65 karma · joined October 5, 2016
I agree that the $100 fee in no way guarantees proper identity verification has been done. I also see plenty of other unnecessary barriers the fee creates, I just don't know what an obvious alternative would be.
"Discrimination is the act of making unjustified distinctions between human beings based on the groups, classes, or other categories to which they are perceived to belong.[1] People may be discriminated on the basis of race, gender, age or sexual orientation, as well as other categories.[2] Discrimination especially occurs when individuals or groups are unfairly treated in a way which is worse than other people are treated, on the basis of their actual or perceived membership in certain groups or social categories.[2][3] It involves restricting members of one group from opportunities or privileges that are available to members of another group.[4]
Do you think Apple giving 100MM specifically to "Black and Brown companies" is unjustified?
From the two citations: [48]
Prejudice, discrimination, or antagonism directed against a person or people on the basis of their membership of a particular racial or ethnic group, typically one that is a minority or marginalized.
[49] a belief that race is a fundamental determinant of human traits and capacities and that racial differences produce an inherent superiority of a particular race
Apple choosing to invest in educational and entrepreneurial programs for "Black and Brown companies" is in no way antagonism towards other races nor is it an indication that Apple supports a belief that any race is superior.
Nearly any file can be set as an executable, and surely a bash script set to launch an un notarized app in your application folder wouldn't magically bypass the gatekeeper security prompt for that app.
Anything that doesn't follow all the steps here wouldn't be notarized, I'm assuming. If a whatever.o file compiled from whatever.cpp with gcc from the command line wouldn't have any viable way of just being uploaded to a website for notarization. You'd have to at least go through all these steps. As far as I can tell, you'd need to have xcode to do that.
If they notarized your prefpane/daemon combo, I suspect they would notarize quite a few .app applications that are properly signed and not malicious or frowned upon use cases.
Serato control vinyl might have found a happy medium or even surpassed what is capable with certain aspects of cdjs or the state of late 70s through early 2000s vinyl djing, but its not cheap, and it doesn't replicate the discovery or ownership aspect of purchasing vinyl.
It would quite odd to me for someone to say that serato control vinyl is a nostalgia oriented product.
If inventing the future, one must be able to discern what is a classic or canonical element of the past that is presently underutilized. One could make the argument that Elon Musk is nostalgia obsessed with what has been resuscitated from NASA technology that had been mothballed.
Perhaps Descartes was absurdly nostalgic for the times of Euclid, and Einstein absurdly nostalgic for the times of Descartes. Will we lapse into the dark ages if all the Amiga 500s stop working, probably not.
If we lost the knowledge of how to tech how the ARM architecture works, as was the case prior to the raspberry pi but after the death of the Acorn line of RISC PCs, we would have lost quite a bit of important knowledge.
I think a computational language would emphasize how you perceive the program in your head, or at least experience reliably how other people do AND do the thing (unlike state diagrams and flowchars, uml etc)
There is no defacto computer science orthodoxy, so let's take that definition as a shred of a candidate orthodoxy to think inside.
Let's go further back in time with the this 'device' oriented way of defining. Through correspondence, we can see Newtons contributions contain the https://en.wikipedia.org/wiki/Fundamental_theorem_of_calculu..., but this was an emergent process. The method of the 'grunt work' of this process apparently had two camps, Abacists and Algorists, and he was more the former. https://en.wikipedia.org/wiki/Abacus#/media/File:Houghton_Ty...
From the Oxford English dictionary it gives the origin of Calculus to be "Origin Mid 17th century: from Latin, literally ‘small pebble (as used on an abacus)’."
It's easy to imagine a layperson watching Newton fiddle with his abacus and presume he was doing whatever pragmatic bean counting prevalent to his time. Most people wouldn't spend time pondering and inventing new usages of the abacus. So, it's easy to imagine we'd do the same with Stephen.
As a device, the abacus doesn't conform to the initial definition of a computer (on its own it is neither a system or device that in and of itself can carry out a sequence of operations).
