Vacation!
Jon is on vacation this week on the west coast, so don't expect any insight (or email responses!)
« April 2008 | Main | June 2008 »
May 24, 2008
digg this
del.icio.us
reddit
Slashdot
Stumble It!May 22, 2008
Ultraportable Ubuntu?
Would you like an Ubuntu to go?
TG: Will you be coming out with a tailored version of Ubuntu for the ultraportable sector?MS We're announcing it in the first week of June. It's called the Netbook Remix. We're working with Intel, which produces chips custom-made for this sector.
Though they're working closely with Intel; with any luck a "lightweight" version of Ubuntu would also be a natural fit for the OLPC (and perhaps Intel's Classmate?). Naturally, the OLPC community already has Ubuntu and other Linux versions and/or window managers running on the XO-1, but further developer support on creating an ULPC/4PC desktop system that can compete feature-to-feature with Windows could be a great asset for the anti-XP/MS/Closed source crowd.
Would you like an Ubuntu to go?
The interview reveals two insightful pieces of how the power of community has shaped both Canonical as a business and Ubuntu as a Linux desktop. First is Canonicals "rather unusual way of picking" their original employees:
I simply read a large amount of correspondence between the developers on one of the projects that is key to the way we do Ubuntu, the Debian project. It's amazing how much jumps out in terms of the way people think, the depth of their experience. So open source is not only a great way to develop your own talent and skills, but it's also a great way to get a job, and a great way to go looking for people.
An innovative way to find and pre-screen for the exact style and skillset you are looking for. The details of the deal with Dell are a fantastic vision of the strength of a community to focus on a feature and push for it:
We found out about it after it was a fait accompli. [Dell are] very much a numbers-driven company. They asked their users what they wanted to see. They had a lot of data and that data pointed to us. That was a little unsettling, because we didn't have a relationship. But it was a significant step up in our corporate profile. It will be very interesting to see what we're able to do with companies like Dell, which are aimed at a wider audience. That's my number one challenge: how to make the Linux desktop something that you want to keep on your computer.
Both good ideas to keep in mind for the OLPC community as we face changes and/or outright removal of Sugar, the move to a closed-source XP-based OS, and the general change of the guard at 1CC.
Project HA-T1093
No, it's not some early George Lucas film, it's the IADB project title for the "Pilot of the One Laptop per Child Model" in Haiti that Wayan gave a great overview of at OLPCNews.com:
In a direct contraction to Nicholas Negroponte guidance at the November, 2005 IADB meeting, where he told Ministers of Education that "To do a pilot project is ridiculous!," the IADB is not only piloting OLPC, they're also going to have objective testing on the efficacy of a one to one education model
It's probably not an accident that the World Bank's InfoDev just published a Handbook on Monitoring and Evaluation of ICT in Education Projects...
I'm interested in the hard costs of the project, which weighs in at a total of $5,100,000USD. You can trace the evolution of the project back to the original November 2007 project summary and watch the numbers dance around as they move towards the final plan of operations, signed in March 2008. Between November and March they realize the need for school servers (at over $1,000USD each), energy and security solutions, as well as increase their setup, implementation, contingency, and measurement costs, (contingency spending alone goes from just under $16k to almost $260k).
Since the 5.1 million total doesn't shift, these costs are taken out of other parts of their implementation plans. IADB's laptop purchase goes from 7,000 to 3,700 (OLPC is providing another 10,000 XOs presumably from G1G1), training and content gets cut down by $200,000, and maintenance gets $30k shaved off.
The end result is focusing on 13,200 students plus 500 teachers (the November plan was for 19,000 students, but didn't have laptops for all of them or any teachers embedded in the budget) in forty communities. The student:teacher ratio would be a respectable, if not quite believable 26.4:1 . I'm creating a spreadsheet to hold all of this information with some formulas to sort it out that you can look at yourself.
