manholio

I wonder how soon we will have microarchitectures that are purely the result of some deep learning algorithms optimizing for performance and compilability of high level languages. We would be able to prove them correct - that they correctly execute the code and have no exploitable bugs - but we will have almost no idea how exactly they achieve said performance.

I agree, there's no reason to think large haul electric trucks will have a better efficiency margin over internal combustion variants than in the case of passenger vehicles. On the contrary, there's reasons to doubt it, for example large diesel engines are highly optimized for efficiency unlike small consumer cars they are bought by businesses which think in TCO.

So probably the real margin over diesel is about 50% cheaper in fueling. But you have to take into account route restrictions due to limited range, time lost charging etc. Not to say it won't find its place in the market, but the margin and advantages of electrics are not yet so clear cut, it will take decades to reach dominance.

Luckily, there are people doing Blu-ray digital backups for the disks we own, in open formats unnecumbered by DRM, then store them in large file distribution systems.

That's why you use the homework as a final step, after screening CVs, technical interview etc. Taking a burn from time to time from a candidate not really looking for a job yet willing to go through hoops is surely more reasonable for a company than burning all your candidates by the equivalent amount, no?

Or you could just pay people for their time at the prevailing rate, > 50$/h cash.

Here is your homework, we expect someone at your level to spend no more than two days on it, here is your $800, have fun.

This is not a story about tenacity or an overloaded court system. It's a story about deliberate bad governance to keep corruption money flowing in an impoverished country.

The political leadership could easily find solutions to stop such abuses of the public legal system, and direct resources to high priority cases with large public impact. By deliberately allowing such cases to bog down the justice system and waste limited public resources, they ensure rule of law cannot operate effectively. It's a typical story.

> But as long as users can still withdraw into a non custodial address, they have the option to escape to that neutral base layer.

Using blockchains for money is just about the most absurd solution a society can come up to the problems of financial centralization. Money is a form of power and the solutions to power problems are always political.

We need a financial system that protects individual privacy while reducing the options for money laundering, tax evasion, bribes, fraud and all the types of crimes motivated by money. There is a inherent contradiction in these requirements, the hope that a private actor or some non-accountable peer to peer algorithm can find a good political compromise is an ideological pipe dream.

It supports exactly the format it needs to so that you can purchase DRM encumbered ebooks off Amazon. I hope that clears it up.

> I hate that phrase, it rarely pans out with anything that makes any sense.

That's just the evolutionary loop of contrarian views. Of course the large majority will be crap, because that's why the mainstream views are dominant.

But sometimes, even if short of an infosec revolution in immunology, which was clearly a rhetorical stretch on my part, some contrarian view will make people think in a different way or challenge them to defend mainstream views and really think really hard about why those views are what they are. It seems by the bulk and quality of the responses that I have achieved such an effect, so the whole exercise was not so non-sensical.

> And even there, the reason why we have a placenta and are mammals is likely because some organism was infected by a retrovirus and learned to form syncytia and learned a knew trick from a virus. Evolution can take a seeming "fault" like that and turn it into a strength on the long-term.

I think this is the crux of the argument right here. Evolution is satisfied and will even favor a certain number of informational security compromises if it induces variability that has (even a small chance of) improving long term fitness. But the perspective I was approaching the problem was short term host security - the thing we, as rational members of a given species living in the present day care about, because we know full well that the vast majority of compromised individuals will not survive. So we are likely to make a different short term tradeoff, even if it hurts the long term fitness of the species. Modern medicine and social security systems already do that to a large degree.

More generally, the perspective I wanted to present was that of intelligent design versus evolutionary fitness, that a blind watchmaker might leave gaping design holes for a malevolent rational attacker. I think there is ample evidence that evolution by natural selection is not all powerful and all knowing against rationality (at least if we don't consider rationality just another facet of evolution).

Rust is not only a C++ alternative, it's a C alternative too. You don't have to have methods in your structs or use generics, traits, closures, interators etc.

You won't be able to use the standard library, but it's doable.

