tomhannen
Posts: 17
Joined: Sun Jul 31, 2011 9:27 am

Re: How will this differ from the OLPC project?

Sun Jul 31, 2011 3:55 pm

Raspberry Pi sounds like a great project! I think a blog post outlining the differences between what you are planning, and the OLPC approach would make for really interesting reading. OLPC tried to build a completely new GUI and "interface paradigm" for their laptops, whereas what you are proposing seems less prescriptive. They tried to build a complete low-cost netbook before the netbook really existed, whereas this is a single board solution, etc. They were initially against any commercial/proprietary software being shipped with their platform - until Windows was made available on the later machines.

I'm not criticising either approach here, but I'd be fascinated to read more about the comparisons and differences between the two projects, both technically and from a business perspective.

Good luck with it!

Michael
Posts: 340
Joined: Sat Jul 30, 2011 6:05 pm

Re: How will this differ from the OLPC project?

Sun Jul 31, 2011 5:18 pm

I think the two projects have quite distinct aims but similar approaches.

My understanding of OLPC is to equip every child in developing countries with a basic laptop so there is a smaller delta between the computer facilities available to them and the average child in a first world country.

Those that grew up in the first world in the 1980s and more recently will hopefully appreciate the different Educational world we grew up in that that of our parents (being able to word process or desktop-publish one's report instead of writing it by hand or typing it out; using the Internet to conduct research rather than having to spend hours in the local or County library).

In contrast, my understanding of the Raspberry Pi Foundation is principally to teach computer programming to children in the developed and developing world; and to promote the study of computer science and software engineering. A key step towards that goal is to make sure that the tools needed to achieve this are inexpensive and easily available.

I agree, a blog post - or indeed an article by a journalist - discussing the two projects in more detail would make an interesting read.

tomhannen
Posts: 17
Joined: Sun Jul 31, 2011 9:27 am

Re: How will this differ from the OLPC project?

Sun Jul 31, 2011 5:42 pm

Another thing would be to learn from OLPCs mistakes: was it a mis-step to describe it as the "$100 laptop" - is Raspberry Pi risking bad press if the final price for the device comes in at over £15? Who are the target market - developing countries or school kids in developed countries? Is the idea to maintain the same spec, but reduce the price over time? Or to improve the hardware over time, but keep to a fixed price? etc etc..

User avatar
abishur
Posts: 4477
Joined: Thu Jul 28, 2011 4:10 am
Location: USA
Contact: Website

Re: How will this differ from the OLPC project?

Sun Jul 31, 2011 6:07 pm

From what I've been hearing, the goal is to provide a low cost PC(ish) devices that school could give out to the students. This is half the reason I think it is so important to sacrifice features in order to keep the price down. By keeping the price down it makes it reasonable for the device to go home with the students and just be a part of school supply costs. Indeed, it makes damage or loss to the device largely inconsequential from a budgetary perspective, yet 25 dollars is also high enough to encourage students to act responsibly with them.
Dear forum: Play nice ;-)

User avatar
liz
Raspberry Pi Foundation Employee & Forum Moderator
Raspberry Pi Foundation Employee & Forum Moderator
Posts: 5212
Joined: Thu Jul 28, 2011 7:22 pm
Contact: Website

Re: How will this differ from the OLPC project?

Sun Jul 31, 2011 8:21 pm

We're planning a few blog posts on the philosophy behind the project.

We differ from OLPC in the lack of an enforced pedagogical approach, which OLPC tends towards. Our hardware's less shiny and pretty, but much cheaper, and our observations both in the UK and US and in visits to developing countries have been that old TVs are readily available everywhere, so a specific display device isn't necessary, and isn't something the user has to pay for.

Cost, of course, is a huge issue, and I'm very confident that we'll hit our price point - at the moment that's at the cost of things like wireless chips, which we hope to fit into later iterations of the device. Replacing devices which have borked in some fashion will also, obviously, be much cheaper than the $200 OLPC. Low cost means we can distribute more units. The fact that the device is physically smaller than OLPC will also, we hope, be helpful in penetrating developing-world markets: it means that we can cut down massively on shipping costs, and fit more into a single shipment (we can outfit a whole school district in one box, we hope).

