Slashdot Mirror
Novel OS Drives the '$100 laptop'
jrwr00 writes with a link to a CNN story about the $100 laptop's unique operating system. We've discussed the OLPC's UI before but the article offers a few new piece of information on the project, which is expected to roll out this year. From the article: "The XO machines are still being tweaked, and [OLPC UI] Sugar isn't expected to be tested by any kids until February. By July or so, several million are expected to reach Brazil, Argentina, Uruguay, Nigeria, Libya, Pakistan, Thailand and the Palestinian territory. Negroponte said three more African countries might sign on in the next two weeks. The Inter-American Development Bank is trying to get the laptops to multiple Central American countries."
I read this story on CNN first as well, and my first thought at seeing the headline was nightmares about a Novell operating system.
In any event, it doesn't really sound particularly novel to me.
I may make you feel, but I can't make you think.
Drill baby drill - on Mars
"It doesn't feel like Linux. It doesn't feel like Windows. It doesn't feel like Apple," said Vota, who is director of Geekcorps, an organization that facilitates technology volunteers in developing countries. He emphasized that his opinions were his own and not on behalf of Geekcorps. so we have: a) kernel b) operating system c) hardware vendor It doesn't feel like any of those? Wow.
Just because I like to repeat myself every time an OLPC story is posted, I'll ask again: Where are the apps for this platform? Can anybody name one app, accessible to end users (e.g., no recompiling required), that is compatible with the Sugar UI, mesh networking, low-end specs, and other unique features?
A platform exists only to run the apps, not visa-versa. BeOS was a great platform, too. Many excellent gaming platforms have failed, because they lacked apps (i.e., games). Linux desktop is getting nowhere, despite it's technical excellence, because it lacks key apps (i.e., Office). Pull a few key apps from MacOS X (e.g., Office, Photoshop, etc.) and see what happens to adoption.
And all those platforms have far, far more apps available than OLPC (just look at sf.net, download.com or cdw.com). I know OLPC runs a flavor of Linux, but no known Linux apps are compatible with the specs above (Sugar, mesh networking, etc.). Go into a shopping mall and give a random person an OLPC -- what would they do with it? Sure, it has some included apps, but that can't be sufficient to meet the needs of millions of kids with every need and in every environment imaginable.
I hope OLPC works out great, but I can't imagine anyone who has ever designed systems looks at this and thinks anything else but -- great platform, but for what applications?
Look at the OLPC wiki.
If a country buys a lot of OLPCs, say 1M, that's $150M. I think they can throw in another million for i18n.
Most of Sugar, the OLPC's desktop environment, is written in Python. The source is here:
http://dev.laptop.org/git.do?p=sugar;a=tree
I just tried it out, and I am pleasantly surprised! It's amazing how much faster Python is for desktop applications than Java is. Even when using IBM's SWT for developing Java applications, they still feel far more bloated and slower-responding than OLPC's Python-based GUI applications.
I would have expected Python to be slower than Java, but apparently that is not the case. It could be that the layers upon layers that make up Swing really slow it down. Maybe it's time for Sun to take a page from OLPC's Sugar project, and develop a UI framework that is fast and easy to use.
http://wiki.laptop.org/go/Sugar_design_review_3/
Lameness filter is a lameness filter
"it still falls victim to engineered obsolescence"
How do you know? Have you inspected the hardware?
I've never understood the concept, really. How does one engineer a product to work properly through the warranty period, but magically fail when it's out of warranty? Certainly, some manufacturers use inexpensive parts when they think they can, and sometimes those parts fail, but it's hard to imagine that's an intended effect.
Maybe I'm naive.
Why yes, I AM a rocket scientist!
For sources and development information see here: http://dev.laptop.org/git.do?p=sugar;a=summary
-- http://htpc.info-on-the.net/
Fuck charity, we need to change the global economy. If you want to help the poor in the third world then don't give them charity unless they are literally starving. If you want to help you should buy what they produce, lobby your government to write off the debt they made them take on and lobby your government to remove trade restrictions. Your country is fucking the third world in the ass and given you live in a democray they are doing it in your name. You need to stop the fucking, not start the giving.
What you are describing is not "engineered obsolescence" but "engineered failure," and indeed is hard to imagine manufacturer's doing. Obsolescence != failure.
Engineered obsolescence means that the manufacturer's product roadmap is such that the product bought today is superceded by better products in a relatively short timeframe, enticing people to keep buying over and over again.
Momentarily, the need for the construction of new light will no longer exist.
Forgive me if I use the term 'engineered obsolescence' a bit more broadly than I should have. I don't mean component failure specifically, and certainly not with respect to warranty duration.
Do you know if this $100 laptop is upgradable? I'm sure that as the lustful fires of consumerism awaken in these nations' loins, they'll want harder, better, faster, stronger laptops that these corporations will be all too happy to *sell* them, as the OLPC simply doesn't meet the gluttonous standards of a modern consumer. It looks to me kind of like what a drug dealer might do with 'free samples'.
I may make you feel, but I can't make you think.
Linux desktop is getting nowhere, despite it's technical excellence,
1. It's very easy to argue it's getting somewhere because of the variety of distros out there. Just because NetCraft or whatever research name you look to for credibility can't/won't measure or validate the progress means absolutely nothing.
