In other news today...
-
@anonymous234 They've probably got some secret Chrome-only security setting or something they can use as a fig leaf if it ever comes down to it
-
@boomzilla said in In other news today...:
@Dragoon said in In other news today...:
10-petawatt power laser pulses.
Femtoseconds long pulses, mind.
-
This is the moment you've all been waiting for. After this news, the world will change and definitely for the better. Nobody could've ever predicted that this would happen, and now we can only speculate just how amazing the future will be.
-
@hungrier said in In other news today...:
I forget who, what or when, but I seem to recall there was another case where someone tried to submit a bug to some project on HackerOne and they also got screwed
From further down in the article:
It’s certainly the most popular, especially since big names like PayPal work exclusively with the platform. There have been issues with HackerOne’s response, including the huge scandal involving Valve, when a researcher was banned from HackerOne after trying to report a Steam zero-day.
-
@Rhywden said in In other news today...:
@boomzilla said in In other news today...:
@Dragoon said in In other news today...:
10-petawatt power laser pulses.
Femtoseconds long pulses, mind.
Yes. As with everything to do with laser-containment fusion, it's better to ask how much energy was used to get how much energy out…
-
@dkf said in In other news today...:
@Rhywden said in In other news today...:
@boomzilla said in In other news today...:
@Dragoon said in In other news today...:
10-petawatt power laser pulses.
Femtoseconds long pulses, mind.
Yes. As with everything to do with laser-containment fusion, it's better to ask how much energy was used to get how much energy out…
I'm betting they're equal!
-
@Tsaukpaetra I'm betting they're not because that'd be damn close to it being a working fusion energy source; that's the break even point. I'm betting it's still orders of magnitude out from break even.
-
@dkf said in In other news today...:
@Tsaukpaetra I'm betting they're not because that'd be damn close to it being a working fusion energy source; that's the break even point. I'm betting it's still orders of magnitude out from break even.
Well, I mean, I didn't say it would be usable output...
-
@Tsaukpaetra said in In other news today...:
Well, I mean, I didn't say it would be usable output...
We've got those already, i.e., fusion devices that output more energy than what you feed them.
Spoilers
Fusion/hydrogen bombs.
-
@hungrier said in In other news today...:
@Carnage "Piles" is British for "hemorrhoids", hope this helps clear things up for you
So an "atomic pile" would be...?
-
@da-Doctah said in In other news today...:
@hungrier said in In other news today...:
@Carnage "Piles" is British for "hemorrhoids", hope this helps clear things up for you
So an "atomic pile" would be...?
Warm, but uncomfortable.
-
-
@HardwareGeek said in In other news today...:
Unlike in Italy, Iranian officials are refusing to impose quarantines in areas affected by the outbreak. They say quarantines are old-fashioned and that they do not believe in them.
Oh my.
Iran is where it turns into a pandemic then, since when it gets bad enough people will start fleeing, spreading the infection to other nations.
-
Daughter can predict today's weatherIt's actually just the radio:
-
Looks like someone picked the wrong "storm drain" to enter with his rickety little boat:
EDIT: Looks like they had to cut open a grate for this smartass:
-
-
I applaud the effort, but we all know it won't matter for the lawyers.
-
@Dragoon said in In other news today...:
it won't matter for the lawyers.
I hope they also have good lawyers because if they produced every song possible, they also produced a bunch of copyrighted songs
-
-
@TimeBandit said in In other news today...:
@Dragoon said in In other news today...:
it won't matter for the lawyers.
I hope they also have good lawyers because if they produced every song possible, they also produced a bunch of copyrighted songs
TFA:
"Under copyright law, numbers are facts, and under copyright law, facts either have thin copyright, almost no copyright, or no copyright at all," Riehl explained in the talk. "So maybe if these numbers have existed since the beginning of time and we're just plucking them out, maybe melodies are just math, which is just facts, which is not copyrightable."
One of them is actually a lawyer as well.
-
@PleegWat In Germany we have the word "Schaffungshöhe" which legally describes that a copyrightable creation has to be "graspable by human senses" and also be a "personal and thoughtful work".
Anything an algorithm creates thus cannot be subject to copyright as there's no mind behind it.
-
@Rhywden said in In other news today...:
@PleegWat In Germany we have the word "Schaffungshöhe" which legally describes that a copyrightable creation has to be "graspable by human senses" and also be a "personal and thoughtful work".
