Some current free software friendly hardware
- Inicie sesión o regístrese para enviar comentarios
I am in the process of looking for a source of new, modern, 100% free hardware on which to run Trisquel. I'd like to accomplish this by next September (2014).
I am aware of ThinkPenguin which is great. I have purchased stuff there. However I am after the elusive 100% free, open, documented hardware that will run Trisquel.
With this goal in mind, I have started to research things that exist right now and wish to share my findings so far.
quidam in his March 2013 keynote, http://quidam.cc/01-04-2013/talk-libreplanet-2013 at around 33 minutes in talks about an ARM version of Trisquel.
Free accelerated 3D graphics drivers have become available.
http://limadriver.org/
https://gitorious.org/lima
https://www.youtube.com/watch?v=l92S3wg2iSU
Excellent background from the guy who reverse engineered it.
http://ftp.osuosl.org/pub/fosdem//2013/maintracks/K.1.105/Open_ARM_GPU_drivers.webm
Using this GPU:
http://malideveloper.arm.com/develop-for-mali/drivers/open-source-mali-gpus-linux-kernel-device-drivers/ which is very popular. https://en.wikipedia.org/wiki/Mali_%28GPU%29
System on a chip SoC cards, which power tablets, and phones, are quite powerful computers right now.
http://www.allwinnertech.com/en/product/A13.html also A10, A10s, and A20. The A20 is dual core.
http://limadriver.org/Hardware/#Allwinner+A10
http://rhombus-tech.net/allwinner_a10/
This board, also an A10s https://www.olimex.com/Products/OLinuXino/
is part of an open-hardware project. https://www.olimex.com/Products/OLinuXino/ The board is similar to the Raspberry Pi and BeagleBone cards which aren't as free in terms of their hardware. The little cards can do amazing things and cost less than USD $100.
Ubuntu already supports ARM.
Some of the tablets using such hardware are so powerful that with various docks, a free lap dock notebook, or a free desk top, ought to be possible.
These parts and projects get us into the 100% free, open, documented hardware arena, but there is a lot more to figure out. Seeing this though, I am excited. Things exist. Right now.
If anyone knows more, or how to pinpoint, or to expand, what is going on in these markets, or with similar products, please let me know.
I will continue digging.
This is an interesting video using an OLinuXino board:
http://www.youtube.com/watch?v=R0q2e6u_vc4&feature=share&list=PL6-W3FoUyb4-hwrYcNDHPrU9uykcuAgLm
Not sure how to edit here. Forgot to mention that the last video is using Android in that setup. The tiny (older version than A10) card works, and can be made into a desktop, that was my point.
Thanks for the list. There should be an "edit" link under each of your comments that will let you to make corrections.
We are looking for freedom-respecting components to build our own desktops, laptops, and servers for the community education centers. For now, we will replace non-friendly parts in the PCs that are donated or purchased from auctions. New machines have too many backdoor components. :-)
Maybe not all new options are bad. That is what I am interested in finding.
A few more ideas anyway. I have to research this out more to see if ALL of it is free.
People are taking cheap tablets using Allwinner A10 or A13 and are running Debian or Ubuntu on them.
Blob-less hardware decoding on Allwinner A10 SoC based tablet
https://www.youtube.com/watch?v=z__RxeVmYCQ
Debian on a low cost 60€ AllWInner A13 tablet
https://www.youtube.com/watch?v=dXQrRPvUVsQ
Lots more to check out yet, but very encouraging so far. ARM chips may provide "enough" of a computer for basic chores such as web browsing and email and a few other things. Now if it can be done inexpensively on free hardware and with free software like Trisquel....
First time I watched Big Buck Bunny was the other day when Chris mentioned it in a comment. What a great video!
We're having good luck with these old PCs. Someone dropped off an ancient Dell running Windoze 98 the other day. It is a beast, but it likes Trisquel! :-)
Old hardware is great. I just hope there will be enough new hardware that will one day be old, that still offers users freedom.
I'd like to find things that are available now that offer freedom. For immediate use and for later use.
Could you post the video on http://gobblin.se , a MediaGoblin instance?
This are *spectacular* news. (And, most exciting ones - at least, for me...)
That it's beginning to be possible to have (even if just a lower-capacity one) a wholly (and truly) Free System, made of only Free Software and Free Hardware. :)
Let's keep this type of threads alive and updated, so that we can start building such type of systems.
I recommend you try the ARM 64-bit (Cortex-A53 and Cortex-A57) instead of 32-bit.
https://en.wikipedia.org/wiki/ARM_architecture#ARMv8_and_64-bit
To be clear, I admit I do not know much about what I am doing although that hasn't stopped me from doing things in the past. I will learn.
From what little I know, the A53 and A57 aren't exactly here yet. Scheduled for 2014 I think I saw.
It would be nice to find a solid available cheap hardware, right now, that is pretty good and get Trisquel to work.
If 100% free hardware plus 100% free software could be rolled into one, the rest would be tweaking and upgrading from there.
I suppose companies can develop anything so long as they can pay for the process of doing so. That would be an expensive way for me to learn. There are many standard development SoC boards that are quite inexpensive and can be purchased in small quantities, some even turned into products.
For example,
"OLINUXINO is completely open source - including hardware and software, which means you have access to all the CAD files and sources and you can reuse them for your own personal or commercial projects. There are NO restrictions to manufacturing and selling these boards for your own use or resale.
OLINUXINO use a widely available microcontroller iMX233, which costs USD 5.50 in 100 pcs quantity. This means that people can spin off their own boards and manufacture them as cheaply as the processor in TQFP package is easy to assemble by a hobbyist package."
I need to find a source for free-as-in-freedom hardware boards in order to cobble something together. My time and my skills as well as my budget, like most peoples, are limited.
It would be a triumph to get a usable 100% free hardware/software system working within the next year.
If there were such a base to exist upon, then it would be upgrades and tweaks. At least something would exist.
This task must be possible else I wouldn't be driven to try. And so far, nothing has been terribly negative.
