GstpulldnJun 11
Matthew Tift 🏳️‍🌈

Douglas Adams (via @pluralistic) on tech (paraphrased):

1. Born with it? Normal.
2. Invented age 15-35? Exciting, a career!
3. Invented after 35? Against the natural order.

The mind loves to draw lines and call it "reality." Inside is normal and outside is a threat. In yoga and meditation circles, we’d say the tech isn't the problem, it’s the attachment.

Jun 11 at 6:27pmPOSSE Post
Show threadalleycat🏳️‍🌈Jun 11
@mtift @pluralistic Being over 35 means learning the long-term consequences of the tech invented in 1. and 2.
Show threadChris HodgesJun 12
@alleycat @mtift @pluralistic
One thing that's interesting with AI is that a large chunk, maybe even the majority, of the teen-to-thirties group is really sceptical.
Show threadPeluche CeroJun 11
@mtift @pluralistic Yes, but how do we know that the scaremongering grandma is not actually right
Show threadmtc_ukJun 11
@mtift @pluralistic
Yeah, you know, that’s pretty much horseshit. Vast swaths of technology that I use and enjoy every day were invented long after I turned 35. You don’t have to be an old boomer shaking his fist at the clouds to recognise when something is fucking dangerous.
Show threadWesDymJun 12

@mtift I suppose it depends on the individual. For me, it was closer to age 55. At that point, I started to see it all as an endless river of mostly pointless BS. Mostly. There's some good stuff, but the vast majority of it is ephemeral crap no one will even remember 10-20 years from now.

Remember SyQuest drives? CueCat? The Clapper? Hypercolor? Yeah.

Show thread🇳🇿   💉*10 RogerJun 13
@mtift @pluralistic hmm. I first encountered cloud services and Kubernetes at 60, I liked this way of doing things so I skilled up and worked in that space for most of a decade. Kind of over that stuff now, but I'm getting into Rust which is fun, a new languge for my 70s. Not touching AI, though. I did enough of that in the 90s and 00s.
Show threadcyberveganJun 13

@rogerparkinson @mtift @pluralistic I have to say that "devops" was cool - VMs and containers are great - but when K8S and Docker became popular, it was always evident to me that it was overkill for small to medium orgs, in fact makes things a lot harder, more expensive and risky (due to data security problems like the CLOUD act etc.). I felt duty bound to learn it, but hated it.

Edit: particularly when hosted on cloud infra.

Show threadTodd KnarrJun 13

@cybervegan @rogerparkinson @mtift @pluralistic Only if you get your physical servers from someone else. If you run your own bare metal, K8S is great for minimizing the hassles of administering physical servers and streamlining provisioning of VMs.

To me containers are just a VM configured to run just one service, so you can skip all the multi-user and multi-service init crap.

Show thread🇳🇿   💉*10 RogerJun 14
@tknarr @cybervegan @mtift @pluralistic agree. I moved from AWS and Google onto a local cluster. The local cluster was more fun, and we had more control over it. These days I'm doing more standalone things on tiny servers so k8s and even vms are overkill.
Show threadcyberveganJun 14
@tknarr @rogerparkinson @mtift @pluralistic I'm kinda old school - cut my teeth on Sun OS and Netware in the early 90s, so I think that colours my view somewhat. I prefer physical tin but have used jails (on BSDs), OVM and KVM and docker on Linux, VMware, LPAR (IBM iSeries) and a few I can't think of right now. I set up a k8s home lab during COVID lockdown, got it working. Did a couple of docker online courses. Concluded that for most uses, Kubernetes and docker ops are really only beneficial for much larger orgs where you need a whole farm of servers for each role with dynamic scalability. If you just want a redundant web server + db server, it's overkill: you're much better off with proxmox be and a handful of vms.