(webdev Tory) @worldsendless@qoto.org

@veer66 as in, writing directly in zstd instead of plain text? How do you do that?

Very nice! I didn't know about zstd. Trying it out:

Comparison: tar.gz 1.9G 11 min, tar.xz 1.2G 88 min

33 min, 1.3G result, -19 and all cores:
tar -c -I 'zstd -19 -T0' -f cleaned_files.tar.zst cleaned_files/

2 min, 1.7G result, -11 and all cores:
tar -c -I 'zstd -11 -T0' -f cleaned_files11.tar.zst cleaned_files/

Result: -11 compression on the ZSTD was smaller than the gzip compression and also more than 5 times faster. -19 compression was almost as small as the xz compression, and nearly 3x faster.

Great answers here
---
RT @fndriven
I still don't understand how to properly use JS promises in #Clojurescript. 🤷
https://twitter.com/fndriven/status/1539582431744909312

I have a directory of 529k json files, totalling 9.5G. Compression: tgz (zip): 1.9G. tar.xz: 1.2G. The 700 mb savings took my machine 1 hour 22 minutes to compress; tgz was 11 minutes.

Immutant was a wonderful way to deploy Clojure to Wildfly when it started, but RedHat disbanded the immutant team years ago and for at least 3 years Immutant has been incompatible with Java > 1.08, due to a transitive dependency and a dauntingly complex opensource codebase. Now I am maintaining apps that need a total server shift due to the fact that I used Immutant years ago...

https://issues.redhat.com/browse/IMMUTANT-645?attachmentViewMode=list

Curses. Guix made all my file permissions secure. Now nothing works.

Me clawing my way back to griping about gnome after being stuck in Windows for a week: "Yay! I'm a llama again!"

@souldessin

Yeah, we have the different dev profiles. Probably my desire for prod data privately in dev is my real problem. I have thought of having dev environments as close as possible to the prod one, and may have let that thought go too far

@souldessin does it work as far as data privacy?

TIL something I should probably have known for at least a decade: inheritance in SQL! https://www.postgresql.org/docs/9.1/ddl-inherit.html

zypper! yum! guix! somebody help!
---
RT @programmerjoke9
Now what?#100Daysofcode #javascript #programming #dev #linux #java #programming #CodeNewbie #python #reactjs #bugbounty #DataScience #infosec #gamedev #BigData @programmerjoke9
https://twitter.com/programmerjoke9/status/1537783679564521472

We have a project with some sensitive data. We want the main project to be git* open/public, but we want the data to be somewhere secure and private, but ideally still connected to the first repo for those with permissions. I would like them to be able to just "git pull" or clone and get the sensitive data in the right location. Is there a git convention for this scenario, or should I just make a second, private repo and tell them to manually pull it into the right locations?

Sometimes I play around, and mistakes teach me stuff. Today I learned, don't encrypt your main drive if you don't need to. It's not worth it if you have a pretty static location.

In a strange complication, I can't get my machine to see my USB boot disk, made with Rufus. Is it the format of the disk? Is it the encrypted boot process? Why is BIOS "boot from USB" not seeming to do the trick?