Slashdot Mirror
Thin Client With OSS for Developing Nations
FridayBob writes "The BBC has a story on a new, ultra-thin client that a group of not-for-profit developers, Ndiyo, hope will open up the potential of computing to people in the developing world. Not surprisingly, their system uses open source software. The system runs Ubuntu Linux with a Gnome/KDE deskto and OpenOffice. From the article: 'Licences for software are often a significant part of expenditure for smaller companies which rely on computers. But a recent UK government study, yet to be formally published, has shown that open source software can significantly reduce school budgets dedicated to computing set-ups.'"
What about thin clients for models? They regurgitate whatever information you feed them.
Such a bootable cdrom (based on Slackware) is already available from LocustWorld.
Maybe the Ubuntu guys should port it over from Slackware.
Seastead this.
I read their white paper. It's not a diskless boot setup. Rather it sends the screen image over Ethernet.
I have no idea what this story is about. Seriously.
100 pounds!? Don't they understand that by using open source software their total cost of ownership will be much greater than if they used Windows. Get with program, poor people. Make Bill richer.
Why are these cheap entry-level systems always targeted at the "Third World", rather than poor people here in the US? They'd have much better chances of success in our society, already geared for computer-readiness, in becoming popular - or gaining entry at all. Poor Americans have less of a culture gap to close to become computer users, and are much more able to bootstrap themselves into becoming unsubsidized computer consumers like the rest of us. And American products filter out to the rest of the world after they're out of fashion here, so feeding the American poor would eventually feed the foreign poor, too. Without setting up the foreign poor as better competitors to our domestic poor, upon whom we all depend. The products would be easier to produce and distribute. Aren't our own poor people worth helping?
--
make install -not war
I wonder if their FPGA-based design is really cheaper than using a Geode or Xilleon.
The benefits of thin-clients are many. First, the client can be really bare bones (i.e. no HD, minimal RAM, low-end graphics, low processesor speeds, etc) so they can be cheap ($170 + monitor from WalMart or donated machines). Second, to upgrade all your workstations (perfomrance-wise) all you need to do is upgrade or add another server - not hundreds of workstations. Third, to upgrade all you clients (software-wise), you just upgrade the software on a few servers. Managing one or two Win2K3 servers for viruses, patches, malware, etc, beats the hell out of 200 WinXP workstations!
There are other benefits, but these are the ones that have really made a difference for us. Don't get me wrong, thin-clients aren't the answer for everything. There are many situations where you need to have a fully functioning workstation. However, with the money you save on thin-clients, you can afford to get really good workstations, which in turn can be turned into thin-clients when they are needing to be upgraded.
Most of our users simply need a means of doing basic office tasks like word-processing, spreadsheets, email, web-surfing, etc. Those are perfect for thin-clients.
What would I want to have to make it better? Easy. First, get OpenOffice to work properly on a Win2K3 terminal server, It's not real good in a multi-user environment like that (unless I'm doing something wrong - possible). And the number 1 thing that would make it better: can you say "Tiger Terminal Server Edition"?
"terrorism" and "pedophilia" are the root passwords to the Constitution
We have one location with over 200 thin-clients on a 100Mb network. The impact is minimal. With QoS and propper VLANing you can to much more than that. Just web-surfing and email take up more bandwidth than the thin-client traffic on that network.
"terrorism" and "pedophilia" are the root passwords to the Constitution
I have never understood the idea that "third world" people want, need, or have to settle for "miraculous" $100 computers or thin clients. The truth is that in "third world" countries, bare bones PCs that run your choice of Windows or Linux simply don't cost a hell of a lot more than $100, and often less. It's all about what the market will bare. This thin client bull shit is just more of the same non-solution looking for a non-problem. People in "third world" countries that want computers have them, and those that don't know that they "need" them can get REAL computers cheap. And, thin client or not, it matters little if there is nothing to connect them together. You know, like phone lines, fiber, dish, wireless? Think about infrastructure, than give them REAL computers.
"Who are in control, they are not in control of anything - they don't even control themselves!" - Glen Beck
The people you're targeting get paid $50 a month, my friend, and their kids go to a school which is basically 4 walls, a floor, and a roof if they're lucky.
Oh yes, a server and some thin clients is really what's needed there.
Not paper and pens. Text books. Teachers. Electricity (what are they planning to plug these things into?).
The thing about developing nations, is not that they're poor, it's that the divide between the rich and the poor is vast.
At the other end of the scale, here, you have your rich, your ex-pats, etc - and you have your $5,000/term "International School" organisations who have wireless internet, computer labs, international standard teachers, and they don't need this. Nor do the businesses, most of which are thriving, thank you very much.
I'm sitting here next to a 3Tb server in my office and a server room full of Dual Xeons next door reading about how developing nations need some sort of solution for cheap computing?
These people have so lost direction they couldn't find it with both hands and a map.
