I don't think Confluent really counts as "non-LinkedIn team" :)
"Jay is co-founder and CEO at Confluent. Prior to Confluent, Jay Kreps was the initial developer on several open source projects, including Apache Kafka, Apache Samza, Voldemort. He was the lead architect for data infrastructure at LinkedIn."
I never quite got what "it takes x gallons" actually means. Where does the water go? Does it get sucked up by the plant somehow? Does the plant split the water into oxygen/hydrogen and those get bound to some other atom? Does it just disappear into the ground and end up as groundwater? Does it just evaporate and rain down somewhere else? Does it end up going to a river and ending up in the ocean as saltwater?
Assuming the farmers are irrigating efficiently then most of the water is being lost to evapotranspiration aka the plants sweating it out through their leaves. Almond trees let more water escape through their leaves than say a cactus might and that is why it takes more water to keep them healthy.
I still think that water desalinization improvements should see a lot more funding... and that cross-country water pipelines from the areas that commonly flood over, to the southwestern US should be started.
Combined with wind and solar power, this can be used as a fuel source (hydrogen storage)... it's rather short sighted that such efforts haven't been made in this country starting years ago.
So if it evaporates, as in the parent question, what does "requires 1 gallon of water" mean? One gallon goes back in the air, right? It will presumably end up as rain? So the net loss of water is 0, or am I missing something?
Is there some loss that is permanent or slow to recover, here?
It means that one gallon of water that could've been used elsewhere in California during the period of growth went into the clouds and is likely going to rain down a few thousand miles away.
The point of the "n gallons of water per x fruit" is saying that one certain crop is incredibly inefficient with its water consumption, and it'd be better to use it on something else. It's like saying a bloated web browser consumes 500 MB of memory/tab. Your memory isn't gone forever after you use it, but so long as you're using that browser or growing that crop, it's draining that resource away from other applications. You probably only have 8-16 GB of memory to use at once, just like a region only has N gallons of water it can use at any one time.
That gallon of water used to water that tree right now is not available to me to shower, drink, or flush right now.
The water becomes immediately unavailable until it returns to the ground as rain somewhere or someWHEN else.
Sure, looking at the big picture the water is never actually "lost". Until you can plumb my domestic water into the clouds directly, the big picture is irrelevant. All that matters is the water that is available in a potable form for me to drink right now.
In financial systems, we might come to an arrangement that I will pay you Friday for a burger Tuesday. With water, it doesn't work like that. If I don't have water now at the time I need it, there is no use promising me water late next week.
And when it comes to a decision between me having a litre of water today for me to survive on, or saving your thousand-dollar almond tree, I will unashamedly pick me.
In a localized sense, the loss is of the water that's readily available for use in that region. I suppose you could think of it as a sort of "debt," where you measure the loss in terms of energy or dollars required to restore the water at the rate it's being consumed, for instance through desalination.
I suspect it's just simpler to call it "loss of water."
Another way to think about it, is that farming has historically occurred in places where there are decent natural growing conditions: Climate, soil, and water. If you let the water go somewhere else, through evaporation and rainfall, then you might end up having to maintain growing conditions artificially by bringing the water to where there is decent climate and soil. And again, you could measure the cost of doing that as a "debt."
This would be true if they were actually getting rain, and they were using surface water, but that is not the case. They are in the middle of a drought, and they are using ground water, which replenishes very slowly.
The water cycle has a net loss of 0, but only on a global scale. Water that is expired by a tree and ends up as vapour in the air does not precipitate straight out again, unless the air is already completely saturated (ie: it's already raining/snowing). That water will indeed come down again... somewhere.
In general, something needs to change for the water vapour to turn into rain, like changing altitude from passing over a mountain (reducing temperature and pressure), or the air stream combining with a colder one. Deserts tend to be flat and featureless, so there's little in the way of atmospheric change as you pass through them.
An interesting example of this is the continent of Australia - the most populated part is on a thin strip on the east coast as that's where the water and hence arable land is. But the air moves across the continent from the west - in general, the air crosses an entire continent before giving up it's water. It gives up that water because it hits the long strip of mountains slightly inland from the east coast, which changes the conditions appropriately. But until it hits those mountains, the entire overland journey is very consistent for the vapour, so it stays as vapour. There are several mechanisms at play here in dumping the water from vapour into rain - check out wikipedia, I couldn't do them all justice.
I don't particularly enjoy poetry (or poets for that matter), but I always liked reading "Algorhyme" by Radia Perlman, the inventor of spanning tree network protocol
I think that I shall never see
A graph more lovely than a tree.
A tree whose crucial property
Is loop-free connectivity.
A tree that must be sure to span
So packets can reach every LAN.
First, the root must be selected.
By ID, it is elected.
Least-cost paths from root are traced.
In the tree, these paths are placed.
