The chairs are probably more like $1000 after bulk discount for buying hundreds of them,and they'll last 20 years. Having worked in many corporate environments, including FAANGs that seemingly spare no expense, none of them have used these monitors. It's always a $300ish monitor for developers and I've seen some senior executives with $1000 wide screen curved monitors. Maybe a few designers somewhere get a fancy Apple one (although I've never seen it), but these are not intended for the masses.
You can pay well more than $1k for a curved wide screen, but as a counterpoint Dell 34" widescreen curved monitors are less than half that on their website. Most likely because you're not getting gaming features, HDR, speakers, etc, etc. They're pretty stripped. Even so I purchased one for every employee and myself on my old team, the experience was so good we all bought them at Microcenter for home use too, and still they weren't $1,000.
I'm just pointing this out in case anyone is thinking about getting a curved widescreen. The experience is amazing and it doesn't have to cost $1k.
Not that I'm disputing managers make status purchases and spend stupid money on useless things. That much is as true as it ever was.
Ya, FAANGs don’t exactly have nice offices, amenity wise sure but nobody GAF what they look like, there’s probably nerf guns and desk clutter everywhere.
Think somewhere with clients and prestige. Law offices, design shops, consulting boutiques etc.
The bigger issue I've seen is that there's a relatively fixed amount of time required for overhead activities (e.g. compliance training) and context sharing (meetings). Purely hypothetically, if those sorts of activities eat up 10 hours a week, then a full time developer gets 30 productive hours, whereas a half time developer only gets 10. You might think there's room to cut down on overhead, and you might be right, but when the majority of employees are full-time, it may not be the most pressing issue to change.
I don't think the second is realistic, as that might be considered an attack on NATO countries, which is likely not a scale of war Russia is looking for.
While I don't dispute the accuracy of your description, there are some conclusions that I wish we as a society would revisit. By your logic that the researchers only work for the university in a very technical sense (and often retain the commercial rights to their work), I honestly don't see how they're underpaid. If anything I question what value they bring to universities at all, and question how much of the inflation of college tuition is driven by this operating model (I know administration is a large part of the increase, but many administrative jobs are tied to the procurement and management of grants).
Bringing value to the university does not really make sense as a concept.
A university is fundamentally a community of scholars. The services the wider society expects from the community vary from generation to generation, but some things remain constant. The university as an organization does not have any goals apart from survival and prestige. It's the least important part of the university. The assumption is that the scholars deal with such highly specialized matters that generalist administrators have no competence in setting the actual goals.
Universities educate a lot of people these days, because the society believes that learning specialized topics from scholars and researchers is more prestigious than learning them from teachers or industry professionals. The society also gives plenty of research funding to the scholars, even though it's questionable whether the results actually benefit the society providing the funds, especially in smaller countries. It's just that you need research to have a community of scholars these days, and such communities tend to generate a lot of activity around them, with clear cultural and economic benefits in the long term.
Whether a person is underpaid or whether their salary is competitive does not depend on the job they do but on the other jobs they could do. If you want to hire a person who could make $500k somewhere else, the competitive salary is $500k. If you can't pay that (maybe the job only creates $300k of value), you have to offer something else if you want to hire them. Universities have traditionally offered stable positions in a community of independent professionals.
It depends on the grant. Most grants do not let you take ownership of the research. Nearly all publicly funded research contracts require the results to be public domain.
researchers do not retain commercial rights to their work, the university does. Sometimes the university allows the researcher a pittance of the license royalties, or make it easier for the researcher to start a company that licenses the technology.
Friend of mine was an apprentice plumber - was working in a tight space, bumped something and dropped his torch. I'd rather not deal with third degree burns.
When I was interviewing developers from a very large offshoring firm, they covertly had an account manager listening in to take down the questions for future candidates. Each interview, I'd ask a slightly different variation and it became obvious what was going on when a candidate gave a textbook answer to the question I'd asked the previous candidate, including references to tables I hadn't mentioned and that had no purpose in the problem I presented them.
Also had an incident where I interviewed one person but am pretty sure someone different showed up and was completely clueless about the most basic tasks. Within a week I think they figured out that I was onto them and made some excuse to quit about taking a contract closer to home.
Honestly, at this point many American companies have an office of their own offshore, and it's generally considered more prestigious and better condensated to work for those companies directly, so I assume there's close to zero talent left in the large outsourcers.
Large outsourcing companies are not really hired for talent (at least not from what I saw in the investment banking world).
They're hired as scapegoats, and attached to an already failed project. Then they're publically blamed for the failure, while the actual guy responsible gets away with murder (and maybe even hops into a new internal role with better pay).
If I was to rename the consulting/outsourcing industry, I would probably call it the "professional scapegoat" industry.
Indian offshoring companies like Wipro, TCS, etc pay $500 per month for fresh graduates from the engineering colleges, where lecturers are incompetent. In a team of 10, only two are capable, all others are for billable hours for these companies. Some of these newbies can learn and improve skills and jump to other companies in a couple of years.
In the end, what matters is the cheapest labor, and billable hours. Account managers are judged on how much profit they bring.
I presume something must have caused you to start giving slightly different questions, right? You did so because you wanted to see if they were collaborating in this fashion.
What was that "something," if you don't mind me asking?
I generally like to base my interview questions on the candidate's resume, asking questions of various levels of difficulty about the areas they claim to be skilled. I find it gives a better sense of their abilities than strictly asking about the tools/languages/algorithms being used on the hiring project. As such, each interview was already unique to an extent - and I might have noticed something like a person whiffing the easy questions but then getting an advanced one that didn't add up.
Also, the "architect" I inherited on the team seemed totally incompetent in a variety of ways, and insisted on using a fixed set of questions without deviation. To He kept flagging people through who seemed totally inept when I'd dive into their background, so they was also probably a tip off that they figured him out.
This conclusion feels like saying more CPU and memory are better. Seems obvious that more moves allows matching to have more nuance, but I guess cool that someone proved it.
From what I understand, it says that more parameters are good. This wasn't obvious before this paper: you can fit a polynomial instead of a neutral net, but adding parameters wouldn't help with robustness in that case: the polynomial would become more and more jagged.
> Seems obvious that more moves allows matching to have more nuance,
This really has to be balance against overfitting. The key problem in ML is generalization, and lots of things improve training performance while making that worse.
You were so close to getting there - he is a rationalist and, purely my personal opinion, the markets have been behaving irrationally for several years (I would say even the Trump years of fiscal manipulation to pump the markets).
Where I think you're off base is that, while markets can stay irrational for extended periods of years (and have before), eventually people realize the emperor has no clothes and there's always a reckoning at some point.
I hope you're right, and I really mean that, because I have personally made a large bet that the stock market peaked in Dec / Jan. However, I am an eternal pessimist, so I kind of assume my bet won't pay off and this new meme stock bubble economy is ACTUALLY the new normal. God I hope you're right.
