flagrant_taco

The open web is dead if the only option is to base every browser on a Google core. How do specs fit into this if every browser is just a UI shell around chrome? Where is the competition driving innovation of the web platform when the only competitive landscape is how pretty you can make your toolbars and bookmarks?

We need independent groups of passionate people making sure that the web platform is focused on the best end user experience. We don't always have that today, but that vision is completely dead if the web functionality is decided entirely by an advertising platform.

Exactly, in that model the only way to build a browser is to depend on someone else to implement the specs. It's not feasible to build a new browser, at best you can start with what Google implemented, including proprietary features, and go from there. At that point you likely don't understand the codebase fully and are going to be dependant on staying in sync with chromium to get new features.

This model of starting with chromium really has nothing to do with building a web browser, it is just building a UI around Google's browser. There isn't much daylight between that and the often complained about limitation of Apple only allowing the use of WebKit on iOS.

I think this is actually a modern misunderstanding of the original benefits of computers. Computers were originally designed as a tool, they allowed us to work more efficiently but weren't directly taking over doing the actual work. A spreadsheet doid the calculations for me, but I had to provide both the data and the actual formulas to run.

AI, or more accurately machine learning, attempts to take us out of the process. It isn't separating content and presentation, it's separating content and understanding. The algorithm will read in data, compress it, and save only the patterns it learned. By the time the content is presented there is little connection to the original source and what I'm seeing isn't just a separated presentation layer, it's something entirely different like playing a game of telephone.

We aren't going to see browser competition unless and until we fundamentally rethink what the internet should be.

The specifications are simply too complex to foster any meaningful competition. There's too much ground to cover playing catchup and Google has such a large market share that, combined with their data and ad monopoly, allows them to create new specs and features that focus entirely on growing their business model.

Web sepcifications should be simple enough that it's pheasible for small teams to actually maintain a quality open source browser with modest funding. We've ended up falling into the same trap as many other big systems, we lost sight of what the goals and limitations for the web were and instead chased convenience and shiny new features above all else.

There's a big gap between the amount of oilwwe use today and zero oil, we can reduce without stopping.

I think your point leads down the path of arguing that we have too many people on the planet though. If we can sustainably feed everyone without causing unknown amounts of damage to the environment, we will eventually hit a wall and run out of ways to dodge that fact.

To what end? What will more data do when it's inevitably going to be incomplete and largely based on modeling?

We won't ever understand the problem space of the entire ecosystem well enough to make a complete solution. But there are easy things that can be done if we actually cared to, and we certainly don't need to keep piling on more things that almost certainly make things worse.

I think we may have a difference in degree when it comes to what we think may be coming. I don't expect the planet to burst into flames over night, but we have already done unimaginable damage to soil, water, and life on this planet. It can probably bounce back and maybe humans will even be here to see it, but in the nearer term I don't see it as a question of mitigating pain while the systems we have today continue to chug along.

I could very well be wrong, and if core systems don't break then I do understand the goal of just being less miserable today.

Oh companies are absolutely to blame, but so are the politicians we'd need to actually regulate them.

Enforcing regulations requires quite a few steps first, from defining the scope and likely granting new authority to regulators to actually drafting regulations that are complete and not full of loop holes.

Your argument is also based on the premise that consumers aren't able to make the best choices for themselves and must be forced into it by regulators. I simply refuse to believe this, what are we even doing here if we think real reform can only come from all powerful authorities forcing it upon us?

Jevon's Paradox is a thought experiment focused on increased efficiency not decreased demand. History is full of examples where a decrease on demand didn't lead to an increase in ridiculous uses for the technology.

I'm curious, why is this the goal rather than something like a decreased dependance on automobiles?

Any concerns that population growth may mean that a lifetime from now we have 50% electric vehicles but still a higher total number of ICE vehicles than we have today?

Defining standards of living in this way is very short term focused. If we reduce energy demand now but avoid a power grid collapse a decade down the road, is that better or worse for standard of living? And if we really are meant to believe that climate change is irreversible if we don't change course before 2030, who gives a damn about stander of lining for the next few years?

Solar and wind are absolutely limited, both by environmental conditions that change the energy potential and by our ability to store energy. Neither can handle surge use as well either, we can't make more wind or sunshine when everyone turns on their suit conditioner.

Nuclear is at least a viable shirt term solution. Combine that with reducing our total energy demand and we're at least buying ourselves more time to figure things out, that's a much better sort term path for sure.

How do we know these exist of we don't know how to access our use them? Without access we can only estimate the reserves, and without a way to use it we can't calculate the potential energy creation or storage.

We really are screwed off our only hope is blind faith that tech can save us. Technology can help, but we can't abandon the option of decisions that can already be made today.

Creating the Federal Reserve was a fundamental change. Confiscating gold then pegging the US dollar to a gold-backed value was a fundamental change. Getting rid of the gold standard and going full fiat sad a fundamental change.

The economic model, monetary policy, and incentives of each system were completely different. We may not have renamed the currency or the country, but that doesn't change the fact that those were major economic shifts.

Or entire system is based on predictable GDP growth. It impacts nearly every part of your life in some way.

Do you have any debt? Or use money? Or pay taxes?do you have a job? Did you know that oil consumption grows at the same rate as GDP? Do you save for retirement, or hold any investments? Do you have a bank account?

It's all directly connected to the economic system, which is driven by the goal of having steady and predictable rate of growth in GDP.