> But that's not realistic, a lot of simple things need to go through the cell membrane all the time.

But surely something that contains an executable payload (DNA/RNA) needs to be better scrutinized than a simple nutrient. Runtime alterations of the cell code are practically never required, the fact that any code that enters the cell is fair game to execute and even to permanently alter the nucleus "binary"(retroviruses) strikes as a bad practice. A static host-key, unique per individual, would stop all such viruses.

So you cannot shake the impression that evolution does not really care about the information security of any given host. What it cares about is overall success of the gene while allowing for sufficient variation and flexibility - the retroviruses are just a tool among many to induce such genetic variability.

Nothing I say above should be taken as dismissing an exceptionally intricate and developed system, just pointing out the different way a rational attacker would approach the problem and that the defenses against such an attack would be fundamentally different than what evolution had to deal with in the past.

In terms of information security, as viewed by an infosec engineer, biological systems seem to have incredibly weak architecture. Hear me out, please.

If any piece of DNA code manages to find its way into the nucleus, it's happily added to the cell "runtime" and has root access to the system.

To defend against this, biology relies on multiple layers of security, both physical, like the skin, and chemical, like the secret protein password it requires to allow foreign DNA to breach the cell membrane. But both are easily circumvented, for example weak points exist in the physical layer and the protein passwords are static and low entropy, easy to bruteforce by an enemy that already has a botnet of infected hosts which produce random permutations of these keys.

And once infection happens, the body response is sort of like the experience of cleaning a Windows 98 which has been used by your grandma to surf the internet for the last two decades, without the option to do a fresh install (kill the host). It's messy and insanely complex - the sort of thing you would expect blind evolution to generate against a blind, brute-forcing script kiddie.

You just have to wonder what are our chances on this planet if intelligent and rational enemies, be they human or AI, gain the knowledge, tools and motivation to attack biological systems and intelligently design sophisticated attack vectors. Our most precious systems are wide open, folks.

Perhaps some day, to prevent such attacks, we will take control of this complex immune machinery and inject ourselves with highly engineered "anti-viruses" built on the same info-sec principles we build software on.

As an avid reader, I anticipate the feedback as "unlikely to scale, since the centralized productivity gains don't outweigh the diminished flexibility and substantial transport costs".

This is a recurring theme with prefab system he analyzes, they scrape by employing labour arbitrage at a cheap production site, but do not unleash the productivity gains we associate with industrial production.

This becomes obvious when you look at the distribution of construction costs for a finished building: https://constructionphysics.substack.com/p/construction-cost...

The most a prefab can hope to achieve is attack maybe 30- 40% of the total costs in that pie chart. You cannot prefab electrics and plumbing, foundation, landscaping, permitting etc. So even if they crush it and get 100% improvement on their products compared to on site build, that's still a 20% at best overall cost improvement.

Bit with that comes the risk of a new and unproven system, costly coordination issues that are easy to rectify in the traditional method - but if the wrong prefabs are shipped across the country you are dead for a few weeks, since the next stage cannot start.

You should really watch that video, it goes into great details, backed by scientific research, into how and when the 10,000 hours trope works and where it fails.

Creativity and skill are somewhat orthogonal, you can be an exceptionally well read and educated person yet never publish anything. But it's extremely unlikely you will create anything out of thin air without good technical skills. We like to to obsess about outliers like Beethoven, the deaf composer, but in reality the vast majority of famous musicians are recruited from accomplished instrumentalists.

"Vision" is what happens in your brain after you grind 10,000 hours on a skill that is improvable: https://www.youtube.com/watch?v=5eW6Eagr9XA

At that point, you no longer have to grind, because your brain has hardwired decision paths that are automatic and produce correct results for extremely complex problems like "is this character likeable and understandable by the reader", and you can effortlessly add a few high value paragraphs to their backstory. But it's foolhardy to think that level of "flow" can be achieved out of thin air, without writing hundreds of characters before them, re-reading them, getting feedback from other readers, studying other briliant authors with a eye for their technique etc.

