Really cool modular and freedom-respecting computer


If it is a technical issue, prove it.
I did. I don't understand what you aren't getting here.
AllWinner violated the GPL. They have, as yet, not done anything to fix that. This "sin" is still on them (the "sin" not literally being against divinity but the philosophy of the FOSS movement). Therefore, technically speaking, everything they produce is still tainted with that sin.

Yes, it does.
My mistake. Let me ask though, has IBM ever violated the GPL for which they haven't been "punished" or otherwise made it back somehow? If not then it's a completely different scenario. And if they have, then that just further proves the point about how important it is to act otherwise, 'cause I certainly didn't know, and odds are a lot of people that actually care might not as well.

I feel like you're making stronger causal links than are necessarily there.
Again, I said it was an extreme example, almost hyperbolic.
I mean, in this example why would you think refusing the money from the assassin would cause them to stop assassinating people?
I wouldn't think that. If I refuse the money but the assassin goes off to continue killing people then it has nothing to do with me. That's not the point. The point here is that the money was obtained by dubious means and I, as a charity, need to decide whether I'm ok with that or not. I can't pretend that the money is entirely clean. The fact that the assassin will just donate the money elsewhere and refusal won't change anything could sway that decision, but it's still tainted money.

You think they'd have made less money had they not violated the GPL?
Well yes, like you say, some marginal amount. I see what you're getting at though, I'm likely over emphasizing the impact it had. I thought it'd be a short-cut explanation that would make the most sense.
Subtracting out the money aspect, however, it still stands: AllWinner violated the GPL and it would be imprudent to act as if anything they produce is entirely guiltless in that.
 
You think they'd have made less money had they not violated the GPL? I doubt it. Maybe there was a marginal amount of money saved by not going through the effort to keep up with the GPL, but I suspect this was all more a matter of social conditioning against wanting to release source even when they're legally required to and plain laziness. Maybe they think that they'd lose money by complying but it's probably not very rational.

I guess you don't have any proof and I don't have any either, so we don't know for sure.
But I think complying with the GPL is easy enough that if someone doesn't, specially after being told in public (and possibly in private) then they must have a reason.
There's some pattern to it, from what I've read, and it's not been a single mistake.

And I don't care if it is social conditioning. That would be an excuse for all sort of abuses. I don't have to tolerate them. If you're in a culture that thinks copyright is nuts, then
fine for you but if you assert your copyright, keep your source secret, and just disregard other people's copyright, then your social conditioning is not something I want to respect.

It may be very rational to use anything you find, don't give back anything, and just disregard any law or claim by others until a judge forces you otherwise.
It's not what I like but it is rational. It is a logical plan to make money. A despisable one.

Not to say that there aren't cases where money can be made by violating the GPL, I just don't think that applies here. They're a hardware company, their software is in support of their hardware and if anything keeping the software open would make it more likely that others would increase accessibility to their hardware and therefore increase sales.

Being a hardware company is no excuse. Hardware is useless without software, and I'd even dare say that building the software is more work than building the hardware, except
that software can be reused from someone else (legally if you keep your licensing right).
Functionality is in software, hardware is always (as should be) as generic as possible. And circuit designs (IPs) are bought from ARM or whoever else.
And devices live so short than chip vendors need to provide most software already made so that OEMs can rush to prepare the software package.
The software you ship with your hardware is key to success. If you respect all copyrights you'll have many more costs than if you just use whatever you can grab.
If you have to program a video decoder so that your VPU(or the VPU you built with the VHDL from someone else) can play the videos people want to watch in whatever
format is popular at the time, you're going to have more costs than if you download a copyleft decoder, buy a license with source for another decoder, link them and adapt
them to your VPU and ship the compiled library without source because the copyleft authors can't afford to sue you and won't probably stop providing you with more
decoders next year because they publish the sources for the world to see.

Yes. Being a hardware company, publishing sources shouldn't have been a concern, in fact it would make sense even without a legal obligation.
If there was the legal obligation and they didn't there must be a reason. It could be because they had lost them,but that would be a little extreme.
Or it could be because they have to comply with incompatible licenses (they linked copyleft and proprietary software, for example) and they chose to comply with the non-GPL license.
Or maybe there's a epidemic of brain damage in that company that just affects the copyright decision making neurons, but I'm not betting on it.

