As an engineer and self-hosting and self-coding enthusiast, I would agree with a lot of the points. I have spent most of my life in IT advocating for decentralization and democratization.
However, as someone who has had enough experience in the real world to notice how different time and skill constraints lead to different requirements for outsourcing, I think that it sounds elitist. Even an LLM is not sufficient for people who don't even know the difference between backend and frontend or what an API is, and therefore don't stand a chance to craft a proper prompt, let alone properly test the code that the LLM produces.
For context, I could also tell Mr. "Having a fucking website" that they're a hypocrite because they run a blog on Wordpress and have a social media account on mastodon.social. Those who really believe in decentralization run their own stuff, or code their own blogging platform like I did. They don't just brag of how morally superior they are just because they deleted their Facebook and Instagram accounts.
Of course I would sound elitist. And that's exactly how their stance sounds to the average bakery shop owner.
You know that you are dealing with a fascist scumbag the moment you hear "terrorist", "antifa" and "law and order" in the same phrase. Flock is a disgrace for the industry, and this fascist sociopath is a disgrace for mankind.
It depends on how much Firefox enshittifies. If it's just about removing some telemetry configuration from upstream, then a couple of downstream patches will still do the job. If, say, Firefox decides to fully embrace the spyware business model and drop support for Manifest V2 in order to kill adblockers, then LibreWolf will probably have to maintain their own fat piece of logic built on top of Firefox. Keeping it as a soft fork would then be a lot of work (you'd basically have a patchset of tens of thousands LoC to keep porting through different versions of Firefox). And making it a hard fork would be even more work (it basically means that the LibreWolf folks are on their own and they have to maintain their own independent browser).
I did my Github -> self-hosted Gitlab -> Gitea -> Forgejo journey over the years, and I haven't looked back.
Forgejo is great and it's probably going to become even greater once federation is done (having distributed forks and PRs across multiple instances solves the fragmentation problem of self-hosted solutions).
And I lost my trust in Gitea once it spun off a for-profit branch backed by VC money (which was exactly the reason why it was forked into Forgejo).
The only thing I lost from Gitlab is the out-of-the-box CI/CD platform. But I could migrate my pipelines to Drone CI and trigger them via webhooks. Just keep in mind that, depending on the complexity of your Gitlab pipelines, this may not always be an option. Anyway, for me hosting a Gitlab server that hogged up 5GB of RAM to serve a couple of small projects was a big no-no. Forgejo takes 500MB of RAM at peak.
Is it open-source? Is it based on open protocols? Does it support federation?
If the answer to any of these questions is no, then your product is irrelevant.
It's nothing personal. It's just that I'm sick and tired of centralized solutions running on somebody else's computers. At least this product makes its business model clear (a $2/month subscription fee for interacting).
But hey, I'm a guy who used to run phpBB forums on a Pentium 1 under my bed 20 years ago. Back then there were tons of those forums, most of us didn't ask users for any subscription money, there was no centralized gatekeeper, and everybody was happy. All of us used to curate our own small gardens and the model was sustainable at all levels.
Can I just get that back, with some federation protocols on top to allow multi-instance interactions? Seriously, everything else is pointless, it's a waste of time, and it doesn't make the digital world any better. I won't jump from a centralized platform to another, and probably nobody should.
The subscription model is a business answer to a recurrent problem in tech: how do you make your revenue streams more reliable?
If you build the perfect product, and everybody buys the first version, then you'll have no revenue stream left after everybody who wanted to purchase it has purchased it.
And that's where the subscription model steps in.
It's also a way of slowing down innovation (and therefore risk) while remaining profitable. In a world where your revenue is based only on how many new copies of your product you sell, you are forced to innovate and keep selling, or your sales will drop and competition will eat your plate. If you already have a user base of subscribed users instead, it's much harder to attack your bottom line.
So yes, it's a great model from a business perspective, but it's an absolute nightmare from a user perspective.
I personally started paying between $300-400 a month just in subscriptions at its peak (about 3 years ago), and we're talking just of 8-9 services. I decided that it was not sustainable, and that I had to fight back.
Today I reward news outlets that let me pay for content "on the go" over those that put the "subscribe now" paywall in front of my eyes. I am very happy (and proud) to scrape and pirate those services without feeling an inch of guilt. If in order to read a single article I have to start a subscription that involves cumbersome interactions with your customer service just to terminate it, then you deserve piracy.
And, when it comes to other services, I started self-hosting everything. There's plenty of open-source alternatives to literally every paid service that you can run yourself. A Celeron minipc in my house with a few TB of storage connected cost me about $350. After 1-2 months running Nextcloud, Miniflux, Wallabag, Matrix, Ubooquity, wger, Bitwarden etc. on it, it already paid its own price back in terms of subscription savings.
I've been using Synapse as my only messaging service, and Element as my only messaging frontend/app, for months by now.
If you self-host it, then the next logical step is to install bridges for all the protocols you currently use (WhatsApp, Messanger, Instagram, Telegram, LinkedIn...). Or, at least, that's been the main reason for me to install it, as I was getting tired of the lack of developments on the bitlbee/libpurple side. And, since the API is open and we'll documented, it's also become my central hub for automated notifications, RSS feeds, social feeds etc.
So Matrix is now my "one app to rule them all". I've got a channel on my server dedicated to the FLOSS project I maintain, and a couple of people are there, but that's not the point - at least not for me. I use Matrix because of its flexibility and ability to bridge anything to anything else. Who cares if there aren't enough people on my server :)
However, as someone who has had enough experience in the real world to notice how different time and skill constraints lead to different requirements for outsourcing, I think that it sounds elitist. Even an LLM is not sufficient for people who don't even know the difference between backend and frontend or what an API is, and therefore don't stand a chance to craft a proper prompt, let alone properly test the code that the LLM produces.
For context, I could also tell Mr. "Having a fucking website" that they're a hypocrite because they run a blog on Wordpress and have a social media account on mastodon.social. Those who really believe in decentralization run their own stuff, or code their own blogging platform like I did. They don't just brag of how morally superior they are just because they deleted their Facebook and Instagram accounts.
Of course I would sound elitist. And that's exactly how their stance sounds to the average bakery shop owner.