I'm finding counterexamples of this constantly now that I can have an agent rewrite large sections of my codebase that have been sorely needing it.
- Moving to a newer and more modern test library
- Refactoring my data layer so it's easier to read, based on years of organic changes that need to be baked in and simplified
- Porting some functionality to another language to vastly improve performance
I agree with the overall sentiment, but having an agent at my finger tips who can really crank out large-scale, involved code changes is unclogging quite a few backburnered todos lately for me.
- Once print newspapers were no longer a thing, even local news outlets are struggling to stay alive, and are resulting to sensationalism and entertainment as news
- Corporate sponsors retain a huge influence in mainstream news (or have outright purchased it and use it for partisan politics).
- "Social" media resides in (you guessed it) corporate-owned walled gardens.
- Even those willing to speak out are being targeted by federal agencies
Wondering where others are finding great places to learn what's going on, what's actually relevant to me, and what I can actually do about it.
Follow trusted journalists that have a history of superb journalism. You have to decide what is good for yourself. Also industry specific journalists for deeper insight into industry you care about. Many of the greats realize that all the news companies are sinking ships and are trying to establish their own thing before things completely collapse.
Focus on what each journalist specializes in and don't read too much into it when they report on topics that are not their forte (like Breaking points talking about AI). Many journalists stay in their lane but the groups covering all the news don't. I wish all journalists would stay in their lane but this is not the world we live in anymore.
For everything else that you are not willing to invest the time in, just accept you are not going to get great coverage.
My (typically far left) biases are comfortable with the following (these are not all far left)
Zeteo + dropsite for foreign middle east leaning coverage
Breaking points for daily news
Ken Klippenstein, Glenn greenwald for national security state/us government news. Klippenstein is more fun (when he gets an FBI email asking for info but also politely asking to not release it, he goes ahead and provides a download link). Greenwald feels a bit more dry.
Industry specific: autoline (youtube+website) for automotive, semianalysis for semi.
Again this is just my opinion, please decide for yourself.
There's a decent amount of non-profit news. I read NPR a lot and donate to them and Propublica. I think one of the big issues is advertising so news you actually pay for is a lot less likely to clickbait you.
There's something "Very HN" to see a comment in a discussion about literal political censorship of the news media pointing out the horrifying hypocrisy in one of those media sources running... something sort of like advertisement.
Is this what kids mean when they tell people to touch grass? My generation would accuse you of having lost the plot.
Being forced to run advertisements because one political party in particular decided being unbeholden to corporate interests is a dangerous thing for a news organization to be.
Someone already mentioned NPR. BBC also does a great job reporting on US and international issues. New York Times still does strong reporting. And there are local sources too, such as the Colorado Sun, LA Times, SF Chronicle, or SF Gate (obviously I’m in the US).
Not all the legacy newspapers are failing; NYT is doing well. There are other news sources beyond legacy newspapers, broadcasters, local news, and social media. There are wire services (AP, Reuters), insider access journalism (Axios, Punchbowl, Semafor), public media (NPR, PBS, BBC), investigative journalism (ProPublica), digital-first outlets (Politico, Vox), and the growing wave of small, indie , creator-led media (YouTube, Substack, Patreon).
NYT as a news organization is, charitably, part of the controlled opposition. They are not meeting this moment at all, through cowardice or intentional complicity I'm not completely sure.
> 4. If the agent + tools flows get too wily, break it down into smaller domains by spawning sub agents with focused prompts and (less?) tools.
I think where this ultimately goes is a "coordinator" sort of model where the top-level agent primarily decides what needs to happen next and which agent is most equipped to handle that task. This could potentially happen in a recursive fashion (e.g. an agent for each product the company makes, that agent can dispatch to a "frontend" agent or a "backend" agent, etc).
That allows the agents that actually "do things" to maintain a limited context and set of tools, and the managing agents only have to maintain context on what their sub-agents can do.
This seems like a much more powerful version of what I wanted MCP "prompts" to be - and I'm curious to know if I have that right.
For me, I want to reduce the friction on repeatable tasks. For example, I often need to create a new GraphQL query, which also requires updating the example query collection, creating a basic new integration test, etc. If I had a MCP-accessible prompt, I hoped the agent would realize I have a set of instructions on how to handle this request when I make it.
In a way, a Muscle Mem trajectory is just a new "meta tool" that combines sequential use of other tools, with parameters that flow through it all.
