I don't want this to just sound like "this is a terrible idea", but I'm interested -- why not just make a good linux distribution for something like the Nexus 10? I already want to get one of those, and would be happy to dual-boot traditional linux, or even just have access to true linux under the android hub.
Unless there is some serious restriction, it seems easier to get software right, than hardware + software.
Agreed. Historically, it has been cheaper to buy a high volume existing piece of hardware and throw away the pre-installed (and paid for) software than it is to build linux-oriented hardware. The price breaks that come from manufacturing million of units simply swamps the cost savings of not paying for an unused OS (e.g. Windows, Android).
Another interesting trick Canonical has talked about is Ubuntu on Android http://www.ubuntu.com/devices/android - essentially running a virtualized linux on the Android kernel. I don't believe it is available yet (their target market appears to be handset manufacturers, not us) and I have my doubts whether it will be useful enough to break into the general market.
WRT getting the software right, it is not easy moving a desktop-oriented system to a tablet/touch-oriented system. Canonical and others have been working that problem, so it will get better over time. This video (from 2010) http://www.youtube.com/watch?v=nn3m09zkcbo makes it look like it works great. Maybe it is time for me to buy that Nexus 7. :-)
Interestingly, their pictures are of existing Allwinner based Android products that are available for purchase now from any number of sources, and that there are plenty of documentation on getting Linux running for.
So making a good Linux distribution seems to be what this project is for. It's very unclear what they're actually offering - it seems that any donations will go towards work on a very limited set of missing functionality for existing Linux ports for these devices.
Personally I found the page very misleading.
EDIT: Here's details on Linux on the MiniPC they show:
I'm sorry if you find anything misleading. We try to be as accurate as possible without making the page too dry for non-technical folks. For example we list the features of the Allwinner processor as our work will be portable to many devices beyond what we sell. This lead to confusion were people were saying "I wont buy it, it has a resistive touch screen". The devices have capacitive touch screens and the chip has a built in resistive controller, but we are not using it. Our attempt to show the technical merits of the A10 confused many and so we must be careful how in depth we go. I'm happy to accept feedback about points we need to clarify.
I think this most important work will be doing is to setup and maintain repositories of available packages for the A10 and tablet based improvements. Also we also providing device preloaded and maintaining images for people who don't have time, or don't yet have the skills but would like to run GNU/Linux on their device.
I realize you can buy these devices at cut rates directly from China. Since we have to flash and test these in the USA, we have to pay livable wages to our staff. I realize its not great for the bottom line. It allows us to ship with the latest work we have completed without delaying the production run. This will help to avoid running into Chinese new year and a long production delay.
"By using crowd funding to consolidate the orders we will be able to achieve the prices listed in the rewards and upgrades to the hardware become possible."
Then why is the PengStick/Mini MK802 twice the price of Amazon?
If they're actually planning on specific improvements of the Linux distro, I might be willing to pay extra for it... But, yeah, it does seem like they're at the very least not doing anything at all to prevent confusion over what this is about.
Things move fast in China, when we started planning this project the cost was based on a bulk order. In the 6 weeks or so since the retail prices have come much closer the wholesale price. This is typical of Chinese goods once they are available long enough.
So the order quantity is not as important now but we still need the bulk to provide enough funds for the software work.
You're addressing the wrong issue: The price is not the problem. The extremely misleading page is.
And with regard to price: I bought one of the tablets you list from Amazon in the UK for less than your stated price many months ago. A single one, via a UK reseller.
Maybe it would help if they pitched this similarly to http://www.emperorlinux.com/ - taking other people's hardware, making sure it works well with linux, and selling an out-of-the-box linux experience.
Essentially, that's what this campaign is- they're looking for funding to make a good linux distribution for Allwinner A10 systems. The hardware already exists, and it already runs Linux. For example, their PengStick seems to just be the MK802 II, which you can pick up for $50 to $60. I have one of those running Xubuntu. I'm not sure what the tablets are, but you can find candidates at sellers like http://www.w2comp.com/catalog.asp?catid=75148
Getting linux running on these devices requires a lot of expertise on your own part, or trawling forums for builds other have created. What they're proposing sounds like a real improvement-
"While we have a good, running Linux system now for all the devices, there are still some rough edges. We intend to fix up the tablet related features, host a central repository of A10 specific packages and create images and guides to make these devices even better. All these efforts will take time and we plan to use the proceeds from the crowd funding to fund the remainder of the software work and maintenance."
There are already resellers that provide Linux images and substantial support. See e.g. www.miniand.com
In terms of "central repository of A10 specific images", that might be useful, but there's already a wide range of community sites collating information on Linux on A10, without trying to extract cash for it..