Those who try to ship software with their hardware that complies with all due licenses are in a competitive disadvantage to those that simply don't care.
And since law does not seem easy to apply, the only recourse is customer rejection. I'm not holding my breath, but htere's little else left.

But then again, I guess this is why people don't really try to actually apply purely utilitarian standards to their decisions because there's too much they just don't ever know.
I'm not sure I understand you.
To find out and get a fair judgement you need a fair trial with proper procedure, impartial judges, and so on and so forth.
Since one can't ususally afford so much for any single decision one makes, we make do with suspicions, reasonable doubt and guesswork
and leave the courts for things that can result in fines or prison, not just lost sales. So I don't know how much of what is said is true or why,
but since I can decide what to buy on any whim, I may choose not to buy from them just from hearsay accusations and unconvincing responses from the accused.
It's always better that just don't care.
 
I guess you don't have any proof and I don't have any either, so we don't know for sure.
But I think complying with the GPL is easy enough that if someone doesn't, specially after being told in public (and possibly in private) then they must have a reason.
There's some pattern to it, from what I've read, and it's not been a single mistake.

And I don't care if it is social conditioning. That would be an excuse for all sort of abuses. I don't have to tolerate them. If you're in a culture that thinks copyright is nuts, then
fine for you but if you assert your copyright, keep your source secret, and just disregard other people's copyright, then your social conditioning is not something I want to respect.

It may be very rational to use anything you find, don't give back anything, and just disregard any law or claim by others until a judge forces you otherwise.
It's not what I like but it is rational. It is a logical plan to make money. A despisable one.

I'm not in any way trying to excuse or justify GPL violations. I'm simply saying that there isn't much of a convincing argument that their violations earned them money.

I don't take a position against proprietary software (obviously) but I would never support or advocate software license violations. Even outside of legal obligations it's extremely disrespectful and exploitative.

I'm not sure I understand you.
To find out and get a fair judgement you need a fair trial with proper procedure, impartial judges, and so on and so forth.
Since one can't ususally afford so much for any single decision one makes, we make do with suspicions, reasonable doubt and guesswork
and leave the courts for things that can result in fines or prison, not just lost sales. So I don't know how much of what is said is true or why,
but since I can decide what to buy on any whim, I may choose not to buy from them just from hearsay accusations and unconvincing responses from the accused.
It's always better that just don't care.

I'm saying that we take broad ethical stances exactly because we can't determine the best outcomes on individual examinations. There's nothing wrong with that. It's sensible to boycott this company because of their software misconduct. But making objective arguments trying to follow the exact cause and effect of financial transactions between people and company gets kind of hairy. You have to make judgement calls on this.
 
998.jpg
 
And just as fun. :p

Nah! THIS IS PRETTY MUCH LIKE THAT PICTURE. Lowercase long text is just business as usual, everyone knows how to skip them, just as everyone knows the loudspeaker is in your CapsLock key.
Yes, kinda fun. :)
 
I meant the rest of the movie, too, not just that scene, however being noisy is quite fun.

I think I missed the use of caps lock, if there was any, or I may be numb to it.
 
I meant the rest of the movie, too, not just that scene, however being noisy is quite fun.
Ah, the whole movie. Ok, I didn't get you, sorry.

I think I missed the use of caps lock, if there was any, or I may be numb to it.
There hasn't been any use, I think. You're not numb. I thought (because of the loudspeakers) you were implying there had been loud talk, maybe because of length of text or some other customary recourse that escaped me.
I just replied meaning that when I want to be loud I use CapsLock so I didn't mean to be loud here.
So you just meant there was a discussion. Sorry if i've been thick.
 
Klumpen originally posted the pic, and I don't know exactly what he meant with it, but it is very fitting with everyone loudly proclaiming whatever and not listening to anyone else. Don't apologize. I did mean that scene is fun (I guess, it has been so long since I watched those that I don't really remember it), just like "arguing" on the internet (I try to avoid taking it seriously), and also the whole movie is fun, as its watching others argue on the internet, especially if they start to get a bit too serious. Sometimes I learn things on the internet, and that is also fun (like reading a book, taking a class, or watching a documentary film).

