Twenty-four hours with my OLPC

It’s important to underline that everything I am about to say is:

(a) the product of someone thoroughly conditioned by, and accustomed to, a long series of essentially Xerox Star-derivative UIs;
(b) written by someone whose understanding of task and tool is very likely tangent at few points to the perceptual and symbolic universe inhabited by the intended target audience; and
(c) underwritten by a single day’s use of the device in question.

Having said that.

Despite its inclusion of some innovative and useful features, I find the OLPC device’s Sugar operating system poorly integrated with applications (here nicely dubbed “activities”), to the degree that it may well be impossible to evaluate whether the underlying idea ever had any merit. My first impression – and I reiterate, it’s only that – is that many of the applications bundled with the device epitomize everything that’s wrong with FLOSS user interfaces, even when the OS itself has been created by professional information designers.

The indicators of trouble begin the moment you unbox your new device. Ostensibly out of sustainability concerns, the laptop arrives bereft of any documentation whatsoever – save a letter over the signature of Nicholas Negroponte, advising the proud new owner that all the necessary instructions can be found at this URL.

What if you can’t connect to a network in the first place? Tough luck. It’s not quite a brick, but you’re not going to be able to do much of anything at all with the machine. In fairness, of course, the primary scenario of use envisioned by the device’s designers precludes a solitary laptop being unpacked in isolation: there “should” always be at least one other machine within range and able to form a mesh, or at least a trained teacher around to help. Try explaining that to a child, though.

I want to respect the fact that design challenges of this order are extraordinarily complicated, and nothing if not inherently prone to compromise and trade-off. As well, it’s crucial to acknowledge (out of professional courtesy, if for no other reason) that mature designers may make different, but equally defensible, choices at any given juncture. But there are just a ton of bizarre decisions made here, at the physical and interface levels both, and I can’t see any way that they won’t damage the user experience.

Two examples:

– The Sugar OS offers the user three ways to situate and understand the context of their use, three scales of representation hardwired into the UI: Home, Groups, and Neighborhood. These are almost certainly the right levels of abstraction, and even, defensibly, the best choices of nomenclature/labeling. But take a look at the Neighborhood view.

I wonder about the choice to apply a spatial metaphor to the depiction of network access points, when no such spatial organizational schema in fact exists. (Available access nodes are apparently ordered arbitrarily on the screen, each time the Neighborhood view is invoked.) There’s a serious mismatch here between the onscreen representation and the reality of access points, which are, after all, disposed in space around the device and user. If you can’t get a lock on one, the very first thing you want to do is move closer to it; follow this map, and all you’re going to get is frustrated.

– I’m generally a big fan of Yves Béhar’s work with Fuseproject. Historically, he’s had an unerring sense for humane and pleasing form, for finding the warmth in objects whose contours are derived algorithmically. That’s why I simply cannot wrap my head around his choice to closely stipple the XO’s surface. I’m sure it helps little hands keep a grip on the box, but that’s what you do to street furniture, for chrissakes, to keep it from being vandalized. And that’s just exactly how it functions here – except that in context, of course, what’s being prevented isn’t vandalism, but personalization.

Proud owners of new XOs will presumably want to slap stickers on their machines, draw on them, make them their own; after all, that’s what we adults do with ours. Why should kids be denied this very real pleasure…or the additional disincentive to theft that goes hand-in-hand with it?

I’ll leave you with those thoughts for now. I want to get into it really deeply, I do. And I will. Having just committed to the Web two looooong and fairly critical pieces in rapid succession, though, I’m hesitant to saddle you-all with another just yet. Maybe you can just leave your responses in comments, and let me be Good Cop for awhile?

8 responses to “Twenty-four hours with my OLPC”

  1. Abe Burmeister says :

    *cough* Lisa *cough* *cough* Strausfeld

    I’ll take the 5th on this one…

  2. Michal Migurski says :

    Just watching the Yves Behar interview (http://www.scribemedia.org/2007/12/20/olpc/) about this thing, and came to the bit about the customizable “XO” logo. There are 20 different color combinations so you can tell which one is yours, but a sticker won’t stick to it. It’s an interesting example of creating a walled garden for creativity, where the “designer” can look at personalized pieces and get meta-credit for the personalization.

  3. Wayan @ OLPC News says :

    Adam,

    Be sure to check out the XO Jabber service in your deep dive: http://olpcnews.com/forum/index.php?topic=396.0 that way you can enjoy the mesh capabilities of the XO even if there aren’t any XO users nearby

  4. enkerli says :

    While you used this post as an example of “harsh criticism.” I’d say you were, in fact, very thoughtful, here. Not exactly a Good Cop. But honest, reflective, level-headed.
    What’s a bit sad is that this is exactly the kind of review of the OLPC’s XO which was being washed away through the hype. The typical answer was “you just don’t get it because you’re not an elementary school student in Africa.” The fact that you specifically addressed the issue of your “biases” might not have changed anything in your post’s dismissal.
    But now is the time to go back to these thoughtful reviews of the OLPC XO-1. Not necessarily for the benefit of the OLPC XO-2. But for the benefits of diverse design projects to learn from the issues surrounding the XO-1 prototype.

Trackbacks / Pingbacks

  1. Like It Matters - 25 January 2008
  2. OLPC Linkfest « Disparate - 20 May 2008
  3. OLPC Linkfest « Disparate - 20 May 2008
  4. Jeffrey Zeldman - 11 February 2009

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: