utopiah @ utopiah @lemmy.ml

Posts

2

Comments

805

Joined

4 yr. ago

extract information

I don't think so, at least assuming the scheme isn't actually broken... but then arguably that would also have far reaching consequence for encryption more broadly, depending on what scheme the implementation would be relying on.

The whole point is precisely that one can compute without "leaks".

Edit: they are relying on Brakerski-Fan-Vercauteren (BFV) HE scheme, cf https://machinelearning.apple.com/research/homomorphic-encryption

It's a balance, namely you are probably wasting time if you jolt down literally everything... but also what you don't write down and forget, have no way to backtrack, will also waste time.

IMHO it's the process itself that matters, namely that by taking the time to write down, organize, lookup, you gradually have to do it less and less because you are more conscious about what you know, what you don't, and adapt accordingly.

If you do find a better way, based on a tool or not, please do share back!

I don’t know how much compute cost this adds to an already expensive computation.

At that scale and because they do pay for servers I bet they did the math and are constantly optimizing the process as they own the entire stack. They might have somebody who worked on the M4 architecture give them hint on how to do so. Just speculating here but arguably they are in a good position to make this quite efficient, even though in fine if it's actually worth the ecological costs is arguable.

So homomorphic encryption means the server can compute on the data without actually knowing what's in it. It's counter-intuitive but better not think about it as encryption/decryption/encryption precisely because the data is NOT decrypted on the server. It's sent there, computed on, then a result is sent back.

"does this even if you’ve already opted out of uploading your photos to iCloud." damn that's a bit much!

Edit; in this thread, people who miss the point of homomorphic encryption to dunk (sadly often rightfully so) on Apple.

For most users most of the time, the repetition/automation is not the point and ability to write scripts is not the most important thing.

Well if that's the case then it'd be like somebody buying a bike, removing the wheels, and complaining that truly it's not as fast as a car or as convenient as walking. Sure, it's true but... if one is missing the point of a tool then they can't really complain about how "bad" it is.

Honestly I do not know how the CLI is most popularly used. I do have usage data for that (and I'm not sure who might, maybe Ubuntu?) but again, if people are using it to "type fast" then they are wrong.

Regarding memorizing and the problem of this thread, yes it IS a problem but that's precisely why I also commented https://lemmy.ml/post/24395107/15908795 before, namely that someone learning the CLI (namely ... ALL of us, even people like me who have been using it for decade, at home and professionally) should actually admit they are learning and thus rely on tools as they otherwise normally would.

because it’s faster if you can type fast and remember everything.

No. That's just wrong. You only need to type ONCE and you don't have to remember it all, that's why reverse-i-search (aka Ctrl-r) is so powerful. It allows you to search within your command history so you only need to remember few letters of the command (which you can annotate, e.g. commandname parameter #it does this) and can even edit after, e.g. changing parameters.

But, more fundamentally, that is NOT the point of the command line!

The whole point of the command line is... to be able to repeat things, namely to script actions that can be customizes and combined to YOUR unique needs. So it's absolutely not about typing speed or memory. In fact, once I do have a good command, what do I do? I save it as a script precisely so that I do NOT have to type it again. Think of commands as Lego bricks that can be combined to together, build on top of. That is basically impractical with GUIs. Sure there are some tools to automate the click on GUIs but it's unreliable, nor can it be easily shared.

PS: I'm not saying anyone should use the command line over GUIs. I'm not being prescriptive. I'm only trying to clarify what the point of the CLI is.

TL;DR: command line is about combining tools to your unique needs, repetitively and reliably.

You can download StackOverflow, Wikipedia, etc very conveniently, cf https://kiwix.org/

I have a local copy that I bring on travels with me and I keep it relatively up to date. Very convenient thanks to having a single file per source and tools to read it (Web server, GUI, etc).

First of all, we are ALL in the same boat. Newcomers from whatever OS to professionals alike, we all struggle to remember commands and their idiosyncratic syntax.

