stay a while and dwell in the fediverse or are you afraid you might enjoy it?

  • 0 Posts
  • 20 Comments
Joined 1 year ago
cake
Cake day: June 14th, 2023

help-circle

  • Do not expect you can offer this service for a competive price against cloud prices. Caring for a company IT system is a big challenge and requires more work the more users there are.

    For a company this size: make a clear contract. Consider how much time you need for setup/installation, monthly hours for maintenance, monitoring and at least daily(!) backups. Let them choose if they want it with a failover and charge for the required hours and material. Also put in the contract when they can expect support from you, including a clause for a holiday substitute admin (if needed). Then put a pricetag on support hours for holding people’s hands when they “can’t find that file they uploaded a week ago and it is surely a server issue” and put a pricetag on engineering hours for any modifications they might want, like installing any plugins they deem useful for themselves. Hardware prices, traffic, rack space and power should be included as well. Have a good plan for updates, choose your distro wisely, do not rely on autoupdates.

    Play all this through in your head, add up the hours, choose a fair rate and then you have your pricetag.

    Cloud will always be cheaper, because they have their infrastructure already deployed. Building from the ground up is more expensive, but I think it is worth it. Will they?


  • But…isn’t unsupervised backfeeding the same as simply overtraining the same dataset? We already know overtraining causes broken models.

    Besides, the next AI models will be fed with the interactions from humans with AI, not just it’s own content. ChatGPT already works like this, it learns with every interaction, every chat.

    And the generative image models will be fed with AI-assisted images where humans will have fixed flaws like anatomy (the famous hands) or other glitches.

    So as interesting as this is, as long as humans interact with AI the hybrid output used for training will contain enough new “input” to keep the models on track. There are already refined image generators trained with their own but human-assisted output that are better than their predecessor.






  • I see no advantage in switching to VanillaOS (from Archlinux), and Ubuntu’s tendendcy to ship everything in snaps does not make it better. Arch is swift and clean, snapshots (LVM) protect from broken updates, the archive allows rollbacks way into the past, the additional LTS Kernel bridges bugs in mainline. Rebuilding packages is easy with the available toolchain, and you can be sure every package is as vanilla as possible conpared to upstream.

    Hiding complexity from the user just builds another layer of complexity that can break. I can fix an Arch or a Debian, but Vanilla sounds like it will break once you want to add a few sprinkles of chocolate.






  • We use Ansible as well, it keeps all servers happily upgraded and all packages in working order - even the weirdest custom software instances. Nodjs is available as lts packages im arch and it, again, just works.

    I have zero issues with upgrades on desktop and server except once last year when my old Core2Duo notebook I use in the kitchen did not suspend correctly for a whole week until the Kernel bug was fixed. (I ran linux-lts for a week, it was… smooth sailing).

    During that time we had 3 failed migrations of old PHP software to the new Ubuntu LTS and were fighting almightly RHEL because it simply did not provide the packages the customer required - we are now running an Arch container on the RHEL box…

    I know this discussion is a little bit like religion, and obviously luck and good circumstances play a role. We both speak from experience and OP can make their own decision.






  • My vote is Archlinux. Debian is sometimes a little too “optimisitic” when backporting security fixes and upgrading from oldstable to stable always comes with manual intervention.

    Release-based distros tend to be deployed and left to fend on their own for years - when it is finally time to upgrade it is often a large manual migration process depending on the deployed software. A rolling release does not have those issues, you just keep upgrading continuously.

    Archlinux performs excellent as a lightweight server distro. Kernel updates do not affect VM hardware the same they do your laptop, so no issues with that. Same for drivers. It just, works.

    Bonus: it is extremely easy to build and maintain your own packages, so administration of many instances with customized software is very convenient.