HNHacker News
TopNewBestAskShowJobs

barnacs

526 karma · joined December 29, 2014

submissionscomments
barnacs··on Show HN: B-IR – An LLM-optimized programming language
I'm looking for a language optimized for human use given the fundamental architectural changes in computing in the last 50 years. That way we could skip both the boilerplate and the LLMs generating boilerplate.
barnacs··on MCP is a fad
That's probably why a boundary (like MCP) is useful. Imagine maintaining the critical application logic the "old fashioned way" and exposing a MCP-like interface to the users so that they can have their LLM generate whatever UI they like, even realtime on the fly as they are engaging with the application. It's a win-win in my mind.
barnacs··on AI might yet follow the path of previous technological revolutions
> the ability to produce useful output beyond the sum total of past experience and present (sensory) input.

Isn't that what mathematical extrapolation or statistical inference does? To me, that's not even close to intelligence.

barnacs··on AI might yet follow the path of previous technological revolutions
> Thinking is using what you know to come to a logical conclusion

What LLMs do is using what they have _seen_ to come to a _statistical_ conclusion. Just like a complex statistical weather forecasting model. I have never heard anyone argue that such models would "know" about weather phenomena and reason about the implications to come to a "logical" conclusion.

barnacs··on AI might yet follow the path of previous technological revolutions
I guess our definition of "thinking" is just very different.

Yes, humans are also capable of learning in a similar fashion and imitating, even extrapolating from a learned function. But I wouldn't call that intelligent, thinking behavior, even if performed by a human.

But no human would ever perform like that, without trying to intuitively understand the motivations of the humans they learned from, and naturally intermingling the performance with their own motivations.

barnacs··on AI might yet follow the path of previous technological revolutions
But it's only able to answer the question because it has been trained on all text in existence written by humans, precisely with the purpose to mimic human language use. It is the humans that produced the training data and then provided feedback in the form of reinforcement that did all the "thinking".

Even if it can extrapolate to some degree (altough that's where "hallucinations" tend to become obvious), it could never, for example, invent a game like chess or a social construct like a legal system. Those require motivations like "boredom", "being social", having a "need for safety".

barnacs··on AI might yet follow the path of previous technological revolutions
In my view, absolutely yes. Thinking is a means to an end. It's about acting upon these motivations by abstracting, recollecting past experiences, planning, exploring, innovating. Without any motivation, there is nothing novel about the process. It really is just statistical approximation, "learning" at best, but definitely not "thinking".
barnacs··on AI might yet follow the path of previous technological revolutions
To me, it's about motivation.

Intelligent living beings have natural, evolutionary inputs as motivation underlying every rational thought. A biological reward system in the brain, a desire to avoid pain, hunger, boredom and sadness, seek to satisfy physiological needs, socialize, self-actualize, etc. These are the fundamental forces that drive us, even if the rational processes are capable of suppressing or delaying them to some degree.

In contrast, machine learning models have a loss function or reward system purely constructed by humans to achieve a specific goal. They have no intrinsic motivations, feelings or goals. They are statistical models that approximate some mathematical function provided by humans.

barnacs··on A Formulation of the Trilemma in Proof of Work Blockchain
Proof of work essentially allows for a periodic leader election, where the leader has sufficient time to propagate its state update along with a verifiable proof of authority in a permissionless decentralized system.

Decentralized cooperation in society likely wouldn't be permissionless, in most cases you would probably want to assign some voting power per human/citizen/share certificate/whatever. I'm also not sure if decentralized, verifiable randomness is easier to achieve outside of computer networks (for example, source some verifiable randomness from the universe based on a pre-defined algorithm).

barnacs··on A Formulation of the Trilemma in Proof of Work Blockchain
Exactly. Lightning minimizes the use of the globally consistent ledger, which inevitably has to make trade-offs due to the "trilemma".

You set aside some funds on the global ledger and then you can use those funds to transact in a much more efficient, truly p2p manner, without touching the global ledger. Eventually, you net out everything and settle it all at once on-chain, utilizing its features to resolve any conflicts.

barnacs··on A Formulation of the Trilemma in Proof of Work Blockchain
I think it all comes down to relativity and the speed of light.

There is no single, universal, true ordered state (ledger/db). Participants need a conflict resolution mechanism to figure out whose truth is correct. One must rely on a localized consinstent state of some authority (leader/consensus).

barnacs··on Firefox 128 enables "privacy-preserving" ad measurements by default
> if we prefer to outright shut down online advertising

Yes, please. Both online and offline. Advertising is probably the most useless, annoying and wasteful industry out there.