One form factor I toyed with was the idea of a dynamically evolving list of tool specs given to a model or presented from an MCP server, but I wasn't thrilled by:
- that'd still require a model in the loop to choose the tool, albeit just once for the whole trajectory vs every step
- it misses the sneaky challenge of Muscle Memory systems, which is continuous cache validation. An environment can change unexpectedly mid-trajectory and a system would need to adapt, so no matter what, it needs something that looks like Muscle Mem's Check abstraction for pre/post step cache validation
If you create a pod and give someone access to information within it, you probably have to assume that they are going to make a copy of it and that they will share this copy with others.
So what’s the point of going to all the trouble of setting it up in the first place?
Anyone know of efforts to fix this space? Looking for a good excuse to quit my well paid corporate job and make a difference. News today is:
- Infotainment
- Tribal reinforcement
- Irrelevant
- Not actionable (and thus, depressing)
- Making people think the world is getting worse
- Immediate and not well thought out or nuanced
In some ways it's their fault, but on the other hand, the modern reader won't pay for news. Maybe it's largely a business model problem?
It's literally killing people and needs to be disrupted, fixed.
But single reporters seem even more aggressively partisan than corporate. They're often at least fairly explicit about this though - billing themselves as a journalist of the left or of the right or whatever spectrum you want - but there's still no balance.
What balance are you talking about? I've never been able to figure it out.
Are we talking just providing an equal number of sentences from the perspective of each major party? How do we decide which parties get included, here? Do we instead break it up by major ideologies? Again, which ones?
Even if we attempt to go no ideology and just raw facts, the facts you choose to report are a result of interpretation. You could easily construct an entirely factual article that contains no subjective opinions and still be majorly and intentionally deceptive. So not mentioning any perspectives is also not balanced.
What balance is everybody talking about? How is it defined? How we determine what is and is not balanced?
You say it like it's so easy. It's not. The reality is many (maybe even most?) articles - even ones from incredibly openly biased sources - do this. It's often difficult to make many meaningful attacks on a proposed law without mentioning what the intent is.
Unless it's one of a few already highly-covered issues with fundamentally irreconcilable subjective issues where you can pretty much recall all that context with a few words... you have to say what the heck the proponents of the law say it'll do to actually attack it.
However you don't seem to find that balanced. I don't blame you. I don't either.
The problem is the arguments made cannot be equal. They are not the same thing. Even a good faith attempt to make a completely neutral article will very often fall short. Even if the author thinks they did not, others will.
The only 'balance' you can really have is a false equivalence where nothing matters because it's all the same anyway - so why even write an article? It doesn't matter what happens.
Unless you count chasing the bottom line as an ideology, that is.
Fox, to take an extreme examples, isn't really partisan. They have just found a juicy business model where they've learnt to titillate the sensitivities of a certain segment of the population to maximize revenue.
I don't have problem with a biased news source, as long as I have enough news sources to dig from and can average out the bias from multiple sources.
It's quite cynical and unfair to say neither Fox nor (for example) Jacobin really believe in what they produce, but instead just serve whichever market segments profit them. In all probability, the employees of both believe deeply in their respective missions and in their reporting.
>Fox, to take an extreme examples, isn't really partisan. They have just found a juicy business model where they've learnt to titillate the sensitivities of a certain segment of the population to maximize revenue.
That just describes their motivations as not being partisan. Their product is still highly partisan.
I think one of the foremost solutions to this problem set is Substack. There are a non-trivial number of journalists on there who's pitch is "hey, I don't want to be part of an institution that has bias/that I don't believe in, give me money and I'll deliver long form 'pure' essays to you."
The potential pitfall is that one can end up only subscribing to journalists that agree with ones point of view. I personally think this is offset by the kinds of view points that are attracted to or do well on a space like Substack, but that might just be me.
I've been impressed with what News Literacy Project is doing, particularly in terms of helping high school students recognize the hallmarks of carefully researched material vs. the opposite. https://newslit.org/
Some very important work happening at the centre for investigative journalism, there are opportunities to help for example developing programs to teach digital safety to journalists and sources. https://tcij.org
This is just an anecdote, but when I was called up as first-time juror, I was surprised how much the police and the prosecuting attorney were pushing for something that just wasn't there.