It actually looks like a nice system, that would be ideal for reducing costs in schools and some businesses world-wide, I have NO idea what they're doing thinking they're doing this for the good of the "third world".
If they really want to do something "not for profit", try volunteering for an aids project, a humanitarian project, or a teaching project.
Sometimes I look at my driver - I pay him $65 a month, and I wonder what he would have been if he'd had the education I did. HE would be sitting in this chair, for a start. I could teach him in front of this PC for a month of Sundays, and it wouldn't make up for the fact he has no basic education.
Pay attention to the beef which is OpenOffice. I am afraid that SUN may pull the plug on java, which OO.o has come to heavily rely on of late. SUN could simply change its license. Let's remember that SUN is practically in bed with M$ after having received some big cash ftom M$, and has never criticized SCO for its actions.
I personally advocate the forking of OO.o portions that are GPLed so that we can finally be free. How about that?
I don't know about the 3rd world, but the ideal client for most libraries and schools would be:
1) video, keyboard, mouse, optionally local removable read/write storage
2) operating system, e.g. Linux, with essential utilities, e.g. firewall and antivirus software
3) web browser
4) most common lightweight apps, e.g. low end word processor, and perhaps software specific to the given installation, e.g. front end to a card catalog or other database.
5) remote access to heavyweight, lightly-used apps like OpenOffice, running on a nearby server
with hardware just beefy enough to run the local apps plus a few web browser windows plus a few remote-access windows.
All of this would boot from a read-only, or at least read-only without administrator action, medium, to all but eliminate the threat of malware and end-user malice - reboot and the damage is undone.
Knowledge is how to play a game, intelligence is how to win, wisdom is knowing what game to play.
Too bad a Sun Ray costs more than a diskless PC.
First, the client can be really bare bones (i.e. no HD, minimal RAM, low-end graphics, low processesor speeds, etc) so they can be cheap ($170 + monitor from WalMart or donated machines).
Yep, thin clients are great, when used in the right places. And they have many advantages. But... price isn't one of them. Not yet, anyway.
Where I live, PC's up to around 200 MHz. (original Pentium and below) are effectively free. You want one? Look around, hand over a sixpack of beer, and you have one.
Now with a $170 budget, I can get you a (used) PC that includes monitor, and beats the crap out of any thin client you can find for same money.
How come? Well, we all know electronics today are 'cheap' thanks to the sheer numbers they're produced in. Apparantly in today's world, a standard beige box with off-the-shelf components, is still cheaper (to produce, or second-hand), than a book-sized thin client produced in limited numbers.
For a business, the numbers may differ. If you'd use old hardware in a thin-client like fashion, you might have to hire someone, to manage parts, build and repair boxes fulltime. In that scenario, it may be cheaper to spend $170 once on thin clients, and very little after that on managing the hardware. But the savings here are not hardware, but management costs. Which I think is the advantage of thin clients anyway.
I hope some day (maybe soon) these economics will change, and make smaller/smarter boxes cheaper than equal performing WalMart beige boxes. Because they have many advantages, and I happen to like small+smart boxes. Even if they're still a bit bigger, like Mini-ITX or Shuttle XPC's."Jesus. What kind of box are you trying to buy? $1000 US? Maybe you should set your sights a little lower and realize that not everyone needs a game box."
I wasn't going to reply at first, but then I realised that you genuinely don't get it. The country where I live cannot survive on income tax revenues, because the cash economy is almost non-existent. This means that it relies on import duties, business license fees, etc. for its revenues. This means that things like computers have hugely inflated prices. PCs, for example, have a 40% duty slapped on them. Vendors also add large markups because they pay extremely high business license fees.
All this means that a low-end computer that would cost about USD 4-500 ends up costing not less than USD 1000 when it arrives here. Is it clearer now?
And stop calling me Jesus. 8^)
Crumb's Corollary: Never bring a knife to a bun fight.
Here in our university(UP), we've been using LTSP to create thin clients. We run a powerful server (2ghz, 1gb ram) and it can host up to 20 Pentium computers.
What's nice about the thin client setup is that once an application is loaded, it boots really fast on all the clients. For instance, we start OpenOffice on the server and it boots with a second on a client.
Another advantage with this setup is control. Since all the clients run on the server, we can restrict access and prioritize security.
Cheap and abundant net cafes. People in the developing world don't need to play quake. If they can rent a pc for some period of time, it helps them communicate. Just like the communal pay phone at the general store worked decades ago. Abundancy means they don't have to walk a long distance.
When I lived in New Delhi, brown outs were a frequent problem and surges were a real hazard. A colleague of mine returned home once to the smell of ozone and burning plastic and found that a surge had left a scorch mark where his fax machine used to be. Any kind of electronic device required a special kind of expensive current regulator to protect against spikes, and even with one you were no better off if the power ever up and died. A desktop system in this sort of environment would be a mistake. Using laptops is a huge advantage, since it lives on its own power supply and really doesn't care much (or so it seemed to me) about brown outs and surges. So... Go MIT!
