While nobody hated the iPad, by any means, the iPad was edged out by some key feedback, said Joel Handler, Hillsborough’s director of technology. Students saw the iPad as a “fun” gaming environment, while the Chromebook was perceived as a place to “get to work.” And as much as students liked to annotate and read on the iPad, the Chromebook’s keyboard was a greater perk – especially since the new Common Core online testing will require a keyboard.
Another important finding came from the technology support department: It was far easier to manage almost 200 Chromebooks than the same number of iPads. Since all the Chromebook files live in an online “cloud,” students could be up and running in seconds on a new device if their machine broke. And apps could be pushed to all of the devices with just a few mouse clicks.
Hillsborough educators also tend to emphasize collaboration, and they found that Google’s Apps for Education suite – which works on either device – was easier to use collaboratively on Chromebooks.
I’m shocked – shocked! – that a device with a keyboard is more useful in educational settings than a tablet.
Why ever buy overpriced iDevices for school?
Its literally public knowledge that Apple have biggest profit margins -> schools pay large percentage of total for nothing.
(Go to Apple own quarterly reports)
So ?
On the other hand I can see appeal of netbooks/chromebooks. Delivery of content over net is really great thing from maintenance standpoint. (No need to do anything even on serious hw/sw upgrades!)
And this way kids may use whatever devices they have at home too.
Add to it that schools rarely have any serious IT stuff. (IT teachers can be for example dual music/IT teachers …….)
So easy of use is important, but easy of maintenance and administration is even more.
And that price tag…
Only problem I can see (for both iDevices and ChromeDevices), are state egzams… which usually relay on specific software, like MS Office or OpenOffice. Both of which wont run on any of the devices. (And pls do not give me that s***** about MS efforts. Same name =/= same application!!!!!)
Most kids with iPads through schools are on schemes where they pay them off over the school year. This was the scheme the company I used to work for had. The kids got inclusive insurance, so any valid damage was included in the monthly cost. The hardware was incredibly cheap because Apple threw massive educational discounts at the schools. The device ended up costing less over the period than it would have done to buy up front and also insure.
We also offered more traditional laptops (netbooks) and most schools went with iPad. No idea why. No one pushed a specific technology, it was just whatever was whatever the educators chose.
In July I went to a one-week seminar, and one of the presents they gave us was an Android tablet to take notes. The first day I didn’t have my laptop or paper to take notes. I used the tablet they had given. The second day, I left the tablet in my hotel room and took my laptop for the class. I just can’t get used to typing on a touch screen. And I tried.
Because all the data lives in the cloud it can be monitored and snooped upon by the school.
No ‘selfies’ etc etc etc will be allowed.
The school, aka big Brother, will be watching everything their pupils do.
Of course the school likes that. Who wouldn’t eh?
Of course. That’s what a school is supposed to do.
Note that this is no different for any device – desktop, laptop, or tablet.
Sadly common core required California among many other states to LOWER educational standards, testing standards and overall bring the entire school system down to the lowest common denominator to qualify for any of the grants by the federal government.
Federal US government – making people smarter by making the tests easier since 1979. =P
I can’t tell if that was ironic or not.
Schools aren’t meant to be substitute parents. Maybe watching everything a child does is a parent’s job, and some say even that is going too far.
I can’t tell if that was ironic or not.
Schools aren’t meant to be substitute parents. Maybe watching everything a child does is a parent’s job, and some say even that is going too far. [/q]
It’s not being a parent, it’s watching their network for suspicious activity. If the student is at school they should only be doing school work. Same thing when you’re at work.
Then parents would have to attend school with their children.
When a child is in school, the school is responsible for them and their safety. That’s not being a substitute parent, that’s simply taking the responsibility that is necessary for a school system to work.
There was no mention that the school would limit this to school hours.
Why? School children don’t have a right to privacy?
Not if they are using school equipment or property.
So it depends on how the system is setup.
If they’re paying for the ipad/chromebook, and its not on the school’s network,they should expect privacy.
If it is school property or uses school resources, then they have no expectation of privacy.
See all the court cases on locker/backpack searches by school administrators.
I was shocked too, when I was at school my teacher’s regularly used to leaf through my notebooks. Occasionally even tear pages out!
Facists!
