thock.in
I like writing software, even if I don't get to do it much anymore. Kubernetes. Star Wars nerd, Expanse fan, dad, dork.
112 posts
3,494 followers
201 following
Regular Contributor
Active Commenter
comment in response to
post
Hot damn, in under a minute it came to the same conclusion it took me 4 tedious hours to reach. I am a luddite.
comment in response to
post
Hold up. This actually worked. I have wasted a whole day.
comment in response to
post
The point for me was - you don't need pure-good and pure-bad space wizards and laser swords. People are good and evil on their own. Good and bad isn't black and white (very literally in SW) it's a swirling mess. Good people do bad things for good reasons. Bad people do good things for bad reasons.
comment in response to
post
If there's a round two, please consider me :)
comment in response to
post
Shoot. I had the same idea independently, but you all beat me to it. :(
comment in response to
post
Did this happen? I want in.
comment in response to
post
This vid didn't introduce the idea of a worldline, which can move on both axes, but only within its own future light cone.
comment in response to
post
AIUI they are 45 by convention. They travel thru both time and space, so any other angle would be biased towards one axis or the other. Something that only travels through time, and doesn't move (relative to the origin) moves vertically, and something that moves infinitely fast moves horizontally.
comment in response to
post
For me it's the "we will just redefine the coordinates to make a singularity disappear" part that hurts.
I think a little more introduction on things like why light-cones are 45 degrees would help. I think I got it EVENTUALLY, but I have only 50% confidence I could explain it.
comment in response to
post
Not as compelling as Hawking's books, but still super interesting. He spends a good amount of time on Penrose diagrams, but some of it was lost on me until this video.
Some of it is still lost on me :)
comment in response to
post
Funny, I am currently reading Brian Cox's book on this same topic, and the book is featured in this video, and in the chapter I just finished :)
comment in response to
post
The payoff is glorious.
comment in response to
post
I guess I am lucky that I have never had to use Outlook, ever. :)
comment in response to
post
So... Hotmail ruined it for everyone and GMail just made it worse?
comment in response to
post
Secrets in base64 was never about "encryption" it was about encoding arbitrary data in a transport-friendly way.
Please stop perpetuating that myth.
comment in response to
post
This isn't Star Trek, Ciruli. No benevolent Federation is going to step in. Our world today is far more Star Wars than Star Trek, sadly.
comment in response to
post
It's the great shame of our generation. We simply cannot agree on a single medium to communicate (or even 3-5 of them!). They all suck in mostly the same ways, and finding past-information is utterly hopeless.
comment in response to
post
Also, I have a hard enough time finding things. Did we discuss it on slack DM or a channel or a thread? Was it a GH issue or PR? Did GH hide the msg? Or was it on bsky? Or in the hall at KubeCon? Or internal email / chat?
Every week I go hunting for some discussion I KNOW I had and fail to find it.
comment in response to
post
I half-agree. k/k "Issues" is probably 2/3 stale, already-fixed, not reproducible, or not actionable. That said, I think ACK'ed feature requests need a place to live, and "discussions" ain't it.
Practically, it wouldn't change much for me - "Discussions" is just another source of messages to watch.
comment in response to
post
Is it solving a real problem? Like really really? All it did for me is break things that used to work. If I wanted to use a Mac I would use a Mac. All I want is my windowshade, focus-follows-mouse, and clear borders/widgets. Almost every one of those has been ripped away by "progress".
comment in response to
post
We have used a variety of JSON libraries over the years, but moved away from code-generating versions. Better strategy was to mitigate the role of raw JSON. I don't think we wlooked at that one, or _I_ didn't, anyway
comment in response to
post
On Thursday Kubecon Family Fortune (not Feud!) with me and @thock.in is back! With 8 amazing members of our contributor community vying for the ultimate prize of Kubecon Family Fortune Champion (definitely worth putting on a CV) kccnceu2025.sched.com/event/1txCe/...
comment in response to
post
I am still #8 after more than an hour. I think that poor person went home.
comment in response to
post
Somehow reading this is exactly like having a conversation with you.
In a good way, I mean.
comment in response to
post
I can't tell if it is good satire or real.
comment in response to
post
www.reddit.com/r/kubernetes...
comment in response to
post
Yeah, but it works.
comment in response to
post
It's an ad. I wish NewStack didn't post ads disguised as articles.
comment in response to
post
Your mission, should you choose to accept it: become elected HOA president.
We now have AMPLE proof that it's possible for the president of an organization ot dismantle said organization from the inside.
comment in response to
post
OTOH, maybe Kubernetes will prompt cloud providers to support some form of WATCH operations, just to alleviate the load of millions of kube clusters polling?
I'm an optimist. :)
comment in response to
post
Same in k8s with load-balancers. If you have 100 services with LBs, and someone modifies one behind the scenes, we don't know to reconcile, because (AFAICT) none of the clouds have watch APIs and they all have API quota. Polling 100 LBs every 10 seconds = 10 QPS of 99.999% do-nothing traffic.
comment in response to
post
Family member got them young - they went from almost incomprehensible speech to perfectly normal. It turns out they were speaking what they could hear, which was massively impacted by blockage.
Good luck!
comment in response to
post
Nice notes.
I keep coming back to api-machinery as one of the most critical (& interesting) technical SIGs. It must be possible to do efficient server-side filtering of read ops (a "where clause").
Like it or not, people treat the API as a database, and when it doesn't work that way, fireworks.