Not quite. It’s just easier to jump in to mechanics you already know, rather than try to reconceptualise the structure and learn to navigate new pitfalls: oh I need to select a server? Which one is the right one? Oh I need a client, which one? Oh I can’t quote retweet? How do I find interesting people to follow? Is this the right handle in the right server or am I following a bot?

Brill. One of us.

I’ve tried it with forgejo, the recommended implementation involves spinning a temporary vm to run the integration and deployment processes, quite resource heavy and slow comparatively to the vm I have that’s running forgejo.

I think there’s an option to have the forgejo server itself run the commands without spinning up vms, but it’s not recommended due to security considerations as they’re running with the same privileges as the server - not a concern if you are the only developer connecting to a private instance of forgejo but something to keep in mind.

I think a possibility is a series of open source anvil or nixos scripts that you can run on most hardware with minimal changes, in an extendable architecture of some kind to add or remove functionality and they perhaps get maintained by the community or some structure of the kind of Linux distributions.

This could enable people with minimal skills set up and maintain a reasonably useful but secure environment just by changing a few variables.

I can’t imagine any system of influence running an exposed ssh without some further protection from connection abuse like fail2ban.

I’ll believe AI can replace engineers when I see NVIDIA firing them. But like the graphic says, the manager’s job seems a lot easier to replace instead.

Perhaps this was written much earlier than v5.

Maybe this is useful https://tunnelbroker.net/

Are you on an enterprise subscription / office 365 work or school account or something like that?

All business models are aimed at company profitability. Customer satisfaction is an expensive early necessity which you can largely do away with as you become entrenched.

I’m assuming it’s aimed at people trying to avoid tying the hosting IP to the publicly consumable service.

The problem is that most of us have swallowed the ‘competence uber alles’ ideal that school fed us through exams and scoring, when the game really is mostly politics (as in interpersonal relationships). So we are understandably disappointed when the incompetent get promoted through brown nosing or luck, when we should be reevaluating the rules of the game.

No worries.

For those who were wondering:

On the security updates:

Yes they’ll provide some security updates for some time even out of contract. No time frame given, only in relative release numbers:

Our naming convention for releases is: <major>.<minor>.<patch>.

Version -2 of currently released minor version goes EOL. The cadence is not explicitly provided

On not renewing or renewing later:

Yes, jump in any time.

Can I update every second or third year? Will the previous versions receive security updates?

Sorry, I must have been too tired, got nearly all details wrong: 32GB RAM 1TB M.2, USB3.2, BT4.2,WiFi 5,4k HDMI, Gigabit Port, and not a Beelink but a DreamQuest. There’s just the M2 interface disk connected, no SSD.

It’s literally one of those little known brand nuc, tiny box - beelink I think. Total cost $200 or so - it’s been running non stop for the last 3 months without an issue. I don’t think it even has a fan in there.

I’m running a n100 16gb with a 256ssd, 4vms and 4 docker images, it’s pulling 7-9w.

Only if you let them

Have you heard of virtual debit cards? You can’t charge what’s not there.

Also, at least AWS will in fact send you an email when you approach the end of free tour usage.

Having said all that, most devs can host the few hundred visits they might get over a month with a $200 home server and a free CloudFlare cache if they know what they’re doing.

Obsidian also.