Only x86 will should be supported, due to its widespread use (Intel or AMD CPUs or so). Support may be extended in the future to other freedom (or almost freedom) based architectures, such as RISC-V.
ARM or similar proprietary architectures will not be supported—there’s no reason to waste time on them. Supporting the most common proprietary architecture (x86) is sufficient. (Though seL4&Linux supports ARM anyway).
Disagree.
This policy is rejected.
Various architectures have various support statuses and this will remain so.
Contributors are welcome.
I don’t think it’s productive to have detailed project strategy discussions. For comparison, I don’t think suggesting something similar to other projects would result in any policy changes either.
It’s a do-ocracy. People who do work and are sufficiently compatible with the project make decisions.
The only thing with at least a chance to change this would be something similar like this for Kicksecure:
Yeah but it doesnt defy the waste of time on supporting multiple proprietary architectures which all give the same result = dealing with a blackbox, it has vulnerabilities which may not get patched and we dont know how many others still uncovered, requires blobs…etc.
For me i wouldnt bother with this, its similar if i choose bitcoin because its famous/highest in value not because its private or the greatest, but on the same time i wouldnt value other coins having the same issues.
There have been various contributors over the years who contributed support for their architecture, among other general improvements. The additional architecture did cost some of my time, but the overall result was a net benefit due to the other general improvements contributed. Due to this experience, I am not preemptively closing the door for future contributions.
RISC-V is not a get out of jail free card:
Even if the entire stack is open, modern silicone has billions of transistors, so verifying zero hardware backdoors is practically impossible, whether to produce a golden snapshot or at scale across time. Hardware security is out of scope for Kicksecure or Whonix.
Yeah was just an example not an endorsement.
Something better freedom/security than Intel/AMD is a considerable effort for inclusive. An example was POWER9, which was a great and we had alot of discussions regarding it, we even have VM dedicated for ppc64 testing made in that time, but sadly there is no more POWER CPUs coming open as 9 at this moment.
That is only seeing the freedom of silicon itself, not the underlying dependency behind it. That is the entire argument I made in the Whonix Forum, it means nothing if the vendor, such as IBM, can simply mandate the concept of freedom obsolete by releasing Power10, putting proprietary anything within it, and offering an illusion of choice for you to consider. Freedom is constant vigilance, not a self-contained FSF island to retreat away from.
Unfortunately, true. References here:
If only major retailers would start selling these type of products. Sadly innovation is not where it was if you look at older systems the design and modularity was peak IMO. I have fond memories of when computers would instruct you how order replacement parts #'s from the catalog. Today this is not really the case. Repair might be another topic but more and more you are paying for garbage that you don’t even really own.
Legacy hardware might become more valuable given the trend. One has to really question why laptops are moving towards SoC and the NPU’s. Thin client future with OCR (Optical Character Reconition) and ACR (Automatic Content Recognition) thrown in with NPU’s and a flavor of 1984?
Legacy hardware is susceptible to security vulnerabilities over time, that is why every security project is an arms race, not a one and done deal.
Newer hardware has Intel ME and these issues might get worse. Depending on one’s threat model, pre-Intel ME hardware may be considered more secure than today’s or future hardware.
Yeah. Huge issues.