I sincerely hope that we aren't at a point where most people's guiding principles are minimizing their own poorness or misery. Those should be a given and self interest is backed into us, but we're in for a world of hurt if those are the only deciding factors take into account.

Some things are in fact worth dying for.

I'm actually arguing the opposite. We don't know precisely that any of the suggested changes I listed would make a clear impact on climate change. My point is that, regardless of what happens with climate change, we could cut oil use with little to no impact on our daily lives.

We don't actually know the extent of impacts from either our oil use or potential cuts. We have modeling data that we extrapolate out, the issue I'm raising is that we get stuck in a loop of people debating whether the modeling data is accurate, complete, or predictive.

There are changes we can easily make without knowing those answers. If we can make simple changes that at a minimum wouldn't hurt the planet and almost certainly would help it, why don't we? My read is that we simply must not care. We don't need to buy as many things as we do, or to expert those costs overseas. We don't need to travel the world by plane. We don't need the latest iPhone made on the other side of the planet. We don't need LLMs. We do it because we like convenience and novelty more that we actually care about our impact on the planet. And we avoid implementing real solutions by first demanding that we have a complete understanding of the cause and blame.

I'd argue that an economy and society built on the framework of terrible economic inequality is fundamentally unable to fix said problems.

I expect we would have a lot of agreement related to fixing that system to better serve the average person. Even if we have different opinions on exact details, that core principle can go very, very far.

An easier first step before nuclear is reducing our power requirements. This one gets laughed out of the room both because we don't think purple will accept it and because there's a direct correlation between GDP growth and energy consumption, we don't care enough about energy use to give up GDP.

The economic model is going to change one way or another. A model designed around perpetual growth will always fail, and there are already plenty of cracks showing.

A major economic change sounds impossible, but we've done it in the US at least three times in the last century.

It doesn't necessarily have to be done through further government intervention. Simply removing government protections and subsidies for industry would go a long way.

That's definitely not a complete solution (if one exists), but that could be implemented over night if we wanted to do it.

That opens up a whole apndoras box of whether a reduction is approach can lead to a complete understanding of complex systems. I'll leave that for another thread, that's a deep rabbit whole full of mostly theoretical debate (though an interesting one IMO!)

What you're describing is the path to inevitable collapse. If people will never be able to give up convenience or growth then we really are doomed, tech advantages only slow down that runaway train.

I don't actually think it's a fundamental blocker though. People sacraficed quite a bit at home during WWII to support the war effort, we could make similar sacrifices now if we saw our impact on the planet as a similarly existential threat.

To me it seems like the failure is in trying to analyze the problem so specifically that we can define the "right" path then logic people onto it. That will never work, mainly because the problem is so complex that there won't be a complete understanding to share. There will always be assumptions used in modeling and holes in the data allowing for alternative explanations that are then argued over even further.

We all know that using oil at this scale isn't a great idea. And we all know that we could actually use less oil without giving up much at all. If we spent less time arguing over how we got here we could find reasonable next steps that are hard to argue with when it isn't all backed by "science" and "fact" that can be debated endlessly.

The aggregate data is often more useful in commercial use cases, but plenty of use cases need the indivual data as well.

Private investigators, three letter agencies, and any company wanting to send mailers to my new address when I move all need the individual data to target me specifically.

I totally agree the aggregate data is given more value in a commercial market heavily focused on advertising and now training LLMs, my only point was that there are markets that highly value individual data as well.

The debate over exact causes has always felt misguided to me. The planet is extremely complex, human impact on the planet is largely based on estimates and modeling, and at the end of the day an understanding of why it's getting hotter isn't itself a course correction.

There are such obvious first steps we could take without a complete understanding of cause if we really cared. We could easily cut our use of fossil fuels if we weren't trying to balance that with an arbitrary GDP growth target. We could stop incentivizing people and companies to consume disposable products. We could stop subsidizing massive industrial farms and instead focus on eating locally and growing some of your own food.

The list goes on, but having spent decades watching people debate passionately over why the planet is warming all while going further down the same path my only conclusion is that we just don't give a damn. When push comes to shove we would rather buy new cloths imported from the other side of the world, throw resources at new construction and gut job renovations, and will happily fly around the world on vacation rather than stay home and enjoy the world around us.

Oh and don't forget about new tech. Machine learning rebranded as AI is so ground breaking that we have no choice but to produce as many GPUs as we can and throw all the power and water needed to run the damn things 24/7.

I'm not really sure what would be considered cheap vs valuable here so I don't have a great answer there. My point was only that individusl data is collected and sold often, it must be valuable or there wouldn't be a market.

I'd be curious to see how the price compares when selling 10k individuals' data versus aggregate data the 10k people. Presumably if it was cheaper to buy all the individual data I would do that and aggregate it myself.

I'm not so sure that Google's data is even their moat, their monopoly on everything related to ads might be the main thing keeping that so afloat.

Individual data is extremely valuable, they wouldn't collect and store it individually if all they wanted was the aggregate.

Identity-specific data is sold all the time for everything from advertising to credit scores and what would otherwise be called social credit scores of they were run by the government instead of private companies.

Seeing a price up front is just a different business model and localization could be easily handled with a setting on your account. Neither is tech-dependent, both could have worked over SMS too.

Real-time mapping would be tricker if that's really a killer feature for you.

I assume it varied greatly by region, but my experience was that it was extremely difficult to find workers during the pandemic response.

It seemed like a combination of fear and restrictions, plus free money at least in the US. I saw hourly rates for physical labor jobs go through the roof, double the pre-response rates and still couldn't fill positions.