> Humans have a tendency to ascribe intelligence to how well spoken a person or thing is
That’s true. I’m fluent in German, but there’s still a difference between me and a native speaker. I’ve often seen my ideas dismissed, only for the exact same point to be praised later when a native speaker expresses it more clearly.
Probably a diversification play and a play to see out bigger contracts. If you've worked in the FEDRamp space, you may be aware that Wiz (last a checked, a year or so ago) is one of the few and possibly ownly player certified to operate in FedRAMP Medium/High deployments operating with the technology it does (eBPF instrumentation).
There is a reason for hoarding data: it’s an asset on the balance sheet. So long as it is legal to liquidate data for cash, there will be incentives to collect and keep it.
Or at least make it a liability on the balance sheet rather than an asset. Sure, you can store as much user data as you want. Oh, what's that, if it leaks you owe each user $10,000 under the new law?
Well, in the case of a company trying to market to you, it literally _is_ their business. It makes them money.
The problem is that we have markets where we:
- Incentivize organizations to pursue profits at the expense of everything else, which includes social good and civic rights
- Rarely hold bad actors accountable (and almost never in a timely manner)
Which means, given enough time, we're always going to trend to whatever makes the most money. Targeted advertising makes money, and will continue to do so unless or until we collectively decide to make it a greater risk to profits than it is today.
I'm still salty that I can't use em-dashes anymore for fear of my writing being flagged as AI generated. Been using them for years—it's just `alt+shift+-` on a Mac keyboard and I find them more legible in many fonts compared to the simple dash on the typical numpad.
It's so sad to me that good typographical conventions have been co-opted by the zeitgeist of LLMs.
LLM fatigue is real. It's not just em-dash — it's the overall tone of the writing that clues people in. But if your viewpoints and approach are unique, your typesetting won't raise suspicion of machine-generation, except in the most dull of readers. Just be you and it will be fine.
If you'd like more tips on writing I'd be happy to help.
Magical signal panacea searching is ultimately fruitless. Other ways to make bot interactions more difficult, there are policy and technological obstacles that could be introduced. For example, require an official desktop or mobile app for interaction. And then for any text copy-pasted, demarcate it. And throw an error message for any input typed inhumanly-fast. Require a micropayment of like $0.10 to comment. While these things would break the interaction style and flexibility for a lot of innocent human users, these would throw big wrenches into some but not all vulnerabilities of bot interactions.
Before the wide adoption of Unicode in mainstream operating systems, quite a few people used -- (two ASCII minus signs) to differentiate between a hyphen and a dash (of either pedigree), and some people used -- in emails and online where a dash was required.
Most think that it came from TeX, which had -- (for an en dash) and --- (for an em dash, although I don't think I have ever observed it out in the wild outside TeX), but in fact, the habit well predates TeX and goes all the way back to typewriters where typists habitually hit two hyphens in a row to approximate an em dash. The approximated em dash was described in hard-copy manuscript preparation rules such as The Chicago Manual of Style.
So, if you have ever used a typewriter or TeX, you can claim an even richer than 20 years’ heritage of using the em dash.
I'm exactly the opposite. It'd been on my todo list for years to one day learn the difference between the different dashes. I kept putting not doing it.
Then came LLMs, and there was so much talk of them using em dashes. A few weeks ago, I finally decided it's time and learned the difference. (Which took all of 2 minutes, btw.) Now I love em dashes and am putting them everywhere I can! Even though most people now assume I'm using AI to write for me.
In a lot of ways, it feels like this is simply a fight for recognition that the Mac keyboard supports emdashes.
This wouldn't be an issue if mobile users or Windows users were exercising it too, but it's just Mac owners and LLMs. And Mac owners are probably the minority of instances where it is used.
i've always used double dashes -- because i once i setup a osx shortcut to change those into em-dashes, but i never bother to setup this again in other computers.
so now, i just use double dashes for everything.
