This entry is part two in the series, "The XO Files: I Want to Believe in the XO" Read Part I about the Laptop Project / Education Project disconnect here.
The XO Files: I Want To Believe
Part II: The New 4PC Market, and its Failings
The OLPC XO is a path-breaking, jaw-dropping piece of technology. And not just any traditional, consumer-focused (faster, shinier) way, but in specific and strategic areas that make the laptop perfect for developing world situations where it might be damp or dusty, the sun might be your light source at school, and you probably don't have reliable electricity at home. It happens to be that those same constraints also produce technological solutions that make the XO attractive to a certain set of users who want a no-frills, but highly functional laptop (like world travelers), as I mentioned in Part I -- it's lightweight, rugged, and low-power (solar chargeable), but powerful enough to connect to faint wifi, play movies, or review digital photos.
These features come with drawbacks - the long battery life required a processor that's simply not as fast as common laptop processors. The cost constraints meant a smaller screen design, a "hard drive" that's microscopic in comparison to most, and much RAM. Still, the laptop still performs admirably - or at least well enough - for most usage.
Unfortunately, unless you happen to be a top-notch software developer, or were willing to pay twice the cost of the laptop to participate in the frustratingly opaque and slow Give-One-Get-One program of 2007 (The XO's children starts up in November), you just simply couldn't get an XO - they were intentionally and adamantly not commercially available, despite the strong demand and buzz on their features. Naturally, markets abhor demand as nature abhors a vacuum, and existing niche company "netbooks" and stepped in, with bigger companies rolling out "mini-notes" soon thereafter.
The scions of OLPC
The biggest legacy of the OLPC project may be to have created a market for more portable laptops, offering functionality that even the most powerful smartphones won't provide due mainly to their size, but focusing not on clock speed but travel-friendly features - the "4P" Computers featuring (long-lasting, low-wattage) Power, (high utility) Performance, (lightweight and rugged) Portability, and (low) Price. Lilliputing and Gizmodo have both been discussing this:
Nicholas Negroponte and his little laptop that could certainly sparked a revolution. Without the OLPC, it's unlikely that we would have the Intel Classmate PC, the Asus Eee PC, or any of the dozens of cheap ultraportable consumer-oriented laptops that make life worth living.
The Economist also weighs in as well on the market that the OLPC opened up:
Hardly any models costing $500 or less were available when the XO burst onto the scene, but now there is a wide selection of such machines, from familiar makers such as HP and Intel, and from relative newcomers such as Asus and Pioneer Computers. By raising the very possibility of a $100 laptop, the XO presented the industry with a challenge. [...] All of these new machines are being aimed at consumers in the rich world, who like the idea of a computer that can be taken anywhere, as well as being sold for educational use in poor countries. The $100 laptop has been a success--just not, so far, in the way its makers intended.
Market Failures?
With this vibrant new 4P laptop market, why do we even need the OLPC Foundation anymore? The Economist hints at the exact problem - the new 4P laptops are consumer-focused. Lacking is the attention to developing-world situations and the ruggedness, ease of repair, long lifespan and low cost those require. This is the second trap of talking about the technology itself. Not only is the project confused by the lack of distinction between being an education project, but by talking tech (like, admittedly, this post has done itself), you start down the path of commercial laptops, and the market dynamics of dealing with customers easily swayed by larger numbers (faster processors, bigger hard drives, more RAM, bigger screens...), and are less interested in accepting the trade-offs that make the laptop useful outside of industrialized parts of the world.
I posit that there remains a niche -- but still significant -- consumer market for the XO, more or less as-is, which could bolster the goals of the project by providing a larger community of developers, content creators and engaged users. The next segment delves into the missteps of the original and current distribution plans, and suggests revisiting the XO "deployment" strategy.
Every article that comes to the conclusion that there's still a big demand for OLPC computers makes a critical assumtion: THAT THE SOFTWARE ACTUALLY WORKS. Good battery life, suspend/resume, trackpad that doesn't wig out, usable gui, simple things like that.
Until OLPC can actually produce such a machine, it's all speculation.
Good battery life: check
suspend/resume: check
trackpad that doesn't wig out: mrm, better
usable gui: check. The included Sugar is excellent for the target market, and a wide range of other environments (Amiga, Ubunto, XFCE, etc) are available.
I like how you complain about software in all caps, and then give counter-factual hardware whines as supporting "evidence."
You sound like you're having a lot of problems with your XO. Sorry to hear that. Mine work's like a charm, and I've kept Sugar.
I too have kept Sugar and continue to enjoy it - it works at least as predictably as my XP computer at work. Low bar, admittedly, but...
Biggest market failure in 4P:
MS Windows XP needs too much HW and is too expensive.
Simply put, the computer market is (still) a monopolistic market with a main player that has both a near monopoly on OS/Office software and $26B in cash to defend it. Essentially, MS is using "incentives" to "limit" the sell of Linux preloaded computers. In general, OEM get "joint marketing" deals if they are MS only which seriously reduce your XP/Vista licensing costs. Preloading Linux not only increases your net licensing costs seriously (wiping out your margin), you also will not get the latest updates and images "in time". Many are the ways that a monopolist can make your life hard. And MS have been convicted a dozen times for actually exercising these practises.
