Such format isn't really necessary, because http has protocol level compression that can get rid of text-based format inefficiencies. And in case you need it precompressed there is a compressed svg format - svgz.
I agree that svg is not dying, it's still growing.
Probably the biggest thing is to separate edge servers from backend servers, running DNS and reverse http(s) proxies (aka load balancers) on the edge servers. This would allow you to achieve high availability on the internet facing layer and below. Use DNS routing and DNS failover to steer clients to specific edge servers and to avoid unavailable ones. The more edge servers you have from more independent hosting providers, the higher availability you can get. They can be cheap VPS servers. Failover between backends then can be handled on the edge servers on the reverse proxy layer. Backend servers won't even need physical IP addresses, they could tunnel to edge servers and be located anywhere. You would need to handle connectivity problems though, you can start by using fail fast approach with very low connect timeout and low timeouts and such to quickly and seemlessly switch between backends. I don't know of any off-the-shelf solution that can handle connectivity problems well, so at some point you will have to write your own tunneling proxy, but before that you can survive on nginx, it can mark backends as failed and avoid sending requests there for configurable amount of time. Start with at least 3 edge nodes from different hosting providers and 2 locations for backends, could be a single location if you can get 2 independent ISPs there.
> Like, would dynamic typing come off looking better if the selection of dynamic languages weren't dominated by warty messes like php and ecmascript?
It already looks better if you consider that dynamic typing is more expressive and such languages take much less code to do the same thing and less time. This is not something that can be researched via commit messages, but it is rather obviously true, so the claim from the paper that "static typing is better" is also obviously false and is way out of paper's reach to even claim. But then again this is CS and such claims that go way beyond data and author's expertise are super common in CS papers. Personally I stopped taking CS papers seriously a long time ago.
All it takes to avoid that is looking into commit histories of projects in each language and figuring out how they mark bugfixes. If it's suspiciously reliable you would know you made a mistake classifying bugfixes somewhere or your sample is too small or the programming practice in the language is unusual and needs closer look.
> The first step should be to show that the github dataset can be used to say anything about quality and productivity of things like language choice.
The dataset can be used to say how defect rates compare across programming languages, that's important given that in CS and PL research these things are usually claimed left and right without any evidence. What the dataset can't say is how that translates to quality and productivity, as that would require a lot more than just commit messages. But at the same time it's something people already have some rough ideas about, like that people writing in low level rigid inexpressive languages are less productive than in high level flexible expressive languages.
> You risk being without a caching layer for a large period of time if you need to restart Redis, where Memcache, due to the lack of persistance is ready right away. Honestly I'm not sure it that still an issue, but I believe it is.
The bigger issue here is starting an empty (cold) caching instance and overloading backends because all the requests are MISSes and go to the backends. Persistence allows you to avoid this problem. Even if you run caching servers in some geo distributed setup, persistence will still help by reducing unnecessary geo bandwidth.
It's pretty hard to find even a single use case where in-memory caching server without persistence is better than the one with.
There isn't a guide to learn how things fail on platform, compiler, version, options combo. Probably the only things that force you to systematically research and accumulate such knowledge is by writing a language VM in C or use C as an intermediate representation. Even writing a C compiler itself won't teach you that.
No, the author's idea is ok and he's mostly right on the cost part. Configuration and the numbers are a bit off and unrealistic, you won't get such low 95% availability per site due to other economical and technological constraints, you'll get at least 99%, but probably closer to three nines per site and 64 out of 96 won't be necessary at all (something like 8 out of 12 could be enough). Dropbox designer is just ignorant, biased and conditioned to US market and environment, but appeals to authority, so people upvote his bad comment. I do storage too, on smaller scale than Dropbox of course, not in the US, but it is distributed and the cost is already lower than what you see in the title.
Should be obvious, because they leak through the sides noticeably when you breath out and if you are sick and coughing spunbond will just get wet and nothing will go through it at all, only through the sides, spreading it everywhere as if there is no mask.
