Slashdot Mirror
Project To Build Dual-Booting Linux, Android Tablet For $100
SternisheFan sends this quote from Ars:
"It likely won’t be as sleek or fast as a Nexus 7 or Nexus 10, but a new tablet running both Android and Linux is in the works for open source enthusiasts and lovers of low-budget devices. PengPod tablets, made by a company called Peacock Imports, will dual-boot Android 4.0 and a version of Linux with the KDE Plasma Active interface for touch screens. But in order to reserve a tablet for yourself, you'll have to contribute to the company's crowdfunding project on Indiegogo and hope enough money is raised to begin production. 'Our goal is to build a powerful, True Linux Tablet, one free of Google and Android's restrictions, at a reasonable price,' the PengPod IndieGogo page says. 'If you're a Linux fanatic you probably ended up getting an Android phone. 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? Too often, this is not so. That is why we set out to find a way to run real Linux and all the software you really want.'"
One Linux Per Contributor?
--- Mercutio was right.
For $100, i really hope its not based on the Maylong 150....
KDE4 runs just fine with 512MB RAM and modern Android devices have even more. If you'll read the thread you've linked, you'll see that it's an openSuSE issue which has a lot of pre-enabled KDE components that are not needed on a tabled (like semantic desktop).
According to various tests on recent distributions (use your Google-fu to find them, it's not difficult), the memory use of KDE is not great, but not nearly as bad as Gnome 3 or Unity. However, it would seem that xfce or LXDE would be better choices for low memory devices.
In general memory required for Ubuntu 12.04 and 12.10 goes: Ubuntu (unity) > Kubuntu (KDE) > Xubuntu (xfce) > Lubuntu (LXDE). Similarly, memory required for the different desktop environments on Fedora 17 is: Gnome 3 > KDE > xfce > LXDE. That's without the akonadi stuff which was affecting the KDE example you linked to.
Those who can make you believe absurdities can make you commit atrocities. - Voltaire
Why dual boot when you can run both simultaneously since both run on the same Linux kernel? Kind of how Windows 8 runs both WinRT apps(for tablet use) and desktop apps simultaneously. Best of both worlds, use the Android apps when you want to use a tablet, and then switch to KDE apps for real work, all without messy rebooting.
This space for rent.
OSX forums are just as bad. Hell I think all forums are full of twats trying to prove their self worth. OSX forums its more along the lines of 'you wanna do something different? IDIOT. apple didnt design it that way'. -- happy OSX and linux user.
Came on, if you really want dual boot, let put it as option as much of OSes possible.
I have an Ainol (*snicker*) NOVO7 Elf II which I paid $120 for (on sale from $140, and free Fed-Ex 3 day shipping). It has 1GB DDR3 RAM, 16GB internal memory, G-Sensor, good display, 5-6 hour battery, and a Dual-Core 1.5GHz ARM processor.
It runs Android 4.1 and I can run Ubunu 12.10 from the sdcard with almost everything except the touchscreen because of no drivers for that, or the Mali-400 GPU.
What they seem to planning here doesn't seem to be all that impressive considering my chinese brand tablet can do all that. Truth be told it may not be open-source like they want, but the kernel sources for Ubuntu are obviously available, and the company has released the Android kernel they use.
Ainol is also one of those companies that churn out tons of Android tablets, and they seem to be doing fine. A sub $100 tablet doesn't seem like such an achievement.
Basically, I don't see what the appeal of this project is aside from mabe extended support, but even my device has a good community behind is releasing custom ROM's and constantly keeping it up-to-date and applying fixes from the hardware manufacturers.
Comparing the desktop KDE with the meant-for-tablets plasma active is not exactly fair. Can't assure for KDE plasma active (not tried it yet), but pure linux running tablet interfaces (maemo, meego, webos) in the past had good user experience (at least for me), specially with up to date hardware. And you don't have so far away the rest that comes with linux, from the system or shell to compiling or adapting for it things for other devices or environments, or, well, have plenty of user interfaces to play with if you don't like one in particular (even Sugar could be a valid one)
the only problem with XFCE4 / LXDE is that they are not designed with touch input in mind. KDE4's Plasma Active is designed for smaller screens and touch input, and generally the smaller screen devices it is designed to run on don't have quite as much RAM so a lot of the more memory intensive eye candy and other goofy crap is turned off by default while retaining the KDE power utilities - Konq / Dolphin spank the pants off Thunar / LxFM in terms of features. Thunar still hasn't gotten the "Open Terminal here" working to actually open the terminal in the directory it was selected from as of the last time I tried it ( ~2-3 months ago) it opens the terminal in your home dir.
To err is human; effective mayhem requires the root password!
I always thought the same. Switching between OSes should be more simple, it just need a better boot management: How about a function built in both OS, that saves the memory just like as it was hibernate and switch by memory content? A modern firmware/BIOS should offer this possibility.
Actually... yeah - why should you need to dual-boot when Linux Desktop and Android both use the Linux kernel?
W..w..W - Willy Waterloo washes Warren Wiggins who is washing Waldo Woo.
