<e​ekk:matrix.org> Considering the rising popularity of blockchain use and storage, and understanding how SSDs' cell lifespan works, it is surprising that no company has developed an SSD with massive storage that can only be written once. It's theoretically possible, but I guess the demand isn't totally there yet.

<n​aphtha:kyun.host> most worm devices nowadays have artificial rewrite protection and could theoretically be rewritten

<n​aphtha:kyun.host> why do that when you can just get a devcie that you can rewrite if you want and enable write protection in software

<g​fdshygti53:monero.social> Storage than can be only write once would cost the same price to make than normal TLC/QLC cells afaik. So I don't think they will make that.

<g​fdshygti53:monero.social> Plus that would need new drivers and possibly new FS.

<g​fdshygti53:monero.social> We use to have storage that could be written only once, back in the 70's. It's what ROM came from

<g​fdshygti53:monero.social> Actually PROM

<g​fdshygti53:monero.social> ROM where mask ROM (so the data was written while they where making the chip, while PROM could be programmed one time)

<t​idux:matrix.org> we also used to have CD-Rs in the 90s and 2000s

<1​23bob123:matrix.org> Burnt one yesterday

<g​fdshygti53:monero.social> It's not the same thing.

<g​fdshygti53:monero.social> But yeah, you could use similar tech if you want.

<g​fdshygti53:monero.social> Have fun with the horrible seek time.

<g​fdshygti53:monero.social> These had a purpose, they where invented while the HDD where sub 500MB in size... And there obsolete. And yes, I did burn a few of them too ~2 weeks ago.

<g​fdshygti53:monero.social> I did not remember if I wanted TAO or DAO so ended up having to reburn them :(

u​nkn8wn69 can you explain it to me like i am stupid? (i am)

can it generate a new mnemonic seed?

<e​ekk:matrix.org> In flash storage, dividing a cell doubles the bits available for writing data. However, each division reduces its rewrite capacity by half. The one-time write capability is merely a side effect of this theoretical SSD, highlighting its exponentially massive capacity.

<g​fdshygti53:monero.social> matrix.monero.social/_matrix/media/…ero.social/bUXCAHeCWVDkSMjSzaoMingX

<g​fdshygti53:monero.social> I don't think so, at least not the way they work.

<g​fdshygti53:monero.social> These ssd use "charge trap", while you can indeed store more bits per cell and could maybe write more per cell in the future (lowering the number of write cycle by doing so), They have another problem, with time, if not connected, they do lose the data. It's why it's recommended to connect SSD at lease a few time a year if you do care about the data. They are usually certified for<clipped message>

<g​fdshygti53:monero.social> 1 year as far as unpowered data retention.

<g​fdshygti53:monero.social> What you need is a different technology.

<g​fdshygti53:monero.social> Not you might think that just leave it plugged on will prevent that but no.

<g​fdshygti53:monero.social> The ssd do rewrite the data that is setting for long time to "reset" that timer if we can call that

<g​fdshygti53:monero.social> These drive do there own maintenance themself, in a transparent manner, while you are using your computers

<e​ekk:matrix.org> The idea is to actively use it instead of just storing it away. Its potential massive capacity can effectively address storage issues arising from high demand over many years, especially in a scenario where people are adopting the Monero blockchain. This would also bring benefits to other blockchains.

<g​fdshygti53:monero.social> Once a cell is writen it is writen.

<g​fdshygti53:monero.social> Data decay, plugged or not.

<g​fdshygti53:monero.social> But while it's on, it silently refresh (relocate) the data

<g​fdshygti53:monero.social> Once a cell is writen it is writen.

<g​fdshygti53:monero.social> Data decay, plugged or not.

<g​fdshygti53:monero.social> But while it's on, it silently refresh (relocate) the data then properly erase the old location the data where

<g​fdshygti53:monero.social> iv.melmac.space/watch?v=YtBysgPOKx4

<g​fdshygti53:monero.social> That explains nicely how it work

<e​ekk:matrix.org> So this theoretical SSD can potentially alleviate storage challenges for full nodes during periods of high blockchain demand. Full nodes typically remain plugged in, facilitating the efficient relocation of aging data.

