HNHacker News
TopNewBestAskShowJobs

zeotroph

508 karma · joined August 2, 2012

submissionscomments
zeotroph··on Robust and Efficient Elimination of Cache and Timing Side Channels (2015)
This seems to require user input

"In our solution, developers annotate the functions perform- ing sensitive computation(s) that they would like to protect."

and assumes that the cache timing is used to gather info on a secret function, so it does not consider the usecase of dumping everything a process can see?

zeotroph··on More details about mitigations for the CPU Speculative Execution issue
Related question about spectre and the cross-process(!?) mem read possibility: the attacker "influences the victim to execute the gadget speculatively.", and then the "areful selection of a gadget" (i.e. in the victim process) can be used to read arbitrary victim memory.

How is that possible? I can understand an attacker reading its own memory, but foreign one via a common mapped gadged section... that is just incredible!

zeotroph··on More details about mitigations for the CPU Speculative Execution issue
See https://millcomputing.com/topic/security-on-mmooo-processors... and https://millcomputing.com/topic/meltdown-and-spectre/

The Mill has portals (possibly) gating everything and making microkernel design easy and fast, so Meltdown would not work. Spectre, not so sure, there are claims the NaR (not a result) option might prevent it.

I bet this will come up in the next talk, looking forward to it.

zeotroph··on Intel Issues Updates to Protect Systems from Security Exploits
This branch prediction inside a "gadget" needs to be mapped into the address spaces of both attacker and victim (i.e. shared lib). But even if the attacker can get this code to run on both sides and examine the dependent timings to find out underlying values, the victim code can't be tricked to sequentially read-by-speculation all its internal data.

Does the instruction cache get poisoned as well on a mispredict and carried over?

zeotroph··on Intel Issues Updates to Protect Systems from Security Exploits
Damn, that is indeed possible. And I thought I understood these conceptually, now I have to read the papers again.
zeotroph··on Intel Issues Updates to Protect Systems from Security Exploits
No, Spectre can only read its own memory, which makes it boring for programs, but interesting for stuff running interpreted code because it can suddenly peek outside, e.g. javascript.

Meltdown goes one step further in that it can read kernel memory (but not memory of other processes either, unless it was explicitly mapped in).

zeotroph··on Intel Issues Updates to Protect Systems from Security Exploits
Using a thread to keep mispredicting as desired is just one (though faster) possibiliy. It can be done in single threaded mode as well.

Also doing that on purpose would again worsen cache performance.

zeotroph··on Eelo: A Google-less Android alternative emerges
Or an android which can properly box apps, i.e. "Yes you can have access to my phonebook (because if I refuse you won't run at all..) but the container you are running in just gives you dummy entries", or not allowing internet access for certain apps.

This is all possible on desktop linux, having that possibility on a rooted Android would make me switch back.

zeotroph··on Firefox Quantum
> I’m sure Firefox 57 will also get a fair share of sour feedback and comments written in uppercase.

I might be part of that crowd (though in lowercase) because none of the WebExtensions mouse gesture addons work on macOS/Linux anymore due to this outstanding bug: https://bugzilla.mozilla.org/show_bug.cgi?id=1360278

No official dev seems to look into it, and the community patches from the tracker only fix the Linux case.

zeotroph··on Zune Diehards
I am regularly on the lookout for one which plays opus/flac plus the usual other formats and has much more than 4G storage, but I fear this type of gadget will die out before long.

Or hopefully the Arduino and Raspberry type of hardware gets small enough that I can build my own (with which I will probably never be allowed on an airplane).

zeotroph··on Zune Diehards
Tactile buttons. A touchscreen is impossible to operate without looking at it. I assume most of us MP3-stick users have listening habits which requires us to change the track or volume every few minutes, if not more often.
zeotroph··on Show HN: A simple, fast and user-friendly alternative to find, written in Rust
I actually forgot why exactly I switch from rg to grep for binary files, but after some tests I found the use case grep can but ripgrep can NOT do:

  $ grep FOO *
  Binary file libBar.so matches
  foo.txt:file with FOO in it
rg/rg -u/rg -uu/ ignore libBar.so completely, whereas rg -uuu spits out binary garbage.

