citation_please

This is a very challenging view-point, and I appreciate the effort you've put into backing it up. I think my biggest hesitation of accepting its results is the difference between annual income and wealth. By this study's standard, I'm nearly upper-middle-class, but I have a negative net-worth due to student loans. If my employment status came into question, I would be very quickly impoverished, and would have to rely on many social and governmental safety nets. Can this really be considered rich?

Maybe I'm an outlier as an individual in my mid-20s, but it's enough to make me question the definition of class on annual income.

Then Juul should be put next to the nicotine patches and gum in the pharmacy, and be marketed as a medication. The FDA heavily regulates substances that can be abused when not used as medication (things like meth ingredients).

Frankly, letting a company aggressively market a highly addictive compound to an impressionable segment of society because of a single-use excuse ("it can be used as medication!") is a bit disappointing. It also reads as astroturfing because it's such a limp talking point.

But this is somewhat an admission of using it as medication, which is far from what Juul is being marketed as. It should be next to the nicotine patches at the pharmacy. Can you even buy nicotine patches in bulk? Substances that can be abused for something other than their primary use as medication are heavily regulated by the FDA (think meth ingredients).

Huh. Well let's assume causality points in only one direction, genetics->tissue->ability, then I'm screwed and there's nothing I can do about it. Let's assume that it at least points partially in the other direction, ability->tissue, now I can keep my tissue healthy by exercising.

Since I don't know for sure which direction causality points, I'm going to take the path of maximal self-determinism and try to stay fit as long as possible.

This is probably heavily affected by at least two things:

1. increase in healthcare costs and how health insurance is bound to employment

2. increase in wealth concentration and a shift from cash compensation to asset compensation (Bezos is only absurdly rich if he liquidates his Amazon holding instantly)

So, I would say your portrayal of hourly compensation is disingenuous as well. All of these statistics probably need to be calculated as medians instead of means ("compensation per hour" sounds suspiciously like "total compensation / total hours", which is a mean) in order for it to come close to accurately describing the situation of the median American, because wealth concentration has skewed the mean American into something a lot more optimistic than one would think.

Made me think: if we cut corporate taxes, raised the minimum wage, and reduced means-tested welfare, where would we end up? Same position but with less bureaucracy?

Something tells me that the welfare safety net has more societal benefits than a wage floor, but right now we don't really have a respectable form of either compared to 20 years ago.

Well you're lucky I read this comment right as my morning coffee kicked in; it really depends on your taste I guess, but I'll try to be as broad as I can. Most of the sets I've heard/been at were not recorded, so syndicated radio shows from the time are the best approximate of what would have been played.

* Eric Prydz at Ultra Music Festival 2014 https://soundcloud.com/eric-prydz/eric-prydz-live-ultra

Lots of more-recent converters to the Prydz fanbase point to this set as being the "A-ha" moment of getting his music. He somehow manages to fit in a lot of different-yet-cohesive moods into 60 minutes, and the final song is a (live?) mash-up of two of his classics that is really, really clever and satisfying.

* Above and Beyond BBC Radio 1 Essential Mixes

* 2004 https://soundcloud.com/aboveandbeyond/r1-essential-mix-of-th...

* 2011 https://soundcloud.com/above-6/above-beyond-bbc-radio-1

Take your pick of 2011 or 2004 depending on whether you want to tackle more-poppy, modern 128BPM or a more classic, lo-fi acid-influenced 138BPM. I think the bootleg of Massive Attack's Teardrop from 2011 is a very special track, can't really speak highly enough of it...

* Yotto at ABGT250

* SoundCloud https://soundcloud.com/b-rizzzle/yotto-abgt250-live-at-the-g...

* YouTube https://www.youtube.com/watch?v=jtL1Ff0T6IM

I included a YouTube link because the audio doesn't do the gorgeous setting justice - sunset at The Gorge, Washington State, if you can get past the interesting rave attire that Americans seem to feel compelled to wear at festivals. Fine Day is an absolute classic, very airy and day-dreamy; bonus for the recording capturing the crowd singing along.

* Undercatt at Watergate https://soundcloud.com/undercatt/undercatt-watergate-berlin-...