I think they understand it, but maybe they believe that advice is misleading. If you are enjoying it and having fun, then you are in your comfort zone, and that means you are improving at a much lower rate than your potential. This is true for any activity, from physical fitness to academic pursuits.

If you are in zone dominated by System 1 thinking and System 2 is on pause, then you are not building skill. There is nothing wrong with having fun in life for things that are not critical to you, but your core areas are developed with effort and focus.

I have not met a single successful self-made person that found success without struggle and focus on their craft.

"-" still works, "+" was bastardized when Google launched Google+ but you can get similar functionality by surrounding your term in quotes, preventing it to be dropped.

I don't dispute it's a philosophy, but it's an extremist version. It's like having the choice between asceticism, living alone in the desert eating critters, versus extreme hedonism, gluttony and materialism. But most people reject both.

What I'm pointing out is that we could imagine a happy compromise where 99% of the goals of free software are achieved, that would not only allow developers to make a comfortable living, but also benefit the vast majority of people who use software without understanding technical complexities.

If the software:

- has source available, and anyone can study it and audit it;

- allows anyone to create, for their own use, derived or custom versions;

- allows anyone to make a business selling derived versions within the license terms (for example, in the time-bomb period the users of your patch must be licensees of the original owner);

- guarantees that the vendor cannot lock you in, and the software will become completely free after a certain date;

- generates resources for software development that would otherwise not be available, so that functionality could only exist in proprietary, closed source form,

... then, I belive, the loss, for a limited span of time, of the ability to use it gratis is a very good overall compromise. We all understand and accept the idea that work must be paid and that a price tag designates something valuable, that is worth making and paying for. Don't castigate money ideologically, use them to achieve the philosophic goals.

Somewhere, somebody pulled a dirty trick on OSS developers and moved the goalposts to "and you must now renounce all material possessions and live on pure ether".

> If neither of those are a good fit, premium support, merchandise and ultimately donations might be an option.

And what if neither of those are a good fit? Will you concede that there is a real world out there where open source software competes with proprietary and closed source systems? And that you cannot compete with corporations eating passion alone? That you have to have a monetization system in place that can generate at least a non-negligible fraction of what the closed source competitor can?

There is nothing unethical in having a commercial contract with your users whereby they pay for the features you implement. The typical ways this is done leads to unethical outcomes (lock-in, secret source etc. etc.) but by itself a commercial contract between willing and informed adults that don't harm people around them is not unethical.

The extremist stance "free as in beer from day 1" is an ideology that, it should have become clear by now, is actually hurting the freedom of the public at large, who is pushed into proprietary cages set up by ruthless profit-maximizing businesses. There should be a better way where developers are paid, high quality software gets written, and users are free. There is nothing fundamental preventing such a world to exist.

It would not be open source for the duration of the time-bomb, clearly, that's the point of creating an exclusivity period.

After the time-bomb, it's undeniably open source: users have all the freedoms granted by open source and they have the option to release their derived version with a zero days time bomb, effectively equivalent in all respects to GPL. What GPL does not grant, and such a license would, is the ability to create a non-zero commercial exclusivity period, during which the software reverts to commercial-with-source-available status.

What I dislike about BSL and related is that on the Change date the software typically reverts to GPL, which prohibits others from creating their own BSL competitor to MariaDB.

What I would like is to take the MariaDB 4 year old source and release my own database with better features, then compete in the market place with MariaDB, and have my own commercial exclusivity period, as I have explained elsewhere in the thread: https://news.ycombinator.com/item?id=32339945

To use the BSL in the sense I am referring here, the Change license must be BSL itself, with flexibility to allow a new change date, at the new developer's discretion, no longer than a set limit (4 years for BSL). If want to make the code free-beer, I set the change date to be the same as the release date, but the next day every other developer has the option to fork my BSL code and release their own BSL time-bombed version for up to 4 years and make money of it. Users who don't want that can take my code and enjoy all the freedoms that GPL would grant. [1]

Otherwise, if my only option on the Change date is GPL, which is very restrictive to me as a developer, I have zero incentive to work on certain types of free software that can't make money off support.

