Back in May when OLPC announced the XO-2 the organization also indicated that an XO-1.5 would be released in spring 2009. To date the only official information regarding the XO 1.5 is that it's supposed to have fewer physical parts and cost less than the XO-1.
Now while re-reading the OLPC community-news update from the beginning of August I stumbled across a paragraph which I had somehow missed before:
Mirror mirror on the wall...
Technology
Robert, Chuck, Kim, Michail, Richard Smith and Darah met to discuss and understand upcoming hardware changes and their business implications. OLPC is planning to migrate from Marvell's 88W8388 wireless chip to the 88W8682. The newer chip offers a 50% improvement in power consumption and has more onboard memory which will allow us to implement better mesh algorithms. Additionally, we are preparing a transition to a new supplier and a single-mode touchpad. Both changes are expected in mid-Q1 2009.
Given the timeframe it seems reasonable to assume that these changes will be integrated into a XO-1.5. While it seems impossible to find more information about that Marvell 88W8682 wireless chip a 50% reduction in power consumption certainly sounds like quite an improvement.
With regard to the touchpad I'm sorry to see the dual mode functionality (regular trackpad and graphics tablet for drawing and writing) go but admittedly most people won't miss it since it never really seemed to work anyway. One thing that I was wondering about is whether integrating a multi-touch touchpad such as on the MacBook Air or eee PC 900 would make sense on the XO?
Talking about touch (and thereby moving into the land of speculations) I'm certainly not the first person to think that an XO with a touchscreen would be pretty damn cool. Interestingly many people who first get to use the XO assume it to have a touchscreen anyway. I guess that's due to the icon-driven and apparently finger-friendly Sugar UI (no small menus, no tiny scrollbars, etc.). Also a touchscreen would make the e-book mode really usable and turn the XO-1.5 into a full blown tablet-PC similar to the Classmate 3 design.
Keyboards are so 20th century!
However I haven't seen any indications of this actually happening. And even though Sugar could quite easily be adapted to be a cool touch-driven interface it would probably take quite some development time to offer a great and coherent user-experience. So while I would be super-excited about an XO-1.5 with a touch-screen I guess we'll have to wait for the XO-2 for that too happen. Unless, Pixel Qi's (former OLPC CTO Mary Lou Jepsen's company) "new screens that will be ready next year" comes with touch-functionality that is.
Another change that seems very likely for the XO-1.5 is the move from the current DDR-1 to DDR-2 RAM. OLPC's Jim Gettys recently mentioned:
AMD is building boards to test auxiliary circuitry allowing the Geode LX to support higher performance DDR2 SDRAM chips. We eagerly anticipate the results of their tests.
This comes as no surprise since DDR-1 memory is an outgoing technology and today DDR-1 memory modules are already more expensive than the newer DDR-2 counterparts. I'm not sure whether such a move would also mean that the XO-1.5 gets more RAM than the current design but it certainly seems like a possibility. This talk about auxiliary circuitry also leads me to believe that the XO-1.5 will stay true to AMD's Geode LX platform and not switch to an ARM based processor which some people expect to happen. Again, I think the XO-2 is a much more likely candiate for such a move.
The last item is the Flash-based mass-storage and there I expect a move up from the currently available 1GB. Microsoft indicated that the upcoming dual-boot XO would have a 4GB NAND storage so again it doesn't seem unreasonable to expect similar changes on the XO-1.5.
Anyway, enough day-dreaming for now, it will certainly be interesting to see what happens over the next 6 to 9 months.
This is beginning to be a problem if OLPC wants to stay as a non-profit education / development project; it's hard to make the case that a country should buy the XO-1 if the XO-1.5 is coming out soon, and the XO-2 by 2010, with vastly improved features. Those are short time frames in commercial laptops, but two years is closer to one pilot cycle project in development. Naturally, there's going to be a need to upgrade the XO hardware over time and redesign it to take advantage of the current cheapest and best components, but this seems premature?
Unfortunately it looks like http://wiki.laptop.org/go/Marvell_microkernel has been a complete failure, and not for lack of people willing to do the dev work.
Disappointing.
So far at least OFW became Free as a result of the OLPC project.