Feel I need to pinpoint usable hardware and then obtain some. Also feel like I'm not feeling it yet, not there yet. But hey, at least there is some hardware that looks ok. The next thing is to go a bit deeper and see what else might be available and then begin to narrow the field down.
I don't want to get locked into ARM chips, although they are certainly available and cheap.
From your link, https://en.wikipedia.org/wiki/ARM_architecture "…in 2010 alone, producers of chips based on ARM architectures reported shipments of 6.1 billion ARM-based processors, representing 95% of smartphones, 35% of digital televisions and set-top boxes and 10% of mobile computers."
These chips will become more and more powerful much faster than Trisquel will bloat. As just a little more ARM power and storage space become the minimum norm why wouldn't free GNU/Linux software show up as an option?
The A53 and A57 seem to be pretty good computers powerwise. Plus,
I don't see ARM in the DASH group member list. That looks promising too. http://dmtf.org/about/list
There are problems with the currently available solutions such that we haven't moved forward with anything. From insufficient ram to non-free software issues. It may be possible to do a custom board with more ram though. A future project. I'm sure an expensive one at that too.
I agree with regard to a full machine as we currently know them. Something a programmer may want.
However for regular folks looking for a "free" experience, the ARM + Mali continue to look good for a simple way to connect hdmi perhaps to a TV and to web browse or check email, take a few notes and so on in a free manner.
Is there anything not free on this board below? It sold out fast (Most of the things I've been finding beyond this are sold out too. Hint.) The other model has no flash memory.
https://www.olimex.com/Products/OLinuXino/A20/A20-OLinuXino-MICRO-4GB/open-source-hardware
CPU A20
GPU Mali 400
VPU CedarX
The free reverse engineered Mali hardware acceleration driver isn't 100% there yet, but it is working and soon will be better. CedarX has had improvements too.
As a very basic free computer, it could work for some of us.
These Olimex boards are created for "fun" apparently by the engineers at the company. Paying projects come first where they are manufacutured in Bulgaria. The company seems small. https://www.olimex.com/About/ Don't know about the price to make a special board, doesn't seem huge volumes are necessary.
The following format at the link below is quite interesting. They use an A10, which can be replaced with an A20, just plug it in. Forum says Mali is the GPU. Never know though. WiFi is probably non-free but I don't know for sure. Watch the videos. This is amazing to me because I envision such a thing as being 100% free, running Trisquel.
https://www.miniand.com/products/Mini%C2%A0Xplus%20TV%20Box%20H24#specifications
They sell these for USD $59.00!
If "free" parts could be counted on, wow. A10 or A20, ok, Mali ok, WiFi maybe ok, probably not, but could be if properly specified. The point is, these products are almost free right now and they are cheap. Getting more powerful too. A Debian version exists for this. How far away are we? Really?
I remain excited. It's not that these products can't be built or would need to start from scratch. Just need the right parts and configuration. Not saying it is easy, only noting that standard options are available, just need to bring them all together into the format we want.
The wifi isn't free. That can be fixed easily via the USB ports though and the other motherboard. The major issue is the insufficient ram. With a custom version of this board though it might be doable. The other issue is there aren't any cases for these boards. Something else that would be an issue if your talking a finished product vs's a toy to play with. There may be other issues we don't know about as well. I have an engineer here who loves to play with this sort of thing. However he is busy working on less interesting projects that will hopefully bring in more money currently.
In a few months we might have the new web site done and then there might be some time to play with this idea of putting out a completely free desktop system. I'd have to run some numbers though and evaluate demand.
Chris,
There is to be a new and good looking box sometime in October according to a post on their forum. A new board too, A10 though. Cheaper. They seem to be trying.
There are many unknowns for sure. The parts get so close to being right sometimes. Among all the things I've been looking at, there may well be a batch that are just correct as is one day, but then probably gone the next. It's a moving target. Having a special board would be the way to go.
I know you offer the free audio players, it won't be long until similar devices could be running a full GNU/Linux operating system. (They already are, just not "free".) And it's not just the software on a stick, but the entire computer. At low cost. Free hardware and free software ought to be an option for that kind of thing one day before long.
Great for the customer, but hard for a business to put much into it if there is low volume, I understand that.
With the introduction of smart phones and tablets, and now these PC's the size of a USB stick for USD $60, not only are computers changing, but the way people use them are too. As are expectations. Free software ought to be in there somewhere, ideally. Unless perhaps the focus might remain on regular computers that do "jobs". With servers that makes sense, but so called "desktops" are really changing. Where will Trisquel fit in. More and more computing no longer happens at a desk. :)
That has been an excellent resource for me.
I am not familiar with wiki's, can anyone just add things?
There is a another board that seems to be like the others listed there.
It is sold by a company in Australia with connections to China. It probably ought to be listed?
https://www.miniand.com/products/Cubieboard%20A20%20Dev%20Board#specifications
I note the company's website says, "We are an Australian company with offices in Canberra, Australia and Guangzhou, China who have a great interest in the proliferation of cost effective, powerful, and unrestricted hardware, coupled with free open source software. We always ask "will it run Linux?"."
Wikis range from wide open for everybody to strictly restricted. Libreplanet is open but you do need to register. Wikis are wonderful collaborative tools.
I will see about registering. Thank you.
You wouldn't want to add anything to the list unless you're absolutely sure it's free.
Maybe there was a reason to leave it off the list. Then again, maybe no one was aware of that particular cubieboard, or no one took the time to add it yet.
Truth be known, I would really rather leave wiki updating to others and concentrate on what I am already doing, looking for existing sources of free hardware. I wouldn't want to post on the wiki unless there was a typical or expected reason to do so.
If there are high standards that I should know, I would rather leave it for someone else to decide whether or not something belongs on the list.
The Cubieboard A20 seems no worse than the OLinuXino already there.
The few definitions seem as if it would be ok.
The ability to work stand-alone - no reliance on other computers being available --YES
Libre boot firmware (BIOS) --YES as far as I know, most of these A20's use U-boot, but honestly I don't know for sure.