We now see that a 4P computer (Acer Aspire one) with XP is MORE expensive in Victoria than a lower spec machine with Linux:
http://brendanscott.wordpress.com/2008/09/10/aspire-one-victorian-education-discriminating-against-linux/
The cause? Government subsidies:
"Details of exactly how such steep discounts can be given are unclear at the moment. There are suggestions that this too-good-to-be-true deal is brought to you by the Victorian Department of Education and Early Childhood Development, which is apparently paying a subsidy on the XP loaded versions (and, of course, by the gormless taxpayers of Victoria on whose behalf the Department is spending the money - cheers!)."
So why should a state in Australia subsidize a US monopolist?
Btw, in the Netherlands, the Linpus version costs 299 Euro, the XP 399 Euro (you might get the latter down to 349 Euro).
The 4P market is still subject to severe monopolistic distortions. The only reason ANY Linux systems are still available is the complete inadequacy of MS's OS offerings.
Winter
I agree with Winter that Microsoft is heavily corrupting netbook manufacturers to artificially limit Linux expansion. The new ASUS eeePc is also a very current example of that. Here in Hungary, lower-spec linux models cost MORE than higher-spec Windows models. + the availability of Linux models is limited so if you need it fast, you have to go with the Windows model. It is disgusting to see this.
The XO would be a HUGE commercial success even with the current look (green) and software problems, IF generally available at max $250.
It seems that even Sugar 8.2 (released soon) could make it at least usable. If there was an additional, easy setup choice for the user to allow running XFCE with some real-life software pre-installed (Firefox,Evince,Flash,Java,Abiword,Gnumeric), that would make the XO a VERY good deal for a lot users.
And we wouldn't have to worry about Microsoft stepping in with a cheaper model since the price would be still very low for that.
It is even WORSE:
http://apcmag.com/Content.aspx?id=2683
"Redmond’s new XP mini-note licence dictates a 1GB memory limit to protect the high-profit position of Vista-powered notebooks"
So that is why you cannot buy a LINUX mini notebook with more than 1GB RAM.
Talking about market FAILURE.
Winter
Winter says "So that is why you cannot buy a LINUX mini notebook with more than 1GB RAM."
I don't see the connection here. The article specifically says "...APCmag.com has learned that Microsoft is dictating that vendors limit their mini-notes to 1GB of RAM if they want to install XP." which means that companies can of course offer Linux configurations with more than 1GB! But then the question really is whether a Linux-powered netbook needs more than 1GB of RAM?
sola, I agree with your comments regarding the software. I've been running the latest joyride builds for several months now and I find them to be a significant improvement over the current stable versions. The re-designed home-view and frame, sugar control panel, activity-update gui, great power-management, updated activities, etc. are really cool! Of course there are still areas which require more improvements (e.g. Journal) but the overall progress over the past 6~9 months has been impressive and things are definitely moving into the right direction.
As to running a 'more traditional' desktop on the XO I hope to find some time over the weekend to get Fedora 9 setup on an SD card per dsd's instructions at http://www.reactivated.net/weblog/archives/2008/08/regular-linux-desktops-on-the-xo/
@ChristophD
"I don't see the connection here. The article specifically says "...APCmag.com has learned that Microsoft is dictating that vendors limit their mini-notes to 1GB of RAM if they want to install XP." which means that companies can of course offer Linux configurations with more than 1GB! But then the question really is whether a Linux-powered netbook needs more than 1GB of RAM?"
- If they sell large numbers of >1GB laptops, MS will question their licenses to make sure they didn't mix up the OS'. So whenever X offers laptops with 1.5GB with Linux, MS lawyers will swarm over the plants and demand proof they did not accidently sell one with XP.
- With a Solid State HD, you cannot use swap memory. All your memory must be physical. That does affect Linux applications too. You might also want to use RAM disks io SSHD.
Winter
Just give it up: its a netbook, ok?
You might think your smarter than others but it doesnt really matter, youre too late..
Netcast is a much better description than podcast but no matter how much Leo Laporte tries, its still gonna remain a podcast even though you dont need an ipod.
Say 4p to anyone and you will get a blank stare.
Quick, name the first company to sell an mp3 player?
Now name me the first to sell a touch screen phone?
Chances are you cant, nor do you care.
That's technology. Trailblazers rarely succeed or get credit.
Im happy that the poor kids in the world get to use an 8 year old OS which will introduce them to the wonderful world of virus, trojans, malware, anti-virus, ad-aware, spybot, avg updates and so on.
I have family in brasil and the when the tech push is finished next year, there will be 50,000 computers labs serving the 50 million students of brazil using KDE desktops on their own homespun Linux variants. Freedom from monopolies and proprietary lock-ins as well as controlling your country's own technological path should not be overlooked.
Btw, a chinese company is releasing next month a 98$ Linux netbook based on an ARM processor (or is it Mips?).
The 100$ is a reality.
Something that has always amused me; flash memory writes are limited. Probably getting better as flash technology improves, but using your flash drive as a swap partition is bad. It reduces the life of your flash drive considerably. Linux users have been saying this since flash drives first came along.
And then Microsoft came up with "ReadyBoost". What?!!