I like the fact that in the UK and US, *parents* are going to be less antsy about a kid experimenting with and potentially breaking the device (although breaking it will be a very hard thing to do - we are engineering for robustness). Replacing a £15 machine is a very different kettle of fish from replacing a family PC - or an OLPC.
Director of Communications, Raspberry Pi

lkedves
Posts: 4
Joined: Mon Aug 01, 2011 5:52 am

Re: How will this differ from the OLPC project?

Mon Aug 01, 2011 6:42 am

I totally agree.
This device is something I was dreaming about, and I am sure this approach has a very important message.
- The generic, USB-based modularity means that this thing is compatible with and can reuse the existing "old" hardware (instead of replacing them in the name of "development" with no real reason - like an OLPC);
- Do not be fooled by the numbers. Ten years ago we at Cygron Ltd. had written and run Comdex winner data visualizing and mining application on computers that were ridiculously weak in ALL parameters compared to this gadget. OK, data amount has grown since then - but this is not a data miner, but a "personal network endpoint".
- This price tag also means that production needs high but now available technology with a very-very little resource requirement. This is the next generation of information technology, because this IS truly scalable to the whole planet.

Sorry for being this excited - but I have waited for this thing for sooo long...

caminati
Posts: 23
Joined: Fri Jul 29, 2011 5:23 pm
Contact: Website

Re: How will this differ from the OLPC project?

Mon Aug 01, 2011 10:10 am

Quote from liz on July 31, 2011, 21:21
I like the fact that in the UK and US, *parents* are going to be less antsy about a kid experimenting with and potentially breaking the device (although breaking it will be a very hard thing to do - we are engineering for robustness). Replacing a £15 machine is a very different kettle of fish from replacing a family PC - or an OLPC.

I am delighted to hear about your attention to robustness.
I am realizing that I am in love with this project because I see it as the closest implementation of a bare modified harvard architecture, with the smallest possible amount of additional hardware.
For the user it is getting more and more difficult to tweak the hardware of its devices, because they're getting so integrated.
Your response is to enclose the computing core, on which the user has no mean to put his hands, to the smallest atom, and leave the rest as modules he can choose and possibly tweak to build its system.
In this view, robustness of the central computing core becomes vital, being it the only fixed and unfixable part of the system.

Moreover, the smaller the thing the more likely it is that it will be carried around and be exposed to hazards and a broad variety of possible stresses.

So, I update my Raspberry mantra to: modularity, price, size, robustness.

User avatar
Lob0426
Posts: 2198
Joined: Fri Aug 05, 2011 4:30 pm
Location: Susanville CA.
Contact: Website

Re: How will this differ from the OLPC project?

Sun Aug 07, 2011 7:04 pm

Ikedves has the right of it! Power is in the eyes of the beholder. Marketing in the USA always goes towards selling you a scale of performance that is usually way beyond your actual need. They will talk you into buying an i7-960 with 8gigs of ram a 1 terabyte HDD and 1gig video card so you can have the best performance on Twitter. Believe me I can make twitter work on a whole lot less than that, anyone for trying an old 386/20 MHz with puppy Linux? Me and my brother built a bulletin board (anybody remember those) with a 386/20 and a 286/20 through a lantastastic serial network with two modems and two CD roms for the bulk of storage running fireBBS if I remember correctly. The power needed is what is necessary to the intended task. Of course I do have an i7- 920 myself. Lol!

Correction: the software was spitfire BBS.
512MB version 2.0 as WordPress Server
Motorola Lapdock with Pi2B
Modded Rev 1.0 with pin headers at USB

http://rich1.dyndns.tv/
(RS)Allied ships old stock to reward its Customers for long wait!

eben
Raspberry Pi Engineer & Forum Moderator
Raspberry Pi Engineer & Forum Moderator
Posts: 85
Joined: Sun Jul 17, 2011 11:54 am

Re: How will this differ from the OLPC project?

