Consider pair work (a generalization of pair programming):
Spin up a VM for you and the candidate to SSH into. Start a shared screen/tmux session. Work on some sysadmin task(s) together. Switch between driving/typing and navigating/directing at regular intervals (15 minutes is typical, though a 30 minute interview would require a <10 minute interval).
You'll have to come up with a task (or set of tasks) that deals with the skills you need. The task(s) would ideally be very difficult for an LLM or search engine to provide a complete answer for; but they shouldn't be gotchas, since that might be false negatives.
Test your task(s) with other people in your company first, so you'll be comfortable when doing it with a candidate.
I can't say that this will 100% prevent cheating. But, you will surely learn a lot about the candidate's habits, in terms of reasoning and communication. If the candidate pastes commands, then that's a possible indication of LLM-usage; if the candidate doesn't use tab-completion, then they might be typing text in from somewhere else (e.g. an LLM, a guide).
I enjoy using the command-line, so I suspect I would enjoy this sort of interview.
Good post. I've definitely experienced this, though a sub-form of it, namely my non-conformance to a sterotype of an ideology I am associated with. The example in TFA doesn't appear to involve any stereotypes, because Bailey is talking about people in general. [0]
> hypocrisy by association
You had it right in your title and elsewhere in your post. Bailey isn't accusing Khetan of hypocrisy, which is a difference between one's words and one's actions; the example in TFA is about a difference between a group's words and an individual's words, which is contradiction; actions-vs-words is not being discussed. It's also not self-contradiction, since Bailey isn't accusing Khetan of contradicting his own earlier statements. [1]
This reminds of the phrase, "stay in your lane": "stay in" here means "speak according to a certain viewpoint"; and "your lane" is the beliefs of some group. "Stop disagreeing with people similar to you," is a ludicrous thing to say. It's even more ludicrous when "people similar to you" is "people in general"; there's billions of "people in general"; that group disagrees on every topic known to man and dog.
"Contradicting humanity" wouldn't sound pithy, but that's what Bailey is accusing Khetan of doing. To your point, Bailey is accusing Khetan of "contradicting your group", which is undoubtedly a form of the Association Fallacy. [2]
Bailey is also engaging in performative contradiction, by demanding that Khetan agree with "people in general", while Bailey is not himself agreeing with "people in general", because that is an impossible task.
[0] Applying a stereotype to all people would be absurd, since a stereotype is about a sub-category of people; a stereotype is meant to highlight alleged differences between categories of people; to highlight the differences between A and A would be absurd -- there are none.
[1] Yes, Khetan is a member of "people in general", but Bailey is comparing Khetan's words to the words of "people in general minus Khetan" -- instance versus class-minus-that-instance. Okay, maybe Bailey isn't thinking that deeply about this, hence why TFA needed to be written.
[2] I could have lead with this, but I don't have enough time to re-write this.
>This reminds of the phrase, "stay in your lane": "stay in" here means "speak according to a certain viewpoint"; and "your lane" is the beliefs of some group
What? 'stay in your lane' means some variant of 'don't comment on stuff you don't know about' or 'mind your own business', it's got nothing to do with speaking in unity with a group.
This sounds like a quasi-Gnosticism: the human body is lower and human rationality is higher; the lower as inhibiting the higher; the best for humans is to transcend their bodies by some means (albeit un-stated in TFA). It is a tempting view, if only because the human body seems frail in comparison to rationality.
It is a QUASI-Gnosticism, because it is a form of rationalism, which is incompatible with Gnosticism; Gnostics valued knowledge as a means, not as an end; they wanted to transcend every aspect of the human body, which includes the rational faculties.
Lovecraft certainly isn't a Gnostic, since total transcendance of the physical world would include (as stated above) transcendance of his cherished rationality. And Lovecraft isn't a quasi-Gnostic: his pessimism is paramount to his stories; do any of his characters ever escape the horrors of his stories? Gnosticism and Xenohumanist's quasi-Gnosticism/xeno-humanism are both fundamentally optimistic; the promise of transcendance is allegedly possible for anyone who acquires relevant secrets (in Gnosticism) or uses the relevant means (in xeno-humanism). I doubt that Lovecraft had room in his heart for such optimism; that would require him to be open to the possibility of the transcendance of Irishmen and blacks.
> All photographs or videos you have seen have been placed there for you to see them.
