This is so unreal. Resembles so closely the novels we read in high school. The Giver and Nineteen Eighty-Four come to mind. Prisoners feeding rehearsed lines to the media about 'enjoying' their incarceration.
In my AP European History class, we just now studied the post-war effects and especially the relationship between Poland and the Soviet Union. To stumble upon this essay now makes it all the more interesting; a primary source of sorts to complement the top-down view of my textbook.
It's exciting to see such an incredible rate of advancement; I can't wait to see what transistors at 7nm and less will make possible. I can still remember when these same components were measured in millimeters.
Chips that are just like the ones we have now but 30% cheaper? I don't see how this is going to open up whole new technologies that weren't possible at 10/15nm.
Maybe not, but better battery life and eventually cheaper hardware on mobile and low power devices, expanding the range of usefulness in the developing world especially, is no small thing.
One of the biggest advancements due to this kind of tech has actually been throughput for radios. By using more sophisticated / denser encodings, we can scale the same 4G tech to even higher speeds. Think 1gbit/s speeds for mid-market phones. This is effectively enabled due to the smaller node sizes also being more power efficient, which is required for mobile phones.
The advances are less in your face, but still there.
It's easy to think we need faster processors when the architectures we use, both software and hardware, are incredibly inefficient.
Any modern processor is much faster than the RAM its programs and data live on, hence the multiple levels of cache inside them. Multi-cores add the complexity of keeping the memory consistent. Multiple threads per core put additional pressure on caches while our OSs assume all processors see a unique and consistent memory image (which requires keeping caches consistent across cores). Our most common software doesn't run on GPUs before extensive changes. Only mobile platforms are exploring asymmetric multiprocessing with a single ISA now.
I don't think we need faster single-thread performance. What I think we need is to adapt our software, which is designed to run on ridiculously fast copies of ancient personal computers, to run on computers that instead of mimicking a successful product of the 80s resemble more what we can do now. We need software that exploits the SIMD units (predicate bits that prevent branches are your friends there), that runs well in multiple cores and OSs that can deal with memory inconsistency between cores (maybe using write-through instructions for shared data and write-back for process-local stuff). We need software that doesn't need complicated instruction reordering or speculative execution.
> due to Amdahl's law, it's always better to have a single faster processor than multiple slower ones
This is not what Amdahl's law says. Amdahl's law gives you a theoretical maximum speedup given a program with a sequential component and a given number of processors. So as long as that speedup is higher than your faster processor is faster then it does make sense to have multiple slower processors than one faster one.
There are many products that don't see the light of day because they can't meet their COGS target. Cutting a huge chunk by 30% could open up a bunch of amazing functionality.
I think the point is that the rate of advancement has already slowed (18mo=>3yr) and is slowing even further (5nm by 2025). The treadmill Moore's law is basically over. There will be significant advancements (AllAroundGate, Wafer Stacking), and we might even see a speed increase again, but the steps are more orthogonal and less all improving. The last thing to keep improving is $/transistor, while MHz and power long since (2005=>2010) stopped getting better with each node. However, if the CFOs ever figure out that the return doesn't meet investment ... that will be the end (at least in the west).
I would say the rate of advancement is kind of plateauing. From the p4 to athlon 64 to c2d the gains were quite amazing. I guess in mobile you still see advancement, but do you really notice the difference between a snapdragon 801,820,835?
I do not understand the dislike this community harbors for AMP. I personally really enjoy the system; whenever I'm searching for any type of article on my phone (Android), I prefer AMP pages, because they load faster and are far more responsive than some of their more bloated counterparts.
It's not just about responsiveness. About half of the AMP pages I open on Safari on iOS literally do not scroll at all, instead scrolling the top bar with the fake URL on it, so as a user AMP is literally breaking the web.
I think the frustration here is seeing the decay of the open web and the rise of AOL-esque walled gardens in which your Facebooks and your Googles own the access methods, and every other third party is subordinate to them.
These guys at Google are abusing the web in order to prevent you from navigating away from google hosted content (I'm a technical user and it still took me way too long to find that menu in the corner), in much the same way that Facebook apps make it hard to escape their respective walled gardens.
I'd like it if people made non crappy websites instead of amp sites that have slightly different behavior than normal websites. Amp adds nothing of value compared to just writing a non bloated site
The fact that you say android explains everything to me. I'm sure Google made sure it was a decent experience on their operating system. But it's clear they didn't give two seconds thought to iOS, and haven't bothered to improve it since it was released.
It really does make google search feel broken. Many of the top results no longer "work right".
Ignoring all the other issues of who is in control and whether it's a walled garden and all that other stuff… On iOS it's a terrible user experience. That ALONE would make me hate it.