We could have pull-only databases of businesses, products and services instead. Ideally, with independently verified, fact-checked information and authentic reviews. Realistically though, this kind of objectivity would probably be infeasible to enforce and maintain. But even if we allow for misinformation, paid rankings and whatnot, the point stands: any such database should follow a pull-only model, users access it voluntarily to search for products and services and it's not an unsolicited broadcast to everyone everywhere all the time.

barnacs··on Hallucination is inevitable: An innate limitation of large language models
Think of it this way:

Intelligent beings in the real world have a very complex built-in biological error function rooted in real world experiences: sensory inputs, feelings, physical and temporal limitations and so on. You feel pain, joy, fear, have a limited lifetime, etc.

"AI" on the other hand only have an external error function, usually roughly designed to minimize the difference of the output from that of an actually intelligent real world being.

barnacs··on Sent – Simple plaintext presentation tool
Might as well just xdg-open files in a directory in alphanumeric order by name as a "presentation" at this point.
barnacs··on Why hasn't economic progress lowered work hours more? (2016) [video]
I think what's more relevant here is Marx's theory of alienation.

In order for people to work less hours at their job there would have to be strong local communities and a kind of socio-economic framework to support constructive, productive and satisfying activities outside of their day job. Things like improving or fixing up the neighborhood or your own household, political activism, caregiving, teaching, citizen science are all examples of things that one could do not for money but for the benefit of the community or the betterment of society. But due to the capitalist mode of production people are alienated from each other and the product of their labor so much that they think of work as a commodity to be traded for money only. And rather than doing these things as a way of finding fulfillment and building communities, people contract others to do it for them for money.

barnacs··on Did Bitcoin just prove it can't scale?
Sure, LN alone might not be enough to serve the daily transactions of billions of users. Still, it proves how layers built on top of the blockchain can improve the efficiency by several orders of magnitude with some reasonable trade-offs. LN is probably just the first of such layers.

Things like sidechains, cross-chain atomic swaps, various block space usage improvements (eg. Schnorr signatures) are also in the pipeline right now, all of which could further improve throughput.

So while LN may not be the silver bullet, it makes me hopeful that at least from a technical perspective, Bitcoin will be able to scale to become a global currency.

barnacs··on Did Bitcoin just prove it can't scale?
> How is this expected to work with 7 billion people using it for every tiny financial transaction?

https://lightning.network/

tl;dr: a network of off-chain, transitive payment channels. two on-chain transactions (to fund and settle channel) allow for ~unlimited, ~free, ~instant transactions across a p2p network of nodes off-chain

barnacs··on We didn’t lose control of the Web – it was stolen
It wasn't stolen, we're giving it away. And not just the web, but our freedom.

If you're using services that support surveillance capitalism or you are working on such products, please stop. Thank you!

barnacs··on We should kill the 40-hour work week
> Large corporations need hordes of people doing boring work in a reliable and predictable, easily manageable manner.

That's what machines are for. We need to embrace automation.

> Society needs what large corporations produce

Nope. Large corporations spend loads of money and manpower to make people think they need what they produce.

> People need the wages these corporations pay.

Not if they resist the manipulation to buy all the useless stuff. Not if automation makes everything cheaper.

> The 40hr work week is going nowhere.

Oh, but it is going. The sooner we realize this, the easier the transition will be.

barnacs··on We should kill the 40-hour work week
Don't rely on laws to fix this. Simply don't sign employment contracts that don't work for you. I think a lot of the HN crowd can afford that.
barnacs··on Bitcoin hits 3-year peak, nears record high on U.S. ETF approval
I don't understand why the segwit softfork and a block size increase would have to be mutually exclusive.

As segwit generally improves the protocol (makes it more efficient, fixes a transaction malleability issue, increases capacity and gives way to proposed future improvements) and doesn't require a hardfork, I think it makes sense to apply that first.

And if with segwit and all the improvements it enables the capacity still doesn't prove to be enough for the future growth, we can hardfork to dynamically increase the block size.

But for now, we need to do something. Why not do the thing that's simpler to put into effect and provides more benefits?

barnacs··on Time to Encrypt the Cloud
In my experience, the only interest of most for-profit corporations is maximizing next quarter's profits. If they can get away with raising the prices to the extremes to achieve that, they will. If they have to offer their services for free and balance that by exploiting their users' data or attention, they will. If they have to shut down some services or even "pivot" the entire company to a completely different domain, they will do that.

I don't see how any of that is aliged with the users' interest. Of course, they will also spend quite a lot on marketing, trying to make you believe the opposite.

Also, in addition to the business model, the whole entity of a for profit corporation is rather transient. And when we're talking about an infrastructure for storing, relaying and accessing information, that's definitely not a good thing. I don't think google, facebook, dropbox or github will be around in a couple of decades, due to political and economical reasons. It's hard enough to keep information around the face of technological advancements (new hardware architectures, protocols, algorithms, data formats, etc), so we really don't need these additional obstacles.

barnacs··on Time to Encrypt the Cloud
I don't think it matters much what the business model is. As long as software services are provided by centralized, for-profit entities, exploitation won't stop.