Here grep's default behavior is what I expected: I want to know which binary files contain some symbol, details are less important (yes, I could use objdump or smarter tools, but grep is good enough 99% of the cases).

rg -u should list the matching binary file, maybe rg -uu should walk a few (printable!) chars to the left and right of the binary match and -uuu should just assume it is text, i.e. the user is right and the auto-guess wrong.

zeotroph··on Show HN: A simple, fast and user-friendly alternative to find, written in Rust
> Not to mention the "opinionated" choice to ignore a dynamic set of files and directories

That has bitten me a few times when using `rg`. Sometimes (but not often enough to remember the switch) I want to grep a lot of binary .so files for a string and am left wondering why nothing was found before just replacing rg with grep. Just calling `TOOL PATTERN` without switches is the most intuitive thing, and there I agree with `fd` to make `fd PATTERN` the default. Unlike BSD find which always needs at least a `.` to function.

> chained patterns.

The -or/-and and are indeed nice, but the -prune logic should get an update. Maybe when fd grows from 80 to 90% of use cases.

zeotroph··on Why is it so hard to swat a fly?
Also, what is the scientific name of the killer fly? There is no such article on wikipedia.

And some more references on how compount eyes see mechanically, as opposed to chemically, would be interesting.

zeotroph··on Why is it so hard to swat a fly?
I thought that most (all?) eyes only see motion, just as you do not feel the weight of your clothes anymore (or the presence/absense of your wallet in your back pocket) after a while. But reading up on microsaccades[1] - tiny involuntary eye movements which I thought fixed that in mammals - that does not seem to be the current understanding anymore.

1: https://en.wikipedia.org/wiki/Microsaccade

zeotroph··on Why is it faster to process a sorted array than an unsorted array? (2012)
I was quite certain that this was the famous post with the railroad junction picture before I followed the link. This was submitted 3 times already (but only sortof gained traction once):

https://news.ycombinator.com/item?id=12490893

https://news.ycombinator.com/item?id=14459549

https://news.ycombinator.com/item?id=12272428

I am not against resubmissions, but if it is an older post the year should be in the title.

zeotroph··on How big oil will die
Indeed, even when using fossil fuel as the energy source for both a gas-powered and an electric car, the well to wheel efficiency is 14-33% vs 42%, respectivly.

Source (2013): http://www.asahi-net.or.jp/~pu4i-aok/cooldata2/hybridcar/hyb... (it also claims a 42% efficiency for diesels on highways)

zeotroph··on Travel the path of the solar eclipse
"3/4 or 1/2 the maximum day-night temperature difference" says eclipse2017.nasa.gov[0], space.com has more details on that[1]. If this drop means the air cools below its current dew point you get fog. Seen and felt it twice with partial eclipses (both of which just happened to be where I was, so no loss).

[0] https://eclipse2017.nasa.gov/typically-how-big-temperature-d... [1] https://www.space.com/37201-solar-eclipse-temperature-drop.h...

zeotroph··on Travel the path of the solar eclipse
> clear day

That is not enough though. Since the temperature drops during any eclipse the air can suddenly hold a lot less water and you have fog ruining your view. So a place with lower humidity is the second thing to consider after the cloud cover itself.

zeotroph··on Show HN: Algorithm Cookbook in Rust
A graph with the members first, next, and endpoints is not exactly self explanatory, something better than a "A compact graph representation" comment would be nice.

Especially since the ownership model of rust makes the classic graph representation of edges owning nodes impossible.

zeotroph··on Social Cooling – How big data is increasing pressure to conform
What is missing is a government actively pushing this scheme - which China is doing with their "social credit score".

An intermediate step would be selling of derived data to anyone, not just companies interested in hiring you but really anybody.

zeotroph··on Ask HN: What do you want to see in Debian 10 (“buster”)?
This (and checkinstall) are stopgaps, ideally the .deb creation process would get an entirely new packaging toolchain, with the minimal arguments: dependencies, build-dependencies, sourcefile/url/repo, buildcommands.

Then a linter could tell you that a changelog.Debian.gz, man-pages and debtags are still missing, but for internal redistribution that is enough for now.

And can failed build commands continue from where they were interrupted? Always cleaning everything is terrible for iterative exploration of the package creation process. The final package should be marked as tainted of course, but always starting a minute long compile process should be skippable.

zeotroph··on Quantum Approaches to Consciousness (2015)
Indeed, Making a decision« is not a coherent concept., I should have written the quoted sentence as "making a 'decision' means you will then randomly (but weighted) fall into one outcome". Whether that source of randomness is QM or just deterministic chaos* (say, it depends on being somewhere a second early or any other butterfly-factor). The point is, some decisions might always have a p=1.0, others only a p=.8. And in the not-QM case that is decided by other factors in advance, so the closer you get to the point-in-time itself, the closer it gets to 1.0.

The point is that these different probability values are there, and could be a faux-will. E.g., no matter how bad your commute was, you will be kind to your coworkers (high p), or if your mind encounters new information, you might change your mind (p ~ .5).