" XO-2 Laptop, launching in Q1 2010, is an update to the original XO laptop that's designed for low-cost computing for kids in developing nations, will carry two 16-inch by 9-inch touchscreens and eschew a keyboard. It opens like a book and can either be held vertically for reading, or horizontal for laptop computing. When used horizontally, the bottom touchscreen displays a keyboard for typing."
Sounds awesome and the current product looks really bad in comparison. No reason to buy it, really.
I'll wait.
Babara
My worry with the advent of the XO-1.5 is that those of us with XO-1s are going to wake up someday soon and discover that the next version of the software is not compatible with our machines. What plans are in the works to support EITHER of the network chips in future releases? Or that the OS will continue to stay within a 1 GB flash footprint?
Microsoft (with far more resources) never promises that the next OS will run on the current hardware. I predict that OLPC will quickly move to the 1.5 platform and will hardly, if at all, test the code for compatibility with the 1.0 hardware. Leaving all of us early adopters with a bunch of doorstops. While all the great features that were promised last November at G1G1v1 time to be implemented only on the newest platform.
@jon
i hear what you are saying, and i agree with you. the thing is though..the XO is an underpowered machine. they havent really sold too many as it is...so perhaps this could help more than hurt them.
i personally thought that they would have waited until they got the machines >100 and then upgrade it. the original goal was obviously 100 bucks, so once they got there i figured it would be fair game to keep it up there for a while while they gave new models a hardware imporvement over the previous versions. i guess though that they are so far away from that that they decided to jump the gun a bit.
(shrugs) i hope it works out for them. my only problem/issue is that i wish they would let me send in my current XO for a substancialy reduced cost XO 1.5. I think alot of people like myself would like a program like that, but thats one of those things that would in all likelyhood require the XO group to become a for profit business
also i'm sad to hear that the new monitors from pixel qi wont be ready in time. man oh man though...if the touch screen feature was added then it really would have be a XO 1.5 with new touchscreen capabilities, using even less electricity, more ram, and more ssd space. obviously the other things are alot more reasonable/plausible, but i think the touch screen would be the real icing on the cake.
i guess we'll see what happens. XP isn't on this, and the there has been little official word on the xo 1.5, so things are still up in the air. will it get a touch screen? probably not. but who (other than NN) really knows for certain what the xo 1.5 will really be like. until the curtains are lifted, i think its best that we all keep our fingers crossed and hope for the best.
Jon said "This is beginning to be a problem if OLPC wants to stay as a non-profit education / development project; it's hard to make the case that a country should buy the XO-1 if the XO-1.5 is coming out soon, and the XO-2 by 2010, with vastly improved features."
I think that very much depends on how OLPC plays its cards. It wouldn't be hard to imagine the XO-1 and XO-1.5 be available in parallel with countries obviously having to pay more for an XO-1.5 due to its improved feature-set. (The same thing could be true for 2010~2011 with the XO-1 being faded out and the choice being between the XO-1.5 and XO-2).
I've always believed that if a country asked for let's say 10K machines with 2GB of Flash instead of the default 1GB (or let's see have a blue instead of a green enclosure) then Quanta would be happy to oblige. Taiwan's IT industry is very flexible in that regard. So I think in the long-run the "one size fits all" approach will be going away.
David Wallace said "My worry with the advent of the XO-1.5 is that those of us with XO-1s are going to wake up someday soon and discover that the next version of the software is not compatible with our machines."
Unless the XO-1.5 were based on a different processor-architecture (read: non-x86) I don't think that this is going to happen. Especially since there are ~400.000 children using the current XO-1 and countries such as Peru and Uruguay would be *seriously* unhappy if OLPC were to drop support for the current hardware. Things like supporting 2 different network chips is also trivial. So all in all I really don't expect this to be an issue.
I highly doubt anything as drastic as a DDR1 to DDR2 change will even happen in the XO 1.5. What we're looking at is similar to what game console manufacturers do as time goes on. Minor changes and removals to deal with supply issues and problematic parts and to make it cheaper to manufacture. In this case, that means the touchpad and maybe the wireless chipset. Presumably, the rest of the changes would be completely unnoticable, not only to the user, but to developers too (except OFW and EC devs perhaps).