A free, open source, peer to peer infrastructure is the only way out.

barnacs··on Ask HN: How can I volunteer my software skills online?
I think too much specialization in the name of economic efficiency is harmful as it causes alienation and apathy.

If say this lawyer is required by a law firm to work on cases that end up harming the poor and then donates some of their salary to help the poor, is it really net useful?

If this lawyer spends an hour volunteering at a soup kitchen, he learns about the process, the people, their circumstances and needs. It may be something that he can then apply to guide his professional work. Or he may spot and improve some legal inefficiencies in the operation of the soup kitchen. Or he may be reminded that he can actually donate his own excess food instead of throwing it out.

Generally, I think we need more communication and cooperation across professions and "social classes", not less. The experience, perspective or education one brings to the table may be just the simple thing the others need to improve.

barnacs··on $5 Showdown: Linode vs. DigitalOcean vs. Amazon Lightsail vs. Vultr
Another potential catch is that they will graudually throttle your cpu speed if you load it too much. I guess this is not unreasonable for the price, but they don't mention it anywhere on the site and market it as "guaranteed" or "dedicated" cpu, which is definitely misleading.

If you want a budget vps and willing to do your due dilligence I recommend looking at lowendtalk[0]. If you check out the the customer reviews and the quarterly top provider poll results, you can often find offers comparable to the bigger providers for a fraction of the price. (I'm not affiliated with the forum in any way)

[0] https://www.lowendtalk.com/

barnacs··on Chrome 56 will mark HTTP pages with password fields as non-secure
10/18/2016

> Distrust certificates with a notBefore date after October 21, 2016

Just in time!

barnacs··on Why we should all have a basic income
Those may be the very same people who play music, build stages or give water to overly wasted kids at summer festivals. Or they might fix your motorbike better and cheaper than your local mechanic if you asked them to.

Or they might not. Either way, I wouldn't consider them any more harmful to society than those probably millions of people working very hard to shove ads into peoples faces ever more efficiently, or those bankers and accountants looking for loopholes to effectively steal from others. Or corrupt politicians whose only concern is filling their pockets. Or I could name a bunch of other mostly useless leech industries.

My point is, I think you are judging these people from a faulty perspective very much ingrained into our society, where one's usefulness to humanity is primarily judged based on how much money they can get their hands on either by stealing from others or selling their freedom to someone willing to pay for it, regardless of what value they actually contribute (if anything) as long as it's considered legal and they do so as part of the so called "workforce".

UBI would fundamentally change that by allowing people to spend time doing things they consider useful, rather than working for whatever gets food on the table, however useless it may actually be. Of course, for that to work, people need to be open to other people's ideas of useful and the currently dominant view of measuring usefulness in GDP or whatever may become obsolete.

barnacs··on Why we should all have a basic income
Are you suggesting the majority of those people are completely, utterly useless?

Useful contributions don't necessarily have to be intellectual. They may be able and happy to provide art, joyful company, social skills, physical assistance (eg. getting groceries, moving equipment) or other support to those in need.

barnacs··on Chrome 56 will mark HTTP pages with password fields as non-secure
It may take way more than 5 minutes.

My experience with let's encrypt so far:

- the name of their tool was changed form "letsencrypt" to "certbot", breaking my cronjob

- for daemons that try to access the cert/key as non-privileged users, additional fiddling with permissions is necessary, which may even be overwritten on cert update if done incorrectly

- when the certs are renewed, daemons need to reload them. This means that ideally, you need to detect when a renewal actually happens (as opposed to an attempt), keep an up-to-date list of all daemons that use the certs and possibly completely restart them, dropping all existing connections (some daemons just don't support a live reload)

I'm not saying these problems are unsolvable, but may take way more than 5 minutes and I, for one, opted to renew my startcom certificate for another 3 years instead.

barnacs··on Why we should all have a basic income
I believe basic human nature includes curiosity. Living in a society means one will naturally seek the appreciation, acceptance and respect of their peers. I think these alone will eventually lead most people to contribute something useful to our world one way or another.

Mentally and/or physically exhausted after a long day of performing a meaningless job, or unable to get one and living off welfare, despised by the rest of society, no wonder one tries to escape reality by playing video games or watching tv excessively. Actually, escaping into 'work' is no more healthy.

But when everyone enjoys the security of having their basic physiological needs satisfied by default and not knowing what the hell you're supposed to do is a feeling widely shared and accepted by society, the situation is completely different. It's an enormous community building force. Additionally, we could build or designate community places and workshops where people get together to converse, build, play, create, cooperate to further help everyone find a direction, or find themselves.

← PreviousPage 2 of 5Next →