No components that can not be replaced by a person with a screwdriver that require drivers with binary blobs to function --YES. Again seems like the OLinuXino.
Usually wifi adapters and graphics adapters are the issues here - you can help with this if you are a hacker with appropriate skills... Mali GPU OK, no WiFi listed. Shows dongle as an option. So YES.
Not make an insane amount of noise or have a form factor that is aimed at the server market --YES. Seems OK.
http://cubieboard.org/ would have the detail.
***EDIT: Reading about something else, I found one source that says cubieboards are NOT open source hardware. http://linux-sunxi.org/Main_Page Scratch the Cubieboard. Still seems to pass the Libre definitions though, but this stops it for me.
These board communities vary in size. (Non-free) Raspberry Pi huge, (some non-free) BeagleBone quite big, (Free or just non-free WiFi) OLinuXino, not bad but maybe small, (seems free) Cubieboard community may be larger than OLinuXino. Presence in many places.
I am still searching out things at this point. Haven't really focused down to anything in particular. It looks like ARM is a big deal. Or could be. I'd like to check other architectures too. Not sure what might be available by comparison, if anything. I am just reporting on what I am finding as I go along. Hopefully building some interest in the general idea of obtaining some real, true, free hardware.
I am not in academia or part of the computer industry. I am not corporate. I am not in government or in non-government. I apologize if my methods aren't neatly squaring up with the norms common in those fields. I don't know what others expectations are here, I am doing what I can.
There is an overall sense here at the Trisquel forum and FSF too although I haven't been there much, that there is a need for free hardware. But also that it will be very costly and take quite a bit of time to actually get it into peoples hands.
I have a different view that free hardware exists now and we haven't found it yet. Also that free hardware can be produced at low cost in low quantities and rather quickly. This is what my gut and experience tell me. I am following that and am doing what I am doing here to provide a resource or background so that if anyone joins the hunt, or perhaps later wishes to get up to speed, well, they'll be able to do that.
I am hell bent on finding one or more 100% free hardware options on which to run free software (ideally Trisquel [which also needs adapting] within the next year. I feel strongly about this. This must be done or tried and I am doing what I can. If I fail, so what. I'll leave 10,000 pieces behind me. It won't all be failure. Although so far, I see absolutely no indication of failure.
This is not about money, or raising money, it is sweat, it is effort. It is a DIY do-it-yourself approach using pocket money.
Can anyone see as I do, those 3 guys who reverse engineered the Mali GPU?
http://ftp.osuosl.org/pub/fosdem//2013/maintracks/K.1.105/Open_ARM_GPU_drivers.webm
All the talk for how long now about GPUs not being free, and then 3 guys do it. Tremendous progress in one year. And they had to start from scratch, from nothing. One of them is 16 years old! I've said it before, this stuff is not that hard to do. It just needs to be done. People need to step forward. Pieces will fall where they will.
ssdclickofdeath thank you for letting me get this out. Your statement made a lot of sense and I am not arguing that, just getting a few things off my chest.
I'm not the best person to talk about "open source hardware" although its not an issue as far as the free software community is concerned. While the designs being available are nice it isn't something you can tweak and produce more of. It takes serious money to do even in small quantities. There is a huge difference between the two.
That said the Cubieboard might be something to work off. I believe the Mali400 has a reverse engineered driver for the graphics. It's not ready for prime time from what I've been told... but... it's a start. The system has 1GB of DDR3. Not adequate in my mind to run a modern desktop environment although could be "good enough" should the desktop environment be "built" (chosen) for the hardware. It's got a dual-core CPU so... thats good (enough?) I think.
Some other things:
10/100, HDMI, USB, & SATA ports: means its got potential to be a "real" desktop in my mind.
The optional wifi dongle amongst other items are probably not free software friendly though. Fortunately we already have a solution for that. I'm also not sure about the 10/100. Probably is... but who knows (I've always been surprised the wifi in these types of devices are non-free when we have a multiple options in the desktop arena, from PCI, to USB G, an older USB N, and now a newer USB N). I'm not well versed (getting more versed in it though) in embedded / ARM / etc world.
Could this be something a completely free GNU/Linux desktop could be built from? Possibly. More investigation work would need to be done. One thing that I have to say about most of these boards though is once you start investigating they tend to fizzle out as far as being “good” boards to build off.
Chris,
Some quick points.
1. As to cost. Once a source is selected, then we'll know for sure. These little boards were born in small quantities as side projects. Admittedly a 100% free hardware computer will likely be a specialty product appealing to a narrow group which in the grand scheme of things means low sales volume.
In such a scenario a low volume producer (a specialty or small board maker) is all that is needed.
2. The Cubieboard is very interesting, lots of support there too in their community, but it is not clear that their board is easily available to turn into a product. The Olimex board, which to seems to be very close to the Cubieboard in its main parts, is industrial quality, and everything is 100% open and available, at least that's what they say. The BeagleBone cards are neat but are not free and with them, there are hoops to go through should one wish to make a product based on their board.
I am new to the finer points of free hardware and free software. How far does one go? How deep? I mean the Cubieboard seems "free" as far as I can tell, however it apparently is not "open hardware". This would seem to say one cannot look at its intimate details. This would not be a good thing if it were software so is it good when it is hardware? I would say no.
Stallman isn't 100% clear on free hardware as far as I can tell. Sometimes he is ok with non-free things baked in that cannot be changed.
Also, hardware has become a more important issue to him. A surprising chunk of his LibrePlanet speech was about the need for free hardware. https://trisquel.info/en/forum/rms-talk-libreplanet
I've noticed that people who write things that are very close to hardware, I mean programmers, need as many, and as clear, details as they can get. It would seem the more open the hardware, the faster and better the programming. So I lean toward that.
3. The difference between a "modern desktop" and "good enough" won't be a question within a year. About every 6 months there is a new board. Always a power improvement, but sadly not always an improvement toward freedom.
Faster chips/boards are available now. But then they use GPU's like PowerVR, which is far from being known or open or free. The Mali GPU is "the one". That guy, Luc Verhaegen, needs help. http://libv.livejournal.com/ But even so I'd expect he'll have useable base code before this year ends.
