2,031 karma · joined July 30, 2010
There are other advantages — low fixed resource costs, global memory management and scheduling, no resource stranding, etc. — but the core intent of gVisor is to capture as many valuable semantics as possible (including the file system semantics) while adding a sufficiently hard security boundary.
I’m not saying moving the file system up into the sandbox is bad (which is basically what a block device gives you), just that there are complex trade-offs. The gVisor root file system overlay is essentially that (the block device is a single sparse memfd, with metadata kept in memory) but applied only to the parts of the file system that are modified.
I’m responding to what I believe is the intent of the comment, but I will also point out that on a literal level it doesn’t make sense. Whether something uses VMX and friends or not is a red herring, as gVisor also “rolls it own” VMM [1] and certainly makes use of VMX and friends.
[1] https://github.com/google/gvisor/tree/master/pkg/sentry/plat...
Process isolation is not the only tool that you have to build a secure architecture. In this case, capabilities are still being limited by available FDs in the first process (as well as seccomp and the noting namespacing and file system controls), and access to FDs is still mediated by the second process. There is no such thing as “being able to take access away … to a file that was already opened” as this is simple not part of the threat model or security model being provided. You still need to be diligent about these security mechanisms as well.
The idea that Google has given up and just does fuzzing is nonsense. Fuzzing is a great tool, and has become more common and standardized — that’s all. It is being added to the full suite of tools.
I believe the oft-cited reason cash is more expensive is primarily due to shrinkage, which is on the same order as the credit card fees (unless you’re a family-run business or have only amazingly trustworthy employees). Cash-only discounts can be just as easily explained by the “flexibility” cash provides — you can’t assume the conclusion as a fact. The fact that most cash discounts are greater than 1% generally supports this idea, since a business offering a 3% cash discount definitely earns less than taking plastic. (But the accounting “flexibility” is worth more than that 3%.)
Are you saying this is not true, and cash shrinkage is less than the fees?
(Obviously there are some consequences to this, wherein you may find yourself with heap-based allocation in hot paths that you didn't expect. But in terms of productivity, I think that not caring and having no distinction certainly reduces one thing that the programmer has to think about.)
I think there would be analogous design decisions that remove various low-level concepts in many languages. For example, Python ints are arbitrary precision, which removes a whole set of concerns from the programmer. Rather than being "weird", I think this is just optimizing for a particular kind of programmer productivity.
> Publishers' relevance on, say, Google, relies partly on sponsored contents. They pay Google for this.
I understand that Google has optimized to avoid the user clicking away. Obviously, this has been done to keep eyeballs and revenue but it is also arguably in the user’s interest as they accomplish their goals faster.
The logical response for this is not to force Google to pay for listing and summarizing links, but to evolve your own business model. Say by moving towards richer, more thoughtful content and analysis that can’t be trivially summarized and might even support a non-ad-based model, such as subscriptions or donations. (I pay for quite a few subscriptions that host content like this.) The basic news business has obviously become increasingly commoditized, and I don’t think anything Google does is going to change this.
The Wikipedia example is actually supporting the viability of good content being a defense for commoditization, not the argument that Google needs to pay content creators to survive. I don’t think there’s any evidence to support the idea that Google summarizing in a one box has harmed Wikipedia, as the Wikimedia foundation has continued to substantially grow its revenue every year and is at all time highs. Arguably, this is because they host content that is sufficiently rich that they are bound to have a decent click-through rate; whereas many news publishers are basically rewriting AP releases with little value add.
> They should be if Chipotle was virtually the only restaurant in town, and the only way for you to get food, which is essentially what Google is in the search space.
It is a strange proposition that if someone doesn’t like the only restaurant in town then it has a moral obligation to close its doors. It’s not like you have an obligation to go there; in this analogy consider opening a new restaurant as trying to do direct ad selling, but you can just as easily “cook at home” by shifting your business model. It might be different if the business was critical and had total monopoly (e.g. water, power) but I’m not sure this holds here.
Just as the restaurant is not required to cater to your specific tastes, I don’t think Google should be required to subsidize a specific publishing business model. It feels like a very narrow form of digital Luddism. Now maybe they are somewhat parasitic in this case, and would be smart to figure an more mutually beneficial arrangement (just as the restaurant would be smart to cater broadly to local tastes), but I wouldn’t go so far as to say “should” or expect any mandate to do so.
You claim that publishers are paying Google and Meta to be in these aggregators. But the aggregators are the problem and should spontaneously shut down? If the model sucks, publishers should stop paying.
The argument you’re making is basically equivalent to saying that Chipotle burritos are making me fat, therefore Chipotle should close all their locations instead of you know, me taking responsibility by not eating there.
It sounds like you would find Self [1] interesting (along with the visual programming environment).
[1] https://en.wikipedia.org/wiki/Self_(programming_language)
See the map here, for example: https://natural-resources.canada.ca/our-natural-resources/en...
Regarding the nanosecond point — I don’t believe that’s how information works, and there should be many obvious problems with the idea of an infinite information channel not to mention the obvious practical ones (propagation variability, lack of a reference point, etc.). There may be some optimizations, but generally the frequency (or frequency bandwidth, which is where the generic computing term comes from) determines the information capacity, and phase modulation doesn’t magically change this (it is actually what is used in many radio systems).
It's not directly equipping the audience with professional skills for a scientific career, but it is generating interest in science, and indirectly encouraging people to expand their own STEM skills and knowledge. I'm sure that dozens (hundreds? thousnads?) of people have gone into mathematics at least in part due to an interest generated by the likes of numberphile, etc.
I don’t believe it is precompressed as it hasn’t been processed by the visual cortex yet, no? Aren’t the optical receptors simply an artifact of the “sensor design”? E.g. if the refractory period of an optical receptor is 100x that of the neuron (or you simply need to cover a certain area, as you probably have tons of receptors attached to a single neuron outside the fovea and a small number per neuron inside the fovea), you’d hook up 100 optical receptors per neuron to use its full capacity. I think this is less compression and more combining a bunch of low information channels into a higher information channel.
All we really care about here is the amount of information reaching the brain not what your physical eye is capable of receiving, so I think using the nerve makes the most sense. There’s an interesting direct analogy: we don’t really care about the number of CCD sensors in the camera that took the image, we only care about how much information is in the video coming from the camera.
I think we could reasonably say that if an optical nerve has 1mm neurons on average, and they can fire at 250Hz at the most, that’s 250mbps or ~31mb/s per eye of uncompressed data as an upper bound.