back
user profile

dijksterhuis

4,117karma·1,383submissions·March 13, 2019
about
data/ops/infra/foss/distributed systems/rust

sometimes makes weird machine music. sometimes makes machines do weird things. sometimes makes music machines do weird things.

recent activity (1,383 total)
comment
> And this point isn't even up for debate. Present value calculations essentially prove that a copyright long-tail provides minimal benefit to a corporation. Especially since long-tail sales a…
7y ago·view thread
comment
A) You’ve made a massive assumption here in regards to the sale price. Sales prices vary depending on a multitude of factors. There’s no single sales price that exists for one album/track/et…
7y ago·view thread
comment
You’re trying to argue that just because the music happens to be there that it has no relevance to the fact he makes money or not. I get that. Fair use applies for cases like holiday videos shared b…
7y ago·view thread
comment
What about overdubbing? That’s the trick the majority of the industry uses. ... > Asked the market vendor to turn the radio off? > Not happening for a regular person. And here were get to the cr…
7y ago·view thread
comment
> These copyright holding corporations are just blindly optimizing for maximizing their grasp on any content whatsoever. This is what the reality is: These copyright holding corporations are just o…
7y ago·view thread
comment
He made money from his content. His content included a recording of a copyrighted work. Therefore he made money from that recording. It’s that simple. He could have: * scrubbed the audio * removed the…
7y ago·view thread
comment
For the “structure long messages” bit, i was taught to use the SCRAP mnemonic: Situation: x has broken Complication: dev is on lunch Resolution: wait for dev to return from lunch and request invest…
7y ago·view thread
comment
The degree of material gains from including a copyrighted musical work don’t matter. It’s a blanket thing. The copyrighted work has been included in content that is monetised. The guy is making money …
7y ago·view thread
comment
The problem is that he is making money of the back of the recording of the musical work that has been registered as copyrighted material. That’s literally the whole point of copyright. Someone is maki…
7y ago·view thread
comment
This is actually similar to how it works for traditional TV and Radio Broadcast in copyright societies: (Total revenue of license period x peak/non peak weighting x duration of song played) /…
7y ago·view thread
comment
I agree with the 10 seconds problem. But that’s a problem with Youtube’s implementation - nothing to do with fair use, incidentals or copyright law. Edit: having replied to another comment, I think I’…
7y ago·view thread
comment
WRT copyright claims: “Abused” in a similar way to how YouTube existed for (something like) 10 years without paying a single iota of money to anyone for copyrighted material? I’d call that abuse of co…
7y ago·view thread
comment
Actually fair use falls firmly on the side of copyright holders when a claim is made, until it is disputed in court. The provision for it is to take it to court. That is the only option. There is noth…
7y ago·view thread
comment
How it was recorded has nothing to do with copyright law (in this case, mechanical rights are a whole thing). A musical work was publicly performed (via the radio). He recorded that musical performanc…
7y ago·view thread
comment
Not quite true. Incidentals under Fair use are entirely dependant on the context of the created content (at least for the UK). News/educational - likely okay as incidental. Everything else is up …
7y ago·view thread
comment
Fair use is entirely dependant on the context of the content. It is never really a “clear case”. It’s decided on a case by case basis by the courts.
7y ago·view thread
comment
The transition you mention for Python is actually not that high. Error handling, edge cases, unit tests, type checking, packaging can all be performed in Python. Type checks are the biggest bug bear. …
7y ago·view thread
comment
I was thinking that as I read it. Korg volcas would have been a much better comparison.
7y ago·view thread
comment
Volcas were a proper source of excitement from everyone I knew when they came out. A drum machine for £110! No way! And a keys synth?! AND A BASS SYNTH?!? Cheap, versatile and had range to the sounds …
7y ago·view thread
comment
The company I worked for previously had access to a big dataset that was “anonymised”. There was one poor individual who was the only person who lives in a certain post code area. Ended up having to g…
7y ago·view thread
comment
Depends how many people in the same post/zip code have the same birth month, are male/female, etc. E.g. If there’s 1 female in post/zip code XXXX born in 09/79, doesn’t matter if y…
7y ago·view thread
comment
This is going to be cryptowars all over again isn’t it...
7y ago·view thread
comment
Here’s a great idea - let’s invent a time machine to enable us to have that foresight! Oh wait, hold on, now we can see what that person is going to do next Tuesday. Let’s stop him doing that thing. S…
7y ago·view thread
comment
I miss the late 00s early 10s YouTube recommendation system. I reckon there was more graph theory based on actual videos watched.
7y ago·view thread
comment
Ever head of copy & paste? Won’t make them “useless”, maybe inconvenient for non-security minded individuals.
7y ago·view thread
comment
my bet is on capsule networks, Hinton is usually on point with his stuff
7y ago·view thread
comment
Ok, it seems like there are some crossed wires or missing context here. Also, widely off topic. I never said anything about the term machine learning. Check my bio, see what I’m working on. Fully awar…
7y ago·view thread
comment
The only nnets I mentioned were “classical” as a purposefully over simplified example. Yeah, they can model any function, but historically they were used for probabilistic density functions (if I reme…
7y ago·view thread
comment
I despise the term Artificial Intelligence. This is all PROBABILISTIC MODELLING . Nothing to do with AI/AGI/whatever. The computers aren’t thinking or learning. It’s just modelling fancy …
7y ago·view thread