BUT, what you appear to be saying too is that developing countries can choose only three of the four items you list. That's bunk. Education is central to long-term development and while e-learning is mostly bullshit, computer science and engineering education needs computers.
Also, you'd be best to check your assumptions about the "Third World" at the door. There is a world of difference in the infrastructure and education needs between Burundi and Papua New Guinea on the one hand and Mexico and Kenya on the other. There are many 'Third Worlds' and few of them are of the sort you are probably imagining.
Why is everybody here focusing on price and not on total cost of ownership and environmental costs, which are the real point here IMO?
13-4=54/6
The solution that eventually occured was that IBM donated a number of G40e laptops (thank you guys!) so we were able to put low-power, fancy computers out in the field.
Now the crux of the issue to me, and this is something which I've brought up earlier, so bear with me, is the question, Now What?
I've got a 2.8 gig 802.11g machine with 512 MB of RAM sitting here, doing what?
The hard part is making it useful. Not many people out in the villages enjoy reading slashdot regularly, so we have to find useful things that they can use these beasts for.
Essentially, what we did was to create an information portal, data was downloaded every day over a CDMA 1x connection, and presented in a form which was accesible to people. Weather forecasts, crop and vegetable prices, information about government schemes, employment opportunities in the nearest town and so on. If you want to know more, then drop me a mail and I would be happy to give you full details. Better yet, if you are involved in something similiar, please do get in touch.
Now the technology part is cool. I designed it to work completely in our favourite browser - Firefox - *ducks*, I used CSS to make sure that when you print out the information it's in an easy to read form. Also, since Open Office, FF, the Linux distro itself a number of other applications have recently been translated into the local language (Tamil) it has been easy for the people themselves to use it, rather than needing either an external person, or to have to painfully learn a new language.
Just to quickly respond to the infrastructure part, India has been really good at providing communications infrastructure at a grassroots level. Every village is linked with a 2 mbps pipe, and wireless internet using CDMA is fairly easily available. This is a god-send for us, who want to put an IT project in, without having to build this stuff up from scratch. I speak from experience in Indonesia, where we had to transmit using VHF. Fuck, that hurt.
Now sub-$100 machines are good. But, like someone else was saying here, the people themselves are NOT going to be buying this. It's more likely to be governements, NGOs and the like who do bulk-purchases and then provide them in conjuntion with various other schemes. Remember that in many parts of the world the annual income is less than $350. This is equivalent to somebody paying about $13'000 for a computer in the US (if they earn about $40k, which I assume an IT manager will).
The technology is cool for us. How useful is it for them?
R.
I like the nick. It speaks volumes for your comment. :)
There is a Universal Life Value Check it
This is a clone of the SunRay solution (see here). With SunRay's you use a smartcard to access, which has the cool feature, that you can pull your card from the thin client, walk over to another (to discuss something with a colleague, to give a preso in the boardroom, to go home even!!) and plug your card into another SunRay, where your session continues as if nothing had happened.
Browsers shouldn't have a back button!! It's all about going forward...
And, FPGAs are really useful as an implementation, not just for prototyping. It can cost up to $30M to set up an ASIC line, not counting engineering or the $100K+ for the tools ...
Yep, some FPGA's are cheap enough that they can be used for both prototyping and production. But comparing one of the least expensive FPGA's in the world with a near maximum "up to" price for ASIC development is not useful. And yes, I'm an FPGA engineer as well (waiting impatiently for 7.1.2i to be released), so I'm not saying this for my love of ASIC's.
Competitive ASIC's can be developed for WELL under $1M (probably even well under $500k for a good number of devices) including tools and NRE/prototype costs. Still doesn't remove the fact that it takes considerable volume to come out ahead against a $7 FPGA. Against a 2VP50, at closing in on $1k a piece, it's a completely different story.
-- PGP keyID: 0x4C95994D
NX is great for running over slow connections (sub-cable modem), but it doesn't seem to drop frames if the CPU can't keep up... I don't know though, I'm not an expert in this stuff.
Anyway, it looks like the big advantage of the Ndiyol is that they've done a lot of work to come up with a custom protocol and/or cheap hardware set-up that actually work out of the box for $170. I noticed that they ship with 2MB of RAM. Also, they plan to move the whole thing on to one chip... that should cut cost significantly.
If they can get full screen video streaming + sound to work, and provide a connector that lets me plug in a remote control, I'm putting one of these in my living room. Assuming that it can run in hot environments, it would outperform all of the sub $2,000 systems that I've been able to find on the market these days.
(I've been trying to find something that can run as a mythtv frontend, and be silent, small and stable in a 95F room. There are probably some systems out there, but it is really hard to find out the operating temperature range or noise level of prebuilt computer systems.)