A well-architected software project is one where the difficulty of adding a feature is proportional to the delta in functionality you want to add. Small tweaks should be easy, and big new features that were not part of the design should not be too scary due to the existing code.

May 31, 2022 · 3:42 PM UTC

11
58
4
331
Replying to @gdb
Small changes should not break anything, big changes should add new functionality without breaking old ones
Replying to @gdb
Can you share an example to drive the point home ?
1
Replying to @gdb
Which only seems to happen through blind luck, or I knew the change was coming.
Replying to @gdb
Not sure why, but I feel that many of your latest tweets are generated by GPT-3 🤷‍♂️They are too ... to the point 🙈 Especially after I read generated tips (on to be productive ) here:
Artem @avysotsky
31 May 2022
As an engineer, I'm terrified of OpenAI. It seems to me, that it can handle a long tail of very simple programming tasks. Mindblowing
1
Replying to @gdb
Somewhat, but how do you form an unbiased view of the measure of the proportion of the delta of adding a new feature? It would have to be compared against other architectures, issue here is adding a new feature to my project website is faster than say to a space rocket.
1
Replying to @gdb
“The Existing Code” by Stephen King.
5
1
18
Replying to @gdb
This. I've had a few instances of engs trying to convince me a suggested (more complicated) architecture is better because of various reasons, and I always push back with the same point of resistance - if it makes eng work take longer it can't be considered a better architecture
3
Replying to @gdb
Greg, are GPT3 and DALL-E well-architected? (A serious question)
2
28
Replying to @gdb
Modular with simple core such that it is fast to add and possible to estimate the impact of the feature addition prior to the final commit
Replying to @gdb
simple things should be simple, complex things will be complex but not impossible
Load more