Android is taking over Linux when it comes to graphics drivers. Wrappers have been created to use Android binaries and most Linux distributions seem ok with that which is not in favor of freedom. It seems most distributions believe they need to do something "now" to keep GNU/Linux in the game and remain an option other than Android which has taken over probably way too much of ARM and device markets. So they'll use the wrappers and run, saying they'll fix things right later. But this of course is a bad idea long term, "real" code needs to exist, no matter the timing.
This is also a long way of saying that once the GPU can really be used instead of the CPU for graphics on these boards, there will be very noticeable improvements in speed/power even on the older ARM boards.
4. I want to do more research on the Loongson. Beyond that if anyone can point to other processors that are worth a look, I would appreciate it.
x86 is quite known, I'm not sure there is a "free" way in? Doesn't seem like it, although I haven't looked into that area much. Seems like if freedom were possible there, there would be more examples.
ARM is big and growing. A nice thing about ARM is that the company began almost like these board makers in that they've always had (or maybe used) a strong community base to develop their products.
Since they do not manufacture anything, only license, this too is different from the x86 world. As mentioned in earlier posts, many independent factories produce ARM licensed products for many uses. When all is said and done, there will need to be a Trisquel or gNewsense ARM version at some point. I would expect that the "Replicant" phone project would have to be ARM based right now.
At the March LibrePlanet, Ruben was all for ARM when the time and conditions were right. http://quidam.cc/01-04-2013/talk-libreplanet-2013 (around 33 minutes in.) The conditions are getting very close to that now, but that's my view. If 100% free hardware were here, and if it is was ARM based, maybe that would be enough.
ARM is fascinating and I urge everyone to learn more about it. Since it is used in phones, tablets, TV's, and other "things" it is here whether we like it or not. Why not have it be as free as possible? Seems to work very well too, except in many implementations it is not free….but in others it appears to be!!
5. If you mean by "fizzle" that with the speed of change in the tech world, by the time a board is identified as "good", everyone has already rushed to a new one which isn't, and the old good one is discontinued, that has me concerned as well. However I see a couple of possibilities.
For example, Olimex seems pretty small, stable, open, whatever. Their philosophies on parts and price seem good. They could make more boards, old or newer, I don't see them as dropping and running to the next big thing. But that is just guesses based on reading, I really don't know. It will be something to find out if an ARM board is "the answer".
Also the Australian company Miniand has a thing for "unrestricted hardware" and for asking "will it run Linux?" when they create products. Maybe there is a way to work with them somehow too. They have an office/history in China. Plus Mimiand is already making some very close to free computers right now. A few tweaks and maybe they could have a free one.
And I'm with you. It seems everyone drops the ball on WiFi. Everything is good, check, check, check, then the WiFi! Nope, not free.
There are a lot of things to check out yet. I am happy to see your interest in these possibilities.
Stallman's position on free hardware looks clear to me. As far as I understand, he considers that, today, the freedoms defining "free software" cannot be applied to hardware because no individual can build her own hardware; even less copy some existing hardware. The fact is: we all depend on mass-produced hardware. This will remain so for years or, more probably, decades. Therefore, rms does not bother to consider this problem that has not arisen. When and if it arises, it ought to be understood and solved with the technology that will be then available.
Now, rms cares about "free software" up to the drivers and firmware. That is why he cares about "documented hardware". The documentation is self contained in free drivers/firmware if the manufacturer provides them. Otherwise, the free software must be able to develop those and documentation helps a lot. If no documentation is available, the free software community has to rely on tedious reverse-engineering (a great contribution to achieve). Whatever the way the free drivers/firmware are obtained, the hardware is then, according to rms, freedom respecting: anybody can use it running only free software.
You have his perspective perfectly stated.
"Whatever the way the free drivers/firmware are obtained, the hardware is then, according to rms, freedom respecting"
But, isn't there a (constant) concern, among Free Software adepts, with security and privacy, also?
I mean, isn't that something that is implied in the concept of Free Software, itself? (The need for security and privacy, that can only be obtained with free and "open source" software?)
If so, you have to take into account the existence of "hardware trojans" that can be possibly embedded into motherboards and microchips. (And, one way to counter that would be the idea of documented/"free" hardware.)
I mean, shouldn't we take the concepts inherent to, and that are behind the idea of, Free Software, and apply them also to the hardware?
(the following is the concern I'm talking about:)
"'Elbit' flash memory chips, allegedly designed at Kir Yat-Gat south of Tel Aviv. The unique feature of the Elbit chips was that they worked on ambient electricity in a computer. In other words, they worked when the computer was turned off. When combined with another newly developed chip, the 'Petrie,' which was capable of storing up to six months worth of key strokes, it was now possible to burst transmit all of a computer's activity in the middle of the night to a nearby receiver - say in a passing truck or even a low flying SIGINT (Signals Intelligence) satellite."
--- http://www.fromthewilderness.com/free/pandora/052401_promis.html
"CELL PHONE (FBI can listen to you when phone is turned off)": http://www.youtube.com/watch?v=0G1fNjK9SXg
I thought this would be a discussion worthy of a thread of its own, so I decided to create one, about it, also in this forum.
So, if anyone wants to answer what I said, please do it here - https://trisquel.info/en/forum/million-dollar-question-concerning-hardware-we-use - instead.
Thank you.
In reference to statement 2. I'm pretty sure his speech wasn't about "open hardware". “Open hardware” has a specific meaning. It was coined by a different group that is not really against free software, but may not entirely understand, or agree with the free software movement. RMS takes issue with its naming. Obviously he doesn't like the word open to begin with. However the “open hardware” movement is about the designs being completely available. To RMS this is not an issue. Users can't do anything with those designs anyway.
RMS's speech was referring to producing free software friendly hardware, and not to the designs to such hardware being completely available. His concern is the source code to all writable areas containing binary code be available. The reason he is concerned about this is because there hasn't been enough cooperation from industry to produce free software friendly hardware and the more hardware that comes how the more hardware that has firmware. 10 years ago there was significantly fewer chipsets dependent on non-free firmware. Keyboards wireless devices, CPUs, etc didn't include little bits of code. Now they do.