What interests me is to see how my cost estimates from November 2006 stack up. I calculated for a 5 year total implementation process, and was using some slightly different numbers (the laptop cost $148 then, for example), which in some ways overlaps with the IADB budget. I tried to honestly extract a "first year" budget from my old numbers and re-arrange the IADB budget to group it into the same categories. the November costs are from the original November 2007 project summary, with the per-laptop costs just dividing by the ran numbers of laptops in the plan (not by the projected numbers of students and teacher recipients, which was a larger number). The March numbers are from the final plan of operations, signed in March 2008, with the per-laptop costs being the total divided by the number of laptops (they fixed the number glitch), and dividing the server costs among the laptops. The 5year projection is not just multiplying that by five, as most of the costs are designed to be one-time implementation costs, but tweaking where necessary to cover maintenance and so forth. The Jon5yr cost is my original guestimate, and the Jon1yr extracts the setup costs and divides the rest of the recurring costs by 5:
| Cost Calulations | |||||||
|---|---|---|---|---|---|---|---|
| November | Nov per-laptop | March | Mar per-laptop | Mar*5yr | Jon1yr | Jon*5yr | |
Training and content |
$696,336 | $40.96 | $490,743 | $35.82 | $35.82 | $27.60 | $138.00 |
Implementation |
$735,664 | $43.27 | $1,355,827 | $98.97 | $98.97 | $108.00 | $108.00 |
Maintenance |
$100,000 | $5.88 | $71,200 | $5.20 | $25.99 | $7.40 | $37.00 |
Hardware |
$3,568,000 | $209.88 | $3,182,230 | $232.28 | $282.17 | $149.00 | $689.00 |
...Connectivity |
$180,000 | $10.59 | $170,880 | $12.47 | $62.36 | $1.00 | $541.00 |
...Laptops |
$3,196,000 | $209.88 | $2,581,480 | $188.43 | $188.43 | $148.00 | $148.00 |
...non-laptop hardware |
$192,000 | $11.29 | $429,870 | $31.38 | $31.38 | $0.00 | $0.00 |
Sum |
$5,100,000 | $300.00 | $5,100,000 | $372.26 | $442.94 | $292.00 | $972.00 |
You can view the entire spreadsheet to see what went where in detail at this Google Spreadsheet of the OLPC/IADB costs with the formulae visible to see how I constructed my numbers (and the linked PDF files have the original arrangement of the costs if you're so interested).
Now, IADB doesn't break down connectivity very well, so it's unclear what part of that is installation and what part (if any?) is service provision, and it's this figure where my estimates appear way off. IADB calculates $3,000 USD/school for "connectivity" . According to a February Community News email:
Michail had a conference call with SES-Americom. They agreed to provide the (C-Band) space segment and internet termination (via their Maryland teleport) for our upcoming deployment in Haiti.
So I'm going to guess that that's $3k for a VSAT installation with donated bandwidth; with no maintenance budget or money for a paid subscription if the bandwidth donation ever dries up -- does the community have any insight on this?
As for the other "line items", I was $10 over per laptop on total implementation costs, and $2 over on maintenance costs, not bad. If I presume that an Internet connection is free of cost, my original number 972 less five years of Internet access (coming in at $541 using the UN dial-up costs discussed here), I end up $11.94 cents higher than a 5-year projection of the IADB's numbers, (not accounting for the change in the laptop price from $148 to $188).
So if anyone can provide some on-the-ground information about bandwidth costs in Haiti (all the ISP websites I found were broken in one way or another); we can see what the ongoing costs of this project will be.
May 16, 2008
MS on XO: It's so bad you have to laugh
Check out this video of James Utzschneider from Microsoft's Unlimited Potential team walking us through Windows XP on the OLPC XO (Video via OLPCNews.com
XP on the XO
So the good news is it boots faster than Sugar; (1:05 into the video) Good going, folks. Too bad you have to cram in an SD card to make XP and Office work -- so that makes it really difficult if you ever want to upgrade to a larger SD card, view photos from your camera, or any of that, presuming it won't successfully boot without the SD card (but maybe they squeezed the XP Operating system into the onboard NAND flash drive and the SD card holds Office? That'd make more sense, so it's probably not true).Update Unsurprisingly, it's not true. From James' blog, emphasis added:
As I have posted earlier, we had to write multiple custom drivers and a BIOS to get Windows to boot from an SD card in order to do the Windows port to the XO. This is the initial implementation customers will be able purchase when the product RTMs and will be a "Windows only" XO that Nicholas Negroponte himself has described as running "really fast." Customers can also choose to buy the existing Linux/Sugar XO. Longer term, the OLPC plans to write a new BIOS and increase the amount of flash storage on the XO to support a "Dual Boot" option that would enable children to use either Linux or Windows on the same machine. This is fine with us as long there continues to be an excellent Windows experience on the XO.
