But his decision to switch from Linux to Windows was and is disasterous for OLPC. The machine just doesn’t have the specs to run Windows well, even with the costly (for a $225 machine) extra RAM and storage put in to make Windows work at all. Many people quit at this point. Negroponte had recruited volunteers by introducing his vision of a fully-open educational machine running open software. So they had less than 0 interest in working on an underpowered Windows machine. The individual buyers also did not want an underpowered windows machine. I think he could have had a successful middle ground by offering a distro where you could run Sugar, *or* run openoffice+firefox for more conventional usage This would not have alienated the open source backers he recruited, or purchasers.. but I think would have appeased the bulk purchasers who “wanted XP” (keep in mind they are not getting XP OLPC boxes to run other XP apps on… the machine has Office only, no room for more software to be installed, and Windows XP Starter addition doesn’t support file shares so they couldn’t load an app off a file server either.)

The OLPC could have been the Kindle. It could have changed the world if only the people who managed it had understood more about the world.

(1) OLPC is near bankruptcy, and is just offloading anything that could plausibly be done by non-staff members for free. This is what I would like to think.

(2) NN decided a long time ago that XO 2 would run Microsoft Windows as the default / recommended by OLPC option, and he has been moving the composition of the organization to not oppose that announcement when it comes. There are some facts that would support this.

a. With the change is management away from Walter Bender a while ago and the recent layoffs, the number of staff members that care enough about user freedom to quit if an announcement such as this were made has moved from the vast majority of employees, including top on-site management, to a much smaller number of people (approaching zero). The successful outsourcing of XO 2 hardware design would allow OLPC to lay off some of the last remaining people who really care about the freedom of users.

b. Looking at http://trac.laptop.org/query (blocker bugs) is scary. The majority of “owners” are people who were among the laid off or otherwise no longer getting money from OLPC – from memory, at least jg, cscott, erikg, dilinger, walter, djbclark, mstone, mbletsas, mchua, and mako.

c. Microsoft has previously surprised the OLPC staff, who initially thought Windows would not run well at all on the XO 1, by going from vague rumors of a port to having a port that worked shockingly well. They did this by throwing lots of engineering time at the problem (I believe more than the entire paid OLPC development staff as it was before the layoffs). There is no way to know that Microsoft doesn’t currently have a large number of people working on a coherent educational suite for the XO 2 to replace Sugar, or to somehow embrace and extend parts of Sugar when running under Windows. And since changes have already been made to the boot firmware to allow it to boot Windows in the XO 1, there isn’t a good reason to believe there will be any hardware-level reasons Windows won’t also work with the XO 2.

d. John Gilmore’s post in reply to “Free Software Foundation Files Suit Against Cisco For GPL Violations” on the olpc devel mailing list –
http://lists.laptop.org/pipermail/devel/2008-December/021841.html – and the “OLPC needs to comply with the GPL” bug – http://dev.laptop.org/ticket/4265 – point to the difficulty of continuing with the OLPC’s Digital Restrictions Management (DRM) “anti-theft” measures if running under GNU/Linux. Running Windows would solve this “problem” (said with sarcasm).

If you have been developing for the OLPC XO 1, these would be my (personal) suggestions:

(a) If you have been working on Sugar / Activities, continue doing so – that project has been split off from OLPC at http://sugarlabs.org – I’d suggest licensing any new activities you create under GPLv3, to combat current and any future XO tivoization. Sugar is generally useful, and can be used on non-XO devices, such as the education-focused gdium liberty 1000 – http://www.gdium.com – or any other GNU/Linux machines.

(b) If you like hacking on operating system level stuff, take a look at http://wiki.gnewsense.org/Projects/GNewSenseToMIPS – a project to port gNewSense, one of the most freedom-respecting GNU/Linux distributions (based on Debian and/or Ubuntu), to the mipsel architecture, so it can run on the first generally available laptop that will be supportable by free software with no binary blobs (including the wifi, unlike the XO), the lemote yeeloong – http://www.lemote.com/english/yeeloong.html – this work would also help get Sugar running on a completely free, mobile platform (the gdium uses the same processor as the yeeloong, and people are porting sugar to work on the gdium).

Since they are already working together it seems like a logical next step.

The kindle is very low power and has a sunlit readable epaper screen. It just needs to be GSM for the devo world.

It’s a bad idea to try and build a wifi network in the devo world when the GSM network is already in place and fairly ubiquitous.

Also, Amazon would then have the killer student reading tool to sell to the developed world.

He is a loser of the first order. Long long ago I actually thought he was cool.It is sad to see the shell the man really is.

Dozens of support volunteers have dropped out because the support crew is expected to handle basically all of the public relations, even though they don’t know anything and aren’t authorized to speak for OLPC. Hundreds of e-mails from people willing to help go unanswered and are eventually deleted because OLPC fundamentally doesn’t give a damn about anyone who isn’t a direct employee of the company.

The last handful of companies that Chuck Kane ran were all sold off or put under soon after he took the reins. OLPC has an undisclosed number of NDAs with Microsoft, Marvell, Quanta, and other companies. They ignore offers of help from major open-source figures, and then complain about how long it takes to develop software. QA is backlogged by about six months, and doesn’t have a testing regimen in place anyway, and probably never will because it’s run by the same pack of in-the-dark volunteers who valiantly try to provide end-user support. Kane explicitly dropped the education mission of OLPC, and claimed it was now basically about distributing laptops and to hell with what is done with them. Regardless, OLPC (in keeping with Negroponte’s amazing hubris) will only deal with national governments, and the few programs that are getting the machines into the hands of whatever schools want them (the Give Many program) never took off because (again) nobody at OLPC gives a crap.

Basically, Negroponte started out with a great idea, then made every possible wrong decision and turned it into a steaming pile of failure, and it’s pretty damn depressing.

But I’m sorry, I’m not going to get one, give one XO unless I know exactly who is going to receive it, simple.

People care for people they know much more than others, and there is a lot of people that could spend 600euros in their three sons but not 1200.

And not all the world is America, more than 90% of the world population find it very expensive to buy dollars, middle class could live well in their countries but to buy something in dollars they have to work much much more than an american).

Unless they manage to get the community back onside, the OLPC is an interesting piece of history. I hope the sunlight readable displays make their way onto other computers but apart from that, I don’t really care what happens to them.