Youtube would rather be able to track every user and make more profit per person than go big tent all audiences.
True, but consider that Google, Alphabet, whatever, is fundamentally an information broker. All of their services and technologies are simply a means to that end. They have no incentive to go big tent if it means sacrificing their ability to harvest data on individuals and groups.
I’m not saying TrueNAS and ZFS aren’t good. For large enterprise systems, arrays with sufficient redundancy, servers with reliable power management, I can see its advantages, esp. w/ snapshots, etc. I acknowledge that Open Media Vault is mickey mouse in comparison.
I just feel compelled to share my experience when I see people considering TrueNAS for their first foray into building a small home media server, running a z1 array, with no mention of battery backup or power management. ZFS isn’t inherently safer. It’s safer when paired with sufficient redundancy and power management.
When I built my first server, TrueNAS, ZFS, and Raid z1 made perfect sense. And I loved it for the first couple months. Then an update and unexpected shutdown rendered my storage pools unrecoverable. Had backups for most but not all of the files, and spent almost a year of bits of free time here wading in way over my head on highly technical support forum threads & there trying to bring the pools back online. Nothing worked, the array was toast.
I don’t know how tech savvy you are, but here’s the advice I’d give my past self - Take a few weeks to read documentation and play with TrueNAS before filling up your drives with stuff. Peek around in troubleshooting forums, see if the troubleshooting you may have to do is in line with your experience level.
After wiping my drives and starting over, I built around Open Media Vault. It’s less pretty and less feature rich than TrueNas, but it’s also much less fragile in a raid z1 setup and I never worry about it.
The process to log in to the online portal of Outlook is so bad it’s crossed into comical territory. So much friction, only to shunt you to a full screen clippy copilot page.
I’d be curious to know what the usage statistics are for that page. Like, what could a person possibly accomplish there?
I like feedbro too. Haven’t found a standalone selfhosted solution that has the same degree of customization.
I’m trying out freshrss right now and don’t like it. Possibly my issues stem from user error, but, I can’t figure out how to automatically hide articles based on keywords, adding extensions is a pain, and the ui feels large and very in-the-way. By default it truncates article titles, which I find absolutely baffling.
Be careful about mentioning who you bank with on the internet, using an account that anyone can browse your comment history.
There are good RSS reader web browser extensions. Firefox has a few. Check out Feedbro.
It’s not obscure, but, for me, Wikipedia is the ultimate example of the old internet that still persists today.
Free to use, no account required, ad free, non-corporate, multilingual, heavily biased toward text, simple and utilitarian design. Hyperlinks concatenate relevant pieces of information, which serve as the means to navigate the site. The code is very simple (seriously, view the page source of a wikipiedia article). It’s based on the human desire to learn and share knowledge with others, and has remained resilient to corruption by commercial interests that pervert that desire for monetary gain. It’s a beautiful thing.
This is the sort of thing that the old internet could really deliver on. Chances are, a search query could lead you to some guy’s hoodie blog, and he just liked hoodies, and posted honestly about them.
Now, it’s all a mess of SEO pumped affiliate link lists filled with crapware. If the query is even thinkable, there will be AI generated pages stuffed with sponsored links, ready and waiting for you. And with search engines preferring recent results, that’s the type of page you’ll be served.
I’ve had decent luck using marginalia search to seek out some of those old internet type results. Obscurity works as a barrier to corporate infiltration. Plus you get page results that don’t have a million tracking and analytics scripts running on them, which is refreshing.
As a user, ‘privacy preserving attribution’ is unappealing for a few reasons.
It seems it would overwhelmingly benefit a type of website that I think is toxic for the internet as a whole - AI generated pages SEO’d to the gills that are designed exclusively as advertisement delivery instruments.
It’s a tool that quantitatively aids in the refinement of clickbait, which I believe is an unethical abuse of human psychology.
Those issues notwithstanding, it’s unrealistic to assume that PPA will make the kind of difference that Mozilla thinks it might. I believe it’s naive to imagine that any advertiser would prefer PPA to the more invasive industry standard methods of tracking. It would be nice if that wasn’t the case, but, I don’t see how PPA would be preferable for advertisers, who want more data, not less.
As a user, having more of my online activity available and distributed doesn’t help or benefit me in any way.
You’re right to point out the difficulty of preparing installation media.
Also, for the average person, friction will probably happen during installation - possibly having to circumvent safe boot to install and run a new OS (knowing how to enter the bios, feeling comfortable playing around in the bios, knowing how to even disable safe boot once you’re there, not exposing your device to security vulnerabilities by having safe boot disabled), the need for an existing understanding of how partitions work and how the partitions are structured on your specific device in order to test the waters with a dual boot setup on a drive that has data/functionality you want to preserve. Needing to know the ‘what’ and ‘why’ of swap, /home, and /root partitions. These points all came up on a recent installation, and I’m sure they would scare some people off.
Installation will be easy if you have the time, motivation, existing knowledge and/or bandwidth for a learning curve. But not everybody has that.
And that’s just installation, to say nothing of the actual use of the desktop environment, which is not as intuitive as its often claimed to be.
This is interesting. Do you have any thoughts on why someone would want to utilize the epstein data for ML? Like, what’s the point, in your opinion? Just lulz? Or, something else?