Recording audio
It goes quickly downhill from there; at 1:36 in, James shows us how to record an audio file on the XPXO. Remember, in Sugar this means pressing the "Record" activity on the bottom toolbar, selecting "Audio" (it defaults to photos, this one "Record" activity records anything -- photos, video, or audio!), and pressing record -- done. In XP, James navigates through 3 submenus of the Start Menu (Start-Programs-Accessories-Entertainment, for you following at home with your own XP, because when I think "record this" I think programs, then accessories, then entertainment!). So after finding the Sound Recorder, he then has to muck with the custom audio properties (Stereo sound and normal compression??) before recording finally. Right. That's intuitive.Recording video
At 2:20 he loads up Windows Movie Maker to capture video (again, to do this in Sugar, you'd just change from Audio to Video in the Record activity). Again he mucks with compression/quality settings (1/2 MB bitrate and 30 FPS -- really? I just want to press record here). It works and has the standard Windows Movie Maker timeline/video editing capabilities -- providing you have any space to store in or a USB thumbdrive (adding even more to the cost of the XPXO). Besides, the video looks choppy on playback -- probably because too many Windows processes are slowing down the poor XO.
"Sharing"
Speaking of thumbdrives; evidentially he expects teachers using XPXOs to have thumbdrives (at 3:19) and be ready to pass them around their class to share videos/photos/recordings and such. Heck, I don't even let my thumbdrive leave my sight at work. With class sizes of over 30, how long will it take for each student to plug a drive in, have it pop up, copy a video to their desktop (again, providing they have any space left over after Windows and Office), and then finding the "Safely Remove" icon in the taskbar, clicking it, and correctly selecting the thummdrive and not the Windows SD card, and then passing it to the next student. Sharing a video becomes an all-class-session activity. What happened to using the mesh?
Power
Putting the laptop into the tablet configuration in Windows seems to switch it to the no-backlight screen mode (4:00); which I hope is not automatic if a child wants to, I dunno, read a book at night in a house without any other light source? In no-backlight mode, he claims you can use the laptop for 20 hours, which I find hard to believe, but if Windows isn't supporting the mesh network and therefore the wifi is also turned off, it's remotely possible. I watched full-screen video with wifi off on a flight recently and it lasted the full duration of the two and a half hour movie, plus some time left at the end to play the Implode activity (my secret XO addiction) before having to turn off all electronics for landing; so in full, CPU-sleeping screen-off mode, it probably could last that long; maybe us Sugar users should turn off wifi and see how long a backlightless Read activity can last?
Wifi
At 4:50 he shows us how to access a wireless network. Now, as a guy who often gets calls from parents, friends, parents of friends and friends of friends trying to connect to a wireless network in XP, I can safely say that configuring wifi on XP is one of the most confusing tasks ever to be standardized. No mention of support mesh networking, which may mean that the laptops are not connected to even a local network once they leave the access-point connectivity of the school (if there's even good connection at the school; my experience with Jamaican schools built with lots of rebar, cinder blocks, and metal roofing played havoc with omni-directional wifi ranges).
Security ?
Not mentioned in the video of course is the dire need for security software -- anti-virus, anti-spyware, anti-malware, anti-phishing and so on that's suddenly very important if you're releasing XP+IE machines to people who haven't developed a callous shell of cynicism and doubt when approached by Nigerian 419 scams, "Your computer is infected" flashing malware banner ads, and the like. By the time you load all of this up, the low-power computer will slow to a barely-usable crawl.