(shit, i wonder when llms will start doing this instead of normal em)
I read a text from the 60s by my grandfather this week and seeing an emdash made the LLM alarm in my head go off... Had to really stop myself before I went all "and you" on him
Office Space jokes aside, you shouldn't. I think it's very important to be yourself and refuse to let people pressure you into changing for no good reason. I am not an em dash user myself, as it's a pain to generate when there's no key on the keyboard for it. But if I were, you best believe I wouldn't change my style one bit. People can accuse me of being an LLM if they wish, but that's no skin off my back.
I switched to semicolons... They look similar enough in use to string things together. I'm sure AI is coming for those too though, and that will be a grim day because those are my last stand.
There are times when an em dash can be used in place of a semicolon, but I don't think that's the usual LLM usage. Instead it's replacing a replacing a comma, colon, or period.
Unless you're talking about restructuring your sentences to allow for a semicolon; that's fine.
For example that semicolon could have been an em dash, but I don't think it's the type that LLMs over favor.
My interpretation of LLM em-dash use is that it's like an aside, which is pretty much always going to be weird if converted to a comma since the punctuation was providing un-relatedness information.
People will accuse of all types of stuff, regardless if you use em-dashes or not. The way I write apparently is familiar to some as LLM-jargon they've told me, I'm guessing because I've spewed my views and writings on the internet for decades, the LLMs were trained on the way I write, so actually the LLMs are copying me! And others like me.
But anyways, you can't really control how people see your stuff, if you're human I think the humanness will come through anyways, even if you have some particular structure or happen to use em-dashes sometimes. They're so easy to prompt around anyways, that the real tricky LLM stuff to detect by sense and reading is the stuff where the prompter been trying to sneakily make them more human.
LLM adopting conventions (typographical or otherwise) is what they do, right? The idea that anyone should then have to change their behaviour is ridiculous, as is the whole conversation, really.
The issue is that LLMs adopt a very particular style that is a mix of being very polished (em-dash, lists-of-three, etc) that is reminiscent of marketing copy, and some quirks picked up from the humans curating the training data somewhere in Africa
If AI was writing like everyone else we wouldn't be talking about this. But instead it writes like a subset of people write, many of them just some of the time as a conscious effort. An effort that now makes what they write look like lower quality
I think this is interesting in that I feel, grammatically and structurally, LLMs often generate _higher quality_ text than most humans do. What tends to be lower quality is the meaning of said texts.
Say what you want about marketing-isms of your typical LLM, they have been trained and often succeed at making legible, easy to scan blobs of text. I suspect if more LLM spam was curated/touched up, most people would be unable to distinguish it from human discourse. There are already folks commenting on this article discussing other patterns they use to detect or flag bots using LLMs.
I mean, yes, LLMs write grammatically perfect, well-structured English (and many other languages prevalent in their training sets). That's exactly why many people are now suspicious of anyone who writes neat, professional-style English on the internet.
I totally agree. When I use em-dashes in my /family iMessage thread/ I get accused of having used ChatGPT to write my reply—my one-sentence reply about dinner plans. Dear Lord.
Funnily enough I've actually started using them a little — it made me realise how much more legible/likable I find them.
(Until a few years ago I probably mostly only saw them in print, and I suppose it just never occurred to me that I liked them in particular vs. just the whole book being professionally typeset generally.)
I feel the same way. I've used em-dashes in my writing forever, and I was always particular about making sure they were used properly (from a typography standpoint with no surrounding spaces).
But now, I have to be so picky about when I use them, even when I think it's the perfect punctuation mark. I'll often just resort to a single hyphen with spaces around. It's wrong, but it doesn't signal someone to go "AI AI AI!!"
> I'm still salty that I can't use em-dashes anymore for fear of my writing being flagged as AI generated.
I've typeset books (back in the QuarkXPress days, before Adobe's InDesign ruled the typesetting world) and never bothered with em-dashes. Writing online is, to me, a subset of ASCII. YMMW.
