I couldn’t remember the name of the movie, and the only other thing I could remember was that the same scene was played in reverse and it took place in a library.

Here’s how my queries began:

[library reverse scene]
[library reverse scene magnifying glass]

That second query actually had the movie listed on the first page of results, but I didn’t recognize it and kept searching. I switched to video search and tried:

[reverse library]
[reverse library magnifying glass]

No luck; back to web search with:

[reverse library magnifying glass]
["shot in reverse" library]
["shot in reverse" library "looking glass"]
["shot in reverse" library "magnifying glass"]

Note that until tonight I didn’t actually know what a “looking glass” was. I had always assumed it was a synonym for “magnifying glass.” It’s not.

See how I narrow and broaden the results by adding and removing terms:

["shot in reverse"]
["shot in reverse" scenes]
["shot in reverse" movie scenes]

At this point I’m trying to find lists of movie scenes that are shot in reverse. Still, nothing pops out at me. What I’m trying to do with all of these query refinements is try to think of words and phrases that may be on the page. Watch how I switch from “shot in reverse” to “played in reverse” since an author could write about the same technique either way:

["played in reverse" movie scenes]
[movie scene reverse]

That first query had the movie listed on the second page of results, but I missed it. If I had thought earlier that “shot in reverse” and “played in reverse” were interchangeable, I could have saved some time with the OR operator:

["shot in reverse" OR "played in reverse"]

Finally, I realized that perhaps the scene didn’t actually take place in a library. But there were definitely bookshelves:

[movie scene reverse bookshelves]

I went through a couple of pages of results and then switched to an image search for the same query, and went through several pages of results there. It’s easy to go through many image results pages since they can be scanned quickly if you know what you’re looking for. Reading… ha!

I tried another image search:

[movie scene magnifying glass]

And then finally thought of another way to describe it. I wasn’t just looking for a scene with a magnifying glass. I was looking for a scene with a magnifying glass that was used as a joke. Still as an image search, I tried:

[movie scene magnifying glass gag]

Almost immediately, I recognized one of the images and clicked on it, bringing me to this forum thread. I found the image on reply #68, which referenced a movie called Top Secret!

Now all I had to do was check my work:

[top secret]

That found the movie, but wasn’t specific enough, so I tried:

[top secret reverse]

From the snippet of the Wikipedia page, I was able to see that the scene took place in a bookstore, not a library. I did a search for [top secret swedish bookstore] and was presented with some video clips confirming that Top Secret! was indeed the movie I was looking for. The whole process took five minutes. I probably should have just known that it would be a Jim Abrahams movie ending in an exclamation point. My favorite gag of his is the pilot getting ready in front of the mirror in Airplane!*

I guess there are a couple of conclusions you could draw from this long list of queries:

1. Even an expert searcher like Wysz doesn’t get what he wants with the first try, so don’t give up on your search too quickly. Stick with it.
2. No wonder Wysz has had to do so many searches. He’s pretty bad at searching.

But what I’d really like you to learn is this:

1. Search is not a solved problem, so there’s plenty of room for Google and others to innovate and improve. A perfect search engine should know what I’m looking for, even if I have a fuzzy memory or my search terms are ambiguous.
2. There are techniques you can learn to become a better searcher. If you’re interested in learning some, one new resource you may want to check out is SearchReSearch, the personal blog of Dan Russell, a research scientist at Google.

* By the way, how do you punctuate a sentence ending with a movie title that ends with punctuation?

I’ve spent the past couple of years trying to find a good offsite backup solution for my personal computers. My needs are a bit more intense than the average user as I am backing up video, which means that not only am I backing up what will soon be terabytes of data, but I have individual files that are in the 20 GB range.

In May 2008, I tried Transport. Unfortunately, I just don’t think Time Machine was developed with such a solution in mind, and didn’t renew my service. I still use the Time Capsule (they shipped it to me after my service expired) for onsite backup.

Last summer, I tried using Mozy. I was never unable to successfully upload one of my larger files, so I cancelled the service and received a full refund. After Mozy didn’t work out, I tried Carbonite. I was able to get a big file uploaded to Carbonite, but as I continued testing I saw my upload speed to the service drop considerably, so I didn’t continue using the service past the free trial. I would be willing to give them another chance, but they still have some pretty big limitations such as only backing up internal drives and not offering any type of site license to consumer users.

The most recent service I signed up for was LiveDrive. It worked well for a while, but recently it’s been choking on large uploads, and I think I need something a little more robust. I’ll probably just let this subscription run its course and not renew.