Conclusions
Sugar had its faults; no doubt about it; but it was clean and intuitive with a core belief of an "unlimited ceiling" of upward development -- Sugar was an adult bike with many layers of training wheels that could be removed; with lots of integrated paths to help do just that with eToys teaching programming methods and the various puzzles teaching slowly-more-challenging problem solving skills. Windows is designed against this, with no programming tools built in, and an almost anti-hacker/explorer/fiddler philosophy that goes beyond it merely being "closed source" to putting up impediments to learning any useful skills. Though Laptop.org currently seems down (perhaps under DDOS by annoyed former fans, or being redesigned with all the "Open" language removed; the Archive.org copy reminds us where the OLPC project was originally headed; and how far it's strayed. Nicholas Negroponte can keep saying that the project has remained "very pure" as much as he wants, and claim that "OLPC remains fully committed to our goal: a completely free and open learning platform for the world's children", I think it's safe to say that no one believes it:
XO is built from free and open-source software. Our commitment to software freedom gives children the opportunity to use their laptop computers on their own terms. While we do not expect every child to become a programmer, we do not want any ceiling imposed on those children who choose to modify their machines. We are using open-document formats for much the same reason: transparency is empowering. The children—and their teachers—will have the freedom to reshape, reinvent, and reapply their software, hardware, and content.
And I certainly don't see how he can with any straight face maintain that "The mission statement of OLPC has not changed in three years"
OLPC Now Microsoft-only
May 14, 2008
Pots, kettles and OLPC fanaticism
Like most in the OLPC community, Ivan Krstić's discussion on the OLPC yesterday left me (almost) speechless, and even Wayan at OLPCNews left it mostly as a repost of Ivan's essay, and slashdot is, well, talking about Australian government issues it seems.
The important parts - 1:1 laptop programs have no studies supporting that they work at all:
As far as I know, there is no real study anywhere that demonstrates constructionism works at scale. There is no documented moderate-scale constructionist learning pilot that has been convincingly successful; when Nicholas points to "decades of work by Seymour Papert, Alan Kay, and Jean Piaget", he's talking about theory. He likes to mention Dakar, but doesn't like to mention how that pilot ended — or that no real facts about the validity of the approach came out of it. And there sure as hell doesn't exist a peer-reviewed study (or any other kind, to my knowledge) showing free software does any better than proprietary software when it comes to aiding learning, or that children prefer the openness, or that they care about software freedom one bit.
Free/Open software is often buggier (by fault of hardware manufacturers) and therefore harder to use, and is not some magic solution to creating constructionist learning using technology
The real insight here is this:
There are three key problems in one-to-one computer programs: choosing a suitable device, getting it to children, and using it to create sustainable learning and teaching experiences. They're listed in order of exponentially increasing difficulty.
OLPC managed the first one, and well, and seems to have opened up interest in the low-cost, "4P" market in general. Deployment and implementation have been, from day one, ignored by OLPC, and I as well as Wayan at OLPCNews have been hitting on the desperate need for implementation plans over and over and over again. It seems like this grand-canyon sized oversight finally got to be too much for some of the core members:
Other than the incredible Carla Gomez-Monroy who worked on setting up the pilots, there was no one hired to work on deployment while I was at OLPC, with Uruguay's and Peru's combined 360,000 laptop rollout in progress. I was parachuted in as the sole OLPC person to deal with Uruguay, and sent to Peru at the last minute. And I'm really good at thinking on my feet, but what the shit do I know about deployment? Right around that time, Walter was demoted and theoretically made the "director of deployment," a position where he directed his expansive team of — himself. Then he left, and get this: now the company has half a million laptops in the wild, with no one even pretending to be officially in charge of deployment. "I quit," Walter told me on the phone after leaving, "because I can't continue to work on a lie."[...] That OLPC was never serious about solving deployment, and that it seems to no longer be interested in even trying, is criminal. Left uncorrected, it will turn the project into a historical fuckup unparalleled in scale.
May 11, 2008
Good will sharing
I read BoingBoing - it provides a steady stream of new and interesting things around the net, and the occasional IT policy tidbit. Lots of people read boingboing as well, giving it a power not unlike the slashdot effect - the ability to direct massive amounts of traffic at a site, taking it down in many cases with the onslaught.
