My advice to him was simple: just pick something and run with it. As Teddy Roosevelt supposedly said, “In any moment of decision the best thing you can do is the right thing, the next best thing is the wrong thing, and the worst thing you can do is nothing.” So we talked about what he wanted to do (build a game), what he had already considered (Java, because that’s what Minecraft is written in), and what criteria he was using to pick something (not terribly well thought out ones) with a goal of landing on a decision.

As we had this discussion, I thought back to how I’ve made these decisions myself, and I realized that once I was done rationally narrowing my options to ones with a similar set of pros and cons, I had a worrisome tendency: to choose less popular options, even if they made my life harder.

How my contrarianism let me down

To give you an example, last year I attempted to go through Michael Hartl’s excellent Rails Tutorial. This is a fantastic free way to learn Rails: it aligned with my goal (learn how to build a web app the right way, rather than cobbling together a bunch of PHP spaghetti), and it would walk me through all the steps required to build and deploy just such an app. But it’s also an opinionated tutorial, in that it recommends you use specific tools as you go along: Git for source control, Heroku for deployments, Blueprint CSS for layout (though the latest version of the tutorial has now switched to Bootstrap).

Being the contrarian that I am, I wasn’t having any of that. I decided that I was going to use Mercurial for source control, YUI CSS to help layout content, and deploy to my own VPS. This would have been just fine if I could competently use all of those tools, but I really couldn’t. The net result was that I expanded the amount of stuff I had to learn exponentially: not only was I learning Rails and Ruby, but now I had to add to that working with a CSS framework, managing my own Linux server and jumping to distributed source control for the first time. Whereas the tutorial gave me the bits of information about its chosen tools I would need to know, I now needed to intentionally ignore them and figure out the corresponding way to accomplish things in my chosen tools.

You might have inferred the results of these decisions from my word choice. Having to spend so much time working on incidental things slowed my progress through the tutorial dramatically, things got busier at work, and I ultimately got distracted. I gave up on the tutorial around chapter 7. And this isn’t the first time I’ve had a similar experience.

Geeks are contrarian by nature

I think I’m not alone in making these mistakes, and I think they’re especially prevalent among people in software for one major reason: Engineers and geeks have a strong tendency to be contrary. Many people end up in these areas precisely because of an unwillingness to accept the status quo in some area, and to fix something most of the world doesn’t perceive as broken.

The recent debate on Hacker News about whether it is a good thing or not that lots of sites are adopting Bootstrap is just one example of this. Just as this tool becomes popular, the programming community begins to push back against it. I guarantee there’s a non-trivial number of people who will now resist using Bootstrap, even if the attention it’s gotten is going to build a strong ecosystem around it that will make it perhaps the best choice for their uses.

Sometimes this might have a great outcome: we’ll get another tool that does something a little different and is perfect for a different set of scenarios and people. But that’s what happens when an expert decides they don’t want to use the tool because it doesn’t meet their needs, not when a newbie (like me) is steered away because “it’s overpopular” and “all sites built this way look alike.”

As we make choices about what to invest our time and effort into, let’s keep in mind that making life harder in one area is a great way to reduce the bandwidth you have to invest in other things. I’ll certainly try to fight my contrarian urges; I hope you will as well.

You can probably imagine the struggles I have finding this range of music on modern subscription services like Spotify or Zune Pass. While music from the most popular French or Spanish artists is generally available, going just a few steps off the beaten path gives me either no matches or an album listing with a sad “unavailable” sign next to any albums by that artist. The situation in stores like iTunes and Amazon tends to be better, but I don’t necessarily want to buy every song right away, and there’s still quite a lot missing.

The utter lack of support for foreign music is frustrating because I know I can simply navigate to YouTube and find multiple videos that will do a perfectly adequate job of letting me listen to songs by most any artist I can think of.

I realize my musical tastes are hardly representative of most people’s, of course. But everything I’ve said about foreign music applies to many indie musicians, particularly ones that may not yet have a deal with a notable record label.

Content licensing is fun!

There’s a simple reason for why YouTube has far and away the largest selection of music online: Unlike the subscription services, it doesn’t need to acquire a content license before it will allow users to listen to a particular song (and view some sort of video). Rather, YouTube relies on two means of getting content:

• Major services that promote artists by uploading their music videos directly to YouTube, like VEVO
• Fans, listeners and creators upload videos of music they like on their own

VEVO and company cover the same general set of music pre-negotiated licensing tracks do, but it’s the second method that gives YouTube’s music collection its incredible breadth. And to keep things legal, Google has added lots of functionality for record labels to monetize music on YouTube, with the ability to display ads on videos that feature a song, block them in particular markets or to take them down entirely.