Anything an algorithm creates thus cannot be subject to copyright as there's no mind behind it.
From my quote, (part of) the idea seems to be not that this counts as prior art for any music, but that this proves that music is math which isn't copyrightable in the first place. I'm not sure that's gonna fly either though.
-
@PleegWat Well, the problem is that this would mean that all written works are not copyrightable either - due to the Work of Infinite Monkeys.
-
-
Well, took them awhile but they finally got it right.
-
-
@TimeBandit Independent creation is a defense, and unlike most people they can point to the algorithm to prove the songs were created without any influence from the prior copyrighted works.That said...
@Rhywden US law dictates "an original work of authorship", and the Copyright office specifically excludes works by non-human authors, such as animals, natural processes, or "a machine or mere mechanical process that operates randomly or automatically without any creative input or intervention from a human author". Even if authorship somehow were attributed to the lawyers who wrote the generator, it'd still fail as it wouldn't rise to the extremely minimal level of creativity required to be "authorship".
@PleegWat The output of that algorithm wouldn't be copyrightable, but it wouldn't invalidate any copyrights. I think they're going for the idea that music as a whole is an "idea, procedure, process, system, method of operation, concept, principle, or discovery" and thus ineligible for copyright -- since it'd be eligible for patenting instead -- but that's not going to work, since songs are individual expressions of those ideas, and expressions are copyrightable.
-
@PleegWat said in In other news today...:
TFA:
"Under copyright law, numbers are facts, and under copyright law, facts either have thin copyright, almost no copyright, or no copyright at all," Riehl explained in the talk. "So maybe if these numbers have existed since the beginning of time and we're just plucking them out, maybe melodies are just math, which is just facts, which is not copyrightable."
One of them is actually a lawyer as well.
What about illegal numbers?
-
@Zecc said in In other news today...:
@PleegWat said in In other news today...:
TFA:
"Under copyright law, numbers are facts, and under copyright law, facts either have thin copyright, almost no copyright, or no copyright at all," Riehl explained in the talk. "So maybe if these numbers have existed since the beginning of time and we're just plucking them out, maybe melodies are just math, which is just facts, which is not copyrightable."
One of them is actually a lawyer as well.
What about illegal numbers?
That is incredibly dumb.
-
@Carnage said in In other news today...:
@Zecc said in In other news today...:
@PleegWat said in In other news today...:
TFA:
"Under copyright law, numbers are facts, and under copyright law, facts either have thin copyright, almost no copyright, or no copyright at all," Riehl explained in the talk. "So maybe if these numbers have existed since the beginning of time and we're just plucking them out, maybe melodies are just math, which is just facts, which is not copyrightable."
One of them is actually a lawyer as well.
What about illegal numbers?
That is incredibly dumb.
While it seems intuitively obvious that "If this picture is illegal, then so is this audio file", overturning the legal logic may be less trivial.
Or not, IANAL.
-
@TimeBandit said in In other news today...:
every possible melody
Seems unlikely. The number of possible melodies is enormous.
Consider the notes of a piano. (There are instruments that can play higher or lower, but let's stick with this for now.) A standard piano can play 88 notes.
One of the defining characteristics of a melody is the interval between one note and the next. Does it go up or down, and by how much?
There is only one possible "melody" consisting of a single note. This is because of transpositional equivalence. Think of a song, any song. It doesn't matter if it's sung by someone with a high voice or a low voice; played on a piccolo or a tuba; it's still the same song, the same melody. Same with our one-note "melody"; you can start it on any note, and it's still the same "melody".
A two-note "melody" can start on any note of the piano and go to any note, including repeating the original note. There are 88² possible sequences of two notes, but many of these are equivalent. Think instead of the possible intervals between the first and second notes. If we start on the lowest note of the piano, the second note can be from 0 to 87 notes higher. If you start on the highest note, the second note can be from 0 to 87 notes lower. If you start in the middle, the second note can be higher or lower, but any possible interval is also one of the ones possible if you start on the lowest or highest note. Thus there are 175 possible intervals between two notes, 175 possible two-note "melodies".
