1> The basic mode - the way the machine arrives – is pretty much all you need for Grandma, if you put Skype on it. There are guides.

2> The “advanced mode” which gives you terminals and a repository to install more software from and so on is a Bad Place To Be. Here’s why: it nearly works. But it doesn’t. You can’t install pidgin without breaking things. You can’t guarantee that if you deinstall things your hardware will still work. Weird stuff – USB audio support – doesn’t work.

3> Putting a real, full Linux on board is, as yet, impossible to do easily or safely. The forums are full of people talking about nearly getting things to work, or pasting four dozen lines of genuinely incomprehensible Unixness involving downloading patches from God Knows Where. Some of these systems seem good and reliable (the famously useful acerfand which turns off the goddamn whining noise of the fan except when the machine gets hot is essential.) But, over all, none of the real distros (Ubuntu, Fedora etc.) are done yet. This is a right pain in the butt.

4> Things which Do Not Work.

* No way I can find to mount HFS+ USB drives, even read-only.
* You can’t boot from the SD card slots (apparently this is a hardware constraint. bleah)
* USB Audio does not work – the relevant kernel modules aren’t there by default and installing them is dodgy.
* No out-of-the-box whole machine bootable backup method. Not such an issue for reasons I’ll note below, but if you try and upgrade something and it screws up, it’s gonna hurt.
* No way to record audio and video together from the web cam. It produces silent movies.

These things are irritations. Full-on Linux distros can handle most or all of these problems easily. The people who’re getting all the hardware to work on those distos are having some issues, and ASUS seems to be being only moderately helpful. There are a few horror stories about proprietary driver issues even under Windows which Does Not Help.

What’s annoying is that it’s close. I guess sometime later this year most of this stuff will work – Acer is shoveling so many of these boxen out the door that eventually it’ll reach critical mass and Get Fixed, but it hasn’t happened yet and I’m hardly a bleeding edge early adopter!

5> Keyboard and screen are excellent. I have unusually small hands and can type faster and more comfortably on the AA1 keyboard than on anything else I’ve tried. My mac now feels like one of those giant joke armchairs. The screen has more pixels per inch than anything I’ve used before – 1024 px in a 9 inch screen – and the additional resolution is really very, very nice. Software, on the other hand, often renders the fonts as it it was on a larger screen, and Firefox has no general setting for “display all text at 125%” although ctrl-shift-plus will increase the font size for any given page. I’d really like a permanent setting of that kind.

6> Storage and media handling. 8 gig of SD isn’t much. Oddly, I’ve found that I have two gig free. Why?

* 8 gig does not permit for music storage
* it does not permit image storage from digital cameras
* it certainly doesn’t allow for video editing
* nothing on this machine will produce a 50 or 100 meg file (hello, Omnigraffle.)

7> You really understand what the Apple Tax is for after using an AA1. You pay for the fact that things Just Work. I can’t stress too strongly how incredibly superior the Mac user experience is, largely because of the Macintosh Mantra which states “Everything Is Either Easy Or Impossible.” It either works out of the box, or it doesn’t work and there’s nothing you can do about it except wait for Cupertino. And that, right there, is the key to the Macintosh experience: there is no “maybe I can try these patches” or “well, if I recompile the kernel…” – no no no, it’s either there, or you’re screwed.

I might type or download a doc or two, but the massive file activities which are so much a part of the Macintosh experience just don’t happen. It’s not that they couldn’t happen, they just don’t. This is very interesting psychologically. Basically, the experience of using the AA1 is entirely different from using the mac. The mac does media. The AA1 just doesn’t. Maybe if it had the 150gb HD version it’d see more like a media machine, but fundamentally the software isn’t there. And my sense of it is that the absence of mass storage is actually a critical feature for the use case I had – a machine which was basically being used as a second monitor, and to partition my personal life (and multiple chat clients) from the Serious Work on the Macintosh. In a lot of ways, the AA1 is a textual experience. I read on it. I write on it. I listen to internet radio. But I’m not creating movies or audio. I’m not ripping, mixing and burning. I’m not making giant charts in Omnigraffle because, well, there is no OG and Inkscape isn’t cutting it. A mac is a different kind of tool at a really fundamental level and I did not realize how huge the gulf in my sense of personal possibilities was until I started using the AA1 every day. There is no Garageband, there is no iMovie, there is no Photobooth. And those things matter. If you’d never seen a computer work The Macintosh Way way, you could easily assume that the AA1 is What Computers Are, but it’s only a fragment of the potential of the artform.

