(webdev Tory) @worldsendless@qoto.org

My realization is the past year is that unbounded size is bad, whether orgs, code bases, government, anything. Define your boundaries. Choose your limits
---
RT @tom_geraghty
It strikes me that, with a few exceptions, the larger an organisation is, the more "work" it takes simply to work there.

The overhead of communication, training, bureaucracy, and constant re-orgs means that less and less individual time is spent delivering valu…
https://twitter.com/tom_geraghty/status/1517055405280698368

Easy troll -- "HTML is 4-letters!"
---
RT @scottmcnealy
I know this one.
https://twitter.com/scottmcnealy/status/1516983361750794240

Code is a liability, though I love it. The poetry of code is in its scarcity.
---
RT @searls
Spent time this week validating it was safe to delete a bunch of reporting code that (1) worked, (2) was correct, (3) I spent a lot of time creating, and (4) I was personally proud of. Nobody was really using them.

Why delete them? Because code is a liability and not an asset.
https://twitter.com/searls/status/1516806149940686856

A big "amen"
---
RT @AliciaRaeburn
A remote work benefit I love the most is the lazy morning. I don’t set an alarm. I don’t fight with crowds or traffic.

I walk my dog and roll into the day. Why in the world would this be seen as negative?

Like, am I lazy or do you just hate yourself?
https://twitter.com/AliciaRaeburn/status/1516771481895776261

@Parienve @turak
Wow. Motherload of the ideas I've been looking for! Let me see if I understand the workflow right:

1. create Pass location, in which
2. every file represents one password
3. git-control the whole location, so eg /Pass/.git

Is it a decision of tomb vs git? The tomb bit is the one I'm having a harder time piece together. Tomb encrypts entire directories, right?

@Parienve @turak passtomb and passwordstore were both new to me. Thanks for the references!

Check out this Meetup: Clojure Web Development Evolved (by Dmitri Sotnikov & Nik Peric) https://www.meetup.com/London-Clojurians/events/283500451 #Meetup via @Meetup

@turak Totally agree about not including sensitive information in code repos. These would be private (re: not even Github/lab) repos. But I have essentially avoided git for these files because it simply doesn't work. Is there any kind of internal solution that would maintain public encryption, but allow git-like reviewable synchronization of encrypted files?

I would like to use git to back-up, vc, and sync everything crucial on my text-based system, but some content, most notably my passwords file, are gpg encypted. This breaks git because reviewing changes essentially produces comparisons of gibberish. Does anyone know a solution for syncing/VC of encrypted data, which won't be only passwords?

https://www.jamesshore.com/v2/blog/2005/microsoft-gets-tdd-completely-wrong Nice old article on Microsoft getting TDD wrong in ways that many others have mistaken, too. Essentially, they suggested "waterfall-style TDD" which is a contradiction of terms. #TDD is meant to be fast and iterative. It is an absolute joy in #Clojure Cider (and presumably other toolings and languages), where you re-run the failing test with every re-evaluation, watching for that joyful green light to say that now you've passed. https://www.jamesshore.com/v2/blog/2005/microsoft-gets-tdd-completely-wrong

@skyblond so detached heads should be thought of as "read only", where other branches are ready to be changed and pushed?

Check out this Meetup: The Secret Art of Storytelling in Programming (by Yehonathan Sharvit) https://www.meetup.com/London-Clojurians/events/285388305 #Meetup via @Meetup

@skyblond but why not just make it an ordinary branch? Why headless?

Why does #Git have a detached head state? This seems like nothing but a "gotcha" to make you frantically search for lost code when you thought you were working on someone else's branch.