Mon Aug 08, 2011 10:05 pm

Quote from Lob0426 on August 7, 2011, 20:04
Ikedves has the right of it! Power is in the eyes of the beholder. Marketing in the USA always goes towards selling you a scale of performance that is usually way beyond your actual need. They will talk you into buying an i7-960 with 8gigs of ram a 1 terabyte HDD and 1gig video card so you can have the best performance on Twitter. Believe me I can make twitter work on a whole lot less than that, anyone for trying an old 386/20 MHz with puppy Linux? Me and my brother built a bulletin board (anybody remember those) with a 386/20 and a 286/20 through a lantastastic serial network with two modems and two CD roms for the bulk of storage running fireBBS if I remember correctly. The power needed is what is necessary to the intended task. Of course I do have an i7- 920 myself. Lol!

This is what MBA courses refer to as "value innovation". As you say, incumbents tend to compete with each other by piling new features into their products, to the point where the capabilities of a device vastly exceed the needs of some sizeable fraction of the consumer base. At this point a value innovator can win by offering just what those consumers want at a lower cost. Southwest Airlines is the most famous example of this.

It seems pretty inevitable to me that someone is going to value innovate in computing; the big question is whether the resulting platform is closed (like iOS or Android) or open (like Raspberry Pi). Fingers crossed the good guys win :)

User avatar
Lob0426
Posts: 2198
Joined: Fri Aug 05, 2011 4:30 pm
Location: Susanville CA.
Contact: Website

Re: How will this differ from the OLPC project?

Mon Aug 08, 2011 11:18 pm

If the "innovator" uses open source, then how are all those lawyers going to pay their bills! It seems like the new profit "margin" is sue everyone else out of business.

A question was asked in another thread about whether you were worried about a company stepping in and under pricing your project. I don't believe that will happen here. Investors put way too much pressure on returns for any company to take up your challenge, a $25 computer! Another factor is their push too a wider market and larger profit margins.
512MB version 2.0 as WordPress Server
Motorola Lapdock with Pi2B
Modded Rev 1.0 with pin headers at USB

http://rich1.dyndns.tv/
(RS)Allied ships old stock to reward its Customers for long wait!

User avatar
abishur
Posts: 4477
Joined: Thu Jul 28, 2011 4:10 am
Location: USA
Contact: Website

Re: How will this differ from the OLPC project?

Tue Aug 09, 2011 12:09 am

Another question is, does it matter if someone undercuts the R-pi? Don't get me wrong I'm extremely excited about the r-pi! But their goal is the promotion of computer literacy, if another group somehow manages to put together another device that's similar and at the same time somehow cheaper... well I wouldn't suppose the R-pi team would be thrilled about not having their device used, but we'd all still be happy that the greater issue of computer illiteracy was being addressed.

That said, I don't really see anyone undercutting the R-pi team. There's only two avenues of people out there who would "compete" with r-pi. The first are businesses, such as Alix or Begalboards. Both of which are fairly similar yet cost a minimum of 4 times the cost of the r-pi for their lowend stuff (which, hey, they're a business so that's necessary). The other are other charities and the only way I could see another charity undercutting the r-pi price would be if they used slower/older components :)
Dear forum: Play nice ;-)

jamesh
Raspberry Pi Engineer & Forum Moderator
Raspberry Pi Engineer & Forum Moderator
Posts: 28358
Joined: Sat Jul 30, 2011 7:41 pm

Re: How will this differ from the OLPC project?

Tue Aug 09, 2011 11:31 am

I think it extremely unlikely anyone could make a board this powerful this cheap at the present time Obviously performance increases as per Moore's law, so expect more power at the same price in the future, but not the near future.

RaspPi has a distinct advantage with Eben working at the chip manufacturer, and being on the design team! (I work there too, but in software, but we will have to do any Pi work in spare time!).
Principal Software Engineer at Raspberry Pi (Trading) Ltd.
Contrary to popular belief, humorous signatures are allowed.
I've been saying "Mucho" to my Spanish friend a lot more lately. It means a lot to him.