Incentives are everything, and these groups of people get paid to put people in jail. And it shows.
I had a similar experience on a grand jury in new york (where you decide whether or not a felony case should go to trial at all) and was shocked to see a prosecutor come back 4 times on the same case, each time with a slightly less serious charge (ie, starting with level 1 felony assault and ending with some lesser assualt charge). So even after we the grand jury decided there was not enough evidence for the felony assault case to go to trial, they attempt it over and over again, with the same evidence, though requesting a slightly lesser charge, desperate for a different result.
And female gamers can be stalked to their home addresses by obsessed people. Anonymity has some advantages for marginalized social groups as it lets them avoid predators better.
edit: Actually I'd assume a lot off harassment of people via other channels would happen including SWATing, emailing employers to get them fired, false reports to get them banned on other sites, contacting spouses with fake evidence of infidelity, etc. Only the victim loses anonymity, the attacker keeps it. Competitive gaming brings out the worst in some people.
I feel like there's a need for a privacy-protecting attestation service/protocol. Fully anonymous discussion boards/social networks are tricky because of obvious sockposting-related problems. Another poster mentioned a Voat mechanism for non-username identification, but it sounds like the old *ch concept where some identifiers are hashed s.t. the same poster gets the same ID across some subset of posts (time, topic, etc). That's an ok way of trying to maintain integrity without publicly deanonymizing participants, but is vulnerable to all kinds of technical workarounds. Anecdotally, dealing with ban-evaders in a forum context got a lot harder when CGNAT'd/v6 mobile IPs started getting common.
So, the next step is attesting your identity directly to the site. But, who's going to give some random forum their government-issued ID? Despite SV firms recently trying to normalize that, hopefully fairly few. Facebook can get away with it, but only because they're already Facebook and benefit from the trust implied by scale. Beyond that, what happens when a user says something impolitic and brings down the mob? Trust the mods to not unmask the user? Meh.
If site operators could ask a question like "is this person a real person, do they already have an account with us, do they leave in geographical region X" and a prospective user was able to obtain an attestation that those qualities are true without that attestation revealing any additional information (as well as being cryptographically secure), you could do some interesting things.
The downside is the attestation service then becomes a point of attack for those interested in unmasking individuals. Is it a private company that must answer subpoenas? What if the government wants the proverbial pen register or the organization gets subverted in some other way? Dunno. The ideal is neither side (attestor or service) knows more than it has to - service doesn't know anything about the user besides the handful of agreed-upon facts; the attestor doesn't know anything about the service. Obligatory buzzword: seems like a potential application of zero-knowledge proofs.
I'd say it wouldn't help much. You're building one small scene that's designed to be viewed from a relatively small area or path. I highly doubt they're building anything that doesn't appear on screen. If you were to walk about that scene in Unreal... I'd imagine it's the digital equivalent of a fake old west town.
That has to be a consideration. But I don't know how much it would really speed up production of a AAA Mandalorian game. Some... maybe a 6 month head start on a 4-5 year game.
It would definitely help make the game environments higher quality and be a cost saving to the studio.
It could be really cool to have in-game art using the exact same assets from the actual film. Even whole scene cinematography could be taken from scene data used in a movie.
I can imagine a consumer experience using high-end VR (say an attraction at Disneyland)... take the assets and cinematography from a scene in a movie, digitally recreate everything that WASN'T already digital from the scene, and then you could relive a scene in VR, with the bonus that you can navigate it in realtime and see it from different perspectives. This would be especially adaptable for franchises like Avatar where everything in the film is already composited in 3D.
On a darker note, you could combine all this with things like the Body Transfer Illusion, take some kinda psychedelic and star in your own horror flick where gruesome things are done to your own body in VR. Good times for the whole family!
I can't tell if this data means Utah got better at welcoming females to JS engineering or there's a point where they stop trying around five years. :) As a member of the JS community I hope it's the former.
There's also someone killing it that's only been working in JS for three years.
Just by people I know on twitter, there's alot of great women from accounting and project management that have done bootcamps and made the switch recently.
- Moving to a newer and more modern test library
- Refactoring my data layer so it's easier to read, based on years of organic changes that need to be baked in and simplified
- Porting some functionality to another language to vastly improve performance
I agree with the overall sentiment, but having an agent at my finger tips who can really crank out large-scale, involved code changes is unclogging quite a few backburnered todos lately for me.