My brain is failing to come to terms with the math required for an analytical solution to the number of possible longer melodies. There are 88n sequences of n notes, but many of these are equivalent. Alternatively, there are 175n-1 sequences of intervals between n notes, but many of these are invalid because they go outside the range of the keyboard. Trying to eliminate the redundant/invalid ones has broken my brain, so I wrote a (terribly inefficient) Python script to exhaustively count valid (very short) sequences. There are 22969 three-note and 2679775 four-note "melodies" that can be played on a piano. The script ran for hours and was using almost 20GB of RAM trying to count five-note "melodies"; it had found almost 179000000 when I killed it. And it continues to grow exponentially; by the time you get to 8 or 9 notes, and that's still not really much of a melody, you're up into the quadrillions. (My math on that was bogus, but I think I'm at least close on the order of magnitude.)
And all this ignores rhythm, which is also an important component of a melody. Each note can have 10-ish rhythmic values. Like the pitch intervals, some combinations are redundant (same relative values but doubled, halved, or whatever), and some don't make sense when used together, but still it's another factor to add into the exponential explosion of melodies. It also ignores Eastern, African, and other non-Western scales, jazz "blue notes", avant-garde microtonal and other melodies that can't be played on a piano.
Note that the vast majority of the possible melodies considered above are absolute rubbish as melodies. They make Sturgeon's Law look extremely optimistic. But the claim was that they had generated all possible melodies, not all good ones.
Good or not, the number of possible melodies is so large that they could not have come anywhere near generating all of them. They might have generated a complete subset that satisfies a set of rules for creating a "good" melody (chances are, if you Google "how to write a good melody", you can find several such rules for things like stepwise motion and leaps, tonal center and whatnot). Given such rules and an initial portion of a melody under construction, the rules may suggest perhaps a half dozen "good" next notes, instead of 88. This might lead to a feasible number of melodies that can be generated algorithmically, but it's vastly different from the "every possible" melody claimed.
-
@HardwareGeek TFA:
To determine the finite nature of melodies, Riehl and Rubin developed an algorithm that recorded every possible 8-note, 12-beat melody combo. This used the same basic tactic some hackers use to guess passwords: Churning through every possible combination of notes until none remained. Riehl says this algorithm works at a rate of 300,000 melodies per second.
12 subsequent notes picking from 8 gives 68 billion melodies, though some will be equivalent. I don't think you need to go the full 88 tone range to cover every significant melody, but I don't think you can get away with less than a full octave.
-
@Rhywden said in In other news today...:
@PleegWat Well, the problem is that this would mean that all written works are not copyrightable either - due to the Work of Infinite Monkeys.
Technically this means nothing is copyrightable, as everything can be digitized and is thus just a number that can be enumerated. See also, T-Shirts with the numbers +/-1 the AACS key. ()
But this one has a bit better chance to fly, as the number of "every possible 8-note, 12-beat melody combos" is low enough that you can feasible enumerate all of them (and they supposedly did), which isn't true for the work of Shakespeare, or a scan of the Mona Lisa.
-
@topspin said in In other news today...:
@Rhywden said in In other news today...:
@PleegWat Well, the problem is that this would mean that all written works are not copyrightable either - due to the Work of Infinite Monkeys.
Technically this means nothing is copyrightable, as everything can be digitized and is thus just a number that can be enumerated. See also, T-Shirts with the numbers +/-1 the AACS key. ()
But this one has a bit better chance to fly, as the number of "every possible 8-note, 12-beat melody combos" is low enough that you can feasible enumerate all of them (and they supposedly did), which isn't true for the work of Shakespeare, or a scan of the Mona Lisa.I think the takeaway here is a creative work involves picking a specific example from a very large set, and that example not being 'close to' a different example which was previously picked from the same set by someone else.
The question then becomes threefold:
- How large must the set be, for the work to be creative?
- How does one define the distance
- How much distance is required for the work to be suitably different
I am not sure whether anyone materialised the entire set in the past is at all relevant to the answer to these questions.
-
-
@Rhywden said in In other news today...:
I'm all for scaring some proper behavior into them, but dragging them off to juvie seems a tad over the top.
-
@topspin said in In other news today...:
@Rhywden said in In other news today...:
@PleegWat Well, the problem is that this would mean that all written works are not copyrightable either - due to the Work of Infinite Monkeys.
Technically this means nothing is copyrightable, as everything can be digitized and is thus just a number that can be enumerated. See also, T-Shirts with the numbers +/-1 the AACS key. ()
But this one has a bit better chance to fly, as the number of "every possible 8-note, 12-beat melody combos" is low enough that you can feasible enumerate all of them (and they supposedly did), which isn't true for the work of Shakespeare, or a scan of the Mona Lisa.This whole exercise reminds me of those "proofs" that look like they prove something inane but turn out to rely on dividing by zero somewhere.