Well, when you get right down to it, the kids are in SCHOOL….to LEARN….so teachers checking up on what they do, exercises, labs, etc etc are a part of that. Checking Facebook, Instagram, and all the other stuff people do online is NOT a part of schoolwork, so the argument about Big Brother does not apply here in any sort of way. I have seen the software and systems schools have used to interface with their iPad setups, and it allows the teacher to check up on their work, see where they are making mistakes, and then work with the student to help them understand, learn and do better. I suppose this new Chromebook system does the same thing somehow.
Edited 2014-08-06 10:13 UTC
My 6th grade daughters were creating blogs as part of their school curriculum as well as working with online calendars. There are other online skills that could easily be considered crucial in today’s educational system. I agree that Facebook and other pure social sites don’t qualify, especially for the younger kids (under 13, depending on local laws).
In my neck of the woods having at least one online class is now required for graduation in many schools.
shotsman,
Is there really any expectation of privacy when you store stuff on the school’s devices & network?
This is nothing…what’s truly creepy is when the school is literally watching what their pupils do even at home.
http://boingboing.net/2010/02/17/school-used-student.html
Edited 2014-08-06 15:18 UTC
Teachers love this stuff. saves them having to leave their chairs.
http://www.lanschool.com/lanschool/features/ios-student
http://www.lanschool.com/lanschool/features/chromebook
If you want to post your selfies on Instagram then do it at home. School is for learning.
Quoting Tom:
^A<>
Actually, I am not surprised.
As soon as one has to enter blips of text longer than 120 characters, any device with a physical keyboard becomes preferable over a tablet.
A tablet is a handy device for interchanges mostly in the “to user” direction. However, it is somewhat limited in anything more complex than a few buttons/short text for the “to system” direction. Mastery of a subject is best demonstrated by expressing abstract thoughts (after cognitive processing) using text.
This message, even much shorter than an assay type answer, is around 850 characters long and was entered using a physical keyboard!
Whoosh. You missed the sarcasm on that one.
I think your sarcasm detector is malfunctioning.
*of course* a device with a keyboard is more usable by students than a device which hasn’t
Edited 2014-08-06 12:34 UTC
It’s been a while since I’ve read an article on OSAlert – one that seemed interesting enough – so my sarcasm detector failed to activate.
Nevertheless, what seems obvious to some would not be to others. In the event that some school administrator or trustee encounters this article in some distant future, he/she may not grasp the reasoning and remain within the hype-fog of device suppliers targeting the education market.
What is Googles agenda here? I mean they don’t make money on any of this and I am sure they are not doing it to be nice?
And that’s what worries me because that means Google is most likely collecting data on all these children. From their locations to the things they look at etc.
Not sure I want my child watched that much and data collected on them for future marketing etc.
And for which OS that would not be true?
Debian.
Is Google the one pushing this? It sounds like a decision made by the school. Google’s agenda is to make money off those computers, directly or indirectly. That a school ended up choosing them is not Google’s “agenda”.
Google and Facebook are already getting loads of data about children without specialized hardware anyway, so there isn’t any agenda that is specifically for school use computers.
That’s actually a gross misconception with the education version of the ChromeBook. First children are protected under COPPA, CHILDREN^aEURTMS ONLINE PRIVACY PROTECTION ACT. Children don’t actually use their names when logging into a ChromeBook but a student ID, for example; 123456.[schools name]@gmail.com. Second, ads are disabled and there is no way to enable them, even with the admin panel, feature has been completely removed. So when the student visits Google Search, they only see search results, nothing else. Thirdly, Google doesn’t track students, nor do they scan their email boxes.
Here is some more information.
http://googleenterprise.blogspot.ch/2014/04/protecting-students-wit…
http://www.google.com/edu/privacy.html
https://support.google.com/a/answer/139019?hl=en
Edited 2014-08-09 07:21 UTC
But, but, it’s not a real OS! You can’t get any real work done, when the internet goes down. Chromebooks can’t actually be useful can they?
Can there be a cue? My impression is most people are apathetic to ChromeOS. Or are you doing a reverse psychology thing to increase ChromeOS’ mindshare? Like Fox News’ “Secular War on Christmas” thing?
Chromebooks are actually useful little devices. However, there’s a huge caveat to using them: you need oodles of Internet bandwidth to use them … especially if you use any of the collaboration features of Google Apps for Ed.