I think I just figured out why "reality" tv is so popular: It is a lot like internet discussions, plus you get to see the people get pissed off.
 
I did. I don't understand what you aren't getting here.
AllWinner violated the GPL. They have, as yet, not done anything to fix that. This "sin" is still on them (the "sin" not literally being against divinity but the philosophy of the FOSS movement). Therefore, technically speaking, everything they produce is still tainted with that sin.
This is a philosophical objection you have to Allwinner as a company. Please don't pretend that its products are technically tainted because of this objection you have. I have already argued against the philosophical objection to supporting companies that do bad things by pointing out that we don't really have a whole lot of options, and all companies producing low-level computer components are doing very nasty things, usually much worse than GPL violations.

My mistake. Let me ask though, has IBM ever violated the GPL for which they haven't been "punished" or otherwise made it back somehow? If not then it's a completely different scenario.
Yes, it's a completely different scenario. It's worse. Do you honestly think that we in the libre software community are more concerned about GPL violations than maliciously designed hardware and DRM?

Realistically, we in the libre software community don't frown upon license violations in any strict sense. We frown upon proprietary software. The GPL and other copyleft licenses are just a tool to make certain proprietary software development illegal, so we can suppress it and make libre software comparatively stronger. GPL license violations that don't lead to proprietary software are only frowned upon because they're illegal and that illegal behavior could affect other users.
 
This is a philosophical objection you have to Allwinner as a company.
Yes! It is! Because FOSS is a philosophy! All arguments about this are by necessity philosophical in nature.

Do you honestly think that we in the libre software community are more concerned about GPL violations than maliciously designed hardware and DRM?
No. I have said, several times in fact, that most people don't care, shouldn't care, and that there is absolutely nothing wrong with it. I just hate the idea of pretending like it doesn't have any relevance at all. They violated the GPL, and this device, RYF or not, is a product of, no matter how small a sliver it may be, that violation, and you need to make peace with that.

It's worse.
It is not worse because at no point in time have I or anyone else made any argument about anything they have said. It has absolutely nothing to do with anything being discussed here. The entire argument revolves around the GPL violation. If IBM did not violate the GPL then it is entirely irrelevant to the argument.

Please don't pretend that its products are technically tainted because of this objection you have
Define "technically" for me, please? Because you are obviously not using it the way I am.
When I say they are technically tainted, I literally mean "in the strictest sense, using the facts we have, in the most pedantic way" the current product is tainted with the proverbial sins of it's forebears.
NOT anything to do with the technology.

I have already argued against the philosophical objection to supporting companies that do bad things by pointing out that we don't really have a whole lot of options
I. DON'T. CARE. if there aren't a whole lot of options. I have never, at any time, under any circumstance, advocated not supporting this product. I have said several times over there is no reason to not support this, that it is fine to accept previous GPL transgressions in order to get this if that's what you want.
The one and only thing I am against, the only thing I have been arguing against, is to not pretend like this device, regardless of it's final RYF certification, does not contain at least some taint of those transgressions.
 
I. DON'T. CARE. if there aren't a whole lot of options. I have never, at any time, under any circumstance, advocated not supporting this product. I have said several times over there is no reason to not support this, that it is fine to accept previous GPL transgressions in order to get this if that's what you want.
The one and only thing I am against, the only thing I have been arguing against, is to not pretend like this device, regardless of it's final RYF certification, does not contain at least some taint of those transgressions.
What exactly are you advocating for? Who is pretending that the computer card does not have an Allwinner SoC? Who is pretending that Allwinner has not been guilty of GPL violations?

All you seem to have is some tenuous claim of "taint" because of Allwinner's unrelated "sin", and a dubious claim that Allwinner would somehow be incapable of making SoCs if it hadn't violated the GPL.
 
Not violating the GPL would have been very easy:
Either release the corresponding software under the terms of the GPL, or, if other obligations prevent you from doing that, don't release anything at all.

It's not like someone forces them to do business.

