I think the author misdiagnoses the problem, and the proposed solution simply hides the centralization instead of removing it.
The reason AWS is expensive is not because of IPv4, or the datacenters. It's mostly in their software/managed offerings, and the ability to quickly add more servers. If you are a "serious company" and you don't want to pay AWS or a similar company, renting a rack and colocating your own servers (either within your premises or in a datacenter) is doable and done by lots of companies.
I disagree that certificates have caused centralization, and they're not something separating the haves and have-nots and are in no way comparable to having or not a mainframe. HTTPS becoming pseudo-mandatory didn't push people into having their own (sub)domains, which is nowadays the only requirement to obtain a certificate. It already happened out of convenience.
The other point of centralization mentioned is DNS, which tailscale doesn't avoid at all. MagicDNS still relies on the ICANN root, as does the tailscale control plane. And if all you wanted was a free subdomain, there are plenty of people offering that.
If you are behind CGNAT, tailnets aren't particularly less centralized, as traffic has to flow through the DERP servers. I doubt tailscale can keep providing these free of charge when the volume is in the tbps instead of the gbps.
I agree that tailscale (and similar solutions) help in the last remaining case, which is accessing your computer that is behind a NAT. I even think they could reach the dozens of millions of users. This is, in my opinion, not enough to claim the title of "the new internet".
Actually I'm not against the suburbs, I simply agree with the article (just a bit) that the classic suburb model have issues because you can't buy anything nearby and so you need long trips just to buy some milk, that's is. I'm definitively against the city model, because IMVHO there is no way to turn a city into a "better city"m a city is just "a tool" you can use and than dispose and rebuild another, witch is practically next to impossible and so costly that we can't do on scale.
BTW if peoples living in the suburbs like them there is no issues at all, I do not denigrate anyone of them. IF some would like to live still in single family homes BUT also with some shops around without being in the city than the solution it's easy in nature or adapting a suburb, that's because suburbs and empty space could be changed a small step at a time, while cities can't.
That's is. I'm definitively not pro-urban as the article author.
- Discovery. Finding related papers in the same field using similar approaches is almost impossible.
- Reproducibility. Almost no paper e.g. in the AI field is reproducible. Try to implement it yourself, and you'll find out that they faked almost all outcomes and argued it's your fault because you don't have the same datasets.
- Objectivity. A lot of papers desperately need to publish a successful story, and are bullshitting their way out of failure in hopes to get more funding this way. Why are failed ideas not publishable? Pretty insane, tbh, that elsevier and other conglomerates think this is how objective research is done.
In computer science, I don't even bother reading a paper that's not on arxiv. If it's an AI/LLM/ML related paper and it doesn't have a github repo, I'm assuming it's a useless paper, unless I hear from other sources that reviewed it. It's just not worth the effort.
Additionally, specifically in Computer Science, I'd argue that using math to describe an invention is too abstract. We need a standardized foocode programming language that can be more efficient to communicate things. Just look at the backpropagation paper, describing two nested loops, basically.
At the turn of the 20th, eg 1900s, most tax was paid by corporations and resouce taxation. At least it was in Canada. An example, a "temporary" personal income tax started in 1917 to fund WWI in Canada. Corporations where hit too, but my point is that tax revenue came from elsewhere. Resources.
Over the last 100+ years, slowly, primary taxation has shifted from corporations* and resources to individuals. One can see why this makes sense, as policy with respect to trade has shifted as well. We've become more global, as eluded to, and the least up to this century the goal has been on "reducing trade barriers". So tax has shifted to "consumption tax" (gas tax, sales tax, sin tax), and personal income tax, as it is harder for people to be in multiple tax jurisdictions.
At least, this is how I see it. Your citizens live here, own property (your municipality gets income), have to eat/do things locally (consumption tax), and can't claim they live elsewhere easily, for the "average Joe" is just going to have a simple tax structure, and it's known where he hangs his hat.
I did a cursory browse through few sections of Chip Huyen's book (namely the CV module), and I think the questions are on the easier end. Normally, I would face some more depth (and equivalently as a tech lead, similarly ask more than surface-level questions to potential hires). If you have gone through an introductory ML course like Andrew Ng's CS229 or CS230, these questions seem obvious and trivial to answer.
While you are correct, you are also not reading it correctly.
(d) refers to "obscene or child pornography".
While some pornography is obscene under US law, in general pornography is "indecent", not obscene.
See Sable Communications of California v. FCC, from when the US tried to regulate dial-a-porn, and the courts decided the original law was overly broad.
Only (b) refers to "indecent communication" (and only in the context of "commercial purposes"), and it specifically requires a telephone.
If I understand the history, that was from the 1989 amendment to the Communications Act of 1934, known as the Helms Amendment, and the courts decided that was sufficiently narrowly tailored to not affect 1st amendment rights (see the court judgement in Dial Information Services v. Thornburgh).
It appears, based on my shallow understanding of the history, that I can set up an automated recording such that if you call it, it plays back a recorded sex chat which is indecent but not obscene. And I can do that without requiring age identification from those who call.
I might have to inform the telephone company that I have done so, so they can let people who opt-out of being able to connect to such numbers from doing so, but otherwise it's my first amendment right.
Which circles us back to yet another analogy which seems more appropriate than your focus on the sale of pornography in physical stores.
Look up the Dutch tax return scandal where the Dutch tax arm of the government (‘IRS’) used machine learning to identify fraud but it turned out to be very racially biased and it uprooted thousands of families with years of financial struggles and legal battles.
> start a low-margin, multi-million dollar capex business and start building their own cars and tvs
(Sorry, I missed that part.)
