Linux has been ready for some time within various educational programs, but maybe you are referring to relatively early education curriculum in public schools? The general anecdotes I’ve heard from teachers within a variety of grade levels in the USA (mostly elementary and high school levels, but some doctoral engineering/scientific as well) convey that the largest hurdles to overcome are:

1. Teaching the teachers. Teachers are usually very smart and capable, but are often chronically overworked, overstressed, and underpaid for their labor. They have limited mental bandwidth in learning new tech workflows while having the added obligation of teaching these workflows to students which may be at an attention/interest deficit.
2. Challenging the status quo at the administrative level. Schools often receive incentives, grants, steep discounts, etc, for installing certain types of hardware or software packages. The software baselines of some schools are restricted at the district level; many public libraries are restricted by the city/county. Perhaps the best approach here is to install Linux as a “secondary” option (similar to how a smaller number of e.g. Macs may be installed in a computer lab comprised mostly of Windows computers) until it’s more widely adopted.
3. Advocating for equivalent Linux support for popular proprietary software. This is especially true for the creative design community, such as graphic design and professional music production. Adobe is usually the target of criticism here; Linux does not currently hold enough market share to capture Adobe’s attention while their patrons usually have unwavering brand loyalty or are unwilling to make any tooling/workflow compromises as to maintain their livelihood.
4. FOSS-friendly awareness campaigns. Showing people that they can remain productive while not being at the mercy of Big Tech. Not using public funds for private industry.
5. Feature parity case studies compared to proprietary options.
6. Overcoming the stereotype that Linux is only for techy people, shrouded by gatekeepers, or subject to drama/infighting.

_{sad richmond noises}

In the meantime, the Liberty phone seems to be the closest option for a US-made smartphone. While not entirely comprised of US-sourced components, the PCBs are manufactured in California, as well as device packaging and assembly.

April 10 update: Right on time, the author of the OP’s linked article has now published an article on the Liberty phone.

If anyone remembers the Mycroft Mark II Voice Assistant Kickstarter and was disappointed when development challenges and patent trolls caused the company’s untimely demise, know that hope is not lost for a FOSS/OSHW voice assistant insulated from Big Tech…

FAQ: OVOS, Neon, and the Future of the Mycroft Voice Assistant

Disclaimer: I do not represent any of these organizations in any way; I just believe in their mission and wish them all the success in getting there by spreading the word.

There is also a NewPipe flatpak available on Flathub.

The Librem 5 offers hardware killswitches on the side of the phone, and the Pinephone offers hardware killswitches in the rear of the device after removing the backplate.

Purism is slowly correcting course regarding refund policies, but meaningful improvements take time. Also, the letter from “Purism” to Rossmann came from a Purism volunteer who had no authority to publicly represent the company.

I used to work for the U.S. Department of Defense and can confidently approve of massive defense budget cuts and merging of several military branches. This is only a single and relatively minor anecdote, but it is a small piece of a much larger problem and is one I can share from personal experience:

I used to be the government lead for a highly successful defensive capability that only consisted of myself and 2-3 defense contractors. We outperformed several long-standing projects that had 10x the staff, 100x the budget, and had been around for approx 10 years without going operational (“operational” in this case meaning that intelligence analysts are authorized to provide actionable intelligence derived solely from the tool). My team released 3 operational releases within 1 calendar year from the start of contract.

I don’t say this to disparage the staff of the other project(s), but rather to highlight how the government can afford to cut long-standing under-performing projects and become more lean and efficient. The government funding allocation is often in the realm of $300k/yr for a single FTE. Multiply that by a team of 20-30 that works on a project that is shelfware after 8-10 years.

My same project was approached by numerous branches of the US and FVEY military community. Branch A offered tons of money to put it on a ship; branch B offered even more money to put it in the back of reconnaissance aircraft or fighter jet; branch C offered money to make it man-packable for ground troops. US taxpayers already paid for this capability once (my team and myself) and we made it as unclassified (i.e. disseminable) and modular as possible (it was literally designed to run on a general host computer running Linux), yet each branch was willing to fork over tens of millions of dollars for something they could have installed on a $2k computer using some internal software repository. And that’s what I suggested they do.

Again, this is just one minor anecdote. How often does this happen where taxpayers are forced (being that they have absolutely no control over how the defense budget is organized) to pay for the same (perhaps MUCH more expensive) tools e.g. 5-10 times because military branch A, B, C, etc, want their own flavor of the same thing? Why does the military often have pissing matches of authority when there is so much overlap between some of them? Take away their stick by taking away some of their funding, and force them to share and cooperate.

It’s a valid point that it could potentially create some confusion when a user assumes that everything in Signal is secure. Unencrypted SMS threads could contain an open padlock icon and even an ominous red window border, but someone inevitably will not understand the difference.

However, my frustration has been how both convenience and security is reduced by removing SMS from Signal.

Many people will continue to use SMS for a variety of reasons, necessitating the use of an additional app. So now we have people continuing to communicate over this insecure protocol, but with the additional target vector of potential vulnerabilities in the supplemental app.

Imagine a world where we can adopt a scalable, secure, open communication protocol where users can use whatever app they want. Imagine humanity moving past the diaspora of special-snowflake chat apps and on to better things.

I don’t use a dedicated app at all to use Lemmy on my Librem 5 daily-driver. I access it via Firefox. To do this, I created a dedicated Firefox profile for Lemmy and created an app icon for it in the app gallery. More details here.