Club instead of festival or Essential Mix. Again, ends on a very classy, satisfying track that rewards you for sticking through to the end. A bit more techy and minimal than my other choices.

* Sasha at Watergate https://soundcloud.com/last-night-on-earth/sasha-presents-la...

Same club, one year earlier. Again, more techy and minimal. The track around 28 minutes, "Trigonometry" is a perfectly drawn out with a very intense climax, and absolutely deserves the "progressive" moniker - unfortunately it's the radio show and Sasha talks over the track hah. This is only a middle snippet of the full 4 hour set.

* Lane 8, Anjunadeep Edition 28 https://soundcloud.com/anjunadeep/anjunadeep-edition-28-lane...

Something a bit poppier, but deep. Just like the Eric Prydz UMF set, a lot of Lane 8 fans point to this radio edition as being their "A-ha" moment. Very, very classy and restrained, but with some darker, more emotional moments.

* Armin van Buuren, Amnesia Ibiza 2008 https://soundcloud.com/rave_on/armin-van-buuren-live-armin-s...

If you can handle modern (post-2000) trance, then this is definitely one of _the_ sets to listen to. Really runs the full gambit of the genre as it was in 2008, starting slower and poppier, and ending with a very intense final two hours.

Hmm, I was hoping for something more contemporary that I could relate to, but was disappointed that it focused on pieces written centuries ago...like most formal musical theory.

Anyway, this passage stuck out to me:

"""There is something that still rings true of Mattheson’s general idea. We do tend to associate some musical features with being uplifted and others with melancholic reflection, both of which might afford a certain subsequent pleasure to listeners. Just think of how we use music in our everyday lives: some tunes help us to work out or to get something done, while others allow us to cry."""

Composers aren't the only ones who have ever tried to manipulate the crowd. I'm not sure how pop stars/rock bands etc. plan out their sets, but one thing that any DJ[0] worth his salt pays attention to is "harmonic mixing"[1]. Taking things "darker", or trying to bring up the energy, are common strategies taken into account when planning a harmonic transition. I've personally bore witness to plenty of incredible sets that take you from the very top to the very bottom, and it's funny to me that the same ideas are masked behind hoighty-toighty music theory terminology that is used almost exclusively in reference to...music that was composed hundreds of years ago.

[0] By DJ I mean someone who does more than weddings, and certainly doesn't take requests.

[1] https://music.stackexchange.com/questions/14291/what-is-harm...

Hmm, electronic genres are definitely clustered around 128, 140, and 170 (+/-2), and those are all reasonable heart rates when you're partying...But a "human heart beat" can be literally anything so I'm not convinced.

Well I'm not the one who does resume sorting, so usually if it makes it to my desk I know there's a good chance they're qualified. If you give me (personally) two resumes, one with a master's degree from almost anywhere, and one with two years of experience doing /exactly/ what we do, I'll choose the latter first. At this point, master's degrees don't carry the weight they used to in my mind based on (1) people I've interviewed (2) my coworkers and (3) my friends and acquiantances.

I would argue that abstaining from voting in a formal setting is significantly different than not "voting" in modern elections. There's significant historical evidence of voter suppression, including suffragist movements. Voter motivation is not something that I think we should brush away lightly, because it has an impact on democracy just like voter manipulation and voter suppression.

Adding onto this, every physics PhD I've talked to has really impressed me with how much problem-solving they tackled during their stint in academia. You really said it succinctly with "a physics student could find themselves having to learn practically any technical skill".

Running and extending numerical simulations with supercomputer clusters is just something you have to learn to do in order to solve your problem in some cases, apparently.

Their ability to code is usually sub-par, but they ask the right questions about the data, which is critical in the data setting that we work in.

We don't work in advertising / marketing / business analytics, which is very easy to have an intuition about; and we don't strictly work with images, which, again, isn't terribly difficult to have an intuition about. So a strong scientific background is actually a major plus for us.

I could see that if we were doing purely deep learning image classification or advertising prediction then the physics degrees would be less useful, but thankfully we don't.

Probably the wrong fit if the industry experience isn't right, in which case they would be on a fast-track to a senior dev role in the non machine-learning part of our company.

