Luker88

15 ms·

151.
by Luker88 5y ago
This might be a very big thing for other reasons too ...What does this imply for projects like Github's Copilot?
152.
by Luker88 5y ago
I said this in the other copilot threads too, but don't forget that copyright is not the only protection there is. Lots of countries (USA first) have tons of software patents. Apache and GPL have clauses to protect the project and its
153.
by Luker88 5y ago
Like it or not, it seems like: * most people here are unhappy * most laywers will say it's fine (it very probably passed MS ones) I can understand that. Copyright was not created with AI/ML in mind, even as a random stray thought.
154.
by Luker88 5y ago
So... Putting the (imho) big licensing problems aside, what about the software patents? Apache and GPL have patent protection clauses. Does this mean that anyone using copilot might somehow get code that implements something patented, but p
155.
by Luker88 5y ago
As someone interested in all of cryptography, I cringe every single time when we call blockchain "crypto". Sure, there's cryptography in it. But you don't call a mail or http server "crypto". There's simil
156.
by Luker88 5y ago
well, there's a lie in the middle though: "Bitcoin miners as a flexible load" I'm just going to guess that miners will not take seriously the idea of "you just have to mine only during the day" The return of mi
157.
by Luker88 5y ago
The discussion points link to the github of the research https://github.com/QiushiWu/QiushiWu.github.io/blob/main/pap... It has yet to be published (due next month) How about opening few bug reports to c
158.
by Luker88 6y ago
I agree bitcoin should switch. I also think it's never going to happen. Bitcoin is as static as it will ever get. Arguing for proof of stake will only get the internet mad at you, not much technical discussion is even possible. It coul
159.
by Luker88 6y ago
> clearly crypto mining is emitting using a lot of power and some of that is CO2. But the competition is fiat; backed by fighter jets and nukes; literally just 2 points: * using crypto will not make the military stop existing * switching
160.
by Luker88 6y ago
So yesterday in HN there was a link to a medium post on the same thing, but limited to Proof of Work instead of the whole cryptocoins. Got flagged pretty quickly, but there was a link to a petition inside with links to various problems of p
161.
by Luker88 6y ago
bitcoin can still work on PoS, we should stop thinking it as a single, unmovable, un-updatable thing. A switch to PoS like ETH2 is not going to make it lose value, decentralization and might even be an opportunity to go beyond the ridiculou
162.
by Luker88 6y ago
Projects can switch to proof of stake. no mining, people don't lose money, still decentralized. I specifically pointed out that it's not about banning bitcoin. Make bitcoin a PoS. You can keep trading, you can keep being decentral
163.
by Luker88 6y ago
But should there be other reason to ban it? Not just bitcoin, but proof of work coins in general: * power usage * environmental issues * worsening chip shortage Do we have technical reason for proof of work? Isn't proof of stake as saf
164.
by Luker88 6y ago
I am even getting partial to licenses that are basically (A)GPL, but only for non-commercial. Or the kinds where the software is automatically (A)GPL after 2-3 years. The community does not lose anything and your projects can't get exp
165.
by Luker88 6y ago
Last month they even managed to fully open parts of the modem, and do calls without the closed blob for the firmware! https://twitter.com/biktorgj/status/1357217053745250304 (edit: caveat: they might not be able t
166.
by Luker88 6y ago
Well, we can all agree that the solution should not be "we'll just require everyone to change all their code through this library" static vs dynamic is a linker (and LTO) problem, we can't start asking everyone everywher
167.
by Luker88 6y ago
> Rust isn't against shared libraries per se, but it doesn't have a stable ABI. I'd like to point out that this effectively means "it's not against shared libs, but it will actively ignore anything needed to work
168.
by Luker88 6y ago
While I appreciate the fast moving pace, I have to admit that it is hurting two big things: * long term support * choice, in particular in building shared libraries. While a lot can be argued on the first point, the community seems to be ac
169.
by Luker88 6y ago
The project is really interesting and I think I am going to try it. It looks like it's text only? would it be possible to add a very filtered, low-bandwidth audio stream since the phone can do all the calculations? I have no idea on th
170.
by Luker88 6y ago
> But I respectfully decline your advice and take the opposite stance that it is my duty to not devote my time for years to staged academic exercises but instead work on real problems I have no idea what particular university or course y
171.
by Luker88 6y ago
Does anyone have an up-to date comparison protobuf vs captnproto instead? ( https://capnproto.org/ ) I am starting a new project and it looked really interesting, but I still have to read comments from people who used both
172.
by Luker88 6y ago
Tried it (rpi4), working. Even managed to encrypt the SD. Can't get nixops to work from my x86_64 notebook to deploy on the rpi though, I gave up on that. Nixos is really nice, But I find that the declarative conf still has a lot to ir
173.
by Luker88 6y ago
Worked with both, only on the opensource version. I agree that scylla shows a lot more promise than C , but we hit (different) instabilities and gotchas on both All has its place. I am not saying C or scylla should not be used. I was just
174.
by Luker88 6y ago
TL;DR: Pay lots of attention before choosing C* or scylla. only use the most basic stuff, don't work on near-realtime. More pifalls than the number of seconds those software have been out. Every time I see people talking about cassandr
175.
by Luker88 6y ago
Sorry but... what is this? I read ( very quickly) the paper, then the article. First of all, all the new names for things that are basically shards are not helping. Then all of this is very domain limited to logs, with only append and trun
176.
by Luker88 6y ago
> A few people suggest re-licensing from AGPL back to MIT. Others say going with unlicensed. We'll think about that. Don't go unlicensed. In most countries it means "all rights reserved". Let alone modifying: no linki
177.
by Luker88 6y ago
I am a fan of the thread-per-core model, but I do not share their views on the data sharding per core While it increases the data locality, I have seen a few software following this sharding model (notably scylla) that work really bad once
178.
by Luker88 6y ago
So instead of having "primitive1 = rsa" you have "age1 = rsa". It's the same thing. You are just bumping the protocol version faster, and confusing people because protocol 42 will be just another crypto change, but
179.
by Luker88 6y ago
current crypto is not stable. at all. challenges for new standards are underway, AEAD standardization is relatively new, ECC is no longer quantum proof, we still have no definite idea on lattice and so on. Even CHAHCA has multiple evoluti
180.
by Luker88 6y ago
The article basically says that it's bad to support multiple choices (crypto primitives) in a protocol. > If you need to enumerate the options you support, it means that not only you support multiple ones, which is already bad, but

More ›