zingar

13 ms·

241.
by zingar 10mo ago
Yup, tolerating inappropriate tech debt for too long leads to bad results. I see this going wrong all the time and I thought it was mostly due to being surrounded by bootcamp grads given nothing but year after year of pressure and no exampl
242.
by zingar 10mo ago
There’s another side to this I see - and have been guilty of committing - where “it took so long because I found some code that felt wrong and while trying to fix that I found something else that felt wrong and while trying to fix that…” at
243.
by zingar 10mo ago
I’m shocked to hear how different your experience is from mine. I very often see good code and bad code and the difference is very clear, even years into a project.
244.
by zingar 11mo ago
The author’s position seems to be that “actual partition” in this sense is a very unusual case for most cloud applications.
245.
by zingar 11mo ago
Can you spell this out for a newb like myself? It looks like you’re saying that a read replica that can’t be read from is still useful, but that doesn’t sound right. What is its usefulness if I can’t read from it?
246.
by zingar 11mo ago
With I assume full size display, keyboard, and full access to permissions. These are the real bottlenecks.
247.
by zingar 11mo ago
Caveat: all this is on iOS: The only reason I want emacs on my phone is the one thing I don’t have: I want my org notes to be on both desktop and mobile. But syncing files across both has been dreadful, even in paid apps: duplicates everywh
248.
by zingar 11mo ago
Same, I think this one post may have cured me of a life long (unrealized) obsession with working at FANG.
249.
by zingar 11mo ago
What are your servers and or people to follow? My mastodon timeline is a wasteland
250.
by zingar 1y ago
What’s the difference between the claim that they’re making and what you say they haven’t done?
251.
by zingar 1y ago
Your comment implies that it’s obviously not this spec that they compare against. Could you spell it out for the ignorant like me? What about that config makes it definitely not the thing that is 86x slower?
252.
by zingar 1y ago
E2E tests in a high ratio to other tests will cause problems. They’re slow and brittle and become a job all on their own. It’s possible that they might help at the start of debugging, but try to isolate the bugs to smaller units of code (or
253.
by zingar 1y ago
Big +1 to customizing emacs! Used to feel so out of reach, but now I basically rolled my own cursor.
254.
by zingar 1y ago
Nitpick: magit for emacs is good enough for everyone whom I’ve seen talk about it describe as “the best git correct” and it is completely free.
255.
Falsehoods software teams believe about user feedback(thoughtbot.com)
1 points by zingar 2y ago|0 comments
256.
by zingar 2y ago
I mean tests of any kind. The main reason I see for tests not existing is because the engineers don't have the skill to write effective tests fast enough (and after a while in that state, just deciding that tests aren't worth it).
257.
by zingar 2y ago
Article author here, this is a really good point and I can see how that exchange might undermine the point (or even the entire article) for some readers. I think I understand it as fighting fire with fire: that people who are enthusiastic a
258.
by zingar 2y ago
I do encounter hostility but what I've realised is that it comes from experience of refactoring being done badly. "We made time for this refactoring thing but it made no difference so now we're not going to 'refactor&#x2
259.
by zingar 2y ago
Original author here :) > Is anything ever really just a refactor, though? I think you're making the point that very few worthwhile business goals could be achieved purely by refactoring. Which is true. Some part of the work will pr
260.
by zingar 2y ago
Great point about team mental context! Without knowing the context I would guess that this a bad refactor (away from revealing the intention)... probably including some cases of applying new abstractions that don't fit the real problem
261.
by zingar 2y ago
I suppose theoretically this must have been behaviour change that crept in with the refactoring, that luckily had a good result and not a bad one. I get nervous about things that accidentally work but I definitely relate to being so tired o
262.
by zingar 2y ago
Could you elaborate on this?
263.
by zingar 2y ago
I personally don’t like VS Code although if this works well I might open it just for this feature. I’d still be coding and taking notes in emacs though. Any thoughts on putting most of the smarts into a server process so that any editor can
264.
by zingar 2y ago
Seems like there is a lot of potential here, although a hard sell up against the licensing fees of products from e.g. Jetbrains. I’m sure yours will be better at its one thing than a more generalised product but it’s a tough market. Compare
265.
by zingar 2y ago
Interesting point! You’ve convinced me that something like this done well could actually be more valuable than code generation.
266.
by zingar 2y ago
I don’t think I’d draw a diagram by hand with an explicit “frontend” label, I think I’d most likely leave it without a label. If I had to choose I would leave it labeled as “UI”.
267.
by zingar 2y ago
It’s interesting to see my industry described from a different point of view. It misses the key point of the economics of subscriptions: a single up front fee is a disincentive to the constant maintenance that delivers the single biggest pi
268.
by zingar 3y ago
Is privateGPT what you're looking for? It stores PDFs in a vector DB and answers questions based on that content alone. Edit: Everything is local. This is not training, it takes a standard LLM and gives it extra context based on what t
269.
by zingar 3y ago
Not quite true, you can get magit to work through tramp and you can tell emacs how to run other process that are remote too. But if the status quo remains as it was in 2022 when I last used codespace, it is a lot of configuration and there
270.
by zingar 4y ago
But isn’t copilot itself a fuzzy llm? Many of its solutions don’t seem formally verified.

More ›