In a sense, a recursive BSL is freer than GPL, because I grant my users all the freedoms of GPL but additionally the freedom to create their own timed-bombed commercial release, i.e., to restrict the freedom of others with their own releases. The end users then have to option to use my original free-beer version, or pay for a better feature set, which itself is guaranteed to become free beer in a number of years. This incentive structure ensures a steady revenue stream for the developers as long as they create new features which are desired, but the revenue stops if they try to rely on vendor lock-in.

[1] (of course to do this recursive-style licensing one would need to modify BSL to grant the actual freedoms to run the code, distribute etc., the current text lacks this and relies on an external open source change license after the Change date).

Exactly. The time-bombed license would grant you from the start all of the important freedoms of OSS - freedom to create a derived copy, to compile and understand the program, to freely distribute in source form etc. The only thing it would restrict, for (say) 3 years, is running the software in production without a commercial license, the "free-as-in-beer" freedom.

Such a licensed application would change its "main developer" over the years, for example someone would take the 4 years old free-beer version, invest money and time and create a compelling feature set that would make people pay for their version, after 4 more years someone else could do the same to the new feature set etc. So while the initial developer has the freedom to release under multiple-licenses, once they go for "commercial-copy-left" they set in motion a chain that restrict what future developers can do, forcing them to release their work too as commercial-copy-left if they want to have the 3 year binary exclusivity period.

What we need is an time-bombed open source license, that allows a limited exclusivity period for the author during which they can charge for the software. After that time expires, say 3 years, the software reverts to a free shareable license, that anyone can fork into their own commercial, 3-year time bombed projects.

Dangle the carrot and let working people make a living, but force them to give back and prevent them from building closed source proprietary empires fueled by vendor lock-in.

There is a certain type of software, for example professional tools use by non-programmers - graphic editors - where traditional open source models, support contract or volunteer developers, can't work. If you won't be able to charge for copy, such software will never be written in an open variant.

Except the photo-manipulation suite, they are not done by software.

You have to declare you own the intellectual rights of the underlying code, a thing GPL does not grant. You can of course dual license your own code but you cannot own code released by someone else under GPL, and you cannot publish it in the App Store. It would be an absolutely trivial change for Apple to make their store friendly towards opens source and there have been multiple protests about this.

I think projects like VLC request some sort of copyright waiver from contributors so they can relicense in the future, or simply ignoring the TOS, and exposing themselves to summary deletion.

I was referring to the terms and conditions Apple imposes on developers that wish to publish software on the App store.

Again, what Linus does and doesn't is not relevant for a discussion about morality - unless you are starting a religion centered around him as a moral model.

Fyi, the App Store license is deliberately incompatible with the GPL.

In regard to proprietary hardware, it is my assertion that Apple and Amazon are essentially making the same move: throw money they obtained by dominating their respective markets into closed hardware development that will allow them to lock into their market dominance in the future by limiting consumer choice. You might disagree but there is nothing fallacious about this line of reasoning.

If you are looking for a fallacy take your argument from authority: if Linus does something then it cannot possibly be wrong. Which doesn't even get the point I'm making above, where I clearly contrast developing FOSS for the MacBook which is clearly a good thing (what Linus is doing), with buying a development MacBook for convenience, thereby putting your money to work against software liberty.

> MacBooks could easily end up becoming the laptop of choice for Linux developers

That is, if they want to subsidize with their own wallets the dominance of consumer-hostile computing that open source was created to avert.

Except for development such as this here, which aims at liberating users from a closed platform, I would have a philosophical problem to buy such hardware purely for my convenience.

Let's not fool ourselves, when proprietary hardware from platform giants like Amazon and Apple becomes dominant killing open ecosystems, the end of road can only lead to consumer coercion, regardless if a negligible fraction of users would be able to still run Linux on such hardware.