2. Putting together a coherent desktop is difficult to say the least. Your average Linux desktop won't be competing directly with apple/microsoft, but you will find pragmatic IT people deploying them everywhere. No, none of those people have been the subjects in desktop market share research either.
because it lacks key apps (i.e., Office). Pull a few key apps from MacOS X (e.g., Office, Photoshop, etc.) and see what happens to adoption.
This is a well-worn and ultimately invalid opinion. History shows us repeatedly that the switch happens when one platform has something a consumer **really** needs. Making look-alike office and graphics apps is not the answer. The answer is a little deeper. Maybe openoffice.org might have something really great lawyers would switch for. Maybe gimp has features that animators want they can't get from Adobemedia. (filmgimp?)
We know it hasn't happened yet, but it's already begun. Proprietary software companies like Microsoft and Adobemedia will tighten the noose by raising prices and offer progressively less innovation. History shows this over and over again.
http://www.maxineudall.com/2010/02/should-economists-be-sued-for-malpractice.html
The word "OS" is not mentioned in the article.
"Do you know if this $100 laptop is upgradable?"
Do you know that it isn't? Do you know if it needs to be upgradeable? I've got a laptop that's several years old, and I wouldn't even consider upgrading it.
"I'm sure that as the lustful fires of consumerism awaken in these nations' loins,"
OK, holy cow. Could we please dial back the rhetoric a little bit?
"they'll want harder, better, faster, stronger laptops that these corporations will be all too happy to *sell* them"
Yeah, sell them for $100. And these people who may or may not want upgraded laptops either will either buy one, or not. Or they might set up a cottage industry to upgrade their neighbors' computers, thereby earning money. You know, kinda the way the rest of us do it.
I find that a lot of people who argue about the evils of consumerism are more interested in telling me what I should or should not do with my money than actually looking out for peoples' best interests.
Why yes, I AM a rocket scientist!
So basically, computers should stop getting faster so that you won't feel bad because you bought one?
Uh, no.
Computers do not lose capability over time. (Except for Windows machines.)
Why yes, I AM a rocket scientist!
No actually, the closest they'll get is an enhanced derivative version of Abiword.
...they'll want harder, better, faster, stronger laptops that these corporations will be all too happy to *sell* them, as the OLPC simply doesn't meet the gluttonous standards of a modern consumer.Oh good God. The point is they can't afford standard consumer electronics as it is. That's what the whole project was about-- provide a low cost computer to people that can't afford current computers. Great insight there. With out a doubt OLPC will soon be trying to sell the latest core 2 duo laptops to the children of Bangladesh. Hell, they'll probably start a new campaign, One Widescreen HD Plasma TV Per Child (OWHDPTVPC), next, just to sucker those unsuspecting s fools in even more.
Because a major reason for the low price is that they aren't doing the kind of packaging and marketing, etc., they'd need to do for individual sales, the cost would be significantly higher than $100 (or even the $150 that looks like it will be the "early adopter" cost) if it were sold to individuals, without any excess to put toward a subsidy.
That being said, OLPC is looking at making a somewhat more expensive and capable derivative system for sale to the public in the US and other advanced countries, but its a secondary priority.
Bloat, and dependencys. This should be a very MINIMALISTIC system, with out any bloat like XML. Furthermore the main document-exchanging will be between those laptops.
I do beleive he was assuming the previous poster, and then you, were trying to say something with some possible relevance, and attempting to devine what that might be. So rather than getting all upset he guessed wrong, perhaps you could enlighten us?
"engineered obsolescence" certainly implies some intent; Do you suggest a $100 dollar laptop, or any laptop, could possibly be designed such that it would not become obsolete?
quote FTA: ""In fact, one of the saddest but most common conditions in elementary school computer labs (when they exist in the developing world), is the children are being trained to use Word, Excel and PowerPoint," Negroponte wrote in an e-mail interview. "I consider that criminal, because children should be making things, communicating, exploring, sharing, not running office automation tools.""
It is my impression that the whole idea of creating a brand new interface is to escape the eternal upgrade spiral. On the surface, they do away with folders and mainstream OS vendors, but consider how this affects the entire paradigm of computing. In a few years these people will be old enough to work in an office (not saying they will, it's just a possibility), and set me tell you, I think they're not going to *want* to touch Windows, MacOS, or KDE/Gnome with a fire poker -- it's too messy. They won't want to work on their computer, they'll want to work on their *tasks*. You and I can do a lot more by donating to charities or 'adopting' a child through a group like World Vision. Great! By all means, if you are so inclined, fund and donate all you like!
As you state in a later post, hardware failures are a different topic; that's mostly a question of build quality and durablity. While it is to a high degree possible for a manufacturer to skimp in this department, and thus encourage more purchases, it's not my impression that the OLPC project has chosen this path -- quite the opposite.
"Good news, everyone!"
In fact, there will probably be a fork of Edubuntu with a name like Subuntu, Sedubuntu, or OXubuntu, unless the devs figure out how to fit it all on the same CD anyway. In that case, different users logging into the same machine can have different default sessions. Those who feel confined by Sugar, who make the effort to learn the desktop paradigm, can use GNOME, KDE, etc.
[100% ISO 646 Compliant]
SVM, ERGO MONSTRO.