No that would be impossible, because its part of an alternate reality you just happened to conjure up.
Things are what they are, and that is what's on offer.

Historically speaking computers have been, worse, and are, in the case of intel and AMDs recent offerings, also worse. That is the origin of Free Software, once you start hiding schematics and not disclosing source, you develop a need for a position on the matter, because its problematic.

Please do tell us if the computer you are posting from is any better.

GPL-violations are the source of some major wins for the free software community, and the GPLv3 license.

You can take Allwinner to court over matters Allwinner it if its your copyright.
It would have been possible, knowing that, to pick a SoC from Freescale, for example, but they didn't.
Your options there are to buy it or don't.

Lambasting this project for the faults of Allwinner doesn't fly, they are trying to do something different with it, showing promise where few others are even communicating anything in this regard.
If thats not GNU mentality, I don't know what is.

Two good hacks, one product.

With the modularity you can make your own insert-card, with the A20 one, it is now possible to get a cool freedom-respecting computer for little money.
EDIT:
Except that the whole point of the RYF certification is to promote The FSF's ideological purism. If it were actually about freedom it would be a lot less restrictive.
-Neelix
This is what a implied non sequitur circular argument looks like. You can't guarantee user freedom without restricting the supplier. If you mean to bring up software licenses, the less restrictive ones have historically proved feign in this regard. And no, this is a hardware license. One that doesn't care what kind of free software it runs, copyleft (which the FSF's GPL licenses are) or otherwise.
Sure... That's how the FSF presents it but the devil is in the details. The details of the certification make it impractical for any company which isn't itself specifically focused on maintaining the FSF's ideals to apply it to their products.
And that is what a certification is. You get to like it, or at your leisure, not.
You can make your own certification, or bring up an argument about the contents of the RYF-certification. This is at best a circumstantial vagueness on your part.
If your point is to be more pure than the RYF license, by all means be, complaining on the part of others, people that want it, isnt it.
In any case what the RYF certification is about isn't really relevant to the point.
Then you have no point. If you bring up an argument as to the merits of RYF certification, on the merits of RYF certification, the source of your argument should be the RYF certification.
The thing is that the only people to whom the RYF certification would have any relevance at all are those who care about the FSF's ideological position.
When you bring up FSF ideological position to disagree with, your argument should entail something where the FSF ideological position differs with yours.
The FSF position in this matter is the RYF certification, which is awarded to products that are far worse, case in point, the thinkpad x200 out of the box. Nobody said it was perfect, but if you want to argue that the "the whole point of the RYF certification is to promote The FSF's ideological purism" you should check that the certification isn't actually more pragmatic than the issue you bring up.
Unless you can find a reason why this wouldn't be certified, you only produced irony.

To then argue that those same people shouldn't or wouldn't care about GPL violations by a company so intimately involved in such an important aspect of the device's construction - violations which may well be present in the default OS of the devices themselves if they use a kernel based on Allwinner's kernel - seems both bizarre and hypocritical.
You never proved that "the only people to whom the RYF certification would have any relevance at all are those who care about the FSF's ideological position." so you can't build off it.
It is entirely possible to want a non-spying computer and not give a toss about FSFs ideological position. Like national security agencies using typewriters, for their business of spying on people. Whether this qualifies for their criteria, is for them to figure out.

You can buy only the Dock, and the current insert-card is proof that it works.
Also, there arent any other current RYF machines you can buy new. So in that regard this is the only, while not perfect, option.
-Neelix[/QUOTE]
Except that we aren't talking about unrelated misconduct specific to other products by the same company. The GPL violations are in the software that's meant to drive the hardware being built into the very core of the products in question. That makes it relevant.
See above argument. Also, yes, you were able to point that out, good.
On similar notes, GPL violations are also good, because that gives possible recourse for rectifying them. It is far superior to just closed source, which is legal.
 