Now that this is out of the way, namely that you are not "special" in having difficulties there, a little challenge. How do YOU normally do when you are learning something new? Typically when people do that, e.g. at school or while doing a professional training, they take notes. Are you doing that? I know it might sound "strange" but learning Linux is... well fundamentally learning so yes you can use whatever tools and techniques for that too. So... here are MY notes https://fabien.benetou.fr/Tools/Shell which will probably be rather pointless to you but at least prove that I'm not suggesting something I didn't try first.

TL;DR: yes, use all the tools & tricks recommended here (IMHO with reverse-i-search first) but don't forget to actually take notes!

Right, but to clarify because maybe we aren't talking about the same thing here, the Nebulas Mars II Pro isn't an SBC, it's an Android device (video projector here). So it has to be rooted and install a custom ROM. Are you saying it's available or know how to make one? Because, again, I'd be eager to try, but it's not the same process, AFAIK, as installing Slack, LibreElec or whatever on a RPi or dekstop.

I used Kodi, then LibreELEC then now "just" VLC (booting to it via https://f-droid.org/en/packages/news.androidtv.launchonboot/ rather than home screen) as my content is on my home server relying on DLNA. So... I'm familiar with LibreELEC but isn't it a Linux distribution? Can it be installed on the device itself (not using HDMI) and if so, how?

Wish I could replace Android on my Nebulas Mars II Pro by Plasma Bigscreen...

Indeed, and for details on why https://specifications.freedesktop.org/thumbnail-spec/latest/directory.html

Interesting question. I live in Belgium and... well first of all I don't care for Christmas. I do like to celebrate with family and friends but the religious celebration itself, no. Second I never actually considered it. I do love snow and ice. I recently took on ice skating and... even though I also love the Summer where I can roller blade and skate, knowing that something else is coming is a genuine joy.

So... I can't speak for others but I absolutely love the Winter, from hot chocolate to waffle outside to ice skating, hikes in the snow then relaxing by the file place, there is just so much to look for during that season that ... never dreamt of "a green Christmas".

Edit: I actually had one last year, going to Madeiras, Portuguese island West of Morocco, North of Africa, and... that was fine too. Honestly truth is I don't really care where and how as long as we share a good time.

Indeed, came to add spreadsheets, e.g. LibreOffice Calc or OnlyOffice on NextCloud if you want it Web based. If it's "good enough" to manage the budget of countries, huge corporations, etc and is taught in schools, it's probably good enough for a household.

Come join us at https://framacolibri.org/c/peertube/38 and possibly check what already exists, e.g. https://www.npmjs.com/search?q=peertube-plugin-

I'm not sure I fully understand what you need but I know the forums are quite dynamic.

Interesting question which to be honest I can't really answer myself... but I'd basically inquire by taking the flip-side of https://www.protondb.com/

Namely gamers like me usually check ProtonDB to see what they can play. Here it would be interesting, and I'm 99.99% sure Valve does that already, to check which games do not work and what's the commonality behind them. It means one can then identify the gaps and try to address them.

Still, to venture an ELI5 answer : games are usually build for Windows. Games are using "bricks" like Lego to avoid re-inventing the wheel. Instead of having a health bar, a game developer might use a "health bar" brick. When you have a collection of such useful bricks, you typically call that a library. That library then makes the work of a game developer much easier but not all libraries are made equal. Some popular libraries target only Windows and thus the bricks make assumptions on how the software running the computer, the operating system, works. So... what Valve does is trying to make new bricks to stack on so that game developers don't even have to use libraries they are not familiar with. They "just" use their typical bricks, Valve "injects" in between their new compatibility bricks and voila, unbeknown to the game developer, their Windows game works on Linux!

Yes indeed, I've followed that from afar (as I generally mostly play offline, definitely not competitively) so I hope this will be the final missing piece.

I'm basically just using Steam with SteamVR on the Index, no tinkering in there.

I also tried other things, e.g. Monado, streaming to headset, etc but in practice I prefer to "just" play when I'm playing and for that the Index works great.

Not exactly, clarified ;)