You're right about "research", although I'd hesitate to use the word hardcore hah. We get tons of candidates with ~2 years of slightly relevant work, but rarely do we get candidates with the exactly right relevant work. The "qualified" candidates I spoke of have ~4 years of slightly relevant work, but there are only a couple of them, and they're either out of our hiring budget or currently working for our team.

I'm on a hiring committee for a job with the title of "Machine Learning Engineer". The most qualified candidates we have come through the process are physics PhDs who did a significant amount of coding during their PhD and have been able to segue into machine learning because they were interested.

Our other tier of hireable candidates have been individuals with 4+ years of industry experience, usually with CS PhDs with a machine learning specialty.

The rest of our team came from internal transfers, people who were less qualifed but proved that they would have a positive impact on the output of the team.

Hah. I want to feel this is right. I like to think of my work as the "full-stack" equivalent of the "data science" career path. There's no part of the data pipeline I'm not currently doing/qualifed to do/interested in doing: acquisition, transformation, storage, exploration, analysis, machine learning, presentation & dashboarding, integration, server maintenance & operations...

The "toy examples" require only a very small subset of the skills required to extract business value from an amorphous blob of data.

Huh. This is actually a route I hadn't considered. Thanks for pointing it out! I will definitely consider it as I continue researching my next opportunities...

Yeah I'm definitely familiar with the Georgia Tech offering. But my impression is that it will be a $7000 + $(my_hourly_rate) * (hours_spent) certificate that will only get me past the employers who have a "Select your highest degree level" drop-down on their application form. Is it that much more respected than a collection of MOOCs?

As someone who's also involved in hiring, if I see no industry experience + GA tech online degree it's still on an entirely separate tier than 2 years of industry experience. But that's my bias I suppose, and part of the reason that I'm not the only one on the hiring committee.

As someone who has been producing value in a data science/machine learning role for multiple years, it's disheartening to see comments that I may be blacklisted from positions due to "only" having a bachelor's degree.

Somewhat non-humbly, I was valedictorian at my high school, I triple-majored at a respectable Big 10 school, I actively use all 3 majors on a daily basis, in a foreign country, and sometimes in a language that is not my mother tongue (as an American).

I can't justify spending time and money on a master's degree (millennial wealth problems) where many courses would just be putting a formal, academic spin on ideas that I'm familiar with from a practical business-value-producing point of view.

Any advice on how I can effectively jump off the black-lists?

It's quite reasonable to suppose that there are many reasons why one person would or would not mobilize to vote, and, correct me if I'm wrong, it's common knowledge that the "swing voter" effect is not due to undecided people becoming decided, rather that people are marginally motivated or unmotivated to vote.

So, in the spirit of democracy, where the desires of all those who /can/ vote (not just those who /do/ vote) are respected, it is probably interesting to examine why voters do and do not find the motivation to vote. Hypothetically, if we had mind-reading devices, and motivation was not a prerequisite for voting, which side of the vote would have benefited?

Asymmetrical motivation is the hypothesis that in referenda where Action/No-Action are the choices, there will be more voter motivation for Action rather than No-Action. This very reasonably applies to Brexit.

You're at least somewhat correct, at least judging by the headlines on realclearpolitics, for example, or by browsing politically opposed subreddits simultaneously. But I'm really astounded that you've said that anti-vaxxers are specifically right-wing. I thought it was a bit more like the flat-earth phenomenon where it's unclear who is being satirical, who supports it simply because of some disposition to conspiracy, and who actually believes that they've found evidence more compelling than the near entirety of the foundation of our scientific system - the same scientific system that has managed to create the device of incomprehensible complexity on which you're reading this comment.

Ah, I really appreciate that you've made that connection. Given the current state of politics in the US, I'm afraid if I ever bring it up it'll be perceived as a gimmicky talking point for expanding socialist policies, but literally every engineer (or someone who's tinkered with something) knows that removing components from a system is an effective method for testing its robustness. Even more so, the worst time to figure out that your system is not robust is the exact moment when you need it to be robust.