But the one thing I don't understand is this: how comes people using LLM outputs are so fucking dumb as to not be able to pass it through a filter (which could even be another LLM prompt) that just says: "remove em-dashes, don't use emojis, don't look like a dumb fuck".
Why oh why are those lazy assholes who ruin our world so dumb that they can't even fix that?
I mean, LLMs aren’t making people sniff around for typography as though that’s a reliable proxy for humanity.
Em dashes, semicolons, deftly delving. It’s all just so…facile. We might as well tell ourselves we can tell it’s shopped from the pixels, having seen some shops in our day.
are there really places that a comma, super-comma; or (parenthesis) dont work roughly as well? I find the em-dash mildly abhorrent, even before this all.
This is the first time I've ever heard the character ";" referred to as such. It's always been "semi-colon" to me, is this a region/culture difference?
I'm not saying you're wrong, I find it interesting.
A poster commented that he read parenthetical remarks in an old-timey voice (I’d guess the trans-Atlantic accent). I love that idea. But for me they read almost as if you’re saying them under your breath (or a character is breaking the fourth wall and talking to the camera quietly). I read them but my brain assigns them less importance.
Em-dashes keep everything on the same level of importance in my brain.
Commas don’t feel as powerful. To be fair to the comma I’d probably do this:
Em-dash matches how I speak and think: A halt, then push onto the digression stack, then pop. So I use them like that.
Edit: I accidentally used an em-dash in the word em-dash. Interestingly HN didn’t consider changing the dash to be a change in my text so didn’t update it. I had to make a separate change and take that change out for my dash change to stick.
For me, a sequence of sentences, strung together by commas, is more in line with how I output thought, and better matches what I believe my speech pattern is.
I picked it up from Salinger. I find that if I can't eradicate parenthesis by some other means, or if it's more effort to do so than I want to spend, em-dashes usually replace them without doing any harm and aren't quite so ugly, aside from being useful in other cases. In particular, parenthesis at the end of a sentence are awful, while a single em-dash does a similar job much more neatly and looks totally natural.
I suspect this is partly due to the quality of documentation for Elixir, Erlang, and BEAM. The OTP documentation has been around for a long time and has been excellently written. Erlang/Elixer doc gen outputs function signatures, arity, and both Elixir and Erlang handle concepts like function overloading in very explicit, well-defined ways.
* Largely stable and unchanged language through out its whole existance
* Authorship is largely senior engineers so the code you train on is high quality
* Relatively low number of abstractions in comparisson to other languages. Meaning there's less ways to do one thing.
* Functional Programming style pushes down hidden state, which lowers the complexity when understanding how a slice of a system works, and the likelyhood you introduce a bug
I've seen but haven't used CEL. Anybody with experience with competing tech have any strong opinions? I've used OPA, know CEL used by GCP and Kyverno, but otherwise haven't seen anything compelling enough to move away from the OPA ecosystem.
And even then, I'm not sure it's apples to apples, at least if by Rego you're thinking of OPA. CEL and Rego take very different approaches, with CEL being quite procedural, while Rego is about constraint satisfaction, not unlike Prolog. At $WORK, Rego (in the form of OPA) gets used quite a bit for complicated access control logic, while CEL gets used in places where we've simpler logic that needs to be broken out and made configurable, and a more procedural focus works there.
This guide has aged surprisingly well, but I’d add to this: the above response is about as good as you can get—it is firm, non-combative, and moves the conversation forward.
Don’t antagonize your recruiter. You want them to advocate _for you_ when a prospective employer is drafting an offer. Work with them to give them the ammo they need to make that happen.
Would you say I was antagonizing him with my response? Because he was an in-house recruiter (and not a headhunter) and I got along with him pretty well.
Vogue did a decent overview of this[1] and history is littered with all kinds of examples if you go looking.
1. https://www.vogue.com/article/dark-history-of-the-far-rights...
reply