<g​fdshygti53:monero.social> Yeah, but then your theorical SSD will fill itself exponentially with time as the blockchain grow and age. (Assuming the cells are "one write only", then the old one will be discarded instead of reused so the total capacity of the drive will shrink faster than the blockchain expansions

<g​fdshygti53:monero.social> Just use really big HDD and NVME cache or something like that?

<g​fdshygti53:monero.social> Wait wait for an actually new technology (Write once read many crystals anyone?)

<g​fdshygti53:monero.social> Just use really big HDD and NVME cache or something like that?

<g​fdshygti53:monero.social> Or wait for an actually new technology (Write once read many crystals anyone?)

<e​ekk:matrix.org> Very informative video, it's great watch.

<g​raverdev:matrix.org> Is there a security problem with mining for someone? Like I want to help some group that does stuff I like, and I want to mine directly to their wallet, is that safe?

If you use the official software (xmrig, p2pool) from github, it's safe

<g​raverdev:matrix.org> thank you!

<o​frnxmr:monero.social> try not to use a top 3 pool!

use p2pool ...

So it seems things have gone too far in our beloved Internet and apparently someone is trying hard to brake every part of it, starting from abundant invasive CDN's like Cloudflare to privacy-invading tech like WebAssembly and User Agent client hints (e.g. Sec-CH-UA).

 

WebAssembly was invented and implemented to favor lazy Node and JS developers who don't know how to code nor optimize properly, since their code isn't running anymore even in a browser that was allocated 8 GB of RAM due to loopholes and thousands of unnecessary XHR calls and websockets created at the same time then on top of that garbled and even bloated more with JS bundlers and such.

 

Instead of optimizing their million-line broken code, they have decided to push that new standard called WebAssembly in order to have their code executed faster. People who push such tech are selfish and never think of risk prospects their new "convenience" tech will bring. To provide an analogy on how things are bad right now - you now accept to run unknown binaries in your browsers without questioning them.

 

Take, for example, yesterday's Ledger attack (nitter.net/bantg/status/1735279127752540465#m) that was only easy to find because malicious actors don't yet dominate WebAssembly to obfuscate their code efficiently. Once they do, a random shitty soymilk drinking whitehat from Twitter wouldn't be able to find that code that easily.

 

Interesting how most NodeJS people who have no idea what they are doing want to push insecurity to masses and promote it. I mean, not people like kraih who knows what he do since people like that are itched by years of hardcode perl experience. I mean people born after 2000 who tried to code Python, but switched to Node since coding Python feels like coding Assembler for a space station project (too strict for beginners). Then everything they couldn't

do in Pythin they do in NodeJS by loading 10000 of malicious libraries to their PCs infecting themselves and their circle. Then we have new tech like Rust, also Google-promoted. I am still writing good malware thanks Rust and no AV ever detects my stuff anymore on these juicy Windows 10 PCs that what many people assume "secure" (sup luigi1111 still haven't found my payload? try harder installing more AVs).

 

Why nobody raises a voice against tech like Webassembly?

Are you, old school FOSS activists and evangelists who still know what IRC is, okay with such hostile tech now living freely in your browsers?

 

Real questions that will now make your face red because you know I am right but you are too afraid to raise concerns against it.

joins > calls luigi1111 a gamer > leaves , anyway

Whatever you do, beware of the random shitty soymilk drinking whitehats from Twitter that won't save the day anymore soon

hmm

<g​oodboy:matrix.org> lol

<g​oodboy:matrix.org> less ranting more browser obsoleting commits dawg

the move to write everything in wasm just looks like the move to java applets 20-some years ago

in either case, bad programmers will still write bloated unusable code. nothing changes there.

maybe the scale changes, the sheer number of bad programmers and hapless users.

<1​23bob123:matrix.org> To much to read!

<t​idux:matrix.org> Rust is pretty great though. Sure it's not a magic silver bullet to security but it makes it a lot easier to do high level things with low level performance without going "oops I forgot a null terminator now the Internet has rwx access to my system memory". I was on call for patching Heartbleed for a few thousand physical servers, which was all kinds of fun. Plus as Monero user<clipped message>

<t​idux:matrix.org> s, Serai is also using Rust to give us a pretty sweet DEX.

<g​fdshygti53:monero.social> Yep, guard you again failing to properly manage the memory

<g​fdshygti53:monero.social> But 500 dependencies you did not audit will eventually get you