Now, thanks to a positive review, I’m investigating CrashPlan. I uploaded two large files as a test: one about 6 GB and the other about 22 GB. It took a few days to upload on my Comcast connection which doesn’t offer a lot of upload bandwidth. (Can I get some Google fiber up in here?) After the upload succeeded, I deleted the local files and tested the restore function. Several hours later, the files were right back where I left them, with even the original creation dates intact. Looks like a success. At the end of my free trial, I intend to purchase the family unlimited plan, which will allow me to back up all of my family’s computers without having to purchase an individual license for each machine.

Update on April 13, 2010: I just had my first support experience with CrashPlan. I emailed them three questions about my account, I received a reply the next day. A reply with specific answers to all three of my questions. I am happy about this.

I work for Google, but this is a personal blog post.

Google’s corporate site has a page listing “Ten things we know to be true.” I’d like to highlight two of them:

1. Focus on the user and all else will follow.
3. Fast is better than slow.

I’m a user, and I like the Internet to be fast. Other technology is fast. It takes a second to start a car (even if traffic is slow). You can flip through TV channels. A lightbulb turns on and off at the flick of a switch. Why should the web be slow?

Sometimes I experience the web at slower speeds than I’m used to, and it frustrates me. Checking out a video is not like turning on the TV. There’s a lot of wasted time.

So, what can you do to make things better? If you have a website, please make sure that it’s fast. Think of speed as a feature. If you wanted to view one of my albums with over 100 photos, would you rather use Picasa or Flickr? I’ve been a longtime fan of Flickr (I joined in 2004), but if I want to scan someone’s photo album, I prefer Picasa.

I’m not the only user who values speed. Mozilla (the makers of Firefox), has been gathering hard data about the effect of speed on user behavior and they’ve published the results on their Blog of Metrics. They learned that by cutting their page load time by 2.2 seconds, their conversions (visits resulting in a Firefox download) increased by 15.4%. In case you’re not used to thinking about conversions, put another way this means that since they reduced their page’s loading time, visitors are 15% more likely to download their product. You probably knew you could increase conversions by lowering prices, buying more advertising to send people to your site, making your pages easier to navigate, and trying different product descriptions, but have you ever thought about how fast the page loads? Looking at Mozilla’s study, speed probably isn’t something you want to ignore.

Earlier today, Google announced that it’s now using site speed as a signal in search ranking. If you’re new to thinking about how Google works, here’s a quick overview:

1. Google’s computers get online and download a good portion of the Internet.
2. A user visits Google, types in what he or she is looking for, and clicks “Google Search.”
3. Google’s computers do a bunch of math to figure out which pages to return and in which order. Over 200 factors can be involved in the calculation of where each page should be listed, such as what words are on the page and which other pages have links to that page.
4. Google presents the results to the user.

What today’s announcement says is that the speed of a site is now one of the over 200 signals mentioned in step 3. Why? Well, we want our users to be happy. If they’re happy, they’ll keep using Google and Google will stay in business. Users are happy when the search results are relevant and high quality. As illustrated by the Firefox example above, speed is one aspect of quality that users value, whether they consciously notice it or not.

One final thought though: If you’re a webmaster and are worried that your site’s ranking in Google is now going to plummet because you haven’t had time to make it faster yet, don’t panic. Just think about how much happier your users will be after you get around to making those improvements.

In my previous post, I showed a video that I shot while riding my bike. I think it’s pretty cool that I was able to record two hours of HD footage on a single card. In fact, I only filled up about half of the 16 gigabyte card.

Here’s what the card looks like:

To give you a little perspective on its size, let’s compare it to something. Here it is next to a 2 GB microSD card:

Okay, maybe I should compare it to something else. Here it is on a DVD:

That little card holds the same amount of data as 23 CDs, and it cost less than $50. Pretty cool.

And if you’re wondering what the original video quality looks like from the camera, here’s a segment of the original recording that hasn’t been recompressed.

I went for a bike ride on Sunday. This route usually takes me about two hours, but thanks to my computer, this time it took less than ten minutes:

The video was taken with my ContourHD using the handlebar mount. It’s an interesting angle, but the mount doesn’t hold the camera as tightly as I would hope so there is quite a bit of vibration between the camera and the mount, even though the mount itself doesn’t move at all relative to the bike. On an earlier weekend I tried recording with the goggle mount connected to my helmet strap, but that was a bit of a hack and I look around a lot. I have ordered the vented helmet mount though to see if I want to utilize that while biking (for reduced noise) or skiing. Here’s a picture of the camera and mount:

I didn’t use the rubber bands for this video, but I’ve found that they do help to reduce vibration a bit.