Few sites respond with an attitude as user-friendly and pro-sharing as Gangstagrass which is promoting a rap + bluegrass mashup CD:
HELLO BOING BOING FOLKS - SORRY IF THIS LOADS A LITTLE SLOW, THE SERIES OF TUBES IS A BIT CLOGGED. CLICK HERE FOR A BIT TORRENT DOWNLOAD OF THE ALBUM!
(If you want the torrent, drop by Gangstagrass for the link).
This is totally backwards from what you'd expect. You mean the website doesn't want to sign me up for their email spam? Force me to log in? Track the number of downloads of their album? Instead, they're confident enough in their product that good customers will return to their site at some later date to check out what else they have going on -- but in the meantime, they're creating goodwill and advertising by making sure people who were intrigued enough by the BoingBoing article to click were able to get what they were after. So now I'll listen to it, and if it's good, I'll remember to visit their site again. If it's not, well, they've not lost anything; I wouldn't be a customer anyway in that situation.
This shouldn't be rocket science, but so rarely do I see people thinking in this pro-user/customer/person way that I thought it'd be nice to give GG a tip of the hat.
Also, this is all a very longwinded way to give thanks to at least one person who's always been very pro-person, at least to me, which is Mom. Hi Mom! Happy Mother's Day! :)
May 07, 2008
Scott McNealy video from CGD
Unfortunately; this clip seems to focus tightly on his IP statements; leaving out some of his more relevant popoints on "sharing" - centric resource creation and open standards.
May 05, 2008
Notes from Open Source, Open Education and Eco-friendly: Can Sharing Improve Policy?
Scott McNealy from Sun was hosted by the Center for Global Development to talk about how ideas of openness and sharing work in business and how they can help improve public policy and international development; with a panel discussion.
Apologies for typos; I was taking notes on my OLPC.
standardize on open standards for policy; procurement needs to find an osistandard; refernce implementation; community etc.
economics of eco-reponsibility; 40% of data center cost is power, power per transactionthread (sun very low)
thin clients + virtualization
Introductions: Lawrence McDonald Director of Communications and Policy; CGD
- key cgd ideas; rich world influence dev world beyond foreign assistance; investment, trade and technology (private sector interactions in dev world)
- cgd as pro-innovation thinktank ; tech but not bleeding edge
- firms come to cgd for beyond CSR; Corp engagement whitepaper
Ed Scott; CGD co-founder and board chair
carter admin and BEA founder
- sun based on open (BSD, open firmware board, "birth")
- move to competitive mkt against closed-software cos
Sun / Scott McNealy
- Open Drives Progress: ("sun is about open")
- internet is growing; chinaMobile went from 300M to 450M subscribers; more subs than US citizens; all using internet, mobile as internet access key
- => critical improtancec of standards; common way; same side of the road
- not helpful that there's competition on which side of street to drive on; role for govt for standards
- IP matters "true believer" in patents and copyright done properly to get return on R&D, Sun is 43 worldwide in R&D spending cross-sector
- => critical improtancec of standards; common way; same side of the road
- No one owns language: common language of the internet should be free ("I'd love to own english; I'd even take French") imagine if anyone owned "y" and charged a nickle?
- not the way the computer industry had been or should go
- Sun cause at start (kept low profile)
- Eliminate the digital divide w/o harming the planet
- 70% of globe without connection/google/etc ("sounds good for about 20 minutes but then oyustart to shake')
- people in the us don't understand poverty; sun not ging to solve maslow / water level needs, but ready tohelp once they're at the it level
- Sun hs a cause and mission beyond standard capitalistic reqs. --- mission is to outfit the data centers/NOCs
- strategy - sharing "we invented open source" ; bill joy BSD; NFS, tcp/ip opened from sun and it won; opened tech platform today; ultrasparc now curriculum ref implementation in China
- 19years cashflow positive with openness
- sharing will never "win" but will succeed
- mySQL gets downloads 70k/day (gives a hunting license to upsell); openoffice/stardownload 1M/week (?)
