kamranjon

I've been reading a lot of Don Delillo lately and so I wanted to see how Grokipedia page on him fares.

I found the "Critiques of elitism" section and noticed this sentence:

"Reviews of Mao II (1991), for instance, highlighted the novel's focus on a performance artist protagonist as emblematic of this tendency, with detractors accusing DeLillo of prioritizing esoteric concerns over relatable human experiences, thereby catering to an academic or literary insider audience."

But Mao II does not have a performance artist as the protagonist, that is the book The Body Artist. Which seems like an obvious failure of the AI model to properly extract the information from the sourced article.

Also strange is that the sourced article (from Metro times) just as a passing comment says: "DeLillo’s choice of a performance artist as his protagonist is one reason why some critics have accused him of elitism." - so it would seem that it is being used as a primary source though it is actually a secondary source (which itself doesn't provide a source)

Overall I'm not too impressed and found some pretty predictable failures almost immediately...

You can support sane border policies without also supporting racial profiling, the militarization of our cities and warrantless searches and detention. These two things don’t have to be mutually exclusive, but arguably much of what ICE has represented recently is what many people would consider to be unconstitutional behavior.

Here is perhaps a simpler metric. If the creator of a security focused software is this awful at protecting their own anonymity- why would I trust them in the first place?

If they are revealing their identity so that we can vet their credentials, then it makes sense that they would want us to judge them based on their online persona.

If we believe all of this to be true - it’s not really cancel culture, it’s by design that the creator of SimpleX has implemented this filtering mechanism for their users.

What do you mean signal “pinky swears” it’s not looking at it? It’s E2E encrypted and the code is open source - am I missing something? How would they possibly look at it?

This is pretty incredibly - Fabrice Bellard builds stuff that just has such a wide spread impact - so cool that QuickJS enabled an old system like PS2 to have a bit of a homebrew revival through this AthenaEnv project

I wonder if there is a way to use tailscale to make it easy again?

I wonder if the inverse rule for this is “how to know when quitting is more effective than trying” - because I think that often times, and especially for folks who can be incredibly persistent when they put their mind to something, it’s helpful to have a cost benefit analysis of your efforts and it might just be that in more cases than not, just not trying actually resolves the core problem your trying to solve.

I really enjoyed this article and it really resonated with me, which made me wonder if it is actually an evolutionarily selected solution in a way - like that ignoring something turns out to be a surprisingly effective form of triage for many situations? Obviously the cases where this doesn’t hold are what the article is addressing but I found it fascinating to think about why that approach might be so common.

I think the crux of this is that wealth is equated with success at the expense of all other qualities or accomplishments.

As a result of this perspective, hoarding of wealth has become the norm and even the goal for many.

If wealth means success, then make number go up = success, even if that incentivizes bad behavior. If wealth means success, our role models become degenerates.

This line of thinking is why I think the tech industry has become so morally bankrupt.

I don’t mean to judge because I think many people come to similar conclusions but I believe there has been a concerted effort to equate accumulation of wealth with genius, and to portray this anti-social behavior of endlessly accumulating wealth in a positive light.

I really think this phenomenon should be studied because it will likely lead to some catastrophic outcomes for the country.

This is probably the biggest and most successful example of utilizing non-permissive open source licensing for the public good - I’m curious why so many places I’ve worked have insisted on avoiding using libraries with GPL licenses in favor of MIT licensed projects, while at the same time hosting all of their services on different flavors of Linux.

It definitely seems like MIT is favored by big corps but at the end of the day, they’ll use GPL licensed code if it’s the best option. Which makes me wonder why it’s so demonized.

Can someone with more context explain what this means and maybe the background?

One of the things that the Coolscan did well from a hardware perspective was that it made the transport sprocketless which allowed damaged film to be easily scanned, and it also allowed non-standard formats to be scanned easily as well. I’m curious if they have a sprocket driven system for this or if it utilizes a similar system as the Coolscan - I’ve used many scanners and the Coolscan is still the best/most convenient because of being able to just sequentially scan an entire roll of negatives.