Now that doesn’t mean that in a lot of situations, my AA1 isn’t the Machine of Choice. Small, light, cheap, nearly indestructible. Limited scope allows for greater focus. It’s a machine to Get Things Done on. But what it takes is the discipline not to tinker with the machine to get it to live up to its potential, but to accept it as it is – a little under it’s theoretical potential, but an excellent solution to a specific set of problems.

Computers like this are going to be all over the world – a billion or so units in the next 10 years I think. They will improve, but the fundamental importance of the Macintosh paradigm cannot be overstated. A Macintosh’s job is to be beautiful, not simply functional, and these machines are incredibly efficient, robustly functional, but there’s a missing spark of soul. I don’t know whether better funded and more mature open source operating systems will pick up that spark of soul, but my suspicion is that it cannot be created by programmers alone, and perhaps not by programmers at all. To squeeze the silicon hard enough that it produces honey is the heart of the Macintosh way, and I hope that people don’t lose sight of that genius in their bitching about the lack of bluetooth keyboard support on the iPhone and other such trivia.

You really need to see what Apple has done from the outside to understand how great it is.

Do I wish they’d produce a netbook? Yes, yes I do. Even if the hardware was pretty much like an AA1 or a similar device. But will they do it? No. They’ll wait, until they can do something Amazingly Great, that pushes the parameters of the computing experience, and that is why we hate and love them: the do not give us what we want, they give us what we need.

Individuals with great vision cannot steer Linux very much. It’s going to take the evolution of systems which allow that kind of grand vision to really open up the full potential of open source development models. Democratic selection for recipients of community funded research grants, perhaps. I took a shot at outlining how this might work for a relatively simple task: getting hardware support for the AA1 into the major distributions. I attach that outline below.

I’m really digging hard for a fundamental point here: the market, and the open source bazaar, cannot produce Macintoshes. Yes, the Mac exists in the market, and is a product of the market, but we all know that, at root, the Mac exists because Steve Jobs burns shareholder money like water by refusing to pander to the desire for short term profit, and instead frog-marches the Apple Faithful towards digital nerdvana one revolution at a time. It’s very important to see how important that job is when judging Apple’s failures, because a world dominate by cheap Linux machines is going to be as bad as one dominated by Microsoft, just for different reasons, unless we find a way to inject the Linux process with taste.

The AA1 is a good machine. It’s more useful to me, right now, in some ways than any Apple product. But with it alone I’d be enormously cognitively limited by the available software aesthetics. What my Apple does is indispensable to a particular kind of thinking and that is something I had to use the alternatives to see.

Viva la Revolucione!

PS: and now, could we please get organized to get real distros on commodity netbook hardware? Much as I love my Mac, I’m going to keep using Linux because, yes, it does some things better, like running on bombproof cheap machines, and I’d like the best of that experience too.

PPS: expect edits in the morning, it’s 3AM, and I’m dropping the mac off for vast repairs tomorrow. Case cracks, screen flickers, and now the battery is *dead.* Argh! How’s that for context?

===============

Would you pay $10 for a perfectly working Ubuntu for the AA1
What do you think it would cost to hire some Unix wizards to make either Fedora or Ubuntu work perfectly on the AAI, and package it as an easy-to-install download?

Let me make a guess here: a month of work, maybe split between two or three people. $50 an hour, for a total of $8000. We might be able to get it done for a ton less than that, but that depends on both the difficulty of the problems, and the price of the hackers. Acer sold seven million units in 2008 and is shooting for 15 million in 2009.*

So if we used something like Fundable (http://www.fundable.com/) to set up a pool, we could find an appropriate team to make all the hardware work, sort out the power management issues, make the fan control safe and reliable, make it possible to record sound and video off the web cam, mount the SD cards properly and everything else. If we got 800 people – and out of 7 million that’s not that many – it would be $10 each. If some people throw in a bit more to get the pot started, so much the better.

What do you think? Could we do this? Which distro would we start with, and how would we identify a credible team to do the work?

I’d really like a known-perfect Fedora or Ubuntu (pref Ubuntu) for the AA1, but I can’t see how we’re going to get it without hiring somebody to make it happen. But I think there are enough of us that it could be a few bucks each, we’d support some great open source development work, and maybe encourage other communities to get Linux customized for their hardware too.

Win-win-win. Who’s with me?

* http://www.pcadvisor.co.uk/news/index.cfm?newsid=13264