Last edited:
Who is pretending that the computer card does not have an Allwinner SoC?
No one.
Who is pretending that Allwinner has not been guilty of GPL violations?
No one.
I have said absolutely nothing about either of these things.
The only thing I've argued against is to not pretend that just because this gets RYF certification it is still from a company that has willfully violated the GPL and that it is therefore tainted with that violation, on a philosophical level. I'll even explicitly say "philosophical" there just so there's no further confusion.

tenuous claim of "taint" because of Allwinner's unrelated "sin"
It's not tenuous, it is fact. They violated the GPL, and not in an "oops, let's fix that" kind of way, a willful violation in direct opposition to not just the letter but also the spirit of the GPL. Technically speaking (the technically that means "strictly", "pedantically", "the most exact meaning of"), everything they do must necessarily carry at least some of that burden. I don't know what you want here. I've given a literal description, I've given an extreme analogy, and you dance around both without actually arguing against them. It's like you're not even trying to understand, like you're arguing something completely different to what's being said.

dubious claim that Allwinner would somehow be incapable of making SoCs if it hadn't violated the GPL.
I never claimed that either: I was pushing in that direction because, as I told Exophase, I thought it would be a shortcut to understanding, but I was obviously wrong, and I backed off that tactic. It doesn't change the underlying argument.
 
@comradekingu:
I don't get your point. You said that, let me quote:
Allwinner couldnt avoid voilating the GPL to make their product,
I say in principle they could have - simply by releasing the software under a GPL-compatible license.

Maybe there are additional NDAs or other contracts involved, that I'm not aware of, that have prevented Allwinner from doing so. But in this case they would still have had the option to not release any software or products at all.

So in the end it was their free and conscious choice to violate the GPL. Whether we find that right or wrong doesn't matter. It's a fact.
That fact stands on its own. Whether other hardware is even worse or other license violations have ultimately helped the FLOSS ecosystem is irrelevant, as it doesn't change the facts in the Allwinner case.
 
Sorry for the ambiguity.
I said that as a historical accuracy. They were given their options, and couldn't avoid abusing GPL. They weren't unable, if you will.
What i'm saying is that the Allwinner car crashed firmly in the GPL lamppost, was caught on GPL-camera, and then tried to roll along as if nothing happened.
They would have been able to come up with their own solutions, and legally close them.

Having used GPL software, its easier to reverse engineer the changes around it, which I assume the volunteer effort has utilized in great amount.

If you want a perfect solution, you have to construct the universe, for everyone else, there is reactionism. We would all like Allwinner to comply and help efforts themselves.
 
Having used GPL software, its easier to reverse engineer the changes around it, which I assume the volunteer effort has utilized in great amount.

I don't understand. I thought the GPL violations was like this:
there's a GPL code A. Its source is public.
Allwinner takes A, alters it or combines it with other code from themseleves or others, resulting in code B.
Allwinner publishes binaries for B to go along their hadrware so that their hardware is useful.
Allwinner profits.
Allwinner refuses to publish B source when asked.

Now. Do you mean than B would be easier to reverse engineer because it uses A than a functionally equivalent hipothetical C code written from scratch by allwinner?

But what would you get from that reverse engineering? the A code that you already had ? B-A sources or hindsight ? I don't think getting B-A sources is any easier or harder than geting C sources.

The point is that the GPL is a foundation for a collaborative development model. If companies set the precedent that you don't have to follow it, then all efforts put on GPL code
hoping to get something in return if someone published derivative works are at risk.

For me proprietary software is bad. But if someone publishes it, I prefer it to have written it from scratch than to abuse GPL code. Because abusing GPL code just makes it
more likely to have proprietary code, and removes incentives to publish GPL code. Some people publish GPL code because they believe in it, some publish GPL code
because it is the legal way to reduce costs by reusing other GPL code. I welcome both. I don't welcome anyone that reuses GPL code and publishes the result as proprietary software.

The point of copyleft is that nobody should be made to jump loops to reverse engineer software, they should have the source. Since under current law you can't force anyone
to do so, you work constructively under current law to build a corpus of software that can only be reused by those that publish their source of derived works (roughly). So you don't force
them to publish free software, but you refuse to help them publish proprietary software. GPL violations break this.

So for me free software is better than proprietary software, legal proprietary software is better than GPL violating proprietary software, and GPL violating software is better than... err...
murder, many other nasty things...