Rebranding is a thing. It allows corporations to get rid of past negative associations with the brand, be more attractive to younger audience by not looking too differently from brands they are exposed to, even just remind everyone they still exist and should be considered as a choice, things like that.
> As a Kubernetes maintainer, I WANT to say that a smaller number of larger clusters is generally a better answer.
I'm not trying to nitpick here, but that justification is awful. It goes against reliability engineering on a deeply fundamental level, pretty much guarantees to make already not that reliable things even less reliable. Generally the more isolated entities you have and the smaller they are the less they affect each other and the environment when something bad happens, the faster they can be recovered, the fewer end users they affect, etc. If I remember correctly, this is even how some Kubernetes people justified ideas behind Kubernetes itself that you want to drop now.
People are not isolated from the rest of the world into specific areas, they know what broadband internet is like and this sets expectation of quality. I know of some organizations that use links on that frequency and not working in rainy, snowy or foggy weather always comes up, but they can deal with that, while consumers can't.
99.95% availability (three nines) is something a single datacenter infrastructure could do, it's even typical, but 99.5% is actually pretty low, that's worse than running a service from home with a single home internet connection.
It also requires clear weather, so won't be enough to "wire" a village with reasonable expectation of service quality. At which point there is little reason to go with anything, but a passive optical network in a village.
It's not copyright infringement either. They are basically looking at your user agent features, country or whatever to sell you subscription, it's like upselling or even freemium. It's weird that people even entertain the idea of calling theft or copyright infringement the act of not going through someone's silly selling technique.
I'm sure they can afford to buy rights to important non open source fonts or fund the design of similarly looking open source fonts to ship as aliases. There is no reason for the madness with fonts to continue.
They could ship web browser with all common web fonts to render them from disk without fetching. This would allow disabling of fetching of web fonts without breaking websites. Also could eliminate fonts as a fingerprinting vector altogether if system fonts were ignored as well.
There is an actual privacy community with Tor at the center of it, with plenty of research and serious ideas. And there is "privacy" propaganda that Mozilla and other corporations are putting out, it doesn't do anything good for privacy whatsoever, it's all bullshit. So calling it "fake sense of security" is somewhat appropriate here.
It's just A/B testing happens to align well with being an employee, not caring about the company, the future of the company, only caring to impress your boss short term and then getting promoted or leaving for another company. It's like a perfect selfish short term instrument for that, it doesn't burden you with mid and long term consequences.
But it's actually pretty hard to find something where A/B testing would be appropriate, outside of measuring effectiveness of ads of course, since they are naturally short term and consequence-free.
Well, currently adblocking is already full of such easy to detect overrides, and extension APIs already have plenty of limitations on what they can touch and replace allowing websites to easily bypass most of the overrides, but they do work at the moment and websites don't bother.
Seems like this is where things are going. None of the main browsers respect users enough today to the point that they might not even stay usable without forking.
No, well, maybe, depending on your country. I know some countries have laws preventing ISPs from interfering with content, but even those laws do not apply to technical measures, like DoH, they are still completely free to block it.
DoH can be blocked by IP addresses, DNS canary and probably SNI, while DoT by IP addresses and port number. So "DoT is DoH with a kill switch." is again nonsense.
There is an actual research on this. More than 90% of alexa's top 1 million websites (more like 95% or so) are uniquely identifiable just by IP addresses you connect to when you visit them. This is not a single IP address per website, but multiple, because websites include subresources with other IP addresses to connect to.
On top of that, there is a pressure on CDNs and clouds to make websites stick to specific IP addresses and make them blockable by IP addresses without affecting other customers. So far they have proven to not go against this pressure and even made technical solutions to aid governments to identify websites by IP addresses, in particular by fixing L7 routing layer to block domain fronting (which is important, because ESNI is basically domain fronting, but crappy, and the pressure didn't go anywhere, so there is the same pressure to make sure ESNI doesn't interfere with identification of websites you are visiting).