Well now you know how most of those web pages felt on Android before - web devs tested on mobile Safari and left pages broken and utterly stuttery on Android web browser.
Why do you expect Google provide a good experience in iOS? Heck I can't even connect to my Apple calendar in Android. Does Apple care? Is there a apple mail app for Android? AMP is fine in all browsers except Safari. AMP is great for browsing with 100£ cheap phone when I was in Thailand.
This community hates AMP the cache, AMP the centrally controlled place for all news which Google can easily manipulate or censor if a totalitarian government should ask them to, the AMP that gives pages an advantage in search results only if they let all their content go through Google.
No one's complaining about AMP the web framework — which is the part responsible for the performance and responsiveness.
My parents were born and have lived in Pakistan for their entire childhood up to adulthood. They then moved to the United States, where I was born and have lived for my entire life.
On occasion, I visited my parent's home in Pakistan, although the trips became increasingly scarce as grandparents passed (now all four are gone).
The house has always had domestic workers present, a single family. The first few times a maid, who we called Maacy whose children went to a nearby school and whose husband would work nearby.
Her family was among the lower classes of the country, and her salary brought her and her children opportunities she could never otherwise have had. I remember talking with her in the small room that was her quarters; sometimes I would help her with her work.
On later visits, she had left and another family, whose father served as the household driver occupied the same space. One time, I saw the previous maid attending a gathering that was taking place. She recognized me immediately, sharing an embrace and leaving me on my way.
The situation you describe seems to involve inequality, but it's significantly different than what "Lola" is described to have endured. The decision to move to the U.S. was entirely the author's family's. Lola was promised that she would get paid enough to move back, but the family never kept that promise. When she begged for them to allow her to visit her dying parents, they refused. And then her 5-year visa ran out, and because the family did not deal with that lapse, Lola was now forced to live in slavery forever because trying to send her back would jeopardize the entire family's immigration status. Stupid Lola!
The author is a great writer, and obviously a distinguished journalist. If this article had been written by the real-life version of "Silicon Valley's" Peter Gregory (including the untimely death), I have a feeling sentiment toward the piece would be much different.
You may be surprised, but the United States may be one of your contenders.
I recommend Private Internet Access [0], which what I use as my VPN service.
Ultimately, it's about trust, and it is very difficult to determine which services ensure your privacy and which ones might sell your history or cooperate with prying governments.
US is a contender as long as you're willing to fight the US government in court, and you'd likely have to. The law may ultimately be on your side, but that doesn't mean the US government won't try to drop the hammer hard on you if they really want your data. They know most of the small companies and even some of the big ones scare easily, so the aggressive attitude is well worth it for them, even if they know they're doing unconstitutional data requests.
Switzerland and Germany are probably the best right now, although Switzerland recently passed a surveillance law, too, but I'm not sure how it affects VPNs. Germany has a privacy-friendly Constitution and resides within the EU, which also has strong privacy laws. However, Germany is also trying to limit privacy rights, so this may be a little unpredictable right now. Germans usually tend to care more about privacy so they may fight this, though.
More than anything I would recommend a "zero-knowledge" approach, so use the strongest and most ephemeral VPN encryption and generally try to discard user-related data as soon as you get it. And put it all in your ToS/Privacy Policy so that it's legally harder to lie to your customers when you're forced to not do that anymore.
Although I agree with many of Stallman's points, it is, unfortunately, true that this post will most likely not change anything at all.
What percent of Uber users read Stallman's website or even HN? Probably less than 1%.
The sad thing is the general populace of the United States (or even, the world) doesn't care if their data are being tracked or kept; or if their personal information is being recorded; the work of people like Edward Snowden has done little to change the behavior of most internet consumers.
It is enormously difficult to change the behaviors of these users, probably even impossible.
If you are taking this as an argument for e.g. linux over Windows, Certificate pinning is a good thing though? It helps prevent mitm attacks, as stated in the article.. (Not to say that you shouldn't run linux )
I'm a beginner when it comes to software development (mostly web development), but it seems to me that the majority of complex exploits like this involve some type of memory overflow and subsequent code execution.
Shouldn't there be methods for detecting these kinds of things in source code or more priority given to preventing it in the C/low-level community?
How is it possible to put the malicious code in the correct memory spaces? Unless the attacker had a full image of the memory, I don't see how this can be accomplished.
The second bit of the exploit chain, CVE-2016-4655, leads to disclosure of kernel memory addresses. Once a single memory address is known, you can calculate the random offset of the kernel, and then exploit the third part to overwrite the return address and return into specific chunks of kernel code ("Return Oriented Programming"), whose addresses you computed from the offset + a fixed code location. These can let you e.g. install your payload.