I'm not at all into reverse engineering, but I don't understand how deriving a work from GPLd software makes it easier to reverse engineer the resulting binaries.
Please give me some links or hints. At most it would be best in the hopeful but incertain case than eventually someone forces the violator to publish full sources for the derived work.
Or in case a judge awards damages that are reinvested in free software, or some such.
But the effort in prosecuting it could be spend in more free software if there hadn't been the violation.
Is is the free software community who has helped Allwinner make money and got nothing back, it is not Allwinner who has helped the free software community get any more free software.

[note: arguments here are approximations: for example I know the GPL does not force anyone to publish source if they don't publish binaries or to give source to anyone beyond those who
you give binaries, etc. I appoximate for brevity because I don't think it changes the argument]
 
I like the concept but don't like the hardware.

The processor is slow and the peripherals are a joke.
The laptop case has only one Micro SD-Slot and 3xUSB 2.0 (2 are internal).

If they would have used better hardware it might be a lot more interesting for me.

hiya askarus, thanks for replying - i answer about the peripherals in some detail in http://rhombus-tech.net/crowdsupply/#questions, "I'm missing Ethernet and SATA on EOMA68, what about power consumption?". one correction: there's 2 Micro-SD card slots (one on the computer card, one on each of the laptop and the micro-desktop housings).

basically it's about what one person (me) can design and make on a *reasonable* budget (around $30k over 18 months) that can *reasonably* be claimed to be Eco-Conscious as a result. you can in no way claim that a design is eco-conscious if you spent $0.25m or above on it, can you?

by sticking to a 15 Watt budget i was able to use a bq24193 *single-cell* charger IC as opposed to having a much more complex hardware casework design housing a dual, triple or quadruple-cell battery, as well as having fans, metal and heatsinks to cope with a 20 to 25 Watt budget, which creates a mass compounding "cascade" effect. all that's gone with a simple single-cell charger IC.

if the laptop were to have even *one* extra USB port, it would need to be added by way of an extra 4-port USB Hub (more components), but worse than that it would take the laptop over the 15 Watt budget (2.5W per USB port). that would mean going into dual-cell territory as well as heat-sinks and fans, now you're into mechanical failure territory as well as cleaning issues down the line.

... start to make sense?

regarding the processor: it's efficient! for what it is it's damn good, and it can definitely be said to be "Good Enough" as in the whole "Good Enough Computing" thing - more on this in the white paper on ecocomputing that i wrote, http://rhombus-tech.net/whitepapers/ecocomputing_07sep2015/

but it's missing the point of upgradeability to say that the processor's slow: you can *ALWAYS UPGRADE LATER*. that's the whole point of the exercise. pop out the old Computer Card, put in a new one - $50 to upgrade instead of $300+ and throwing the old laptop into landfill.