In reference to statement 5. I didn't mean that things become outdated too quickly. That is an issue, but not what I meant. What I mean is every time I look into one of these boards/phones/devices they all end up having a major issue. For example the PI has power issues (amongst other issues with the VGA, underpowered, and other parts needing non-free "stuff"). Then there is the mars board, it might be freedom friendly with some work, but there are issues with playing video I think (on this specific board). Then there is that other board. It's got only 512MB and no case. Solvable for sure... but why bother if it only has 512MB? 1GB would be better, and a case, but what major problem have we yet to discover if we picked this up? Another example is the ZTE Open phone. It seems mostly free- but then the wifi is non-free. There are lots of phones and other similar devices running ARM which at first appear like they might be OK. In reality once you get them there are major problems. Ohh and the ZTE Open (FireFox phone) which promotes freedom doesn't even have anything you can download and actually build the OS for the device (license violation?). It appears they might be technically in compliance (download is available, but won't build anything that works for the phone)? Anyway. My point is just that even if we are close it's probably going to take some serious effort on our (ThinkPenguin, community, etc) part to really produce something that you can readily get and use without major issues, etc.
Chris and Magic,
On Stallman, I respectfully disagree with some of what you say.
If I am using incorrect terms re: open hardware, I apologize for that. What I'd like to find is hardware that is the most free, clean, known, i.e., known what is inside, whatever...and it needs to be available for purchase, not discontinued.
This recent speech (link below) [Did you watch it? Even if you heard it in person earlier this year you might want to refresh your memory.], beginning at the 5 minute mark and continuing for the rest of the clip is about worsening hardware issues. http://files.jxself.org/rms1.ogv
Since I don't have a history with this, I can only listen to his words and voice. Seems clear to me that hardware is an increasing challenge to the concept of free software. More so of late.
Magic, "Therefore, rms does not bother to consider this problem that has not arisen." What? His words, this 10+ minutes of his speech, say otherwise.
Chris, "His concern is the source code to all writable areas containing binary code be available." Yes, agree.
"10 years ago there was significantly fewer chipsets dependent on non-free firmware. Keyboards wireless devices, CPUs, etc didn't include little bits of code. Now they do." Also yes, agree.
He continues to favor his Lemote. Apparently that computer is cleaner or freer than other options even now. He satisfyingly held it in his hand. I will look more into the details of that, just so I can know what the apparent standard is.
Chris, I can empathize with your "every time" you find what appears to be a decent board there is a "major issue". Can't give up man. These things are tough. It doesn't matter if all those boards didn't work out. It does not mean new ones cannot work. Rightly the odds are not good, but that's no reason to give into even rightful hopelessness. Perhaps also your standards are higher than mine, or at least, different. I too would like a really good option that covers all the bases and has no major issues. However I would be happy with having something that is fully free, even if it were a little clunky or slow, but that's me. I believe (for some reason that I really do not understand) that it is important for me to have and use that type of system. Existing or living with it means, "it's here." It's alive so to speak. It's an idea that lives. I don't want the idea to be a unicorn that I am chasing after (not saying you see it that way, simply my view). My effort now is to lead to such a machine that I can purchase or build and use...by next September.
Did a very little research on the lemote and more on pluggable computers. Nothing bigger to report today.
RMS is just saying hardware is an increasing challenge because traditional tasks are now being done in firmware and stored on flash or loaded to the device on boot. That's all RMS was saying. It's being done so much that we can't keep up with reverse engineering it all. Reverse engineering is a huge undertaking. One of the things in his speech was we need more people with reverse engineering skills. So it's not been written off the table. He also said there needs to be money invested in producing our own hardware that is free software friendly.
Nobody is giving up. It's just that it's not on the immediate agenda. There just aren't the resources, people, and organization to move things forward. Little steps require huge resource investments by lots of different people.
Lastly. There is nothing that is really 100% free software friendly for which even 1% of the people would consider actually using and that is readily available today. That is even amongst the free software community. The closest thing to a 100% free software system is the Ben NanoNote and possibly the older Lemote. I think the Ben NanoNote actually is 100% free although its missing critical components and isn't a desktop/laptop by any stretch of the imagination. It's missing wireless, is a pocket size device, very old, very slow, etc. I have the Ben NanoNote in my collection. I don't have the older Lemote.
"I have the Ben NanoNote in my collection."
Is it for sale? :)
No,
Although I'll make you even more jealous :) . It's signed by RMS himself.
Oh, this is unbearable! :)
The thing is--blame it on me being new perhaps--the freedom effect of free software is compromised when proprietary or secret software/firmware controls are lurking inside of hardware.
If these features or firmwares were forever frozen in, inside the hardware, and their details were known, free software would be written around it, retaining complete control over the system.
Chris, I'm not sure I'd call the interior functions of todays' hardware "traditional". Some may be, but others are a new form of copy protection and potential or actual spying and unauthorized data collection and lockdown mechanisms.
Common now, not as before, the inside of many chips and hardware devices are not only closed, but also legally secret, and the innards can be updated/changed but not known to most people including the chip makers.
I like what free software represents as it gives me a sense that I am operating in a general, fair, more or less trustable, best honest effort environment. I would also call it "real". Not bloated, not the fanciest, latest anything, just solid good stuff.
When I learned about vPro and DASH, my heart sank. Then there are Snowden's claims too, which continue to be supported by evidence that what he said is true.
Free software is the best thing going right now if one wants maximum personal control over their computing environment. There is quite a bit of passion in this forum over the possibility of non-free software being inadvertently installed (Debian) because of easy connection to the wrong repository.
Many times people have recommended here that if you are choosing to add non-free software, why even bother with something like Trisquel, as the non-free items defeat the whole idea of "free". There are purists here, not that that is a bad thing. In fact I like it.