For those reading the above, wondering about this phenomenon, read Baudrilliard's The Gulf War Did Not Take Place. Even if you don't agree with Baudrilliard's overall thesis, the facts he brings up are still cogent (e.g. a photo from the Exon Valdez spill was used instead of an actual photo of the Iraqi military's destruction of Kuwaiti oil fields). The media has been a critical aspect of war since at least the Falklands War.
The word "decentralized" has become synonymous with "cryptocurrency", especially those cryptocurrencies which promise they will decentralize things unrelated to money production (e.g. data storage, computation, etc). Such cryptocurrencies rarely deliver even a fraction of their promises of decentralization; most of these are outright scams.
It is unfortunate, because there is still great potential for new forms of decentralization, but many are blinded by such shovelware cryptocurrencies.
The actual title of the acticle is "Brad Lander Detained by Masked Federal Agents Inside Immigration Court".
Contrary to the current title here on HN, Lander was not arrested for asking to see a warrant; TFA states the opposite, "It wasn’t immediately clear what charges, if any, the mayoral candidate will face. A spokesperson for ICE didn’t immediately return a request for comment."
If an event is so important to know about, why fabricate such an important aspect of the event in this way?
If we want to stick to the facts: we don't actually have any proof that these were federal agents because they refuse to identify themselves. All we actually know is that Lander was kidnapped.
This.. What is been happening lately is absolutely batshit crazy. Now anyone could mask up and arrest some key witness right from the courthouse posing as ICE agents, regardless of their status and nobody could bat an eye because ICE seem to have some kind of supreme auhority and no law applies to them, they don't need to identify themselves, even show their faces.
He was, in fact, arrested for asking to see a warrant, that is clearly documented.
The claims of assault that DHS fabricated and published on social media and via other channels after the fact to justify it, of which there is no evidence, before Lander was released without any charges are interesting in terms of understanding the current regime's propaganda propensity, but have nothing to do with explaining the events clearly captured on video.
I don't think that's what they were referring to. From watching the video, I assume it was when he grabbed onto the fellow they were detaining and refused to let go.
If you witness police misconduct, you do not have the right to impede the police. Regardless, there was no way for him to form an accurate belief about whether misconduct was occurring. The agents have no duty to provide a warrant to a bystander, even if he is a government official.
The other commenter mentioned "narrative", which is very relevant, because that is an important part of simulation (and your username)
Baudrilliard was careful to point out that simulation isn't a matter of fabrication; to simulate is to obscure the absence of facts, not to create false facts. A simulacrum is a symbol that obscures the fact that it refers to nothing; whereas a symbol, in centuries past, invariably referred to something, real or imagined. The resulting reality (or maybe "mindspace"?) is a construct on top of the real world -- a hyper-reality -- in which every symbol is a simulacrum; the only thing real in hyper-reality is that the symbols hide the absence of facts. This is why, again as the other commenter mentions, we appear to live in a post-truth society; we are fully living in hyper-reality.
>Bad Journalism
The guy who created the Pullitzer prize also co-invented Yellow Journalism.[0][1] There is neither good journalism or bad journalism; it's all simulation.
> the only thing real in hyper-reality is that the symbols hide the absence of facts
What's a fact? Concepts like justice and fairness are fundamentally cultural constructs, and yet they've always been a core concern of human society. Setting up "facts" in opposition to "simulation" is no less a rhetorical narrative than what the article is pushing.
My takeaway from post-structuralism generally isn't that we live in a "fake" reality, but that the human experience--individually, collectively--is deeply complicated.
Baudrilliard didn't assert that reality/facts never existed; he in fact asserted that prior to the 20th century, there was plenty of correlation between symbols and facts/reality. His vision of the hyper-real is that it is detached from reality and it's facts; this is why I included "mindspace" parenthetically as an alternative word for "hyper-reality"; those operating in hyper-reality are physically in reality, but their actions appear to be based on another world, which they share through things like news media.
> post-structuralism
I don't think Baudrilliard can be categorized as post-structuralist or post-modernist, because "Baudrillard had also opposed post-structuralism, and had distanced himself from postmodernism."[0]
Camus didn't consider his views existentialist, but I consider them as such. Likewise, for me authors like Baudrillard and Benedict Anderson (political scientist) have written works that I think well capture the substantive gist of post-structuralism, even if neither saw themselves as part of that intellectual movement (and few if anyone else would relate Anderson to post-structuralism).