User avatar
liz
Raspberry Pi Foundation Employee & Forum Moderator
Raspberry Pi Foundation Employee & Forum Moderator
Posts: 5212
Joined: Thu Jul 28, 2011 7:22 pm
Contact: Website

Re: How will this differ from the OLPC project?

Tue Aug 09, 2011 12:00 pm

I'm with James on feasibility and our good fortune with being able to work with the chip manufacturer - but Abishur is quite right, we'd actually be quite *pleased* if this suddenly sparked a flood of companies making $25 PCs. Raspi is a non-profit, so we're not going to get rich off it; our goal is to democratise computing, not to make millions. And frankly, I quite like the idea of someone else doing all the hard work for us! (Don't see it happening, though: it's a *lot* of hard work - it's been in development for four years now - and I don't see how anyone else would be able to get prices this low and take a profit. We'll be selling very near cost.)
Director of Communications, Raspberry Pi

User avatar
Gert van Loo
Posts: 2487
Joined: Tue Aug 02, 2011 7:27 am
Contact: Website

Re: How will this differ from the OLPC project?

Tue Aug 09, 2011 3:35 pm

The part which worries me somewhat is the investment required. To keep the price down you have to think of batches of 10K. That requires an op-front investment of $250.000 and with a lot of people asking for B's you get to $300.000. You are probably safe for the first batch as there seems to be enough demand. But with making no (or very little) profit the next up-front will not be that much smaller. etc. You have to keep gambling that the demand keeps up or you're left with a big loss. It would be nice if those who can afford it pay a bit more. Would be interesting to offer a +$2 version for organization support and see how many (misers) chose the cheap ones.

User avatar
liz
Raspberry Pi Foundation Employee & Forum Moderator
Raspberry Pi Foundation Employee & Forum Moderator
Posts: 5212
Joined: Thu Jul 28, 2011 7:22 pm
Contact: Website

Re: How will this differ from the OLPC project?

Tue Aug 09, 2011 3:49 pm

Oh, tell me (and the extra mortgage) about it! We do have some investment and other discussions underway which I can't discuss here (but we can chat about it offline when I next see you as long as you sign an NDA ;) ); the business end of things is, however, extraordinarily hard work. It's one of the reasons we're planning on running a buy-one-give-one scheme when we ship, along with plain old donations.
Director of Communications, Raspberry Pi

Cafe
Posts: 62
Joined: Thu Jul 28, 2011 2:22 am

Re: How will this differ from the OLPC project?

Tue Aug 09, 2011 7:46 pm

Quote from eben on August 8, 2011, 23:05
It seems pretty inevitable to me that someone is going to value innovate in computing; the big question is whether the resulting platform is closed (like iOS or Android) or open (like Raspberry Pi). Fingers crossed the good guys win :)

Don't get me wrong, I love the Raspi and laud the effort, but given that you have mentioned elsewhere in these forums that the Raspi includes a few closed source-parts, the swipe at closed-source platforms is a bit surprising 8O .

Perhaps a bit too philosophically extreme, but for the Raspi to be the real "good guys", it will require openness beyond software, but somehow I doubt your vendor will (easily) open up the AP's design. BTW, an interesting contrast is that the Raspi logo contest requires/expects that designs use only copyright-free fonts. So, it Raspi can go that heavy on fonts for designs which Raspi will own, it might need to carefully think about the closed nature of the critical piece (the AP) of its offering. However, having run similar things, I do understand the realities of a non-profit such as Raspi. You need to carefully choose your battles.

Again, please don't get me wrong. I just think that every now and then it is important to parse things very carefully, and feel this is one of those times. :)

User avatar
liz
Raspberry Pi Foundation Employee & Forum Moderator
Raspberry Pi Foundation Employee & Forum Moderator
Posts: 5212
Joined: Thu Jul 28, 2011 7:22 pm
Contact: Website

Re: How will this differ from the OLPC project?

Tue Aug 09, 2011 8:13 pm

I think that *is* a bit philosophically extreme. It's simply not realistic to expect that we could do anything else.