also, if you know of any faster processors that are available, that you would like to see put into EOMA68 form-factor, PLEASE TELL ME! i am looking for more to put on the roadmap because i am committed to this project for the next 10 years. but, let me give you a run-down of some of the ones that i've evaluated already, here we go:
  • Mediatek SoCs: GPL-violating. Require NDAs. Fairphones got into seriously embarrassing trouble over selling a product that they couldn't provide people security updates over. what's the point of having an eco-conscious product if it has to go into landfill because it's dangerous for its end-users security and privacy??
  • Freescale's iMX6: too costly ($36 for the quad-core!!) and it's an older 40nm design using ARM Cortex A9s that runs at around 4.5 watts (excluding PMIC and memory which are about another 1.5 watts), plus start-up requires something mad like 4 amps (!) EOMA68 has a hard limit of 5 watts, so as to not need fans or special heat-sinks.
  • Texas Instruments OMAP4 series: Cartelled. Requires NDAs. If you're ordering a million and above you can gain access to it. Below that you have to be "evaluated" or referred to companies that will charge $0.25m and above for "design" work.
  • Texas Instruments AM series (that's in the various Beagleboards): not good enough. 1ghz Cortex A8 is lower performance than the A20. They're also massively expensive for what they offer. And they use PowerVR graphics which is just... don't do it.
  • RockChip: GPL-violating. Require GPL-violating NDAs. Require $100k "BSP" fees. There is however one person who managed to get round these problems by paying *only* $5k - Tom Cubie. Tom's team designed the Radxa Rock and it has full PCB and CAD files available for the Quad-Core RK3188. It doesn't have HDMI output, but it is at least a quad-core 32-bit SoC and i will be looking at this one, next.
  • AMLogic: GPL-violating DESPITE BEING A USA-BASED COMPANY NOW. AMLogic cost me a large customer by embarrassing me due to their blatant GPL violations. I will never deal with them ever again.
  • Qualcomm: HAHAhahahahaha https://news.slashdot.org/story/16/...o-guide-to-crack-android-full-disk-encryption AAAAhahahahaha https://hardware.slashdot.org/story/15/12/08/214239/qualcomm-faces-antitrust-charges-in-europe
  • Actions Semi: can't get in touch with them (no response)
  • Intel: Require NDAs. No BSPs. And, to cap it all, they just shut down their Smartphone and Tablet division, terminating the production of things like the Z37xx series which were used in those compute-stick PCs (that were 10 to 15 watt anyway!!!). Basically Intel's had it, but it will take a while for them to notice because they have such a large amount of cash reserves. also their SoCs are $21 which goes a long way to explaining why they shut down the division.
  • RISC-V (aka LowRISC): nice idea, but the devil is in the *interfaces* not the *cpu type*. they don't really know how to make a good SoC
  • Broadcom: hypocritical beyond belief. Telling kids that they can educate themselves so far but NO FURTHER because you may run ARM code no problem but if you want to view MP4-encoded media you have to PAY the "REASONABLE" fee of $2.50, what kind of message does that send to kids? broadcom is a greedy company that's only interested in profit.
  • NVidia: the Tegra series is around 10-15 watts peak. The GPU requires something like EIGHT AMPS. there's no way that's going into a 5 watt (absolute max) computer card. Plus they won't let individuals at the BSPs (board support packages).
... did i miss anybody out? :) serious question although it's asked tongue-in-cheek - i really *do* need to know if, amongst this dog's-dinner catalogue of endemic industry-wide failures to Get With The Programme, there is even a glimmer of hope!

but does this give you some idea as to why i'm using the A20 for the first Computer Card? it's $7. not $21: $7. not $36: $7. we've bludgeoned Allwinner to give us the source code. it's well-supported now. it works, and it's extremely efficient. Even FreeBSD supports it for goodness sake! https://wiki.freebsd.org/FreeBSD/arm/Allwinner. Fedora supports it, Centos 7 supports it, Debian supports it, ArchLinux supports it, Parabola GNU-Linux/Libre supports it - even L4Linux and L4Android run on it - that's a Micro-kernel based real-time OS which most people have never heard of - what more do you want?? :)


anyway - that should give you some idea of what's been involved. i *really have* thought about this a lot - it's been five years after all - i've very carefully evaluated both the standard and the available SoCs, and threaded a path that nobody else has noticed, to bring you a pretty awesome concept, making it easier for people to make their own computers, repair them, upgrade them and much more. any questions let me know. and please, i really meant it about the SoCs - i *need* more to evaluate!
[doublepost=1467479955,1467478331][/doublepost]
@onpon4:
Since I agree with WizardStan in all points I will spare you a lengthy answer.
I'm not saying, you shouldn't buy the product. I'm just saying, if you care for FLOSS you should be aware that it contains components that were created by violating the very essence of what FLOSS is about. And a RYF certificate would harm that awareness imo.

sulu: thank you for pointing this out. in my earlier reply (i'm just catching up here) i point out that there really isn't a lot of choice. i'm in touch with Allwinner, i have been for five years, and i can tell you that they're KEENLY aware of the harm to their reputation. but we have to be realistic. think about this: the computer you're using, is it entirely spyware-free? it's an intel or an AMD post 2009 or post 2013 respectively, right? so it's got that "remote management" feature, right? so should you boycott intel because of that? and is it okay to criticise one company writing the message from another company's privacy-violating device?

