Amazon Echo Review

echo1
Amazon Echo, or I guess The Amazon Echo, or as we tend to call it, Alexa, is an unusual product to review because at its very core it is just a speaker with a Voice User Interface to Amazon and some parts of the Internet. It’s Siri or Cortona or the Google Lady but for the living room (or bedroom or kitchen or wherever). It’s so similar a product that some functions seem strange for the living room (e.g., how far is it to San Diego from here? Who was the 33rd president of the United States?), but I find myself often using the timer or asking for the weather forecast, or even, my favorite, asking for a “flash brief” which shares top news stories from the day from NPR. It sounds great too. It has a flat equalization with clean highs and a solid bottom end with neither hiss nor a boomy rattle. If you like a huge dynamic range, like the similar sized Jambox or the larger SoundFreq or Bose systems, then you will find the Echo lacking, but I really like the way it sounds for the music I listen to.

Unfortunately, interpreting music commands is the thing that it doesn’t do as well as everything else which is an irony of sorts. You can access music from four sources: IHeartRadio, TuneIn, Amazon Prime Music, and your personal Amazon Music Library (that $25/year service which I needed for my huge music library); however, there is no way to tell Alexa that you want a particular version or source of music except to utter the precise wording of whatever it is you want. So, if I say, “Alexa, play radio KCRW,” it will play KCRW from TuneIn, but if I say, “play radio KCRW 89.9 On Air” it will play it from IHeartRadio (they are both the same stream of the live station; they are just named differently at each web service). A quirk, but not really a deal breaker. The real problem is that the speech recognition middleware for music from one’s own Amazon Music Library seems like it’s an entirely different entity than the one for the other functions. Jaina can mumble amidst her imaginative play with mermaids and dragons while the TV is blaring, “Alexa, tell me a joke” and the Echo replies with some cute joke, where I can say in a perfectly quiet room, articulating every sound, “Alexa, play the album The Wall” and she responds with “playing Have a Ball by Me First and the Gimmie Gimmies.” As with the radio services, the solution is to give the Echo as much information as possible (e.g., “Alexa, play the album The Wall from Pink Floyd”).

Regrettably, there are two broken aspects to Echo that you will have to live with (and I have told the Amazon service team this a couple times, and they have thanked me kindly but have yet to fix either).  The first is that the Amazon Music Library management software is total crap. Meticulously tagged music ends up with butchered tags when uploaded to the service (many people, not just me, have pointed out that compilation albums rarely end up as one album, but instead, as 20 albums each with its own artist). There’s also no way to import playlists, and there are minimal sorting features in Amazon Music Library for sorting a library to make a playlist. That’s all not Echo’s fault, but it makes what should be easy solutions to some of the speech recognition shortcomings more pain than they are worth. The other broken aspect is surely a bug, but Amazon insists it isn’t. Let’s say you want to listen to every version of the song “Yesterday” in your library. You say, “Alexa, play Yesterday.” She responds with, “Starting with the most recent version, Yesterday by The Beatles.” Sounds promising enough (even if it’s not the “most recent,” but whatever) except that after that one song, it just stops. Maybe it’s the English major in me, but if one says that they are “starting from the most recent version” I expect them to continue to play the rest of the versions. In my second email to Amazon about this problem, they told me to just create a playlist with all the “Yesterdays” in it. Sure, that makes perfect sense, but it also seems weird that a speaker that is advertised as “learning” from you cannot just play the 5 versions of Yesterday from the library. I mean, that really doesn’t take the computing power that, say, looking up how many miles it is to the moon does.  There is one additional annoyance of the Echo, but it doesn’t bother me that much. If Alexa doesn’t understand you, she will often suggest or just start playing another song from Prime Music: “Alexa, play iwresteledabearonce” to which she replies, “I cannot find I rested on a bear’s buns. Here’s Planet of Ice from Minus the Bear instead.” It’s only annoying when you are trying to parse out the magic sounds to get her to play the song you want and she keeps “suggesting” stuff instead.

Despite all of these annoyances, I still think it is a fun product, and I use it more than I originally thought I would. I do think that it is very much a $99 product and not a $199 product, so I wonder when they release it to the public whether they will stick to that price point.

Sparklepony Drama

Bandwidth has been sparkling lately with ponies, an inside joke turned proto-public meme now turned synecdoche for what other people think I do.  Great.  Just to recap the hoopla, the Conference on College Composition and Communication has informally been host to a gamification of conference attendance called C’s the Day.  I know a few of the people who designed the game and fewer of those who continue to run it, and I even played it one year for the experience, although I wasn’t really the intended audience when I did so.  In a nutshell, participants who sign up are given a booklet with a series of quests split into categories such as publishing, networking, and administration.  These quests ask participants to get involved in the conference and not just sit in their hotel rooms grading or working on their dissertations in between conference sessions (although, as I recall, that might have been a quest).  Of the rewards and quest related activities are hand-crafted sparkleponies that, as Jill Morris, one of the designers of the game has said in comments, was a tweet reference a few years ago to the Celestial Steed from World of Warcraft that became a unique badge of honor.  As prizes go, unique is good.  Sure, they could give out pens or a tote bag, but we already have bins of those we never use, and a shot or pint glass is just the type of encouragement I don’t need given the point in the academic year when Cs occurs.  I like the sparklepony.  And while I find some instances of gamification problematic, this is not one of them.    Continue reading

Terms of Play: Essays on Words That Matter in Videogame Theory

terms_coverZach’s new edited collection came out.  There are some good pieces in there.  My favorite, actually, is QQMore from Jon Bakos (it dovetails nicely with my current research on online forums).

I have a piece in there on game mechanics which was really fun to write, but also very challenging.  I started a conversation about it on a listserv on gaming, and it evolved in ways I hadn’t anticipated–it took me what felt like an immensely long time to sort through it in drafting the article.    What does that mean?  I actually kept track how long it took me  (I was sharing this with a colleague who was doing an informal survey on how much time we spent writing a piece).  To the best of my calculations, here’s how long it took (version #: hours:minutes [total pages in document]):

  • Version 1: 11:12 [10 pages]
  • Version 2: 01:39 [11 pages]
  • Version 3: 00:14 [10 pages]
  • Version 4: 07:14 [15 pages]
  • Version 5: 06:04 [18 pages]
  • Version 6: 01:02 [17 pages]
  • Version 7: 12:15 [37 pages]
  • Version 8: 02:33 [35 pages]
  • Version 9: 01:03 [35 pages]

So, 43.27 hours worth of writing (not including reading/research of which I could not really keep track of).  I wonder if that is long?  Or maybe short.  Maybe I should be cranking out an article a week if all it takes me is 40 hours.  Edward Tufte comes to mind: “at the heart of quantitative reasoning is a single question: compared to what?” (Envisioning Information, p. 67).

It’s Out! Rhetoric/Composition/Play through Video Games

coverWhat a great experience working with lots of smart scholars to make Rhetoric/Composition/Play through Video Games.  For those interested in the process, we sent this proposal in February 2012, and we heard back with an appraisal in April 2012.  We submitted the “final” draft in September and then proofs in January 2013.  The book was released today (March 20).  It felt like an eternity, but looking at it now, it moved quickly, and our authors (and my co-editors) were equally adept at getting everything done.