Be mindful that such a program would have to be safer than the situation without. A program on a public repository that isn't used by any distribution, isn't audited, hasn't a lot of comments (and thus eyes on its code) might be a disproportionate risk compared to the default settings of a popular open source distribution IMHO.

Hi there, how about keeping history of past messages? I mean if all participants leave the channel, can they all keep history using e.g. localStorage and when they come back, see what has been shared until now including when they were away thanks to history of that channel from others?

It doesn't have to be though. It could be BOTH convenient AND private. It's only because we, as a society, didn't fully understand the "cost" of "free". We thought it was just so nice to get a good search engine without having to pay. We didn't grasp that it was the beginning of surveillance capitalism. We didn't understand that this business model would be so successful every company, from news ones like Meta, to "old" ones like Microsoft or Amazon, would try to be hybrids, both selling stuff and but also re-selling data to advertisers.

So no it's not a false choice, it's a corner we strategically got pushed into.

I believe, maybe naively, that initiatives like https://uattest.net/ or even https://www.taler.net/ are trying to show that it can be both convenient and private, but NOT while relying on surveillance capitalism which is precisely investing a lot of money to bring the maximum convenience, including free (hard to beat) but at the cost of privacy.

Edit: seems GrapheneOS isn't into UAttest initiative https://grapheneos.social/@GrapheneOS/116200110686604617 but I'm not sure what alternative they propose.

trust that they won’t add any collection without telling people.

It's open source so you can inspect it. If you don't know how to do that you can pay for a 3rd party audit.

Also if it were to be found out, even without being open source via some pack inspection (e.g. using a software that checks if data is being sent to a server, e.g. imagine starting Firefox on a virtual machine then checking if any data goes to e.g. firefox.com) and it were to be published then their entire brand would be dead. So rationally speaking I don't think that's a worthwhile bet.

FWIW not on income but on top wealth, Musk has $792 Billion (ffs...) so ~$1000B and we are 8.4B Earthlings so ~10B. If we were to spread equality his wealth (which I'm all for) it would "only" gives each of us ~$100. Conclude from that what you will but to me it's just a reminder of just how many people we are. A lot.

PS: this isn't about income and it might be totally different there. If you have a better metric and approximation I'd be all ears.

assuming you trust JMP

Any 3rd party security audit that would help on this specifically?

I think that's precisely what this is questioning : is this helping fund critical FOSS?

What if a fraction of that money instead went to Signal infrastructure? Wikimedia? FSF which initially made GNU PG? FSFE? NLNet which supports Delta Chat? Sovereign Tech Fund? etc rather than individuals?

I don't think anybody is criticizing that hard working people contributing to a good project are well paid. I believe the question is rather what's the cost to OTHER projects when there is 1 project, not an umbrella projects which funds others (again like NLNet or the Sovereign Tech Fund).

What model are we reproducing and what's the risk?

FWIW the question isn't new. It happens also with Mozilla with the compensation of its C-suite staff, not the "random" software engineer.

Agreed. That being said

• here is a link to an archived copy of it https://archive.is/qO6pe
• I'm a 404media paid subscriber so IMHO it's worth supporting their proper journalistic work in tech and politics

No and honestly I don't think it matters. Set the age of your OS to 18 (assuming you are 18) and move on. What's the issue?

That being said if you are really interested in the topic and use this as an "excuse" to learn check out https://jsandler18.github.io/ and don't worry if you don't have an RPi to run it, you can use QEMU. After that you can dig into https://wiki.osdev.org/ really a fascinating journey.

I tinker with VLC using and... so far my "trick" is a bit dirty, namely I don't "control" it as much as I killall potential running instances then I start cvlc again with the right parameters. If you want continuity though you might not want that.

Note also that if you plan to do scripting check mpv as it's a bit easier to tinker with IMHO.

Python

Check https://pypi.org/project/python-vlc/ then for Python bindings, that should give you the affordances you need.

It's not about TomsGuide or OP, it's about Securitum (who did the audit) and their methodology. If you don't trust the audit then please out why and we'll see if they fix the problem. It's genuinely valuable.

Checkout https://splintercon.net/ and follow the advice given to political dissidents and journalists.