[ken]

So by this standard, is there a computer that I should use, or is all computing in 2019 unethical?

I see he's using a >10yr old Intel CPU, presumably to avoid the Management Engine, so presumably he thinks it's OK to pay money to these companies in certain circumstances?

[kilburn]

By RMS's standard, it seems that it is ok to reuse old systems (irrespective of which company built them originally) provided you can use them without binary blobs (irreplaceable built-in no-source firmware is fine though). Quoting his "how I do computing page" [1]:

> I use a Thinkpad T400s computer, which has a free initialization program (libreboot) and a free operating system (Trisquel GNU/Linux). It was not sold that way by Lenovo, however; small businesses buy them used, recondition them, and install the free software. This is one of the computers endorsed by the FSF.

Of course I'm pretty sure he would also approve buying and using any open hardware product [2] (provided it can run using only free software). There are some initiatives popping up in this space, such as:

- The siFive HiFive's [3] (Risc-V based), that can get you a fully open, slow system for $3k (1k for the base board + 2k for the PCIe-and-others expansion board).

- The Talos™ (Power9-based) systems [4], that give you 2019-grade performance at ~2-3x the usual price.

All in all, not a pleasant stance to make, but definitely doable for some folks.

[1] https://stallman.org/stallman-computing.html [2] https://en.wikipedia.org/wiki/Open-source_computing_hardware... [3] https://www.sifive.com/boards [4] https://www.raptorcs.com/content/base/products.html

[derekp7]

This is the thing that gets me, is that if a piece of hardware has firmware that you can't replace then it is OK to use. But if they add the functionality to replace the firmware, then it all of a sudden is not OK to use that hardware. Or, is it the case that if the firmware is pre-loaded, but replaceable, instead of needing to download the firmware to the board every time you boot, then that would still be acceptable. Not sure where he stands on that last point.

[kilburn]

RMS thinks that free hardware is a different matter than free software, on the basis that software is easily copyable but hardware is not.

If a piece of hardware contains unmodifiable software, RMS considers it an "integral part of the hardware". It is no different to him than a cog or any other physical piece in that thing.

In contrast, if said software (firmware) is replaceable, it means that it can be copied at no cost, modified, used on other hardware, etc.. It is a fine line, but I can see some difference.

[0815test]

The thing is, "functionality to replace the firmware" is not an unambiguous positive. A firmware that can be replaced is also a firmware that can be bricked, made less useful by a future update, or even subverted entirely by an intruder. So, RMS would like whoever is placing that firmware in there to tell the user the information they need in order to replace it without bricking the device.

[craigsmansion]

It's a point that causes a lot of confusion.

As soon as a piece of software is replaceable, the one who provides you with the software can use its distribution to make you promise to ignore the four software freedoms.

If a piece of software is not replaceable, that point is moot, since the one who provided you with the software has no sway over you regardless. They have no control over you or the device.

[Yetanfou]

Better rephrase that to indicate that they have no more control over you or the device than the non-replaceable firmware allows, this is definitely not the same as "no control". Non-replaceable firmware can ping home, it can be used to monitor users, it can download arbitrary code and run it, it can listen for some trigger to start messing with your data or downright destroy said data, it can start randomly rebooting systems or non-randomly restart all affected systems at a specific time - which may cause brown- or blackouts depending on the power draw and -supply, etcetera.

[slfnflctd]

> presumably he thinks it's OK to pay money to these companies in certain circumstances

When they're behaving in what he considers an ethical manner, or at least offering some ethically designed products, I think so. He's hoping to help catalyze a revolution of sorts regardless, and AFAIK is still waiting for the hypothetical ideal entity to arrive (which probably isn't going to happen any time soon).

I look at Stallman's lifestyle as a form of performance art at this point. It's more than this, for sure, but that's the best way to describe its immediate impact much of the time. However, we will likely be chewing on the thoughts he has brought forth for generations to come.

[fghtr]

https://fsf.org/ryf