We've looked all over the world for open multimedia cores, and they just don't exist. There's no such thing. No chip company out there is going to open up something that has cost them years and millions of dollars to develop (thereby giving every Chinese fab churning out pirated chips out there an instant win), so we, like all other hardware companies, can only licence those cores. Creating an AP is not a cheap or a quick process. You've got huge teams of people, on both the hardware and software sides, involved; it all adds up to a huge cost and a lengthy process, which companies like Broadcom expect to see some upside from.

All the other Raspi hardware is open, which is *vastly* unusual; and TBH, I can't think of a single instance in which the multimedia core being closed is going to inhibit your development on the machine.

You're right; we do need to choose our battles, and this is one which we simply can't win. I think we'd be crazy to try. In an ideal world, of course we'd prefer the core to be open. We can express preferences like this because we are a charity, and we are not trying to make a profit from the device - our goal is to get as many of the things out there as possible, and if it's copied and someone else sells something similar for a similar price, that's great; the situation is entirely different for our suppliers, who are companies with shareholders who do not want copies of their product in the market. As it is, we have a very limited budget, and Broadcom are doing us a great favour in selling the chip to us in much smaller numbers than the several million that they usually require a customer to buy at any one time. They're making Raspberry Pi possible. We can't find any other AP this good out there for anything approaching this good a price. It'd be mad to expect a multinational semiconductor company to make business decisions purely out of philanthropy or out of open-source idealism; as it is, they're cutting us a very good deal. It's not a utopia, and companies like Broadcom have to make money for their shareholders; we are not the only customer for this chip, and to open source it would mean that Broadcom couldn't sell it to anyone else.

I can assure you that our intentions are of the very best. We're big enthusiasts for open sourcing, and any software we produce, as well as all the non-multimedia-core bits of the chip will be open sourced. But in the end, we're operating in the real world, and buying parts from real-world companies. The core stays closed.
Director of Communications, Raspberry Pi

User avatar
abishur
Posts: 4477
Joined: Thu Jul 28, 2011 4:10 am
Location: USA
Contact: Website

Re: How will this differ from the OLPC project?

Tue Aug 09, 2011 8:32 pm

Don't sweat it! I don't think we're gonna flame you for asking for clarification ;) From what we've been told, the desire of the r-pi team is to be open source. The issue at hand is to make the whole $25 computer concept work, they had to use a chip where, if I'm remembering correctly, they may not be allowed to release all the information, or there are complications with doing doing so.

Edit: And there's the official stance from Liz! That's what I get for walking off in the middle of a post :P
Dear forum: Play nice ;-)

Cafe
Posts: 62
Joined: Thu Jul 28, 2011 2:22 am

Re: How will this differ from the OLPC project?

Wed Aug 10, 2011 2:50 pm

Quote from liz on August 9, 2011, 21:13
All the other Raspi hardware is open, which is *vastly* unusual; and TBH, I can't think of a single instance in which the multimedia core being closed is going to inhibit your development on the machine.
...
I can assure you that our intentions are of the very best. We're big enthusiasts for open sourcing, and any software we produce, as well as all the non-multimedia-core bits of the chip will be open sourced. But in the end, we're operating in the real world, and buying parts from real-world companies. The core stays closed.

Thanks Liz for that patient and detailed response. As I said earlier, I do understand the difference between the ideal and the real, and I applaud Raspi for making what thus far appear to be pretty good choices. Also, I completely trust the honesty of your intentions and share your vision of computing education (though our means might differ).

I believe the closed parts are unlikely to hinder the Raspi's use, but as long as key software parts are closed, I will be hard pressed to call Raspi an open software platform. (With the little I know of the Raspi's software, I won't be surprised if it is no more open or closed than Android.) I know our opinion on this matter can differ, but I hope that we can at least agree that caution is needed before knocking other platforms, because such knocking can cause unnecessary distractions.

But then, you likely know all of this. Any way, closed or open, I will buy some Raspis to both support its cause and to evaluate it for use in my own cause. :)

Thanks for this intellectual engagement.

Return to “General discussion”