The people on the forum are trying to use the classic Linux excuses and make the user feel stupid and blame him for Linux shortcomings.
No, the people in this forum are using a classic FUD technique of finding one nasty datum and pretending it's the whole world.
They're also lying about KDE, and being deceptive about the DE which will be used on this tablet. If you're genuinely interested in the system and have read past the mess of disinformation and proprietary propaganda that is today's Slashdot, go to the KDE Plasma-active site and test it yourself for free.
"I've got more toys than Teruhisa Kitahara."
Some people get very confused about the kernel vs. user space applications. RMS said it best:
“Android is very different from the GNU/Linux operating system because it contains very little of GNU. Indeed, just about the only component in common between Android and GNU/Linux is Linux, the kernel. People who erroneously think “Linux” refers to the entire GNU/Linux combination get tied in knots by these facts, and make paradoxical statements such as “Android contains Linux, but it isn’t Linux”. If we avoid starting from the confusion, the situation is simple: Android contains Linux, but not GNU; thus, Android and GNU/Linux are mostly different.“
I live in Shenzhen, and here in China you can pick up these Allwiner based tablets for about $100-$125 USD. My buddy couldn't resist a bargain and bought one a few weeks ago. I was surprised how well it worked out of the box. Decent performance browsing heavy pages, and the all the 3D games I could throw at it ran smoothly. That Allwiner blows the Rasberry’s CPU out of the water.
Initially, I was tempted to get one. Then I started noticing the problems. The accelerometer hadn't been properly calibrated or mounted at the factory, meaning some racing games you have to hold the device at a 20 degree angle to drive straight. When the battery started getting low, I plugged it in to its proprietary charger only to find out the touch screen doesn't work when charging.
Then about a week later my buddy said the screen popped out after he left it charging overnight. Turns out the battery had swollen up. All these issue point to shoddy cheap components and lack of testing and QC. With only $100 to spend, suggest a used Kindle or Nexus 7.
In a couple years, when I get tired of whatever low-end SoC they can get in their $100 tablet, there'll be a couple new generation of SoCs, and either a low-end from the newest generation or a mid-range from the older generation will easily double the performance for the same price.
If the tablet uses EOMA-68 CPU cards, I'll just be able to buy a new CPU card and upgrade the tablet. And then I can put the old CPU card in something else (maybe a plugserver or such). If not, I'll have to buy a whole new tablet, and in that case, why wouldn't I just spend $200 no on one that's twice as fast?
EOMA-68 or go home.
Crowdsourcing is the combination of your uncle borrowing money from your dad for his latest "hair-brained scheme" and a chain letter.
"Well, good luck finding a judge that doesn't run a bestiality site."
Linux is absolutely lightweight and flexible, it's just that KDE and gnome seem to be making a mess of things lately in a stupid attempt to keep up with the Joneses.
I use xubuntu+xmonad and it is blazingly fast and flexible on 4+ year old hardware. I tried using Gnome and KDE but I feel they get in my way more than helping me do my stuff, but I am completely happy with my current setup.
This is putting the cart before the horse. Make something that runs well on _existing_ tablets. _Then_ talk about building a special tablet to run it on. There are a lot of fine candidates out there—there's no reason to waste effort building another one that will deliver half the performance at the same price. A Nexus 7 or a Nexus 10 would be a great platform for prototyping this.
Having said that, I completely agree with your point about KDE. In addition to being a memory hog, it's hopelessly complicated and presents an impossible learning curve to anyone who just wants to figure out why dumped core. It might as well be closed source, for all the good that having the source does.
I'm right now on an HP-mini running KDE-Plasma on 2 Gb of RAM with 6 tabs open in Firefox and looking at the system monitor less than 400 Mb of RAM is in use.
When on a bare desktop with a couple of widgets running it's below 300 Mb, and yes that's of course without the indexing service running.
Would you take the tablet version of KDE you'd get even better results.
Stop trolling in a place with Real Users.
"The likes of Facebook and WhatsApp are free to those whose privacy is of zero value."
Trying to port Linux to Android tablets is a dead end. They will get Mer, OpenWrt, etc running on ONE tablet a year. If at all.
The alternative is to consider Android as a different Unix platform, with its limitations, and port KDE, Gnome, etc to Android. More details here:
http://www.elpauer.org/?p=1191
That path would reach potentially every Android tablet (and phone!). Easy? No. Doable? Sure thing.
Just when will this FUD end?! Oi!
Plasma Active is not Plasma Desktop a.k.a. KDE4 Desktop shell. They share code but they are two different projects so it's not one size fit all, it's the exact opposite where there's one framework with multiple but independant graphical shell implementations.
It is right in the ballpark. Similarly specced Android tablets sell for around $80-$120 at 7" (although they tend to have 1024x600 screens). You can get a single core tablet at 800x480 for a bit over $50.
It is worth supporting these sorts of projects if only to keep development going for tablet interfaces with Linux - or more specifically the applications that run on it. Android is fine in its own right, but what we need is an infrastructure that assumes that apps will be open source and free. For that, we need a repository that does not charge developers for hosting the apps - just like on Linux now.