30 Chromebooks will completely saturate an E10 fibre link (10 Mbps symmetrical). Just turning them on, logging in, and trying to access Google Docs. Our bandwidth graphs just flatline for the hour they’re in use. And they’re not actually usable during that time, the kids are sitting around waiting for things to load.
60 Chromebooks in an elementary school with only an E10? You can’t even login and load the first website.
Want two or more student to collaborate simultaneously on the same Google Doc? You need about 1.4 Mbps *per student* *per document*. And it goes up from there the more students you add to the same document. Don’t even bother trying to get 30 students to open the same document unless you have at least a 100 Mbps connection, and you’ll need more than that to actually type in the document.
Chromebooks are bandwidth hogs. They actually make iPads, iPhones, and iPod Touches appear to be good network neighbours, and they’re simply atrocious when you get more than 10 onto the same network.
We’ve been migrating our elementary schools off their province-managed Internet connections (4 Mbps ADSL unless they were lucky enough to get E10 a year or two ago) to point-to-point wireless links with nearby secondary schools (which have gigabit fibre owned by the district). These give each school 100 Mbps (theoretical; 60 Mbps actual). For the schools that don’t have line-of-site to anywhere in the district, we’re upgrading them to E20 and E30 (20 Mbps or 30 Mbps) using the money saved from the wireless project. Simply to support Chromebooks.
However, no matter how fast we make the school’s Internet connection, all Internet traffic still goes through the provincial gateways (2x 10 Gbps currently). Which means, even if the school has 100 Mbps, logins still make round trips through the provincial gateways and fight for bandwidth with the rest of the province. Same with file saves, and history, and settings, etc. We spent almost a decade now improving the system by keeping everything local to work around crappy Internet connections … and everyone wants to stuff everything “into the cloud” where we can’t manage it, and we all fight for bandwidth.
Local will always trump remote. Too bad nobody can get a teacher to understand that.
That is interesting. Google themselves claim that 0.2-0.5 Mbps per concurrent session is enough.[1]
From reading other reports of the Internet and from opening a Google Docs session in my browser it seems that the actual requirements are even more modest.
A report from Principled Technologies[2] found that, while Chromebooks have much higher bandwidth use compared to Windows notebooks for certain tasks, for document editing this was around 10 KBps (80 Kbps).
If you are seeing drastically different results, it could be that something else is wrong in your network or with the Chromebooks. A closer look might be appropriate.
One big issue for many Chromebook schools was that their Wifi infrastructure was not capable of accomodating several dozen active clients simultaneously. This made it necessary to install one 5 GHz access point per classroom.
The amount of work for managing local clients is much higher. When someone comes along and provides a similar turnkey solution which comes with zero-effort backups, collaboration, central device management and malware protection, then surely they would win some of the business.
Until then, it’s fix your network or bust.
[1] https://support.google.com/chrome/a/answer/3339263?hl=en
[2] http://principledtechnologies.com/Microsoft/Chromebook_PC_network_t…
Internet connections and WiFi often are a total mess.
First of all, it’s usually not about bandwidth these days. It’s about latency: https://www.igvita.com/2012/07/19/latency-the-new-web-performance-bo…
It turns out we were all building and deploying Internet gateway devices the wrong way. A problem called Bufferbloat arose from this.
Deploying WiFi properly is no easy task.
WiFi itself is pretty broken.
Here is a presentation where both get discussed: https://www.youtube.com/watch?v=Wksh2DPHCDI
Have you considered instead of deploying one large WiFi AD, to deploy a bunch of smaller WiFi ADs ?:
https://www.usenix.org/conference/lisa12/technical-sessions/presenta…
Wow, really? That’s your suggestion? Yeah, cause we definitely wouldn’t have thought of that over our 6+ years of doing wireless in the schools. :roll-eyes:
In the elementary schools, we have 1 AP for every 4 class rooms (in a square around the AP). Dual-radio, 2.4 GHz 802.11g, 5 GHz 802.11n (54 Mbps). These are being upgraded this year to dual-radio 2.4/5 GHz 802.11n (dual-stream).
In the secondary schools, we have 1 AP for every 2 classrooms, sometimes every 4 classrooms, and sometimes for a single room (depending on usage). Dual-radio 2.4 /5 GHz 802.11n (dual- and tri-stream). These are being upgraded to dual-radio 802.11ac APs (with the older ones trickling down to the elems).
