I am incredibly sad that TALOS has not gotten the required cash flow, short of a miracle in the next few days.
The coreboot project is pretty much dead in the water without it, the only real choices for further development are either super low power crappy ARM devices or always going to be expensive IBM/TYAN POWER servers, so what do we do?
I am wondering, how come they didn't bark up some government or corporate trees for TALOS funding? AFIAK there are various government agencies interested in secure hardware and assured computing; I have always wondered what the NSA uses for their own computing needs, maybe they paid intel for firmware source code and a system that doesn't need ME to run.
The way things are going: +10 years - Microsoft and Intel have announced the "PrivaSec" initiative, aimed at producing a secure vertically intergrated computing platform where firmware agents prohibit the execution of unapproved programs - protecting your data from unauthorized access.
+20 years - We're sorry, but the GlobeX Trade Agreement and the Secure Communities act of 2035 prohibit the viewing, copying or transmission of this file - Further violations may result in fine, arrest and or the revocation of your work permit and internet operators license.
Thoughts: It seems that so many linux people just don't really care about libre anything, considering that the average linux sysadmin makes over $100K per year the community could have easily funded the project.
These days there are a lot more people with skills, but without the computer enthusiast/hacker culture of the 90's, the kind of hypocritical people who use a macbook, facebook, etc but who chide me for saying that working for the government is not at all immoral.
If I wasn't unemployed I would happily pay $5K for a high performance libre computer, but not everyone is me.
People went nuts for the faux libre purism laptop but talos gets hardly any comparative publicity/hype - why? - "We'll get intel to open up ME one day, we promise!"
On 12/12/2016 03:27 AM, Taiidan@gmx.com wrote:
The coreboot project is pretty much dead in the water without it, the only real choices for further development are either super low power crappy ARM devices or always going to be expensive IBM/TYAN POWER servers, so what do we do?
Maybe we should ask ourselfs do we really need high performance computers?
I am wondering, how come they didn't bark up some government or corporate trees for TALOS funding? AFIAK there are various government agencies interested in secure hardware and assured computing; I have always wondered what the NSA uses for their own computing needs, maybe they paid intel for firmware source code and a system that doesn't need ME to run.
Have anybody approached goverment? They don't take initiative by themselves. I suspect that it wold be possible to get funding from EU. But there is one more problem It's US company that can just decide "We'll make POWER10 with POWER-ME".
The way things are going: (...)
Our hope is(should be) in fully open systems like lowRISC. Invest in making them fast enough to power a basic laptop.
It seems that so many linux people just don't really care about libre anything, (...)
Many(most) 'linux people' are 'open source' not 'free software'.
(...) considering that the average linux sysadmin makes over $100K per year the community could have easily funded the project.
In Germany, US, France... Not in Poland, Ukraine... many hackers there.
If I wasn't unemployed I would happily pay $5K for a high performance libre computer, but not everyone is me.
Me too. If I had work that gave me $60K a year. For me (a student in Poland) $5K is enough to live for a year. Setting a price tag so high reduces size of the community.
People went nuts for the faux libre purism laptop but talos gets hardly any comparative publicity/hype - why? - "We'll get intel to open up ME one day, we promise!"
Because it was viable 'solution': It's amd64: Everything will just work! It's not underpowered: It will not suck! Maybe even more importantly: What people? I highly doubt that those were highly informed people like us.
I think that people don't believe that this would succeed. Including me. Let me ask a different question: Why so many free hardware projects are so impractical? Let's look at novena, it's supposed to be a laptop. But it can't be practical because of it's size. Layout of motherboard makes it impossible to craft a comfortable chassis. It's got built-in fpga that makes it ~$40 more expensive (In my opinion needlessly -- not many people use fpga.). [Lot's of text here]. The pros is that even I could afford laptop at this prize tag (if it wasn't a toy).
I had an idea to make a usable* laptop and talked with people from my HackerSpace. We've given up after realizing how difficult it would be to pull it off. (We would likely be able to make a prototype but producing this is out of reach.)
