Might as well be a REIT, and as its books stand, its a crappy one. I wouldn't touch wework unless it had a dividend. What their CEO has done with inside dealing should be considered fraud to investors.
I really am surprised to hear this. I'm about 250 all in and I haven't heard of many software devs pulling over 200 all-in. This is for Boston and I'm 6 years out of my undergrad with no graduate degree (although I didn't go to college until my late twenties and I have noticed my maturity helps a little). Maybe on average it's the same but data scientist at the right company has a higher upper limit?
I am just going on my experience, but according to glassdoor data scientists of the same years of experience earn about 20k more than senior software engineers here in Boston.
This is an awesome analysis of the situation. Some things I have noticed as a data scientist of 4 years so far:
- Increasingly, data scientist is just being used in place of senior analyst because it attracts more applications.
- At the firms I've worked that are software tech companies, there was an outsized interest in mid-level software engineers wanting to be data scientists, mostly because the career development prospects at that stage are grim and data science usually means a pay bump. This demand has had the opposite effect - software shops are leery of promoting engineers to data scientists for fear of inciting contention among the ranks.
- Building on the data scientist usually means senior analyst, it has also come to mean analyst that can build their sql query into a scheduled ETL or daily process of some sort. You work in pandas instead of excel sorta thing.
- I have personally gotten all my data science jobs from talking about the business side of things. I think engineers approaching the field from a hard-skills perspective is totally wrong. My last technical take home was in a language I had never used before and likewise my execution was shitty, but I was able to well explain the problem, how the data could be used to predict the variation, and how the data science product fit into the business. I got an offer before I left the building.
I'll counter: There have been quite a few instances of FB friends like this in my life. I rarely deny a friend request or go on deleting sprees. If I meet someone doing interesting things, I'll be the requestor even. What happens occasionally is I find out someone I spoke with in person for only a few minutes, reveals over weeks, months, or even years, has very similar interests, humor style, etc. Comments on each other's post increase in frequency, until the fake friend becomes a real one after reaching out via a message that doesn't really come off as odd at all at this point. Who I do find myself unfriending is those I dated briefly that inevitably get married and have kids. Good for them, but time to disconnect. As I grow older, I realize more friends are a function of convenience. Your "real friends" will go on to do things in their life that distance them from you, and seem less like real friends, and your "fake friends" will become real friends when it seems logical. This ebb and flow may seem ugly to some, that would prefer to view all their friends as genuine stakeholders in their life with unwavering caring, and some of that does happen, but there is beauty in friendships being temporal as well.
I really wish we would curb this kind of behavior. Even if this speed is related to the brokerage business and not HFT, it is unfair to retail investors that large businesses have better market access. My solution: exchanges are required by law to process contracts in pulses with strict execution ordering rules. E.g., every 15 seconds process all buy and sells, processing from highest bid and lowest ask first for each security. Basically a turn-based stock market.
Crossing over from thinking something is one way to testing that something really is that way is certainly a line held within the purview of science. I think you are reading my statement too literally.
Well, until it has been confirmed by a study, it's not science, obvious or not. A few years ago there was a major study that proved a relationship between multiple concussions and long term brain damage. Anyone could have made the conclusion that repeated blows to the head is bad but it doesn't become science until we apply rigor to test it. That's a good thing. At all points in history we believed foolish things that we thought were obvious.
Yea, but what are they using to analyze your results? You literally have no idea how robust (or not) their statistical methods are. It's been my experience that 3rd party solutions are notoriously shoddy in this department. They have no incentive to be robust.
You are over-simplifying. Almost all of the metrics Facebook will feed you when you setup an account are on a view-through attribution basis. Yes, there will be some clicks, but if you attribute revenue from Facebook by clicks only, it will look terrible. This is an old problem for anyone that has spent time with display ad platforms. So, you install a facebook pixel on your checkout, and if the person making a purchase saw your ad on Facebook, Facebook's reporting console will take credit for that. Thats view-through attribution and it highly overstates the credit Facebook should receive for making the sale, but there's not much you can do to measure it otherwise. So what do you do?
The answer is, you setup a holdout test. Facebook has a black-box solution for helping with this, but it is a huge conflict of interest to let Facebook handle your test. The thing to do then is set up the test on your own platform. That is a massive undertaking, and any business that doesn't have a significantly sized engineering and analytic team doesn't stand a chance.