With the exception of the libraries during in-service days (when they stuff 200+ staff all with phones, tablets, Chromebooks, and laptops in for an online workshop), wireless bandwidth is not an issue for us.
Internet bandwidth is, and Chromebooks are bandwidth hogs.
I meant multiple APs per classroom, but clearly you have a plan and don’t think that is where the problems are.
On the issue of bandwidth, is it all at the start of the class ? So when they all open their Chromebook at the same time ?
Or does it stay congested, because I’m surprised it’s a problem.
But maybe I know what is going on, you have a bunch of Chromebooks in a classroom and every kid gets a random Chromebook. So they all start downloading their profile-data.
Maybe it would have been better to have a dedicated Chromebook per student. Then probably the issue wouldn’t exists.
Lennie,
It really shouldn’t matter, students should be able to login to lab computers without these problems. I’m not very familiar with chromebooks, however going by phoenix’s posts it seems like a completely stupid implementation. The obvious answer is to keep student profiles & data on a server at the school or at least in the district (Sun Workstations come to mind as a very robust solution in their day). Even if there was sufficient bandwidth, it still seems like a terrible idea to have devices which are continuously reliant on connectivity with google. It makes me wonder how these devices actually got approved?
If I’m correct then they are being used in a way they were never intended to be used.
It’s not stupid. It’s just a generic product.
Lennie,
I’m trying to find out if google offers local chromebook management, but everything I’ve found shows that it is indeed being used as google intended, managed through google’s website rather than under local control.
https://sites.google.com/site/chromebookclassroominthecloud/chromebo…
http://www.google.com/intl/en/chrome/education/devices/features-man…
But it’s still stupid from an engineering point of view, it’s not a solution any competent engineer would have proposed for a school district. Just think if local windows workstations needed to be managed through microsoft.com without local domain controllers, that would obviously be unacceptable, right? Google clearly chose this design because their business model overrode robust & efficient engineering considerations.
Maybe I’m just resisting the inevitable future of things, where everything is in kept in external data silos under third party control. I just feel it’s creepy and dangerous to be so needlessly reliant on google.com (or anyone else for that matter) for data and device management.
Edited 2014-08-08 12:42 UTC
You can also look at it from an other perspective.
This solves a lot of other problems they don’t have to manage anymore.
No local servers, no management of devices.
Fast iteration of new services/features from Google, because they don’t have to wait for people to upgrade their own servers (that is the same advantage as every web companies has. This saves a lot of developer time/effort).
Those are some pretty big advantages.
They can add local storage server feature later, if they really want to.
The solution, as long as they don’t add a local server feature: add more bandwidth or give each kid their own Chromebook then there is nothing large to download at the same time.
My guess is, this serves the needs of most of the schools. When I mention most, I do mean: maybe even 90%
Also remember, if I’m not mistake, there is a leased option for Chromebooks. So adding a few more Chromebooks doesn’t create some large amount of money that has to be paid at the start.
Instead of a local storage server, they could also add a local caching cloud storage proxy server to save bandwidth instead. That would be even better.
Edited 2014-08-08 12:52 UTC
I disagree, all of this would still be possible with a local server, had google provided one. Administration could have been identical. The reliance on google.com is an unnecessary artificial construct. The more devices there are, the more a local server makes sense from a financial perspective too.
Yes, hopefully they do since it was poor to centralize the design around google.com from the beginning. Although I still think google chose it deliberately due to their business model over engineering reasons.
It depends how the bandwidth is being used, I would be surprised if syncing profiles really takes a significant amount of bandwidth compared to syncing user documents over the session. If it’s more of a continuous bandwidth footprint, giving each kid a chromebook wouldn’t help.
And I’d argue it’s a silly requirement anyways, there isn’t anything unreasonable about having classes with communal devices, all our labs at university were setup this way and it was never a problem. Back then a state of the art OC-3 internet connection only had 150Mbps across the campus, but it was irrelevant – the internet could go down and we’d still be able to login and work locally.
Edited 2014-08-08 13:55 UTC
You underestimate the time and investments that have to be made to keep stuff running. And the overhead of managing services.
Look at it as using a website like OSAlert and some locally installed application or on your desktop or phone.
I know OSAlert isn’t getting a lot of new features right now, but let’s imagine it did.