* underpowered but it would be enough to browse simple sites and use ssh -- seems enough for sys admin away from stationary pc.
On Mon, 12 Dec 2016 13:18:42 +0100 Łukasz Dobrowolski lukasz@dobrowolski.io wrote:
On 12/12/2016 03:27 AM, Taiidan@gmx.com wrote:
The coreboot project is pretty much dead in the water without it, the only real choices for further development are either super low power crappy ARM devices or always going to be expensive IBM/TYAN POWER servers, so what do we do?
Maybe we should ask ourselfs do we really need high performance computers?
There are people who really need that performance, e.g. for various (open source) software / hardware development. I can see your point and I presume you don't really need such a powerful workstation -- me neither. But if you are working on big software projects, you cannot keep pace using librebooted Thinkpads or even ARM devices I guess :/
I think that people don't believe that this would succeed. Including me. Let me ask a different question: Why so many free hardware projects are so impractical? Let's look at novena, it's supposed to be a laptop. But it can't be practical because of it's size. Layout of motherboard makes it impossible to craft a comfortable chassis. It's got built-in fpga that makes it ~$40 more expensive (In my opinion needlessly -- not many people use fpga.). [Lot's of text here]. The pros is that even I could afford laptop at this prize tag (if it wasn't a toy).
I think the novena was not really supposed to be a laptop, but rather a mobile "hacking station". It was not targeted at the average linux user who just wants an open laptop.
By the way there still are a few interesting projects in this lower performance class:
https://www.crowdsupply.com/eoma68/micro-desktop
https://www.olimex.com/Products/DIY%20Laptop/
http://www.powerpc-notebook.org/en/
... and of course the upcoming RISC-V stuff.
The Olimex DIY Laptop will be at FOSDEM afaik.
From their website:
"With different main board configurations: first with ARM64 and x86 later MIPS and other architectures may follow."
That sounds promising.
Of course it's an entirely different thing than Talos.
Regarding Talos: This may sound silly and more knowledged readers now probably smile mildly and roll their eyes, but: Maybe it's still possible to create some "light version" of the Talos Workstation? Maybe not with all the features, but still based on POWER8. Maybe RE could team up with the Olimex guys? Or RE could work together with several (professional) volunteers to make R&D less expensive?
Some guys from 51nb recently designed an Intel i7 motherboard that fits in an Thinkpad X61. If you'd calculate US-based R&D, maybe this would be in the range of Talos? Or at least it would sound unlikely to realize in the first place...
Regards,
Merlin
On 12/12/2016 02:42 PM, Merlin Büge wrote:
There are people who really need that performance, e.g. for various (open source) software / hardware development. I can see your point and I presume you don't really need such a powerful workstation -- me neither. But if you are working on big software projects, you cannot keep pace using librebooted Thinkpads or even ARM devices I guess :/
Nothing that could be produced by the coumunity right now can suit everyone's needs. If somebody needs that performance then this would not be for them.
I think the novena was not really supposed to be a laptop, but rather a mobile "hacking station". It was not targeted at the average linux user who just wants an open laptop.
True. But this is actually something that is close to being viable. What is needed is a project that can meet it's target's needs. I think that many people don't need that much power. I've seen many hackers with x201s' with x200 (i use x140e). Some of those people also use more powerful stationary PC-s.
By the way there still are a few interesting projects in this lower performance class: https://www.crowdsupply.com/eoma68/micro-desktop
I doubt that there is a viable use case that will popular enough. After all i can just buy a ATX board suported by libreboot and put an 8 core amd fx.
I've watched talk by mr. Olimex. His focus is to be able to swap only the broken parts instead of the whole laptop. Afaik mr. olimex doesent mind using allwiners. And we know that it's not good for security. Many(most) people who don't want ME are conscious about security.
Modularity is something that makes many open projects unpractical. Modularity is *not* free, it has serious drawbacks. It makes devices way bigger then they need to be. There is a reason why normal projects are not modular.
AMD's version of ME. Yay!