And the collection is just the tip of the iceberg…

While YouTube is unrivaled in its collection, it’s also got quite a few other things going for it. First and foremost, it’s incredibly low friction: I don’t need to sign up or log in to an account, launch an app, or wait for anything to download: I just open the site in my browser, type something in the search box, and before I know it, I’m listening to whatever it was I wanted to hear. Plus, I can instantly share a link to that video with all my friends, and be pretty confident that they will be able to hear the song with the same minimal amount of friction.

YouTube’s existing functionality does a reasonable job of covering other key music site functionality, like the related videos section letting people learn about new, similar artists they might like. Over the past few years, YouTube has also added quite a few music-specific features, like artist-specific “YouTube Mixes” and upcoming concert information.

YouTube isn’t the solution

There’s at least one difficult-to-solve problem that prevents YouTube from being the One True Music Service, though: the fact that it’s a video site, and every musical track is tied to an only sometimes interesting clip.

This is fine when you’re consuming music on a computer with a fast internet connection: just play the video in the background as you go about your other business. But so much of our music consumption happens in other contexts where the videos make YouTube an unlikely option. After all, no one wants to wait for a video to download on a slow mobile data connection when they aren’t planning on paying attention to the video anyway. And that assumes people have a data connection at all. Playing videos also uses more battery life, and most modern mobile devices can’t multitask with videos the same way they can with music.

Ultimately, unless YouTube one day decides to make audio a first-class citizen on its site, that limitation will continue to prevent it from being usable as a music service outside the house.

And what of the music services?

There’s no question that music services offer an experience that is more optimized for straight-up music, and features like upcoming concerts are either already available or easily added.

The collection question, though, is harder to fix, and it has mystified me for a long time. Given that YouTube is successful with a model where the legality of an upload is determined after the inclusion in its playable corpus, why hasn’t a single music service emulated the model? Indeed, why hasn’t Google’s own music service gone down a path where any user-uploaded content is included in its database and playable by anyone else unless a record label desires otherwise?

Perhaps YouTube is truly just a historical accident that the record labels wish they could put back in the bottle. But shouldn’t that historical accident  serve as precedent for the possibility of other sites that follow the same model? SoundCloud is one audio-focused site that relies on user-generated content, but it’s explicitly not focusing on music.

Until someone builds that service, I will have to continue relying on YouTube for discovering and trying out artists, and assorted unlicensed means for actually acquiring music the industry hasn’t bothered to make available to me.

This is not because I don’t love you, but rather because my insistence on using odd random scripts like Chyrp meant that I had to transfer my blog posts into this (WordPress-powered) system manually. That was sufficiently little fun in and of itself that moving comments did not seem worthwhile. My apologies.

Now that this site is running WordPress, though, I can do nice things like using Windows Live Writer to update, and take advantage of the vast ecosystem that surrounds the most popular blog engine. And ultimately the goal of that is to enable me to blog more often—something I’ve committed to for 2010.

Till then, please let me know if anything doesn’t look right.

1. Go to the Bing home page at bing.com. Look at the pretty picture and read the fun captions. Then right click on the search box and select “Create search.”
2. In the Create Search box, you can leave the “Keyword” field blank, or assign it a letter or two to quickly perform a Bing search from the address bar.
3. Click on “Details.” Tick the two check boxes that appear: “Use as default search engine” and “Use as speed dial search engine.”
4. Click “OK”

And you’re set!

I signed up for the pretty cool URL-shortening service tr.im today, mostly because of the statistics it enables users to view about each link. (Yes, there are issues with using URL shorteners, but if you have to use them, you may as well use a good one.) To make using it more convenient, I wanted to add a button to my browser to automatically shorten a URL.

They had a convenient bookmarklet that worked in Opera, my web browser of choice, but when I added it, I realized all custom buttons get the same default icon in Opera, the “new bookmark” icon. I was already using that for my delicious bookmarklet, and I didn’t want two stars I’d get confused between in my menu bar. So I whipped up a custom button using Opera’s “Fit to width” image. The screenshot shows how it looks. If you’re using Opera and want to add it, just drag this link to a menu bar:

tr.im URL