Is this Librem13 fully free this time?

113 respostas [Última entrada]
Chris

I am a member!

Desconectado
Joined: 04/23/2011

OK-

Sure- we can design and manufacture *something* today or lets say tomorrow, because today it is a bit difficult. It's only being designed today. It still has to get the support, the money, the manufacturing, etc. But even once you've got that your still missing a graphics chip, a wifi chip, etc.

We might get lucky with next generation graphics on ARM. Which might leave us with just wifi to fix, but that is still a huge problem. Reverse engineering isn't cheap. Do you have $100,000 to fund one person to work on it for a year? It's a skill set that pays very well. You might think people should work for nothing, but reality tends to get in the way if you want stuff to get done. Most free software is paid for by somebody whose got an interest in its existence and in some way is making money off it. Whose going to make the money off reverse engineering a chip when the people here have evidenced the fact they're not willing pay a premium for something they can get elsewhere for less? Even where you have someone whose young and not tied down (ie children, wife, job, etc), willing to work for nothing, etc someone is still going to have to pay the persons rent, basic essentials, etc. That's $24,000 / yr. It is happening. And this isn't to suggest there aren't people like that. There are. But even if we have $24,000 / yr to cover one person you need multiple people to work on the different issues.

It's a hard problem to solve. We're funding some of this now, but it's not necessarily a money maker. It's not likely to result in a significant return on investment relative to other projects one could invest in. That's what makes it difficult. Even projects that you think were successful (fund-raisers) don't necessarily have significant amount of money to really fund this development. They've raised 500,000, but that has to go into manufacturing the goods. Which leaves little or nothing for actual design, reverse engineering, etc. And even if it did you still need somebody with those skill sets, and there isn't a guarantee of the results.

Look for example at X86 systems. We might have the specs (we don't, but we'll just pretend that we do), we might have everything (by that I mean source code/specs/etc), no reverse engineering necessary, but still can't free these systems (because of digital signatures and digital restrictions). So... yea. It's not easy to solve even if it seems like we're making progress. We are and we aren't. We go a few steps forward and we take a few steps back.

In some respects things probably are better than they were. The X200 for example- but that isn't something thats going to see widespread adoption. It can't. There aren't enough units to make it happen- and even if you could- Francis has said it's a brick wall going forward. In other words its only going to get worse.

Chris

I am a member!

Desconectado
Joined: 04/23/2011

That's kinda the point I'm making. There are people who are *really* dedicated to making this stuff happen. Some people are working on it part time with 2nd "real jobs" and others have muster'd up the bare minimum of resources to survive, but as a result can work on these issues full time. We're funding a few of them because we want the same thing (ie freedom, not for them to barely survive, but hopefully they'll thrive in time).

I can think of 4-5 people who post here regularly or occasionally like this. There are some people who have also done well financially and are also contributing *a lot*. Some are working on stuff part time and some are working on stuff full time. Honestly I think you just have to open your eyes and look around. You can take some educated guesses on who I'm talking about. I'd name names, but don't want to embarrass anybody.

Chris

I am a member!

Desconectado
Joined: 04/23/2011

Actually- yes in reference to the Russian CPU question. It's something like $5,000 a CPU for several year old tech. Utterly unrealistic. Maybe future versions will be cheaper, and a real possibility, but not currently.

As far as the last question. No, not really, I'm not a violent person.

moxalt
Desconectado
Joined: 06/19/2015

> people (men)

*facepalm*

tonlee
Desconectado
Joined: 09/08/2014

I have accepted your arguments on why the x86 notebook cannot be made.
You then fall back to the arm option. I do not like the arm option because I do not believe its performance is high enough. That is why I ask about the risc v.
My questions reflects my technical knowledge.
Lets say you have a risc v cpu at x86 duo core performance level. And required risc v software. You cannot take a x86 mainboard from a notebook and connect a risc v to the mainboard?

If you have to make a risc v mainboard then how big a technical task would that be? Hundred thousand dollars, millions?

You put emphasis on non free wifi 802.11 devices. They are for sale, even if in limited numbers. When they have been made once, maybe they can be manufactured again?

In general, what importance does expiring patents and rights on hardware have?

Chris

I am a member!

Desconectado
Joined: 04/23/2011

I think your missing something big. It's less relevant what the architecture is. What is important is that we can build off it, it be free, modern, and we end up with a design that can be cost effective and cheaply adaptable if something better comes out (ie not ARM, but something else). However I think people are underestimating the power of ARM anyway. While any initial design would likely be considerably less powerful today than what we expect it's where the tech is going that matters. ARM is catching up as the majority of computing devices use it over Intel or AMD CPUs. Most devices manufactured today are not X86. They're cell phones, tablets, and similar devices that are built off ARM.

I think expired patents are not the issue. If you try to build off 20 year old expired tech we're talking about pentium era technology. We've already got more powerful freedom-friendly tech we can build off without going that route. The cost of building off anything that's not being manufactured in huge quantities (ie hundreds of thousands of units a year) just won't work. It would be too expensive to manufacture a mere 100,000 units on a per unit basis to compete with other tech on the market.

Christianity
Desconectado
Joined: 10/09/2012
tonlee
Desconectado
Joined: 09/08/2014

And appelbaum retweeted an endorsement of purism.

Mzee
Desconectado
Joined: 07/10/2013
Mzee
Desconectado
Joined: 07/10/2013

And "Librem 13 Free Software Laptop Nears Funding Goals, After Self-Funding": https://phoronix.com/scan.php?page=news_item&px=Libre-13-Self-Funding

tonlee
Desconectado
Joined: 09/08/2014

appelbaum retweeted librem crowd funding. It will show if thinkpenguin and libreboot are right. To my knowledge nowhere has appelbaum explained his involvement in librem. It will fall on him, if it shows that he attached himself to what is likely a scam and he should have known.
I want to thank thinkpenguin and libreboot for their explanations. For a non tech it is a difficult matter and to gather information is also difficult.

tonlee
Desconectado
Joined: 09/08/2014

Thinkpenguin says, arm is the most viable path. It is difficult to argue with thinkpenguin because he has dealt with the difficulties about free software on hardware. About arm I think he is wrong. I do not think that arm and x86 are different. Both will not release software to provide all free software hardware. Likely they will invent more features that will make it more difficult or impossible to make all free software hardware.

On 32c3 this video https://www.youtube.com/watch?v=H6bJ5b8Dgoc shows efforts to get x86 hardware that is secure for the user. If it can be made and verified, it is great. But what stops intel from making new hardware that would require new security software again?

That is why I think free software people should support riscv. Riscv is all free hardware if people want it. http://www.eetimes.com/document.asp?doc_id=1328620&page_number=1. Lowrisc may one day have a raspberry pi2 performance comparable board for sale. Can't people like thinkpenguin and other promote riscv? I do not know what skills lowrisc is in demand of. And if they are common among computer people. If they are can't an effort be made to provide software?

Mzee
Desconectado
Joined: 07/10/2013

Wow, RISC-V looks indeed like the architecture the free software community has been waiting for. It will take some years to finish this project but it will definitely be a game-changer. I think it would be better to create a separate thread about this topic as this is very important but a bit off-topic on this thread.