Closer but still not it. Somehow I don't see hackers with 15.6" laptops that are open. I think it's because many of those people carry their laptop with them most of the time.
... and of course the upcoming RISC-V stuff.
Not going to be viable for quite some time.
On 12/12/2016 03:22 PM, Łukasz Dobrowolski wrote:
What is needed is a project that can meet it's target's needs.
Well, novena did meet it's target and got funded.
We need something that targets more people. Like sys admins who have a stationary PC + super mobile laptop.
On 2016-12-12 at 13:18:42 +0100, Łukasz Dobrowolski wrote:
On 12/12/2016 03:27 AM, Taiidan@gmx.com wrote:
(...) considering that the average linux sysadmin makes over $100K per year the community could have easily funded the project.
In Germany, US, France... Not in Poland, Ukraine... many hackers there.
Indeed. In my country 20k EUR per year is a good pay, but even with a pay of 60k EUR|$ a year, that would mean paying the whole earnings of a month for something that is still at the crowdfunding stage and may just disappear into nothing.
If you consider that people have to, well, live with that money, that would mean multiple months of savings / disposable income.
If somebody has a family, that probably goes in the range of expenses that have to be negotiated with the other family members.
All of this necessarily reduces the number of people who can even think about partecipating in that crowdfunding, not to say actually decide to put down their money for it.
The coreboot project is pretty much dead in the water without it, *the
only real choices for further development are either* *> super low power crappy ARM devices* or always going to be expensive IBM/TYAN POWER servers, so what do we do?
What is wrong with "super low power crappy ARM devices"? Do you realize that almost ALL Qualcomm/Mediatek/Samsung ARM SoCs for Android smart phones are falling in this category? Or maybe not? And do you know that 99.999% of this "crappy" devices are using U-Boot? U-Boot out of (these days) DENX Software Engineering GmbH (Managing Director: Wolfgang Denk)?
BTW what is your definition of: *"super low power crappy ARM devices"* ???
Floor all yours! :-)
Thank you, Zoran
On Mon, Dec 12, 2016 at 3:27 AM, Taiidan@gmx.com Taiidan@gmx.com wrote:
I am incredibly sad that TALOS has not gotten the required cash flow, short of a miracle in the next few days.
The coreboot project is pretty much dead in the water without it, the only real choices for further development are either super low power crappy ARM devices or always going to be expensive IBM/TYAN POWER servers, so what do we do?
I am wondering, how come they didn't bark up some government or corporate trees for TALOS funding? AFIAK there are various government agencies interested in secure hardware and assured computing; I have always wondered what the NSA uses for their own computing needs, maybe they paid intel for firmware source code and a system that doesn't need ME to run.
The way things are going: +10 years - Microsoft and Intel have announced the "PrivaSec" initiative, aimed at producing a secure vertically intergrated computing platform where firmware agents prohibit the execution of unapproved programs - protecting your data from unauthorized access.
+20 years - We're sorry, but the GlobeX Trade Agreement and the Secure Communities act of 2035 prohibit the viewing, copying or transmission of this file - Further violations may result in fine, arrest and or the revocation of your work permit and internet operators license.
Thoughts: It seems that so many linux people just don't really care about libre anything, considering that the average linux sysadmin makes over $100K per year the community could have easily funded the project.
These days there are a lot more people with skills, but without the computer enthusiast/hacker culture of the 90's, the kind of hypocritical people who use a macbook, facebook, etc but who chide me for saying that working for the government is not at all immoral.
If I wasn't unemployed I would happily pay $5K for a high performance libre computer, but not everyone is me.
People went nuts for the faux libre purism laptop but talos gets hardly any comparative publicity/hype - why? - "We'll get intel to open up ME one day, we promise!"
-- coreboot mailing list: coreboot@coreboot.org https://www.coreboot.org/mailman/listinfo/coreboot
-
Elena ``of Valhalla'' -
Merlin Büge -
Taiidan＠gmx.com -
Zoran Stojsavljevic -
Łukasz Dobrowolski