Plus, AFAIU Baudrillard turned into an angry, cynical, conspiratorial old man, kinda like a teenager who discovers the world is far more complex than the simplified versions he was taught, and then becomes angry at the world for being hoodwinked, as well as at everybody's complicity. IOW, some of Baudrillard insights are powerful, but I don't care all that much about how he chose to make use of them. (That said, the radical and exaggerated way he conceives of and presents things lends much of that power.)
I've never read any of Baudrillard's books, though, just several of his essays.
I read all Baudrillard's books, back when he was relevant in the day. You've not missed much. In fact you seem more insightful than those who do have them all to set.
Everytime news like this comes up, I'm never affected. I'm using uBlock Origin. Is there something unique about uBlock Origin? Are other adblockers easily noticed by YouTube?
TFA mentions complaints from users of Brave browser, which appears to use it's own bespoke ad-blocker [0]. It includes some filters from uBlock Origin [1], but these files are six years old. Are those filters updated at runtime (e.g. from [2], [3], [4], [5])? If so, then why include old filters in the repo? Surely, Brave isn't using six year old filters.
Roko's Basilisk has been replaced by Altman's Basilisk. Where once we feared a computer torturing a digital copy of us (Roko's Basilisk), we now fear a computer eliminating all our jobs (Altman's Basilisk). The former has been forgotten, because losing one's job is one step away from losing one's home, which is one of more serious secular deadly sins you can commit in the 21st century.
I wait with baited breathe to see what people will come up with to replace Altman's Basilisk in ~15 years.
FTFA: "Unless you've devoured several cans of sardines in the hopes that your fishy breath will lure a nice big trout out of the river, baited breath is incorrect."*
> The end result of that was that we had to significantly rearchitect the project for us to essentially inject manually crafted code into arbitrary places in the generated code.
This sounds like putting assembly in C code. What was the input language? These two bits ("Not AI generated", "a feature flag") suggest that the code generator didn't have a natural language frontend, but rather a real programming language frontend.
Did you or anyone else inform management that a code generator is essentially a compiler with extra characters? [0] If yes, then what was their response?
I am concerned that your current/past work might have been to build a Compiler-as-a-Service (CaaS). [1] No shade, I'm just concerned that other managers might read all this and then try to build their own CaaS.
[0] Yes, I'm implying that LLMs are compilers. Altman has played us for fools; he's taught a billion people the worst part of programming: fighting the compiler to give you the output you want.
[1] Compiler-as-a-Service is the future our forefathers couldn't imagine warning us about. LLMs are CaaS's; time is a flat circle; where's the exit?; I want off this ride.
The input was a highly structured pdf specification of a family of protocols and formats. Essentially, a real language with very stupid parsing requirements and the occasional typo.
The PDF itself was clearly intended for human consumption, but I'm 99% sure that someone somewhere at some point had a machine readable specification that was used to generate most of the PDF. Sadly, no one seems to know where to even start looking for such a thing.
> Did you or anyone else inform management that a code generator is essentially a compiler with extra characters?
The output of the code generator was itself fed into a compiler that we also built; and about half of the codegen team (myself included) were themselves developers for the compiler.
I think management is still scared from the 20 year old M4 monstrosity we are still maintaining because writing a compiler would be "too complex".
Spin up a VM for you and the candidate to SSH into. Start a shared screen/tmux session. Work on some sysadmin task(s) together. Switch between driving/typing and navigating/directing at regular intervals (15 minutes is typical, though a 30 minute interview would require a <10 minute interval).
You'll have to come up with a task (or set of tasks) that deals with the skills you need. The task(s) would ideally be very difficult for an LLM or search engine to provide a complete answer for; but they shouldn't be gotchas, since that might be false negatives.
Test your task(s) with other people in your company first, so you'll be comfortable when doing it with a candidate.
I can't say that this will 100% prevent cheating. But, you will surely learn a lot about the candidate's habits, in terms of reasoning and communication. If the candidate pastes commands, then that's a possible indication of LLM-usage; if the candidate doesn't use tab-completion, then they might be typing text in from somewhere else (e.g. an LLM, a guide).
I enjoy using the command-line, so I suspect I would enjoy this sort of interview.
At any rate, I hope you find a good candidate.