Ah, when bullshit jobs come up, sometimes the measure is "if this person disappeared, how long would it take for the world to notice?". This article is interesting to me because it applies a slight modification, "if this task was eliminated, would systems be more efficient?".

In some ways this is a good question to ask, but much like a statement in legacy code, it's sometimes unclear why a rule exists until after you remove it, and then it's too late.

If only there were unit tests for real-life bureaucracy...

HN loves it's metaphors and analogies. This one falls apart quite quickly because the transaction between you and your car mechanic is orders of magnitude more elastic/liquid than that of a employee and an employer. This is why the laws that govern those transactions are wildly different.

Once your car mechanic's single source of income is your transaction, which feeds their children and puts a roof over their family's head (as is the case for many employees), then maybe we can revisit this analogy.

Finally, because your analogy was so weak, I would also recommend to reexamine your opinion on employee rights at large, because they may be formed around similarly broken logic. Remember, the more your neighbors are protected from poverty and destitution, the less likely they are to be caught in a lifestyle of drugs and violence, which also benefits you and your family.

> "...the reason for democratic representation in general is that humans have a hard time giving real weight to others' priorities. They'll help them when they can, but certainly not if it involves real costs and trade-offs to their own, obviously far more important interests..."

I apologise if this draws away from the primary point of your comment, but I thought this was quite poignant. It's a sentiment that manages to convey my intuition about the motivation behind hard-line anti-socialist groups, and also has the added bonus of explaining why "feel-good" one-off donation/volunteering stories in the media are so popular. People like to pat themselves on the back for helping others, as long as they aren't inconvenienced. People like to praise social constructs that help others, as long as they don't encourage constructs that inconvenience themselves.

To be honest I didn't think my comment would get that much attention - and if I did I would've worded it a bit more gently. I think the framework that you've built around this product is really, really, really impressive (the virality and interactivity, for example). But I also think it's a very hard problem that you've set out to tackle. I alluded to it a bit elsewhere in this thread, but there hasn't been nearly as much work put into audio signal learning as there has been photographic signal learning. I'm really excited to see what else can be done, and I'm happy to see that someone is trying to do it. I wish you the best of luck

(But I'm also hoping that this doesn't further dilute the music industry by making music "production" even easier. Pop music is already "algorithmically solved", and other genres aren't as unique as any fan would like to think.)

Oh I definitely agree we have nothing to worry about :) I may have to check you out though! I enjoy live house of many flavors.

But at the same time, photo style transfer is...impressive[1]. It's interesting to me that so many engineers are focused on creating machinery that learns the photographic signal subspace, but there's not too many working on the audio signal subspace.

[1] https://i.stack.imgur.com/Rdoiv.jpg

Seems to just be a beat matcher / BPM adjuster, which is interesting in itself, but quite basic features in any DAW/DJing software.

At great risk of sounding haughty, as someone who's perhaps a bit more into the DJing scene than what's probably the average HN user, all of these sound horrible. Beat matching is the easiest part of DJing, and that's about all that this tool does, so I don't think it's solving any DJing problems specifically.

One thing that it does do is take away the turntable GUI (or hardware) and replace it with a two-field upload form, radically reducing the amount of know-how required to perform beat-matching. That's just really scratching the surface of what a lot of "live" DJs do (as they're called these days), as many are bringing instruments like drums and saxophones (see Big Gigantic[1]), keyboards and vocoders (see Lorenz Rhode[2]), their voice (see Gallago[3]), and traditional African drums (see Black Motion[4]), let alone simple song selection and set progression - "reading the crowd".

[1] https://www.youtube.com/watch?v=TkxMjfYOE-g

[2] https://www.youtube.com/watch?v=vwf2uvWlL_c

[3] https://www.youtube.com/watch?v=wmCFPjwlxIU

[4] https://www.youtube.com/watch?v=zHEc9uA2Zmw

When the tool comes along that can style-transfer two songs in the same way that it can for images, then I will admit that AI is on the verge of taking our music-creating/music-playing jobs. But for now, I'm satisfied that there's a bit more to the depth of DJing than that what an AI can learn.

Fram e Djurmen purßpektev, ets e bet kahmplekäted.