The abacus device seems much more like the contemporary 20 dollar TI scientific calculator. The human has to manually administer the sequence. The system of thought could be arithmetic, algebra, etc, but a contemporary vanilla calculus textbook does not describe a Turing complete anything. Without some explanation of theory to bootstrap the explanation the idea of how recursion/looping works, it's going to hard to even metaphorically mechanize the the idea of sequencing.
Enter lambda calculus and church numerals. With these systems, you can systematically represent mathematical idea with whatever notation you want. Church goes the other way and instead of inventing new squiggles, he gets reduces the 'character set' for his notation, and has really long as expressions, for example, the number 3: λf.λx.f (f (f x)). The algorists were unconstrained by ASCII and unicode so they may have balked at the aesthetics of these expressions, but's not inherently great practice to have ad-hoc invention of notation. Wolframs link to the history of mathematical notations starts to trail off, probably correlated with Church's rendering the need for new squiggles optional.
Contemporaneously, (1930s) Turing contributes the idea of a turing machine, but this is an abstraction. It doesn't actually carry out the sequence of operations. This is after the enigma machine, but ultimately we don't spend much time fussing over who specifically invented the physical abacus, sliderule, digital calculator or digital computer. Its an emergent process, and as a system we don't see a pool of folk communities of "algorists vs abacists" until sometime around the homebrew computer club.
Because of this feels very superfluously flippant that Stephen Wolfram is always coining "new this" and "wolfram that" when we think about Newton leaving others to coin "Newtonian Physics." Personally I can just mentally redact this aspect of his behavior.
So the question is is Stephen actually talking about a new thing that exist? Is a "computational language" a thing? I'd say he's probably right that its coming, but the wolfram language isn't quite yet that thing.
What he is pointing out here is the missed oportunity to have a vernacular, or a text in the semiotic sense, to depict the actual computation aspects of our calcuations. When humans moved the calculator out of our/Newtons brain and into the device as uniform-ish human collective activity, wasn't really happening in significant numbers until the 70s. Signficant as defined by my metric of "the amount of people who used abacus back in the day." Once we let the computer do the task, we didn't force it to show its work. But if the computer carries out the operations, what is it we're still operating manually, why do we continue to interact with it at all?
Well we don't have a language to describe what we're doing and thinking in those gaps, and we're rarely automating state diagrams and flowcharts from our code much less starting with just painting out the state diagrams and flowcharts and letting the computer do the work.
Perhaps wolfram will set the stage for a defacto method towards this type higher order representation aka "computational language." He's definitely trying, but newtons's notations didn't catch on, its up to you to decide if the representation or the premise is more important.
So, to get even more specific with contraction hierarchies, google maps has a feature where you can avoid highways, and/or avoid tolls, and/or avoid ferries.
So in this scenario google has likely done "customization contraction hierarchies" as described here https://dl.acm.org/citation.cfm?doid=2886843
Therefor to answer the original question of this thread, is a* is superseded by contraction hierarchies, the short answer is yes if you have even the slightest amount of time to preprocess a graph (If i'm understanding gugagore and reference material correctly).
However, the answer is no in a more arcane/toy/hypothetical context. In an example of something like a randomly generated massive pac man vs ai ms pac man game where one of the two can can win if he/she reaches a single randomly generated exit first, contraction hierarchies don't really help/apply over a-star. The randomly generated map sort of implies we don't even want to give the computer time to preprocess the map, ultimately its human a-star-ish vs ai a* in real time, while trying to avoid ghosts. Otherwise we could simply give the preprocessed shortest path to ai ms pacman upfront, and tell the ghosts to avoid ai ms pac man. This would seem very similar to the player, but would be far less insightful for someone trying to learn and understand a* vs contraction hierarchies and pathfinding in general.
Also when MIT moved to python vs scheme not to dumb down the CS, but to more emphasize combing and programming complex systems on the fly (robotics etc).