- Eliminate the digital divide w/o harming the planet
- 5 reasons to be open
- lower barrier to entry (free less oppcost)
- increase interop (biggest platform for sun solarisis HP)
- lower R&D (community help for javaphone roi when I is zero; competiting)
- more secure (no more secrets; many eyes)
- Lower exit barriers (zero; how hard to shift from ford to chevy vs english to mandarin, easier to switch)
- Proposal: OpenGov platform for an open source govt tech platform from census/mapping/email/polling/etc.; 26B r&d non-xclusive contribution
- China adopting; germany interested
- increase access for citizens, safety/security, cut costs, drive efficiencies
- Open Source Education -- text, online curricula, community developed, 37k mmbers, (curriki) 11k assets already, FREE, social networking; statistical evidence for testing improvements
- $4.3B/yr spent in US on textbooks - why constnt revisions yearly expenditures
- h1b visa cap -- let\s not let the smart people in; sun and java very immigrant focused
- Eco: innovate; act; share -- CEO (Eco) Douglas; apply transparency and open source to enviro challenges
- openeco.org - tracking tools to compare your org to others; community learnig and sharing; all orgs welcome
- "I am a stunning raging capitalist" proprietary code || central planning; open source is about competition of ideas and democracy
Panel Discussion
- ellen miller sunlight foundation crp
- open govt data; importance of open data? is there a connection?
- philospohically inconsiderate and counterintuitive
- bring philosophy of oss to opengov advocates
- open govt data; importance of open data? is there a connection?
- dave witzel forumOne cofounder; ex-Banker on sabbatical at CGD
- policycommons - moving valuesof oss to education; green; also apply to governance
Lawrence McDonald connection of sun mission and dev goals for 2B under $1/day/ Scott McN: importance of competition on pricing model (MC of software is zero, monopoly rent of ms?) bulk of new net users are getting online using java phones, on 2B phones
LMcD policy research - tech policy unexplored field; only some from IP on dec index; what would a pro-dev US tech policy look like?
- Scott; pro-IP; returns to ragig capitalist Ip matters to get R&D investment; but sharing can expode opportunities with right licensing models; ip and open source not mutually excusive
- phizer; contribute drug dev as open source (ran away)
- ...you can make a profit off of open source licensing
- phizer; contribute drug dev as open source (ran away)
Audience Questions
- bill sabidoff econ consultant at cgd
- collective decisions on standards; how collectively decide what's open and what\s proprietary?
- focus on CostC barrier to exit costs to drive decisions on what's proprietary
- collective decisions on standards; how collectively decide what's open and what\s proprietary?
- goerge ingrham AED
- practical lvl of first 3 stepsforgovt openness/
- attitude adjustment (govt data belongs to people); FOIA is backwards; burden of providing info should be on govt
- rest by executive order to go open
- Scott: mandate a balanced budget to motivate openness; need bully pulpit to focus on openness; procurement; restrictions of govt investment
- practical lvl of first 3 stepsforgovt openness/
- John richards center for stud... resopnsible...
- headlinenumber for savings to go open; ~25% on it budget; prob is barrier to exit from proprietary systems => stuck
- "gazillions"
- why hasn't the market gone already? b/c tipping point/network system/cost of switching + monopoly power; govt has buying and regulatory power
- claudia - policy health markle foundtion; health info sharing
- if goal is privacy; need tranparent tech choices
- can't separate tech choice from policy choice
- Scott: security: who's who + what's what + access
- wiztel: philosphoical agree on openess; how to you get there?
- Ellen: newer institutions more open to open; struggle for older ngos and extrapolating harder for govt; non-profit will force govt to meet public expectation of online openness
- sometimes you're the windshield. sometimes the bug -- world book; card catelog. classified ads -- or wikipedia;oogle. ebaycraigslist -- if you want to compete; ,go clear + transparntand fast
cost trends towards marginal costs of zero; free is the new number
govt and ngos should ride the free wave instead of not curriki in india; s.af.; focus on innovation to avoid high costs
random notes
- tweem hashtag #cgdmcnealy : http://m.twemes.com/cgdmcnealy
- yunus blogging youth project yunus social action group .org
- curriki.org
May 02, 2008
It's full of stars....
Somehow, this is exactly like what I expect it should look like:
Courtesy of Nexus's Facebook plugin
All I can say is wow.