To me, when someone buys non-free hardware instead of free(er) hardware, their freedom is diminished and they ought to realize that and not think that free software alone equals a free-as-in-freedom computing experience.
There is an attitude that it is ok, forgivable might be another word, when hardware is not free, because it is mass-produced and the average person can't build it or copy it, whereas with software, it can be written and rewritten and copied (Not by this average person though who isn't a programmer.) And that point of view makes sense in a software focused organization like FSF or maybe Triquel; hardware is another world.
I have no urge to convert anyone to my views, it takes too much energy. It is hard enough taking the time to write them down. I'd rather be researching and looking for hardware right now.
Finding answers to what quantum gravity has pointed out:
1. can the hardware be used with free software?
2. does it contain any malicous features? (vpro etc)
(Thank you quantum. That is what I mean by "known", how much is "known" about it. I would also add finding a source of purchase or supply or production that is so basic and so small that there is virtually no interest in gumming up those basics with extra locks, weaknesses/back doors, controls, etc. just boring basic hardware that can be used as a computing device with free software. That is what I mean by "clean". A source of supply that is not in the mainstream and never will be.) Note there are companies who produce tables and chairs exactly like those from the 1950s. Same methods, etc. Not a big market, but a living for those producers. Believe me, there are people and companies who have what I seek, I just haven't found them yet.
Chris you are doing a wonderful thing with ThinkPenguin, it's as close as close can be for offering computers that resemble regular mass market computers, only more special because of their adherence to philosophies of freedom and free software. That's great. But they aren't 100% free. Your WiFi usb is, which is working great for me by the way.
I hope you don't think that I don't respect or appreciate your hard work to bring this about.
quantum's point #2 above is huge to me, because we can't know for sure what is inside of hardware with features that are secret, but we can do our best with it. Not exactly reverse engineering in my case, but maybe something like hyper-knowledge if there is such a thing. If 1000 hours goes into the study of point #2, I'll know a lot. And once things are known at that level, the little important things, the clues, are easier to recognize and discern.
If there isn't interest in this sort of thing, that is ok. I am doing this to satisfy my own desire. Things would perhaps move faster with additional help here and there, but if not, I've got time, and definitely an interest.
Others have started projects http://opencores.org/ and have sadly fallen away from updating them as often when conditions change. Note that I am more interested in locating a source than designing or building from scratch as opencores is.
Trisquel and the Free Software Foundation (even ThinkPenguin) shouldn't be expected to do everything. I understand that. Finding the limits and boundaries of these groups is helpful to me. Maybe what I wish to do doesn't completely belong here, or with them, I am aware of that too.
Chris I agree from an organizational point of view that as you say "Little steps require huge resource investments by lots of different people."
Independently I disagree.
One or two people can and often do make a huge contribution when freed from the demands and politics and finances and whims of an organization. Not saying either is better or worse than the other. They each have their strengths. Over the years I've been in various groups as a member and in leadership positions, I am best when just doing my own thing though.
quantum sorry I am not posting this under your post, I want to just handle this in one place with one post.
You said, "I still don't understand why "baked-in proprietary code" is considered as fine by rms; I mean code you can't change without distroying the chip.
Can't this code spy on the users as well? What about a baked-in NSA backdoor?
If something like this is possible, maybe we should think about the initial goales of the "100% free software pc" project again.
Nobody should take such an effort of work for creating a pc which can't even provide privacy."
Exactly!! That is what I mean about RMS not being clear (to me, who is new). Organizationally it may be too much to morph a software foundation into a hardware area, but without decent hardware, the impact to the software has to be significant. He says his list for reverse engineering is growing longer. Perhaps that is the best thing to do with a membership that may be software and programmer focused. Apply them to the craziest hardware and crack, crack, crack.
I just want a true free-as-in-freedom computing experience kind of like I used to have way back when. At one point copy protection was a pain but it was easily defeated. This stuff today is insidious. And I hate spending money on something that supports what I don't want, what I don't believe in, and what I think is wrong. There need to be additional options. Privacy is a given if what's mine is truly mine and is not compromised.
So, first, I can attest to the fact a whole computer will definitely require multiple companies and people to participate.
1. You need somebody working on a distribution, like Rubén
2. You'll probably need somebody to reverse engineer something (like video drivers)
3. You'll need somebody to design something (be it at the board level and/or a complete system level)
4. You'll need to get the cooperation of people within companies
5. You'll need to get a factory somewhere to produce it
6. You'll need companies to distribute it
7. You'll need companies to support it
8. You'll need people to invest in it (factories won't produce your stuff without a upfront payment)
Just to get the ath9k firmware free'd and supported it took at least a dozen people. Many of them played a major part. The point was only that there are many people which need to participate in the proccess before any kind of product will get into your hands.
A computer contains many parts/chipsets and getting them all free'd is going to be a lot of work.
The privacy issue is separate from the free software issue. While you can hide stuff in non-free firmware more easily there is also the possibility of hiding it in hardware. It's better for the firmware to be free though as it does make it a bit more difficult to hide something malicious.
Nobody who understands whats happening with hardware is going to argue traditional functions aren't being moved to firmware. It's been happening near 20 years now (1990s). In the early days these devices would eat up to 1/3 of a users system resources. The goal was definitely to cut costs and not to add DRM or other privacy invasive features. There was a significant savings back then to move to firmware based designs. Now they are doing it to reduce time to release (results in shipping buggy and incomplete products), cut costs (you can see its even further as the flash is no longer included on the boards even in many cases, it's just uploaded to the memory on insertion or system boot), etc.
I didn't say that the inerts of todays PCs aren't providing additional, different, or privacy invasive functionality either. I would say these are a side matter though. What I did say was that traditional functions done in hardware are being moved to firmware.
"What I'd like to find is hardware that is the most free, clean, known, i.e., known what is inside, whatever...and it needs to be available for purchase, not discontinued."
All this is very vague; what is "clean hardware?" or "known hardware?".
There are only two points we should be interested in:
1. can the hardware be used with free software?