With a website:
users don’t need to install anything, all they need is a basic operating system and a browser and the device it runs on. If something isn’t working, you walk over to an other device and similar components are probably there and you can continue using OSAlert.
The developers of OSAlert only need to put their code on their server(s). They don’t have to deal with old less compatible versions installed on peoples computers or phones.
With a locally installed application you get this:
users or system administrators need to do a lot more to keep things working: your data is probably stored on the device, if the device is dead, you have no data. So you need to set up backups. You need to install the application, you need to install updates. Maybe even handle data-migration between major versions and upgrades.
If you have your own servers, you have even more systems to manage and maintain and upgrade and replace/re-install.
The developers of the application need to keep a lot of old code to be compatible with older versions. This can increase a code base by 2 to 3 times, easily (!). The old code usually doesn’t get used as often, it will not get any security reviews. It will also use up extra resources when running on the devices people run it on.
etc., etc.
That is why I mentioned a Google provided caching proxy server appliance would probably be the only solution Google would even consider.
Still, this appliance can be a bottleneck too. So this could still cause a lot of slowdowns.
If each kid had their own Chromebook, the initial sync after login is not needed.
This is what is taking up the bandwidth.
kid 1 uses Chromebook A, all data is copied to Chromebook A.
Later on Chromebook A is used by kid 2, all data of kid 2 is copied to Chromebook A. kid 1 gets Chromebook B, all data of kid 1 needs to be copied to Chromebook B.
Edited 2014-08-08 14:16 UTC
Lennie,
Except that google could have made chromebook device management work the *exact same* way if google had provided a means to use a local server. There’s no technical reason to require everything to be managed by google.com.
I understood what you were saying, however it wouldn’t be a problem with a local server, and I’d still like to see evidence showing that initial login takes the bulk of the bandwidth.
No, I don’t think they see it like that and I don’t either. Are you a developer or system administrator ?
In your case the software needs to be made to run on a certain operating system and it’s version and who is going to keep that up to date ? Maybe even support different operating systems.
Or they deliver it as an operating system image (like an ‘appliance VM’) in that case the operating system image from Google needs to support a diverse set of (virtual or non virtual) hardware.
They also need the software to be able to run in a clustered configuration. Then you have to support different network configurations.
You also need to have support staff, in case things don’t work. And you need to create documentation.
Instead what they have now: The developers for this part of ChromeOS/server infrastructure at Google only needs build on version of their software.
And they run it on the existing Google infrastructure. That already provides all these pieces and much, much more.
This means a lot of code never has to be written (and old code can be thrown out easily) this saves a lot of time, money and effort for them.
Google has a product for which they already do this.
Have a look at what price this starts. It’s called the Google Search Appliance, sometimes called the Google Mini. It’s a machine they ship to you.
“The Google Search Appliance starts at $32,000 for an index of up to 500,000 documents with a two-year contract.”
http://www.informationweek.com/software/enterprise-applications/goo…
You be the judge why the business model and ‘packaging’ for this product is the way it is.
That is my own conclusion based on what was mentioned in the first few comments, it could be wrong of course.
Edited 2014-08-08 15:42 UTC
I’ve had both roles, and both roles tell me that a locally managed system would be better, honestly it’s not that big a deal to run a local server especially when it’s built to be a self contained network appliance. It doesn’t have to be as complex as you keep implying.
Well, why couldn’t Google just dictate the requirements like every other vendor does. It doesn’t really matter if the server runs windows, linux, or freebsd, or whatever, most operating systems already have update mechanisms built in. Google would probably be providing the hardware in turnkey fashion, so nobody would have to care what OS it used. I’m absolutely positive google could figure out how to make it work as easily as a typical consumer network device. If they can’t then maybe they should consider hiring me
They can. They just don’t want to (the usecase has to few users)
The only things we disagree on is the reason why they don’t want to.
22% of school districts and growing really isn’t a small use case though.
http://www.businessinsider.com/22-percent-us-school-districts-use-c…
I guess if you don’t want to question google’s motivation for making schools dependent on google’s servers, then perhaps you will at least admit that there’s a huge risk of vendor lock here.
What I meant is: schools with to little bandwidth and also not enough budget to buy every kid a Chromebook.
Yes, totally agree. Exactly the reason why when someone from Google contacted me if I wanted to work there I said: not interested.
They are an incredible large risk. My top risk is not vendor lock in though, because Google services are pretty open, usually build around standards and they are helping to create standards at standards bodies.