Here’s the route I recorded with My Tracks:

View 2-14-2010 in a larger map

It would be really convenient if there was a standard way to record GPS data over time in a video, like how location data is stored in EXIF for photos. There’s at least one video camera coming out that records location data, so we’ll see how they implement things and what video software starts to support doing interesting things with video and location.

Now that I think about it, it’s also disappointing that location awareness isn’t built in to all still cameras yet. Here’s one way we could approach the problem:

Installing a GPS receiver in every camera may not be practical, especially since GPS doesn’t always work, such as indoors. There are other ways to get location data, but adding WiFi and cellular radios is just going to increase costs even more and may even require subscription fees. What you can do is assume that anyone who has a camera also has a phone, and that the phone has some sort of idea about where it is. You can then pair the camera with the phone (via Bluetooth or WiFi; whichever is easier/cheaper), and then with every picture the camera will ask the phone for its location and store it right within the EXIF data for the image. It may not always have a GPS fix, but location data of some accuracy would be recorded with each photo, which would save the photographer from manual location tagging later.

Sometimes when I’m about to make a really hilarious joke that I just came up with on my own, I search for it on Google just to make sure I’m being original. And then right there at the top of the results, I see that someone else has already used the joke, and this frustrates me.

I’m afraid to search and see if someone else has already complained about this.

If you ever don’t feel like explaining something technical to me, just tell me that it involves long division, and I will stop asking questions.

I just made a quick video showing the three different mice that I use with my computer: a Logitech MX1000, an Apple Magic Mouse*, and a 3Dconnexion SpaceNavigator.

One tip I’d like to add is that for the Magic Mouse, it may at first feel uncomfortable if you are used to a higher mouse. Make sure you lay your hand flat, and it should feel a lot more natural.

And if you were curious about my keyboard, I type on an original Das Keyboard, which has no markings on the keys.

While I was having fun recording stuff, I also grabbed a quick video of what the new YouTube uploader looks like in case you haven’t seen it. As your video is uploading, YouTube starts crunching on it right away and shows you the processing progress via a “flipbook” of thumbnails. Pretty neat, and I’m amazed at how quickly YouTube is able to process the videos even as 20 hours are uploaded every minute.

* I accidentally referred to the Apple mouse as a “Mighty Mouse.” Whoops.

I’m starting to get back into reading books after a long hiatus. One thing that’s holding me back, though, is that I’m not really a big fan of buying a bunch of physical books that I’ll only use once. It feels like a waste, I don’t want to go through the hassle of reselling them or going to a library, and I have limited shelf space.

Obviously it makes sense for me to go digital. The Kindle looks very appealing. It has a long battery life, a display that’s not just another LCD screen for me to stare at, and it has a broad selection of books. But, as far as I know, if I buy a book on the Kindle, I cannot transfer it to another non-Kindle device. So if I have a library of Kindle books and then one day decide I’d like to use a book reader from Sony or buy a nook, I’m out of luck, right?

I’m looking for DRM-free ebooks, not free ebooks. Consider music, for example. If I buy music from iTunes, I can play it on my Mac or iPhone, but I can also play it on my Nexus One without having to sign in to my iTunes account or anything silly like that. This wasn’t always the case (Apple used to sell DRM’d music), but I just tested one of their DRM-free AAC files and I can assure you that it works on my Nexus One. And then of course there is the Amazon MP3 store which I have warmed up to recently. They also sell DRM-free music in a standard format. I think this is very reasonable. I still pay for the music, but once I have it, I can play it on any device I want.

So, my questions for anyone who is reading this:

1. Am I wrong about the Kindle? (And no, “They have an app for X device” is not what I’m looking for.)
2. Is there an existing store with a significant number of titles from major publishers that sells DRM-free books in a standard format? If so, which is it?

Please leave any advice in the comments section below. I noticed that iBooks on the iPad will be in the ePub format. This is a good sign, but I haven’t found any clear answers on whether or not these files will have DRM. If anyone can find a confirmation either way, please let me know along with a link to your source. Thanks!

My initial reaction to the iPad:

Looks cool; I can definitely see myself using it to surf the web on my couch. But I really don’t understand having to pay an extra $130 just so I have the option to add 3G service later.