The tablets they show are almost certainly LY-F1 - it's sold under dozens of brand names and in a few different configurations and a few different cases, but the specs match perfectly and at least one of the pics are of the case that's most commonly used with the A10 version of the LY-F1.
But most all of the A10 based boards are practically identical - I'm assuming most are variations over some prototype provided by Allwinner given how similar they are.
I don't really want to say anything bad about Miniand, I'm a fan of their work and like what they have going, however... Our prices actually compare very well with Miniand once you include shipping. Also the images officially provided by miniand haven't been updated in a while, where as we plan to keep things more up to date.
The Nexus 10 also has the fewest binary blobs of any recent Android device.
Also, if you want an Allwinner based tablet there are/will be many available. The main work would be to find one where you can get the schematics, and also spec sheets for all the peripherals. Some companies might be open to releasing that stuff if you pay them.
Spec wise, the problem is the A10 SoC itself, not the board - the boards for most of the A10 based devices are practically identical and very simple. It's particularly the Mali400 GPU that's posing a problem, and there the challenge in terms of specs might very well be ARM themselves.
That said, there's work ongoing to reverse engineer it (Linux works on it, but without hardware acceleration)
Nexus 10 linux and great keyboard please! I would definitely fund that!
Edit: Das keyboard in laptop form with a battery. When I use the keyboard i'm sitting down for 'serious work' anyway, so a bit bigger wouldn't matter; 20 hours battery life + a real programmers keyboard in a clamshell would rock.
Well, we have an OS that is completely open source. An SDK that can be used on any platform and with any IDE without paying anything. Freedom to write and distribute applications to any device without anyone's approval.
> Hey, it's Linux right? It'll be open, run all the programs I'm familiar with and let me hack around and have some fun right?
Linux is a kernel. No, it won't run the programs made for desktop operating systems just because they use the same kernel, which are likely to offer poor user experience on a tablet anyway.
Yes, it will run the programs made for desktop operating systems, as long as they're recompiled for ARM:
This is an example of Ubuntu running on the LY-F1 (which is incidentally one of the tablets these guys show pictures of, so it's not very clear what they're actually planning on doing for this money other than passing orders on to distributors and chucking in suitable Ubuntu image):
This is a strange reveral of the gnu/linux debate. The linux kernel is not the important component, the windowing system is.
I dont want to run X applications on a tablet, I have tried that before with the n700. I want a fully finger friendly user interface, the one that comes with android works pretty well.
I can understand the appeal of a community owned FOSS user interface, but why attach the linux tag as a selling point.
Because some people would like to be able to attach a keyboard and use it as a makeshift netbook now and again.
And because there's a lot of work ongoing to make X apps "finger friendly". E.g. Ubuntu 13.04 has making Ubuntu work better on tablets as a major goal.
And these devices are not just tablets. The MiniPC (which is just a MK802 or MK803) has HDMI out and full HD.
IIRC the A10 specs are closed (and at a time I've been involved with a project including a A10, I've even been reported the A10 designers did not even have written one...), so this might not be the best SoC to start with, especially to target Linux geeks, promising them openness.
Also $50k of funding seems quite light for this project.
I hope anyway that it succeeds, because it would still be an interesting device.
Really disappointing screen resolution on all of these. Haven't they been paying attention to what's been going on in tablets - at least, in the tablets that people are actually buying?
The tablet form factor is, at its base, a minimal container for a screen. If the screen isn't good, then there goes the whole thing no matter how many other goodies you throw in. Low screen resolution means not good screen.
What's more frustrating is that I share the goals of this device. I want a linux tablet. That is, I want a tablet that will integrate into the workflow I have on my other computing devices, which are all linux.
I don't have very complex requirements for that; I don't need appliKations or Gancy Gesktop Genvironments. I just want a nice high res screen that talks to a bluetooth keyboard when I want, run a standards-compliant web browser, a bunch of tiled terminals, and most importantly Emacs. Android doesn't fit the bill. I can only see appreciating the way Android runs emacs if I was a devoted Buddhist and embraced frequent segfaults as a lesson in patience and discipline.
They're not making the hardware. The units they list are run of the mill lowest-end Android tablets you can pick up on Amazon (or directly from China via Aliexpress etc.) that more standard Linux setups have already been ported to.
EDIT: At least that's what the pictures are of. I have that keyboard case, and at least one of those tablets, already.
There are cheaper ones around too, but ~80 is the cheapest I'd consider in terms of build quality (unless for a child - one of the things you see on a lot of the cheaper tablets is that instead of a glass edge to edge face plate, there's a plastic shell that protrudes out from the screen, and perspex/plastic instead of glass covering the screem, and from look and feel those tablets seem a lot more likely to withstand being dropped - none of these cheap tablets use Gorilla glass as far as I know, after all)
Unless there is some serious restriction, it seems easier to get software right, than hardware + software.