My top risk is: privacy
Lennie,
Well, it could be that chromebooks are popular because schools are underfunded, but I admit it’s only speculation. I agree WRT privacy. Anyways, I gotta run, it’s been good talking Lennie, seems we had the place to ourselves!
Edited 2014-08-09 04:27 UTC
I actually think Docker will solve a lot of the deployment issues I mentioned. So all this should all get easier over time.
I’m curious, do you use some expensive APs, or maybe fairly inexpensive consumer units flashed with your own firmware?
It’s our closer look that we did that came up with the 1.4 Mbps per student, per document.
Our wireless infrastructure handles the load quite nicely. We have just slightly over 3000 simultaneous active connections across the district on our student wireless network during the day (and just over 12,000 registered devices). Very few of our APs have more than 60 devices connected to a single radio (and that’s only in the highschools); most are under 20. Even with that load, most CPUs and radios are under 60% utilisation during the day.
But, it’s not the unicast traffic that’s the problem. iOS devices are the chattiest pieces of crap out there, and it’s all multicast and broadcast traffic. Get more than 10 iOS devices on a network, and Bonjour/Rendezvous and other multi/broadcast traffic spikes through the roof. Get 100+ of those things on there, and there’s very little bandwidth left for unicast traffic.
We’ve mitigated a lot of it via the radios over the past year, and added all kinds of filters to the switches, and we have more mitigation work planned for this fall. It’s just ridiculous how badly iOS devices work on a non-home network. And Apple continues to push them on schools.
We provide just such a solution on the desktop. We run diskless Linux stations that are treated like appliances. The only moving part in these things are the case fans. There’s no harddrive, no CD-ROM. Just a motherboard, CPU, and RAM; everything is onboard. Systems boot off the network, mount all their filesystems via NFS, and run all programs locally using local CPU, RAM, GPU (fully 3D-accelerated), and sound. If a system breaks for any reason, it’s a 15 minute job to replace it:
– power off
– disconnect all cables
– grab spare from tech area in school
– connect all cables
– follow instructions onscreen (which is to call the service desk to have the spare activated)
– reboot
– carry on where you left off
Everything is installed on the server. Everything is managed via the server. And the server is backed up every night, and replicated to a second backup system on the other side of town every day. We have daily snapshots going back 2 years, with off-line backups going back another 2 or 3.
All running Debian Linux and Ubuntu Linux (XFce desktop). It’s actually easier to manage than the Chromebooks, and we provide a lot more management tools to the teachers than the Chromebooks provide. Things like:
– screenshots of all stations in a lab (librarians especially love that feature),
– ability to view the screen of any logged in student in real-time (VNC),
– ability to take control over every station in a lab (reverse VNC; especially useful for the “watch how you do this” portion of a lesson as it removes all distractions)
– ability to turn off, turn on, or power cycle every station in a lab; including the ability to reboot into “exam mode” which is locked down to just the browser and the website for government exams
– ability to login an entire lab using generic accounts (great for the primary students)
– and more
And all for under $200 CDN a workstation. Running full-featured local apps that don’t waste Internet bandwidth unless they actually need to get something off the Internet.
But, we have a very annoying “tech coordinator” that is working against the IT department and convincing schools that “Chromebooks are the future”, and “who needs computer labs when you have Chromebooks”, and similar things. Instead of finding ways to make the existing infrastructure more useful, or finding ways to fix any shortcomings the current system may have.
We have one of the better district-wide Internet infrastructures in the province. But Chromebooks are bandwidth hogs, any way you slice it. If Google would provide any kind of a caching server so that (at the very least) logins were handled locally, things would be a lot better. But, since everything (from logins, to document management) requires a round-trip to Google servers, it doesn’t matter how good our local district-wide network infrastructure is … we’re bound by the provincial gateways that tend to get overloaded (there’s 60 school districts and 160,000 students funnelled through 2x 10 Gbps gateways).
And our network graphs at the schools with E10 connections go from < 1 Mbps before class, to 9+ Mbps flatline once Chromebooks are powered on, to < 1 Mbps at the end of class. So, no matter what the “official” docs say, in reality, in practice, in the field, they are bandwidth hogs.
Surely without a monitor? :p (what kind of hw is it?)
Hey, what happened to the OLPC tablet?