my point is: we have to be realistic and make use of what we've got as a stepping-stone to work TOWARDS what we want. my goal is to get to the point where i see a mass-volume FSF-Endorseable processor being brought into existence. i don't mind how that's done, but if nobody else tackles it I CERTAINLY WILL. i'm taking steps on that path already, starting with this project and pursuing other avenues at the same time. one of those efforts was several years ago, with this: https://tech.slashdot.org/story/12/12/04/1748232/toward-an-fsf-endorsable-embedded-processor

when the A10 came out (and was then upgraded to the A20) i was the person who, behind the scenes, worked indirectly to get the source code out of allwinner. since then, the A20 has become one of the most widely-supported SoCs from a china-based company ever. even freebsd, l4linux and l4android can run on it. very few people have even *heard* of the l4ka microkernel project!

so, at least there aren't any GPL-violating blobs *now* in the A20, and we have the opportunity to talk to them - with some pointed words - about how to proceed. the VP is tearing his hair out, btw. he knows that his neck is on the line, legally and ethically. But he's dealing with a situation where there are multiple investors who have carved out their own "enclaves", they're competing *INTERNALLY* and he needs some serious help to get them in order. so i will be going over to help advise them. but i CAN'T DO THAT UNLESS THIS PROJECT IS FUNDED.

... start to see what's going on?

would you rather make a stand *against* something (by boycotting them for example) or would you prefer to show them that there's a better way?

example: pine64 should never have allowed Allwinner to get away with saying that the A64 boot loader is "proprietary" and has to be kept "secret". they should have said, "well if you're going to commit GPL violations and expect us to sell copyright-illegal products that on import to the USA could risk being impounded at customs and destroyed without recourse to insurance because it's a criminal act, we're going to refund everybody's money". THAT would have got Allwinner's attention. but they didn't do that, did they?

any questions please ask. this is a *genuinely* libre project for the hardware, software and discussion. there's nothing to hide, and nothing to be gained by secrecy.
[doublepost=1467480820][/doublepost]
Except that the whole point of the RYF certification is to promote The FSF's ideological purism.

-Neelix

NO.

i didn't really get it - i knew intuitively that there was a point but couldn't explain it well - until chris encouraged me to apply for RYF Certification on the Libre Tea Computer Card. chris' business model is basically "plug it it, it'll work". if you've ever upgraded a computer that has a non-free proprietary firmware blob for the WIFI, you'll be familiar with the concept of being DISCONNECTED in the middle of the upgrade because the version of the OS is now incompatible with the proprietary firmware, and now you're screwed. *this* is what chris makes money doing: he researches the hardware so that YOU DON'T HAVE TO (and is actively involved in getting firmware released including full sources from atheros for USB-WIFI 802.11n chipsets for example - see https://www.fsf.org/news/ryf-certification-thinkpenguin-usb-with-atheros-chip)

the business model is therefore based around hassle-free products that JUST WORK. he doesn't *get* all the "my computer stopped working because i upgraded it" problems, and has many loyal and grateful customers as a result who are more than happy to pay a bit more money because it's not worth their time to use the cheaper stuff that everyone else buys (and then throws away in landfill when it doesn't work).

the point is: it's ***NOT*** about quotes ideology, it *REALLY IS* about real-world practical products.

when we applied for RYF Certification, Josh Gay explained to us that the problem with e.g. Debian is that the average end-user could "blindly wander" into this non-free "pain" world where hardware could fail (or spy on you) *WITHOUT WARNING THEM*. in other words, right there in synaptics package manager, there's a check-box "non-free". Debian doesn't give you *any* warning about the consequences of simply naively clicking that checkbox.

*this* is what the work of the FSF is about. it's *not* quotes ideology quotes religion quotes any-other-bullshit-you-want-to-make-up quotes

does that make sense?

so the eco-laptop is based around that same principle. EOMA68's tag-line is even "Just Plug It In: It Will Work". there's a whole stack of other benefits as well.

--
http://crowdsupply.com/eoma68/micro-desktop
 
It would be great if the Pyra and EOMA68 communities could work together on software and such. It would also be useful to be able to pop the card out of one device and into another depending on where you are and what you are doing. I know it is a bit late for both projects to change directions to accommodate sharing the same card, but the software side might get a boost if they are both using the same SoC, so I recommend considering the OMAP 5.
 
Back
Top