2. does it contain any malicous features? (vpro etc)
Beside of this, the whole discussion reminds me of one thing:
I still don't understand why "baked-in proprietary code" is considered as fine by rms; I mean code you can't change without distroying the chip.
Can't this code spy on the users as well? What about a baked-in NSA backdoor?
If something like this is possible, maybe we should think about the initial goales of the "100% free software pc" project again.
Nobody should take such an effort of work for creating a pc which can't even provide privacy.
"Beside of this, the whole discussion reminds me of one thing:
I still don't understand why "baked-in proprietary code" is considered as fine by rms"
Actually that's not true. rms does ackgowledge the issue with malicious circuitry:
http://www.gnu.org/philosophy/android-and-users-freedom.html
"The phone network firmware comes preinstalled. If all it did was sit there and talk to the phone network when you wish, we could regard it as equivalent to a circuit. When we insist that the software in a computing device must be free, we can overlook preinstalled firmware that will never be upgraded, because it makes no difference to the user that it's a program rather than a circuit.
Unfortunately, in this case it would be a malicious circuit. Malicious features are unacceptable no matter how they are implemented."
and
"Putting these points together, we can tolerate nonfree phone network firmware provided new versions of it won't be loaded, it can't take control of the main computer, and it can only communicate when and as the free operating system chooses to let it communicate. In other words, it has to be equivalent to circuitry, and that circuitry must not be malicious."
Thank you, Starchild, for the clarification!
In that speech, rms only talks about software. Software to drive hardware but software anyway. He talks about the problem of running proprietary blobs that can, therefore should, be replaced by free software. For instance, at 7'45:
The microcode was just something that was inside of a chip. It might as well have been a circuit. It was not something we had to care about. But when it became something that users were expected to replace, then it changed into software installed in a computer, which is something we do have to care about.
Later in the talk, it becomes then pretty clear that, whatever the way the free drivers/firmware are obtained (even through reverse engineering), the hardware is, according to rms, freedom respecting.
What rms would consider as "free hardware" is hardware the user controls by being practically able to exercise the four freedoms. Today, it simply does not exist. We all rely on mass-produced hardware. What you call "free hardware" is not clear at all:
What I'd like to find is hardware that is the most free, clean, known, i.e., known what is inside, whatever...and it needs to be available for purchase, not discontinued.
The two points quantumgravity emphasizes (+ starchild's precisions) are those rms emphasizes as well. I do not think it has anything to do with "free hardware". It is all about "free software".
There is another aspect to these small boards and devices which I want to point out in this thread. There is a lot going on with ARM architecture and a free software option that supports it would be nice. Learn more here http://www.linaro.org/
[EDIT: Learned linaro was created by ARM, not that it is a bad thing, but something to keep in mind.]
These little boards are not only used as prototypes, hobbyist devices, or as bases for consumer products. They are being used to build storage devices and high powered servers.
AMD has announced that they are launching a new ARM Cortex-A57 64bit ARMv8 Processor in 2014, targeted for the server market. I think they will use 16 boards in this server.
https://www.youtube.com/watch?v=t9utP7BGfrY
The A57 chips on their proposed servers' boards I think are entirely different from the A10 and A20 and likely not as free. However once these little boards are figured out, they will be combined into many different and serious products.
An interesting point about ARM is unlike the x86 world, there are many ARM chip makers.
The four "big" producers of ARM chips in 2012 were Nvidia, Qualcomm, Samsung and TI.
ARM doesn't mandate how things must be, they do not manufacture anything, they just license their architecture, and chip makers go from there.
It therefore must be possible to license "plain vanilla" ARM, without fancy backdoor anything. The equivalent of the plain "goo" that I mentioned in an earlier post in this thread.
I wouldn't be surprised if many chip factories had some sort of blanket base license to produce ARM, perhaps with add on features/costs as extras.
The ARM A20 and A10 SoC chips made by Allwinner are very good, however support is not. This is why many small board makers jump around with their sourcing. It's not always features that drive changes. The point is, there is a real choice of supply with ARM vs x86 which would indicate more freedom in ARM than in x86 if only because there are more choices of supply and of design.
Allwinner has been very popular, but I'm not sure they are in the "big" 4 or "big" whatever of chip makers.
And even if Allwinner has become a "big" chip maker during the past year (I don't know if that is true or false), and if perhaps they got bossy and demanding about how "an Allwinner" chip must be; since ARM architecture is licensable and the chips producable in many independent factories around the world, it seems it would be easy enough and maybe cheaper to find another source of supply (a smaller producer perhaps) more willing to manufacture a free hardware design/chip/board.
Trying to do that with x86 or via an all new chip created from scratch and produced by one supplier would have to be a lot harder and maybe impossible due to costs and legal issues.
Seems like there is lots of flexibility with ARM. Obviously I don't know all the details yet. Just saying things continue to look quite good.
---
Not related to the above info, but wanted to get it posted in this thread, ARM does offer virtualization…but not on all designs.
One question that I have about this...
(And, I'm not sure if this is what "trisq" meant, when he said, in the first post, "Some of the tablets using such hardware are so powerful that with various docks, a free lap dock notebook, or a free desk top, ought to be possible"?...)
Does anyone know how feasible (or easy/simple) it would be (for those who would like to have a powerful system) to build a "cluster" with this type of low-capacity computers?
There is a lot of cobbling going on. In videos, people have stuff laying all over connected with wires and are getting certain things to work. Not really a purchasable option at this time though.
The AMD link video sort of proves the concept in that a server is being released consisting of multiple small boards. I don't think what they will offer will be "free" as in freedom though.
Low power consumption is a driver of using the small boards too.
I've noticed a dead project/product from a guy in Germany, I didn't save the link. He created a boxed storage device using tiny laptop drives and a small board that was pretty close to free. Used almost no power compared to the usual options.
I think Atrix? phones had these "lapdocks" which look like a notebook computer. Screen, keyboard, etc. there is a way to use these (which are now cheap and discontinued) with a small board as a computer.
There are similar things to turn tablets into more of a regular computer. That's what I meant about a near desktop experience. Not sure of freedom issues, probably something won't be quite right.