That is what we need. That is not something «most consumers are not about to start»: it should be a drive to have a few serving the rest.
An easier and (on an important side, fail-safe resilience aside) better solution would be a society that would not have let the situation happen in the first place.
because “income” means 3 different things to the IRS
there are income taxes
and then there are taxes on income, which is a category of earning types, which has subcategories passive income and earned income. but not to be confused with capital gains tax, which is under the umbrella of income taxes, after capital gains taxes have been computed.
it is very easy to reduce income taxes (are you still following) despite high passive or earned income. so yeah they should probably look into them
I’ve had 7 figures of earned income but had negative adjusted gross income, the President even sent me a stimulus check because my reported income was so low. I just blamed the White House and Congress for not differentiating and moved on.
I don't use Facebook but for example KLM/AF at least 5-6 years ago pushed hard "means of contact" as "talk to us at whatsapp/facebook". And they weren't the only one. And the track of thinking was probably "If everyone/millions use it then let's move away from this old pesky email"...
There are groups/communities (real ones, for example building or school) that settled on WA/FB group so while you are not necessarily forced to if you don't there is a high chance that you will miss some crucial info…
I don't use it in any manner for personal communication though…
Converting web pages to Markdown is a common requirement. I have found that turndown does a good job, but it cannot meet the needs of all dynamic web page content. As far as I know, if you need to process dynamic web pages, you need targeted adaptation, such as Google extensions such as Web2Markdown.
In my specific case Bosnia & Herzegovina, but we're far from only ones. They list 46 countries, though I'd describe it more as EU + 19 others: https://stripe.com/global
PayPal, for all its faults, is still a far more universal option, but that means very little if people keep making Stripe-only tools. It has become a habit of mine to immediately look for the FAQ to see if PayPal is supported.
See also The Linux Memory Manager: https://linuxmemory.org/chapters Last update the author sent out was in early July noting that the book is now in editing:
> I am happy to report that I have completed the first draft of the book [...]
> I am now in an editing phase, which may well take some time. Sadly I can't give a reasonable estimate as this will be done in concert with my publisher.
My understanding is that frequentist statistics was developed in response to the Bayesian methodology which was prevalent in the 1800s and which was starting to be perceived as having important flaws. Thus the idea that the invention of Bayesian statistics made frequentist statistics obsolete doesn't seem agree with the historical facts. What actually happened was quite the contrary.
The production limit isn't in the amount of VRAM. They could offer any of their existing consumer GPUs in versions with more VRAM with prices that add twice their own cost for the extra VRAM (but still far less than the datacenter versions) and thereby sell the same GPUs for higher margins.
Ok, I want to publicly apologize for the python thing
I cannot live with myself unless we are fully committed to emacs and common lisp
I don't care if I am banned from IRC or other platforms, I am used to being banned from marketplaces for highly aggressive prospecting
The fact is, until there are COMMON JOBS using MUUPI EMACS and COMMON LISP as the default standard, I can't look in the mirror
I set out to get all my homies out of rotmaxxing and to QUEEN USERS, that's what I'll do or die. Not until all my beautiful birds are free.
https://www.youtube.com/watch?v=QPntjTPWgKE
I have switched to Gentoo as my main Linux distribution for desktops and laptops around 2003. Gentoo still has reasonably good documentation today, but in the beginning its documentation was really outstanding and I have learned a lot from it.
Then, some time more than 10 years ago, perhaps around 2010 or 2011, the Gentoo package system began to have some annoying problems and I have discovered Funtoo at the same time. At that time Funtoo was clearly better, so I have switched to it.
I have used Funtoo for about a decade on a large variety of computers with great success. Then, a few years ago, the package system began to have maintenance problems, like packages that were not updated for long periods, becoming obsolete.
At that time I have switched back to Gentoo, which I am still using.
Due to their great similarity, switching between Gentoo and Funtoo is very easy. Therefore anyone who still uses Funtoo can easily go back to Gentoo with minimal effort.
In any case, I am extremely grateful to Daniel Robbins, the creator of Gentoo and Funtoo, for his work. Like also the work of other great contributors to open-source software, it has saved many years of my life that would have been required to duplicate the results of such work.
> it doesn't even have to link to the correct place.
That's an obvious DMA violation. You can't preference your own service over others, like when you link to the exact pin on your service but just the general area on another.
The reason AWS is expensive is not because of IPv4, or the datacenters. It's mostly in their software/managed offerings, and the ability to quickly add more servers. If you are a "serious company" and you don't want to pay AWS or a similar company, renting a rack and colocating your own servers (either within your premises or in a datacenter) is doable and done by lots of companies.
I disagree that certificates have caused centralization, and they're not something separating the haves and have-nots and are in no way comparable to having or not a mainframe. HTTPS becoming pseudo-mandatory didn't push people into having their own (sub)domains, which is nowadays the only requirement to obtain a certificate. It already happened out of convenience.
The other point of centralization mentioned is DNS, which tailscale doesn't avoid at all. MagicDNS still relies on the ICANN root, as does the tailscale control plane. And if all you wanted was a free subdomain, there are plenty of people offering that.
If you are behind CGNAT, tailnets aren't particularly less centralized, as traffic has to flow through the DERP servers. I doubt tailscale can keep providing these free of charge when the volume is in the tbps instead of the gbps.
I agree that tailscale (and similar solutions) help in the last remaining case, which is accessing your computer that is behind a NAT. I even think they could reach the dozens of millions of users. This is, in my opinion, not enough to claim the title of "the new internet".