Concrete example:

like encouraging people are meaningless

Person has an exam in a week, nobody cheers, and depending how good the coffee is over a week, the weather, etc., p_passing_the_exam is 0.6. So, rolling back time and playing it again 40:60, QM outside the mind, maybe someone making decisions with a geiger counter could be the randomness source. But with some encouraging, p might climb to 0.7 because they outweigh some negative factors. The fact that you cheered someone on, is of course also just a certain probability, but one which you could shape.

* if it is all deterministic chaos and QM is a big lie, then Maxwells daemon could compute it all, but nobody in this universe will every have access to all that data, so it is the same.

zeotroph··on Quantum Approaches to Consciousness (2015)
Exactly that one, thank you two walnut-shaped minds, both thinking alike.

Greg Egan calls this concept (see the linked FAQ there) the Dust Theory, and again

> > How seriously do you take the Dust Theory yourself?

> Not very seriously, although I have yet to hear a convincing refutation of it on purely logical grounds.

it juuuust might maybe true (however unlikely).

zeotroph··on Quantum Approaches to Consciousness (2015)
"There is no free will!" is also used by some radical neurologists to suggest questionable changes to society, some seriously close to human rights abuses. So to counter that other areas of science (or gaps therein) are searched for what seems like free will, but as you said this won't change anything.

I think the concept of free will is needed, and not as a necessary lie to hold society together. I think of it as the number of paths open to you and the likelihood of every one. With every decision you will then randomly fall into one (aided by QM or not). So every person can and should be encouraged to cultivate their pathway network, which is what an open society based on discussions and science already does. Yes, all this is not "free will" in the strictest sense and should probably get a hipper name, but this will serve humanity and humanism more than the alternative of what in the end is just fatalism.

zeotroph··on Quantum Approaches to Consciousness (2015)
Is there an name for this "reality first" or "reality triumphs" concept? The closest I can think of is The Talos Principle, it is the name of a game, but in its context the meaning is, paraphrasing: "At the end of the day, even the mightiest thinkers and philosophers are made of flesh and bone".

On the other end, there is a Greg Egan (science fiction) novel where exactly this does not hold and pure maths can create virtual worlds in some abstract math universe, i.e. not running on a physical computer (though I think one was needed to bootstrap this into perpetual existence). The name escapes me at the moment.

You might be able to weave the consciousness problem into there by following the simulation hypothesis and claiming that the brain (of humans or some smart enough ancestor?) was either created with a shortcut to the underlying compute power or that over the course of evolution (think the FPGA + evolutionary algorithm story) this link developed, possibly independently multiple times. And since most concepts underlying biology are understood, "Quantum-woo" is the last holdout (see Penroses microtubuli idea).

zeotroph··on Quantum Approaches to Consciousness (2015)
> Platonic trap.

i.e. is a believer in the mind & body dualism, an idea first put forth by Aristotle and Plato, and later famously championed by Descartes. Though the "quantum" variant is not as far out as the "soul" one. However neither are supported by science, but the knowledge gap regarding consciousness is still large enough that something might still hide in it.

Though Socrates, Plato, and Aristotle fell into many traps, this is probably the relevant one.

zeotroph··on Maru OS – A complete desktop experience on a smartphone
I am really surprised that the "single computing / storage device" idea which can switch between different screens and input methods seamlessly (and maybe support plug-in CPUs) is just is not happening, though the concept here is a small step in that direction.

Having everything in the cloud seems to have made this somewhat redundant for most users, though I still hope one day I can carry all my data around, of course given proper encryption and backups and the ability to distinguish between safe and possibly monitored (public) displays/inputs.

Or maybe wireless displays are missing so that decision to quickly do a task with a mouse and keyboard leaves out the cable plugging aspect.

zeotroph··on Every neuron potentially has a different genome than those it's connected to
(while most here understands CPU RAM RAII DRY etc., you might want to mentions that SNP means "single nucleotide polymorphism", which are mutations where just a single base pair differs.)
zeotroph··on Every neuron potentially has a different genome than those it's connected to
Neurons practically never die, so the DNA itself will not directly reflect what a decades old braincell stores, though some mutations might influence how or if at all a certain neuron stores something. But since these mutations are random they are closer to defects (the article draws a connection to psychiatric diseases) than features.

But DNA can also be methylated, which is something that also happens in neurons and might be involved with actual memory.

And if you ever want to digitize a brain, then you might have to grab the DNA of 100 billion neurons as well, and there straightforward mutations are easier than just a somewhat transient methyl group here and there.

← PreviousPage 5 of 6Next →