So yea, people look at things like reach, and view through conversions, and it would be very hard not too or they might not have the means to do something more sophisticated. Fake accounts undeniably puff of numbers for these people, and they should be taken to court for not giving a shit about that fact.
I want to preface the statement I am about to make with another statement first: The basic economic theory, if anyone needs a refresher, is that by letting people have a temporary monopoly over an invention, they will be more-so incentivized to pour money into the development of inventions than they otherwise would. The US has historically been the most stringent protector of intellectual property, we have a vastly outsized technology and research sector versus the rest of the world, and most people in this country believe those two things are causally related. I do not believe in that causal relationship as there is plenty of evidence that it is overblown or non-existent. E.g., I believe our outsized technology sector thrives in spite of, not because of, our stringent ip laws.
Now then, onto China stealing tech. This is absolutely happening. They have some of the weakest IP protection in the world. They don't just steal from us, businesses in China also steal from other businesses in China. They steal from everyone. It's in the open. It's not a smear campaign, it's an entirely different cultural and governmental take on how IP rights should be enforced, if at all.
There is no conspiracy here. These are directly two very different views on how IP should be handled.
Both rideshares subsidize about half of what they pay drivers, but their deficit is not nearly half of their total revenue. Take out the drivers, profit tomorrow. Patent good for 20 years. That's about as juicy as it ever gets.
While I would agree with you, driverless tech is in fact the reason behind the valuation and balance sheet deficit. No large investors at this point think they are going to get a dividend paying profitable commodity business.
I'm not quite sure everyone discussing this, including the economist article author, is in the loop with why Lyft and Uber are getting valued the way they are. Wall Street is not paying for a taxi company, they are betting that Uber and Lyft are the best poised to first rollout driverless ride sharing, i.e. one of them will own and monopolize patentable tech that the mass public wants to use. The existing experience is a commodity business where profits go to zero, that's well known and old news.
Safety bars keep you from dying if you have to bail, but bailing is still bad. Basically you want to say "take" and have the weight immediately removed. There is no use in straining yourself or risking injury on a bail; you have the lift or you don't.
Mark Rippetoe, the author, is a broken old man as a result of his methods. He is also known as a loudmouth that gets more attention than he deserves in the community. I would not recommend his methods or book. Although his book got popular among amateurs, it has no place in modern powerlifting and is considered antiquated and rife with bad advice.
Former competitive powerlifter here. Find a team. If you search around, you are guaranteed to find some dungeon gym that has a group that trains together. It won't be a big-box gym. It will have rusty iron. This is where you want to be, this is essential. You need 2 or 3 people to properly spot a powerlifting squat, don't bother doing it alone.
That author fails to grasp the concept of good enough, or optimal marginal return. Subsampling is often done because you can get the level of prediction desired at a much lower cost, both in terms of hardware and employee hours. Those projects that can glean value from full data usage certainly go for it, but always using full data on everything would be a poor decision.
Simple linear regression would work just fine for something rudimentary like this. This using neural nets stuff for everything lately is getting ridiculous.
I specialize in AdTech, mostly building bidding algorithms for emerging clients after years of managing AdTech at larger companies. It's mostly secret sauce, no one really publishes that stuff. It would be bad PR and you don't want the competition to know what you are doing.
Utility is an non-measurable unit with no connection to anything real that economics textbooks use to help people conceptualize basic theory. Capital creation versus rent seeking are concepts grounded in reality with real consequences. Creating money does not necessitate capital creation or increased well being. I suggest reading about the broken window fallacy to understand why you are wrong.
I fully understand this reaction. Crypo is delivering on none of it's promises and leaving an ugly mark on the production side. Transactions are way too slow for retail, decentralization is a joke because the largest mining pools are manipulating the supply with less transparency than national currencies, the value is too volatile to entice people to use it for anything other than speculation, and it is literally the most environmentally destructive modern technology at the moment. If miners moved into my town you bet my ass I would be at town hall telling them to fuck off.
"Mathiness" as the author puts it is one of my main complaints of many academic papers, this is not at all limited to ML. I think we really need to get away from typical math notation and start to present the calculations as code. This is more readable to the modern practicing professionally and the algorithm's degree of parallelization opportunities (or troubling lack thereof) would be more clear.