Eric retweeted
We need to reconsider our priors on code readability.
Humans, compared to agents, are pretty slow at understanding large quantities of text. So in the old days, readability was important so the human could efficiently grok what was going on.
So readability now comes down to:
1. Can a human, if needed, jump in and understand the code
2. Does the code use the same domain language as the human describing the code?
1 helps with observability and the occasional spot checks you need to do on important PR's. 2 helps the agent translate the words you use into code changes.
In other words, readability is now a nicety, not a necessity.
The way agents 'read' code is different - good agents are all about gathering data about how the code works. They'll prod around with it, execute parts of it.
So 'readability' to agents really means 'verifiability'. How easy it is to mess about with that code to see it working.
Let's take this scenario:
- You have a huge module in your codebase that uses all sorts of complex performance tricks
- It has 100% test coverage
A talented human would need an hour or two to get comfortable working in that module.
An agent needs zero time. They verify it with the tests, poke about a bit, and they're good to go.
The takeaways:
- When judging a module (or a codebase), don't think about how easy the code is to read
- Instead, think about how easy it is to verify
- Design your codebase so modules can be tested and verified independently
- Feel free to sacrifice a bit of readability for performance
- Cyclomatic complexity and function size don't matter to agents
Eric retweeted
Announcing mattpocock/skills v1
- Achieved a 63% reduction in token cost for skill descriptions
- Split skills into model-invocable and user-invocable skills, adding /codebase-design, /domain-modeling, and /grilling
- (UPDATED) /writing-great-skills - rewritten from the ground up, encoding my skill-writing best practices
- (UPDATED) /diagnose -> /diagnosing-bugs - now model-invocable, awesome for fixing hard bugs
- (NEW) /ask-matt: a router skill that teaches you how all the engineering skills work together
I don't often see Reddit agree on things. But a small sample set apparently agree with you @tannerlinsley on directives reddit.com/r/javascript/s/Bo… tanstack.com/blog/directives…
Me? I'm just dealing with fomo here and my SPA 😂
5
308
I have found a #programming language I hate more than #python. It's #bash - omg I miss my #javascript.
1
25
I'll be referring to this when I nope out of a bug fix or feature request.
A codebase shouldn't feel like this, but I see it a lot.
unused imports and packages
rampant copy/paste
half-written features
unclear patterns
commented out code
frequent spelling errors
inconsistency
100s of warnings
no validation
swallowed errors
a general sense of disorder
35
Eric retweeted
real men know the true path to success is being servant to all not money/material success
Eric retweeted
who wants a doodle??
rt, comment ● - ●, and follow me so I can dm you if you're chosen. you can unfollow me after, all good
Shout out to @Netlify for being able to swap deploys with the click of a button. Friday production deploys with unexpected bug?
35
Me two years ago to a junior: I don't think you should #useeffect here.
Junior: it's ok, just let me do it.
-- Junior leaves company --
Me one year later: oh, this use effect causes this bug. I'll work around it
Me two years later: aaaand now I can't work around it. #refactor
34
> Anyway, doom-scroll less.
> Look at cute animals more.
Moar cuteness on interwebs please.
#cutedog
Lots of people are pinging me about the "Adpocalypse on Twitch" and wondering if we're ok.
None of our ad-revenue on Twitch has changed because none of the content we talk about has sexual themes or politics. Our ferret rescue is run entirely off ad-revenue and it's gaining income following the normal yearly trends for October/November.
There are likely some channels that get misidentified as having political content due to tags or categories but this so-far seems to be rare. A tag lacks critical context for the kind of content being provided and many people use tags in joking ways or to identify themselves in a particular way. I don't feel tags make a lot of sense for targeting ads as it leads to splash damage because of this.
An Adpocalypse would be damaging the entire platform as a whole. None of that has been seen yet so there seems to be a lot of jumping to conclusions here. If this was happening in a widespread or apocalyptic manner we'd see impacts far outside of isolated instances like we have.
Anyway, doom-scroll less.
Look at cute animals more.
2
168
Key principles for #productdevelopment
Some engineering principles I live by:
✓ Make it work, make it right, make it fast
✓ Progressive disclosure of complexity
✓ Minimize the number of concepts & modes
✓ Most 'flukes' aren't… your tech just sucks
✓ Feedback must be given to users instantly
✓ Maximize user exposure hours
✓ Demo your software frequently to fresh eyes
✓ Sweat every word of product copy you render
✓ You're never done working on performance
✓ You're never done. Software ages like milk, not wine
✓ Visualizing traces of time is the best way to optimize it
✓ Ship frequently and strive to build in public
✓ Errors must have globally unique codes & hyperlinks
✓ Red is not enough to signal "error" (8% of men have red-green color blindness)
26
I've realised that my disdain for #python comes from over exposure to #SpaghettiCode and unfamiliar APIs. As I write my own code and focus on readability and actually using libs, I start to hate it less.
1
20
I don’t ask for permission to do my job well.
I don’t ask for permission to refactor.
I don’t ask for permission to optimize performance.
I don’t ask for permission to make it secure.
I don’t ask for permission to make it accessible.
I don’t ask for permission to write tests.
I don’t ask for permission to do my job well.
1
44












