It's been like 24 hours since the last one. Any new critical vulnerabilities I need to deal with yet?

#infosec

A view from near the #Punggol library.

#Haze #Singapore

Why add encryption to ed? That's what crypt(1) is for.

There is no fog. They grey haze is caused by fires in Indonesia. For once, this is not a climate problem. These are illegal fires started by humans clearing land for illegal palm oil plantations. Not sure if that's better.

Best Feuerschwanz cover?

Gangnam style
Ding!
Gimme gimme!
Blinding Lights

#feuerschwanz #music

I saw the Vasa today!! absolutely cool af as museums go— it has tragedy, comedy, adventure, plenty of history, just a great way to spend an hour

@loke A robot could do it even better, but that would not be interesting to watch. It's the human we love to see. You are free to draw parallels here.

How? This is like a magic trick that doesn't hide anything. How long does it take to be this smooth?

https://youtu.be/EcDJJSKs2o8?is=bM1Ul2AKd8eJ7uKd

Stoibers #Transrapid-Rede hier glorios illustriert, ab 02:39 min:

https://chaos.social/@frumble/117260995831299981

I wrote a blog post:

https://blog.dhsdevelopments.com/new-feature-in-kap-subscript-index

#kap #apl #programming

When I have a break at work, I like reading old computer magazines. Here's a fun letter to the editor I saw.

This is from Zzap 64, January 86.

As usual when I see maths posts with some fun calculation, I implement a solution for it in Kap.

Now, why am I posting this? Well, the Kap engine has primitives for modular exponentiation since they are needed for the prime calculation stuff, but they are not exposed to the language. It would be trivial to expose separate functions for it, but my plan is to make the straightforward code below "just work" by having the compiler automatically determine that the the entire expression is performed mod s*2 (the * means exponentiation in Kap).

So, for now, this thing will run slowly. It takes 1 second to compute the results up to 10000 (that's in the JVM version. It's much, much slower in the javascript version), but then it gets exponentially slower due to the large bigint calculations.

1+⍸ { s←(10*⌈10⍟⍵) ⋄ ((s*2)|⍵*⍵) = ⍵+⍵×s } 1+⍳1000

Run the example

https://mathstodon.xyz/@aperiodical/117365341793871594

Thanks to @christianp for the fun blog post.

Here's the game of life glider, running in your browser:

Glider implementation in Kap (click 'Send' to run it)

Array programmers worldwide are eternally grateful to both Conway and Guy for this great way of demonstrating the use of these types of tool.

This is in relation to the 110'th birthday of Richard Guy who invented the glider: https://mathstodon.xyz/@ColinTheMathmo/117360829737969019

I was making potato and leek soup and the recipe asked for five potatoes cubed.

Even for potato soup, 125 potatoes seems a little excessive...

Pretty good video about cycling infrastructure here. https://www.youtube.com/watch?v=ka8kT_YOtU0

OK, this is actually kinda amusing. Some time during all my testing of anubis to make sure everything is served correctly, my browser is now denied.

Any other browser: Works fine. Accessing the same site using the alternative DNS name with the same browser: Also fine. Restarting anubis doesn't help.

I don't know if I should be annoyed or impressed. :-)

#anubis

Wow... After making that post, I sure got a lot of page requests from Mastodon servers. I'm glad those were not filtered by anubis.

Because of this, I took a closer look at my access logs. Indeed, they are much more clean now, but I see this really strange thing.

This is from the nginx access log for https://kap-lang.org/ (as well as the alternative name: https://kapdemo.dhsdevelopments.com/)

92.209.236.167 - - [27/Sep/2026:13:34:00 +0000] "GET / HTTP/1.1" 200 1168 "https://blog.dhsdevelopments.com/" "Mozilla/5.0 (X11; Linux x86_64; rv:156.0) Gecko/20100101 Firefox/156.0"

The interesting thing is that blog.dhsdevelopments.com is just a CNAME to me.writeas.com. So why does this request show up here? Some kind of bot that is testing things? It certainly can't be a normal web browser, could it?

Because of this, I took a closer look at my access logs. Indeed, they are much more clean now, but I see this really strange thing.

This is from the nginx access log for https://kap-lang.org/ (as well as the alternative name: https://kapdemo.dhsdevelopments.com/)

92.209.236.167 - - [27/Sep/2026:13:34:00 +0000] "GET / HTTP/1.1" 200 1168 "https://blog.dhsdevelopments.com/" "Mozilla/5.0 (X11; Linux x86_64; rv:156.0) Gecko/20100101 Firefox/156.0"

The interesting thing is that blog.dhsdevelopments.com is just a CNAME to me.writeas.com. So why does this request show up here? Some kind of bot that is testing things? It certainly can't be a normal web browser, could it?

To get anubis to work with my website (which is using a rather simple ngnix configuration) I ended up editing things that was generated by certbot. I strongly suspect this will cause issues in the future.