I immediately thought about how this would be awesome if it worked with tailscale - pretty complimentary tech I think.

A fun tidbit about crystal oscillators is that they allowed “un-tethered” sound recording on motion picture film cameras. If both your sound recorder and your film camera are using a crystal oscillator as a reference for their motors - you can sync them up in post without needing them to be physically connected during filming.

“libpng support brings emoji rendering in the base system. Without that, we don’t know what all the AI tools are trying to tell us in the terminal.” <3

Isn’t IBM still operating through a pattern of selling service contracts for sub-par products that do everything in their power to lock their customers in? I feel like they embedded themselves so deeply in some of these public sector orgs that they can’t really orient themselves in any other way than through a predatory business model that lacks innovation.

It would seem strange if that was the purpose since the first photo on the website is ~40kb

“Noone made any money or got any oil from the Iraq War.”

This seems wrong?

Halliburton secured a 7billion dollar contract where they were the only allowed bidders, their subsidiary Kellog Brown & Root (KBR) was estimated to receive at least 39.5billion dollars in federal contracts.

Halliburton experienced a 80% increase in revenue by the first quarter of 2004…

Most of these contracts were “cost-plus” which meant they were guaranteed reimbursement for their costs + a guaranteed profit on top of that.

As a result of the war many oil companies including BP were awarded access and development contracts to the oil fields, with BP extracting well over 15billion dollars worth of oil.

Over 150billion dollars in oil money apparently just went missing or was “stolen”.

Also the US has maintained control over Iraqi oil revenue since the beginning of the war in 2003.

It would seem to be almost rewriting history to make such a statement…

I loved text based adventure games when I was growing up, but I also thought this comparison was incredibly apt and also found it very funny. I’m a bit surprised people are so offended by this article, have we lost the ability to read something with nuance?

He really was one of the first techies to project genius when he really just had antisocial behavior - sort of paving the way for folks like Musk and Thiel. There is such a distortion field around Gates that has been meticulously crafted for years, really since his rebranding after those antitrust trials that ruined his image.

This is interesting! I once trained a t5 model by removing newlines from Wikipedia text and it worked surprisingly well / at the time the context length was the biggest issue.

Another, not so easy to solve issue was conversational dialogue type data, which wasn’t super well represented in the training data.

I’ve always wanted to come back to working on the problem again, because I think it’s very interesting and we will have a bunch of unstructured text as a result of STT models like whisper that do a great job of transcribing/translating but generally don’t format anything.

Does it seem strange that this has 121 comments within 4 hours and somehow is ranked 131st on HN? I would think this would be front page with those type of numbers.

Have you tried moondream yet[1]? The moondream 3 preview model[2], according to the blogpost[3] appears to outperform many frontier models on VLM tasks and does so with a relatively small footprint.

[1] https://moondream.ai/

[2] https://huggingface.co/moondream/moondream3-preview

[3] https://moondream.ai/blog/moondream-3-preview

It’s interesting with the whole quote: “OpenAI has 800 million weekly active users, and putting aside the fact that OpenAI’s own research (see page 10, footnote 20) says it double-counts users who are logged out if they’re use different devices”

The number may not actually be too accurate - but I imagine it’s also paired with what another commentator has said - OpenAI is basically giving their product to companies and the companies are making the employees log in and use it in some way - it’s not natural growth in any sense of the word.

Is being nice equivalent to jumping off a bridge? I think it's relatively simple to comprehend and also harmless. The guy who built this thing is nice, let's try to continue that tradition so that our community doesn't turn to shit.

I suspect, from the responses of the creator here, that this app they are selling is likely violating a number of open source licenses…

What RLCD would you recommend?

Wow this is awesome, I think you’ve inspired me to try and make something similar…

I’m just wondering if anyone that is closer to this space could shed some light on if this $2.5B valuation seems more or less accurate? I have played around with n8n - I just didn’t know if it was ubiquitous/profitable. $180m round seemed pretty huge but maybe it’s really a unicorn?

awesome! I have wanted to do something like this since the deno compile feature came out - great to see someone pulled it off, and last i checked the binaries were significantly less bulky than electron.