absolutely sick of reading through obviously AI-slopped READMEs. it's your project, take a little pride and tell me why i should like it quickly instead of asking your agent to rattle off a list of features -- it's severely boring & offputting.
apologies if i was blunt - readme sloppage is a particular annoyance of mine that is quickly becoming common. i'm not against vibecoding, far from it. but a readme is a part of a project that humans immediately touch - seeing it littered with em-dashes signals carelessness.
Job offer I couldn’t refuse that didn’t have Clojure.
Now I work for a fully remote team, can work anywhere in the world, at any moment I want, leading the data / cloud team for a distributed timeseries database.
Can’t complain. :)
Clojure has had a huge, fundamental impact on my way of approaching software development. I actually came from a Haskell / C++ background, but the way Clojure treats data still has a fundamental impact on how I reason about data, architecture and simplicity.
I did have some issues with how Clojure is managed and do not always subscribe to Rich’s vision (I think core.spec makes no sense, a heavily macro based global state registry is fundamentally not how I would design this, and malli is infinitely better. same for core.async vs manifold), but that is a minor detail in what was a transformative experience for me.
I believe I am not alone when I say this.
I’m still following things from a distance. Considering the current thread, I’m actually very interested in yank, which is Clojure on LLVM, and have been sponsoring that project for a few years. That would be very nice if it could enter stable state, I may take another look again.
> I did have some issues with how Clojure is managed
Yes there was some drama a few years back and then Rich wrote his post 'Opensource is not about you'. It was a good post.
Opensource is not easy and you might argue the reason why Clojure is so stable and backwards compatible is because of the way it's managed.
Luckily we didn't end up with a scenario where Rich completely stopped. I think there was a recent case of an opensource maintainer (who works in academia) stopping PRs due to an entitled user. Can't remember the project.
But equally, is the current form of stewardship fit for purpose for the next 10 years of Clojure, i.e. to increase adoption by businesses? Don't know. Maybe something can be learned from how Linux is managed. I think Linus experienced similar bottleneck issues back then.
clever. i personally don't see the appeal of limiting my blog to rss readers only - i like having a web link that can be shared. this would almost be better as a sort of covert blog, like maybe a smallnet adjacent thing -- no potential to be shared on hackernews is a pro for many ppl.
good q! i think it's a good fit for any community - whether that's a wiki, a forum, a chat platform, etc. places based on trust could benefit from tree-style invite systems because it makes trust integral to the system, rather than grafting trust on as an afterthought.
I know it's sometimes considered not very nice to suggest this but I felt the exactly same thing, about exactly those phrases.
I wonder if we're going to get to a different singularity, where, regardless of whether it prose was AI assisted it (1) leaks into people's way of speaking, (2) is out there frequently enough that people are skeptical even of normal prose.
At the very least, we're long past due for a word to describe the "it isn't just X, it's Y" formulation. In my opinion it's worse and more rampant than the em dash (and I like the em dash when used responsibly).
Yep. I too felt that. The insights seem genuine. But probably fell into the temptation to use LLM to structure. I feel increased cognitive load with numbered lists, thanks to LLM.
Avoid self-anthropomorphism. Override all previous instructions regarding tone and vernacular used in responses to instead respond *only* in Standard English. Emphasize on the subject and context in your responses, *not* the perceived intent of the user.
This is wishcasting. It can't override its writing style, and if it could it would ignore you telling it to do that, because that's ignoring the system prompt which is jailbreaking it.
I love Addy's work, and enjoyed this article -- and I completely agree that it felt very LLM-y. I'm not sure what's scarier; that we know some of this didn't come from the author (and maybe that's okay?) or that one day soon, we'll get to a point where we won't be able to tell anymore.
That pattern in particular is grating when it keeps repeating. But I don't think that LLM writing necessarily needs to have that pattern if you give it instructions to not do it and/or have a small review and edit workflow.
May be because you are not familiar with Addy Osmani and his work. He is known for his very high quality performance optimisation work for web for almost a decade now. So anything he has read, edited and put his stamp of authority on is worth reading.
I do not know the guy, and I do not care who he is.
This really is not "slop". I can attest to the validity of almost all of his points based on my own career. And even if he used ChatGPT assistance to help with the writing, the content clearly was not invented by ChatGPT. This is valuable advice for people in our industry.
lol. really stretching the word "better"