31 years later, we found the flight recorders • Hacker News Board

Re: 31 years later, we found the flight recorders

#91

Post by LoSboccacc » Sun, Jun 05, 2016, 7:04 PM UTC

gsnedders wrote: post #89

LoSboccacc wrote: post #88

Earlier quoted context omitted.

all browser have fixed dpi, at 96dpi precisely. it's part of the css3 standard and it's tied to the physical unit - so if you get 1cm as css width, it's the same amount of 'logical' or 'css' pixels irregardless of actual device density. (and of course if you ask 1inch to css you get 96px https://www.w3.org/TR/css3-values/#absolute-lengths ) you can use device-pixel-ratio and the likes to have an hint on the pixel den…

It's actually in 2.1 as well, the fact that 96 CSS pixels equal 1 CSS inch. But that's not what's relevant here, as we're talking about media queries where we have resolution units, and resolution units give the device pixels per some CSS unit. > you can use device-pixel-ratio and the likes to have an hint on the pixel density, which doesn't protect you from the extremes (small, very high density devices and large lo…

exactly, instead they went on the road of giving physical pixel per css pixel, which means we need to create all combination of kinda valid queries to support future devices (problem isn't THAT bad now, but it's getting worse)

and then you have iframes, and browser zoom

Re: 31 years later, we found the flight recorders

#92

Post by justinclift » Sun, Jun 05, 2016, 8:45 PM UTC

PeterisP wrote: post #85

justinclift wrote: post #35

Earlier quoted context omitted.

If the software that does the sending has vulnerabilities in it, that would be bad. Although we can all think of approaches to try to minimise/exclude that possibility... reality often has ways of showing up things people didn't think of. :(

Why would a transmit-only device expose any vulnerabilities in its software? It shouldn't have the physical capability to receive any remote signals, much less act in any way in response to them.

Depends on how "transmit only" is done. If it's using the equivalent of UDP and no encryption of the data, then you're likely correct.

But if the "transmit only" has anything that receives data - even just ack signals for what's sent, let alone something to negotiate encryption - all bets are off. :(

Re: 31 years later, we found the flight recorders

#93

Post by gsnedders » Sun, Jun 05, 2016, 9:04 PM UTC

LoSboccacc wrote: post #91

gsnedders wrote: post #89

Earlier quoted context omitted.

It's actually in 2.1 as well, the fact that 96 CSS pixels equal 1 CSS inch. But that's not what's relevant here, as we're talking about media queries where we have resolution units, and resolution units give the device pixels per some CSS unit. > you can use device-pixel-ratio and the likes to have an hint on the pixel density, which doesn't protect you from the extremes (small, very high density devices and large lo…

exactly, instead they went on the road of giving physical pixel per css pixel, which means we need to create all combination of kinda valid queries to support future devices (problem isn't THAT bad now, but it's getting worse) and then you have iframes, and browser zoom

> exactly, instead they went on the road of giving physical pixel per css pixel, which means we need to create all combination of kinda valid queries to support future devices (problem isn't THAT bad now, but it's getting worse)

Do you have examples of other device and browser combinations of this being a problem? Let's make this situation better!

(FWIW, these browsers are likely not conforming CSS implementations, because you either must have 1 CSS inch equal 1 physical inch (and 1 CSS pixel is still 1/96 of a CSS inch, with whatever ratio to device pixel that ends up being), or by relating the CSS pixel to "the visual angle of one pixel on a device with a pixel density of 96dpi and a distance from the reader of an arm's length", but hey, they exist so we may as well add media queries to deal with the resultant fallout.)

[Edit: In the Galaxy Tab 7.7 case, the reference pixel, assuming 15.75" viewing distance (which is totally non-scientifically the viewing distance my partner uses her tablet at) works out to be 1/170th of an inch: almost exactly a device pixel. I guess a lot of the problem with tablets is the viewing distance varies so much...]

Re: 31 years later, we found the flight recorders

#94

Post by PeterisP » Sun, Jun 05, 2016, 10:31 PM UTC

justinclift wrote: post #92

PeterisP wrote: post #85

Earlier quoted context omitted.

Why would a transmit-only device expose any vulnerabilities in its software? It shouldn't have the physical capability to receive any remote signals, much less act in any way in response to them.

Depends on how "transmit only" is done. If it's using the equivalent of UDP and no encryption of the data, then you're likely correct. But if the "transmit only" has anything that receives data - even just ack signals for what's sent, let alone something to negotiate encryption - all bets are off. :(

You're thinking a few OSI layers too high. Any such radio communications from similar systems have one-way broadcast as the physical layer.

If you don't include a radio receiver, then there is no capability to receive a single bit, much less negotiate encryption; the system throws out a waveform and the rest of the world may figure out what to do with it if they manage to receive it.

Re: 31 years later, we found the flight recorders

#95

Post by justinclift » Tue, Jun 07, 2016, 7:18 AM UTC

PeterisP wrote: post #94

justinclift wrote: post #92

Earlier quoted context omitted.

Depends on how "transmit only" is done. If it's using the equivalent of UDP and no encryption of the data, then you're likely correct. But if the "transmit only" has anything that receives data - even just ack signals for what's sent, let alone something to negotiate encryption - all bets are off. :(

You're thinking a few OSI layers too high. Any such radio communications from similar systems have one-way broadcast as the physical layer. If you don't include a radio receiver, then there is no capability to receive a single bit, much less negotiate encryption; the system throws out a waveform and the rest of the world may figure out what to do with it if they manage to receive it.

Thanks. :)

Re: 31 years later, we found the flight recorders

#96

Post by jo6gwb » Tue, Jun 07, 2016, 2:24 PM UTC

technofiend wrote: post #82

kylehotchkiss wrote: post #9

Earlier quoted context omitted.

There just isn't adequate satellite internet worldwide to trust that system just yet. Airplanes don't even broadcast their coordinates to satellites in many airlines, despite existing networks likely (ie Iridium) being able to handle the small increase in data (if they kept the packets small - just flight number/lat/lon - and let the satellites figure out where to send and what time the packets come, the total size f…

It doesn't seem to be on Elon Musk's radar (no pun intended) but if his or Amazon's reusable launch platform really does make commodity launches possible then it would make sense for Elon or someone to put up Iridium 2.0. Maybe Google or Facebook will do it to bring internet to the masses.

There is currently a company called Globalstar seeking permission from the FCC to offer satellite based internet services. The stock dropped 60% last week on rumors that they won't be approved but nothing has been confirmed.