It's a stunt, and the copyright fanatics will get excited, most people will roll their eyes and life will go on like it did before.
-
-
See? Every addiction can be beaten!
-
@PleegWat said in In other news today...:
every possible 8-note, 12-beat melody combo.
That's still a bit rubbish as a "melody"; it's more like a phrase of a melody. Take "Twinkle, Twinkle Little Star". Twelve beats gets you "Twinkle, twinkle little star. How I wonder
what you are." Not the complete melody; not even whole first line of the song.Also, there are more than 8 notes in an octave; eliminating chromatic tones eliminates a great many melodies. My Google-fu didn't find many specific examples of songs with chromatic tones in the melody (and thus are excluded from their "every possible melody"), but here's one that may be familiar if you like going to the circus (at least American circuses), Julius Fučik's "Entry of the Gladiators". The opening fanfare includes descending and ascending chromatic runs that span more than an octave (so also excluded on that basis) and hit every chromatic tone, not just the 8 diatonic tones, within each octave or partial octave. When the main melody begins, it stays within a single octave, at least at first, but it uses all 12 chromatic tones in the octave.
Another familiar example is George Gershwin's "Rhapsody in Blue", which you may recognize from old United Airlines commercials, if nothing else. In the famous opening clarinet solo, if the slide up the scale were played as distinct, individual notes as written (instead of a slide, which includes not only all the chromatic notes, but also all the pitches in between — incidentally, it's written as a 17-tuplet of 64th notes, not something you see every day, unless, I suppose, you're a clarinetist practicing Gershwin), it would be a diatonic scale in the key of B flat, and thus could be among the "possible" melodies (except it spans more than an octave; oops) and ends on a B flat. But the very next note is an A flat, which is not one of the 8 notes in the B flat scale, and continues with lots of chromatic notes sprinkled around (but not as many as Fučik). Sorry, George, your melody isn't possible.
So they generated a complete set of melodies that satisfy a simple set of rules, but the rules are too simple and exclude many real melodies.
-
-
@HardwareGeek said in In other news today...:
My Google-fu didn't find many specific examples of songs with chromatic tones in the melody
Someone doesn't listen to 1980's progressive metal.
-
@mott555 No, I don't. I'd gladly use such songs as examples, but I don't know any to search for them. It just further demonstrates my point, though, that there are a lot of possible melodies that aren't among the "all possible melodies" this group generated.
-
@Mason_Wheeler said in In other news today...:
Florida Man thread is
Police accountability thread is https://what.thedailywtf.com/topic/23472/police-accountability-in-the-usa/272
-
NSA violates your privacy, but it's for your security
-
asteroid that has been captured by Earth's gravity,
Probably. It doesn't appear to be a piece of man-made space junk, but the possibility hasn't been ruled out yet. It also appears to be in an unstable orbit and will probably be flung back into interplanetary space in a few months.
-
@HardwareGeek said in In other news today...:
"Probably fly away again in April."
Yeah, I've had houseguests like that.
-
@HardwareGeek said in In other news today...:
asteroid that has been captured by Earth's gravity,
Probably. It doesn't appear to be a piece of man-made space junk, but the possibility hasn't been ruled out yet. It also appears to be in an unstable orbit and will probably be flung back into interplanetary space in a few months.
Did the Roadster not make it further?
-
@mott555 I listened to the song you posted in Song of the Day. I wouldn't normally — it's definitely not my cup of tea — but because you posted it in response to this discussion, I listened for the purpose of analysis.
Yes, definitely an example of a chromatic melody that is not included in the set of "every possible melody" discussed above.
By coincidence, I happened to encounter a couple more last night. "White Christmas" by Irving Berlin, and the Prelude to Tristan und Isolde. (Pretty much anything by Wagner would also fit, but it happened to be the Prelude that was under discussion last night.) Both are chromatic and work really well together in a "what would ________ sound like if ________ had written it?" mashup.
-
@hungrier said in In other news today...:
I don't know if "Canadian telecom giant continues to produce self-serving lies" is really news, per se, but they're doing it again:
IIRC Telus made the same press release on the same day. Rogers was a week late:
The sub-heading is a punchline that writes itself:
The carrier agreed with the other national carriers in saying that the market is already competitive