Most likely part by part would need to be selected and hooked together. Harder to find than to do.
I'm not sure what is happening with this guy, but how do you get to the bottom of this kind of mess? Free hardware and software might have helped if only because potentially one could look deeply inside.
http://www.wilderssecurity.com/showthread.php?s=9eb0179c27cae4ce66ffdc7327006407&t=353112
Thank you everyone for those excellent points. The clarity of my definitions are not good, too much of doing my own thing I guess. Usually I do not have to explain things to a high degree of detail. More often I don't have to explain things at all, because no one cares! haha.
As I've said I will continue doing my own thing. Your points are well taken. One thing still has me concerned though.
Thank you magic b for getting that RMS quote. When he says "The microcode was just something that was inside of a chip. It might as well have been a circuit. It was not something we had to care about. But when it became something that users were expected to replace, then it changed into software installed in a computer, which is something we do have to care about."
Ok. When "...users were expected to replace, then it changed..." what about when something changes the functioning of the hardware and users don't know about it? Is that something we have to care about too in a freedom sense? Will free software provide freedom on such a machine?
Does anyone know how vPro works, or DASH features? Can they be opened/closed/started/stopped/changed without the user knowing? Absolutely. The system administrator can do things and the user doesn't have to know.
Excellent old vPro demo
https://www.youtube.com/watch?v=wlj7u3tOQ9s (start at 1:20)
Back then, there were 15 "out-of-band" features. Does anyone believe there are fewer now?
As a private individual using my own money for my own computer, do I need anyone to "administrate" it besides me? Can I make sure that I can lock out such features? Can I make sure such features aren't there?
I know I do not want such features.
That, plus I want the ability to run free software.
There are still non-corporate secret or spying issues I suppose but those are likely focused on mainstream devices if mass surveillance is the goal.
Non-mainstrean products that are simple or basic and are made by small nearly unknown companies seem less likely to have such junk in them. That is the type of hardware source I will continue to look for.
If I find something interesting, I will post it here. If anyone is able to correctly interpret what I am looking for and knows of a good lead, please post it here as well. Thank you.
"When something changes the functioning of the hardware and users don't know about it", well, there is nothing that can be done against it: we all depend on mass-produced software and, even if documentation is provided, malware would not be described in it.
Now, if we do know about malware implemented in the hardware (such as vPro's treacherous computing), we can at least avoid the devices that are known to implement it... as long as alternate devices exist! Today, ThinkPenguin does not propose CPUs with vPro. However, Intel may soon stop producing CPUs without vPro. What is the solution then? I do not think that, today, the solution can be technical: we will have to convince the manufacturers to respect our freedoms.
"we will have to convince the manufacturers to respect our freedoms."
How should we ever know if they really do?
I can't think of a method which can recognize the existence of circuits with malicious functionality.
Maybe the next step of "vpro" will be implemented silently without anyone knowing - or is already?
Indeed. That is what I wrote:
"When something changes the functioning of the hardware and users don't know about it", well, there is nothing that can be done against it: we all depend on mass-produced software and, even if documentation is provided, malware would not be described in it.
By only accepting free software, we can know of the driver/firmware running on our machines. If malware is implemented in the hardware (or in non-flashable ROM), there is, today, no remedy. We are back to what rms says.
A few new things...
1.) ARM processors also have built-in system restriction features available (DRM).
http://www.arm.com/products/processors/technologies/trustzone.php
2.) Since it apparently is ethically ok to use hardware as is, (Is there such a thing as unethical hardware?) if ARM TrustZone is just another processor mode, can such hardware be used along with free software code to ensure the hardware DRM features for instance are turned off?
Can interacting with these features be used to keep things free rather than locked down?
3.) I suppose there could be (are?) backdoors or secret keys in these trusted features?
http://comments.gmane.org/gmane.comp.hardware.netbook.arm/7461
http://http://trisquel.info/en/forum/secret-3g-intel-chip-gives-snoops-backdoor-pc-access
4.) Also found this project to build a laptop.
http://rhombus-tech.net/community_ideas/laptop/
5.) Still looking into Loongson options.
Interesting way to build your own custom boards quickly and inexpensively, exactly as you want them. Set up fee is waived for now.
http://linuxgizmos.com/crowdfunding-a-custom-linux-sbc-the-easy-way/
I am not sure if there are enough (freedom respecting) options there yet. If anyone notices anything positive/negative, please post and let us all know. I don't formally know how to design a board, but was able to easily place objects etc. https://geppetto.gumstix.com/
Here are many boards http://linuxgizmos.com/category/boards/ and devices http://linuxgizmos.com/category/devices/ to check out. I've only spent about an hour so far on this site but I will be thoroughly going through these leads.
There sure are a lot of existing products here: http://www.aliexpress.com/category/70803003/mini-pcs.html I am checking out these devices looking for the right mix of freedom respecting components.
If anyone has an interest in browsing, reading, picking, noticing, whatever, please update us with your findings.
Every element of the Novena laptop will be open source.
Idea originated here in December 2012: http://www.bunniestudios.com/blog/?p=2686
Novena Info: http://www.kosagi.com/w/index.php?title=Novena_Main_Page
ZDnet Article: http://www.zdnet.com/building-the-open-source-laptop-how-one-engineer-turned-the-geek-fantasy-to-reality-7000018987/
Interesting quote: "...what Huang would really like to achieve is to demonstrate you don't need a multi-million R&D budget to compete in the hardware space.
"One of the things I've been trying to do for the past few years is change the way that people think about hardware and [to get over] that it's not just this impenetrable thing that only big companies can do," he said."
--
Note that a blob-free driver for the Vivante GPU is available and improving:
Wladimir J. van der Laan
https://github.com/laanwj/etna_viv
https://github.com/laanwj/mesa
Looks to be 100% free.
EDIT: Started a new thread specific to this post. https://trisquel.info/en/forum/novena-laptop-will-be-open-source Will keep posting general hardware finds here in this current thread.
- Inicie sesión o regístrese para enviar comentarios