With all the fury of netbooks, XO-2 mock-ups, and dual-boot XO's, you might forget what we expect from the next (real) OELPC: One Educational Laptop Per Child
Obviously, the OELPC device will be a netbook with good ratings for the 4P's: Power, Performance, Price, and Portability. How do the current netbook offerings stack up to what we need?
- Ultra low power
A day's worth of battery charge seems to requite ARM/MIPS chips. With a power consumption of around 5 Watt this should be possible. Such low-power laptops are available today
Everyone has complained about performance since the advent of computers. Simply put, today's slow netbooks were yesteryear's workstations. It beats having no computer and connectivity at all. My idea is that performance can be sacrificed against any other of the P's. However, this does not held for the wireless connections. With bad wireless, the value of the laptop drops fast.
- Toy level price
I think Negroponte was right when he quoted the $100 label. This is in the range of toys and other gadgets and whoever reaches that price will sell (tens of) millions of items. Personally, I think that a price tag of $75 would saturate the human population with netbooks
- Breast pocket weight
Ideal weight would be around 500 grams or lower. 650 grams is already possible.
I think the 4P Computers are a reality now. The current crop of cheap, solid state netbooks come close. And gadgets like the Trendtac 700 EPC are really spot on. Obviously, we want more. Mary-Lou seems to have done it again with her XO-2. Here is my wish list:
- Mesh, and a better scaling one
- I want Touchscreen io Mouse. Touch-pad and mouse are logistic nightmares
- Could we really replace real keyboards with a touchscreen keyboard? That would be great.
- And there is never enough screen space
To stay in the mood, I propose 4C software functionality:
For children, communication is everything. Email, IM, VoIP, video conferencing, message boards, they are all swamped with youths. The fact that you are sitting in your room doing your homework at home doesn't mean you do not need to "talk" to your classmates and friends. In the wider scope of things, education is all about communication.
With 60 children to a classroom (or no classroom) and half-day classes, peer tutoring and project work is paramount to getting an education. In real life, computers often hamper collaboration efforts more than they support them. These children need functionality that makes collaboration transparent and intuitive.
In school, children make things: write, draw, glue, paint. The more different things they can make, the better the education seems to be. So an OELPC should allow more of that and provide the tools for creative endeavors. Simulations (eToys), photographs, movies, audio recordings, music, stories, drawings, and mashing these all up.
Much more importantly, on a higher level the OELPC should allow the design and creation of all teaching materials.
It is often said that computers in education are only useful with enough educational content and teaching materials. So whoever delivers the hard & software should also do the localization and write all the teaching materials. However, publishers do not create, authors do.
It is simply impossible for any entity in the world to create all the teaching materials for all the schools in the developing world. The only thing that can be done is to create the tools and try to organize the construction of teaching materials.
At some places, the idea of teacher-produced learning materials is already going to be tested. It is well known that students already share homework assignments on-line on a large scale. I think that schools should profit from that entrepreneurial spirit instead of trying to criminalize it.
Sugar comes pretty close in all these areas. Traditional software distributions are far from this ideal. Most "tools" are impervious and opaque to collaboration and are completely unintuitive to children and adults alike. Setting up collaborative projects seems to require a degree in Computer Science. VoIP and video-conferencing are almost there, but not yet completely.
So my wish list is:
- Content for teaching should be produced as close to (or in) the schools as possible
- Real Plug-and-Play VoIP and videoconferencing
- Better educational tools. Maybe including SWF (Flash), a graphical IDE for Python, locally run Ajax, Silverlight/Moonlight. Whatever is most productive.
- Better mesh
The real foundation: 4S
For the above to be suitable for children, we need a good foundation in 4S:
A computer for a child should be safe to own and use, PERIOD. There should be no compromises here. There should be no hazards to the child nor should hazardous components be exposed during use, service, or after disposal. The child's identity, privacy, data, and laptop should be protected against inspection, disclosure, theft, and compromise in general.
The OELPC should remain malware-free and protected against intrusions without user intervention. Children simply do not remember passwords nor can they be bothered with security software. System integrity should not only be protected, but also be obvious to the child. There should be no silent rooting or hidden malware
Children will be children and the laptops will be used in hostile environments: Damp, dusty, and hot. They will be dropped and abused. The hardware should be robust and keyboards and pointers should be accordingly constructed (or removed). The software and electronics should protect batteries, memory, and system configuration.
Even robust systems break in the hands of a child. In the developing world, sending it back to the shop is not an option. The laptops should be repairable in the field with a minimum of tools. And we should understand "field" to include open spaces with ubiquitous plant growth, dust, sand, and sun with occasional damp weather.
Children lose things. And they do it often. They also tend to mess up their work with a vengeance. The software solution to this is versioning (or journaling) and backup. Consciously, children will use this even less than adults (if that is possible). So the OELPC should do it for them. Modern file systems (eg, ZFS, btrfs, Hammer) already do this, but these are not (yet) suitable for a netbook. Versioning control and snapshots can already be done automatically based on systems like Git.
What netbooks (or even computers) are up to the 4S's? I only know the XO with Sugar/Bitfrost. There really is no alternative. The only alternative on the horizon is the XO-2 with Sugar/Bitfrost. My Wish list:
- Choice. Why is there no other netbook available that tries to make even a little step in this direction?
Wayan's 4P computing terminology has taken the market, for adults. However, a lot of the so called netbooks actually contain a hard disk drive which defeats the purpose as that worsens all four P's at the same time.
The four C's are basically ignored in current netbooks. I think that is because adults could not be bothered to change their habits of isolated desk & hub slaves. For people with a social live, like children, communication and collaboration (think social sites, YouTube) are way more important than a spreadsheet.
And all these social activities require tools to pimp them (cf, Facebook without photographs, YouTube without movies). For those who have no other computer than the OELPC, it should be all in there. The OELPC is not a second or third computer, it is the only computer these children have access to.
When you give a toy to a child you make sure it is safe and robust. If you give a computer to a child you should watch the 4S's. Current practise in the developed world is that parents (and peers) keep the child safe and her computer running. If things break down, eg, malware takes over and trashes the computer or the OS trashes itself, there are resources to "repair" or replace the computer.
If the child messes up on the Internet, there is law-enforcement to protect her. Such resources are not available to children in the developing world. Parents are not only computer illiterate, they often are simply illiterate. And if something bad happens, the child is on her own.
Looking at the "market" of educational netbooks for children in the developing world I see only a complete desert wasteland with the XO as an isolated oasis.
The XO is ageing. Even though it still beats all the other offerings in 4P, 4C, and 4S, it is too expensive to produce, it still uses too much power, and its hardware needs an upgrade. Given the wasteland of unusable alternative netbooks, how to proceed?
I think one way forward is PixelQi producing the hardware. A $75 dual touchscreen netbook like the mock-up demonstrated as the XO-2 would take the market overnight. At $100 or $150, it would still be a raving success. If Sugar labs could load an updated Sugar with Bitfrost on this XO-2, the XO-2 would take care of most requirements.
There is still a question of the hardware "extras" of mesh networking, robust casing, field serviceability, and cryptographic security in hardware etc. These are requirements that the "market" in the developed world is normally not willing to include. This could be handled by a central entity that organizes the distribution of the laptops, say, the new-and-improved OLPC ;-)