A mesh is made by folks like me,
Then bridges find a spanning tree.
- Radia Perlman
I remember that this seemed very appealing when I first looked into a place to buy ETFs. I then opened the Schwab account because I really enjoy their checking account and they are, as far as banks go, pretty solid and not prone to screwing customers over.
I personally use Charles Schwab (just because I have my bank account with them) and Wealthfront to invest in ETFs.
Schwab is comparable with Vanguard in terms of fees and ETF availability, so I never bothered with opening up an extra account at Vanguard.
I use Wealthfront to diversify my ETFs. They take care of distributing the money I put in the account among several areas of the market (U.S. Stocks, Foreign Stocks, Emerging Markets, Dividend Stocks, Natural Resources, Municipal Bonds and Cash).
They mostly use Vanguard funds. I could probably also do this myself, but I don't really want to manually balance my portfolio every time asset classes go up and down. So far I only put the amount of money in there that they manage for free. If you intend on signing up, one word of caution:
I signed up on their homepage and had 10k under free management. If you sign up via a referral link, you get 15k. I'm still slightly annoyed that I didn't use one when opening an account. I know that it's against the usual HN policy to post them, but since it is detrimental to sign up without one, here we go: http://wlth.fr/196dDW2
(I'd check with my friends first to see if one of them is already on the platform)
I looked at both of them initially. As far as I recall, at least back then, the fee structure for people that initially only have a small amount of money invested (10k or so), Wealthfront was slightly better.
I also like Wealthfront's UI. Not that I'd do a whole lot besides click the "add more money" button and look at the graph. Turbotax import worked nicely too.
There is an entertaining public fight going on between Wealthfronts and Schwabs CEOs about their robo-investor options. Once the big boys get involved I have a feeling Wealthfront, Betterment and others are going to feel the pinch. Schwab has their offering now with zero fees (except for those from the ETFs owned).
Do you have any backtest links that show a small cash position 'is an incredibly bad idea'?
A small cash position is another diversification strategy. As your portfolio increases in value, taking some profits to cash and then having it available on the dips may be perfectly valid. Buffet himself has called cash the never expiring call option.
As far as I understood, target retirement funds will switch the allocation from stocks to bonds once you get close to retirement. With the robo advisors, you'd have to manually move your risk tolerance downwards.
I personally would hate to have that happen in an automated way. I don't want the fund to automatically move over during an economic downturn. I'd rather have it happen while the S&P is at a new high :)
I find that relying on my employer for everyday stuff isn't conducive to building future proof systems: am I alone in finding it hard to budget for expenses when my employer is providing a large chunk of my expenses?
Whenever possible, I prefer taking a cash equivalent and buying my own. Insurance, food, living quarters, public transportation..whatever.
Well employers get discounts for getting a large amount of these expensive.
Food might not be true, but if an employer would give you the exact amount of money he pays for insurance, you would get an insurance a lot crappier, simply because the employer gets discounts.
I am currently paying insurance out of the pocket, it sucks. When I'll reach 30 (no more catastrophic plans), I'll definitely will find another employer with health insurance.
I care about these things specifically because the IRS doesn't tax it. I save about 6k per year because my startup feeds me 5 lunches/3 dinners per week and pays for my public transportation. That's probably ~8k pre-tax dollars.
That's a good question. We are currently considering salary only but in near future we will add job position too. For example, a candidate can say: "Contact me only if some company is willing to consider me for VP of Eng.".
You should consider total compensation and salary vs. just a salary. If you're looking at finance, people will have requirements for both a base salary and the total compensation. Looking at them in isolation doesn't work. Same thing if a large portion of your offer from say Google is in RSUs, that's not cash; but is essential for evaluating the economics of the offer.
In my opinion, you should add total compensation right from the beginning. This is because (1) You would be capturing data from many people (hopefully) and it would be difficult to fix later, (2) The site will be less useful without for many people including me, and (3) Some people will invariably confuse and put total compensation anyways which will introduce unnecessary variability in your data.
On the other hand, adding total compensation from the beginning cannot be too complex.
Things like startup options are basically monopoly money. If the salary is $30,000 but they offer two ten-thousandths of a company which is currently worth nothing, but they plan for it to be worth five billion dollars in five years, is that a $230,000 salary?
Recruiters will post bullshit if there's no downside; to avoid this you need concrete measures that can't be manipulated and can be audited.
Total compensation need not be included as a single number, it may/should be separated into actual factual numbers like base salary, fair market value of stock options / RSUs at the time of grant, cash bonus, etc. probably with a statement that the candidate is willing to trade one component for another to find a suitable mix.
Startups and big companies follow different norms, I am unsure if the latter should be ignored.
You can put a price on perks. Food? Provide something that matches what an urban centre provides for free? €10 per day by 240 days ~= €2,500 extra post-tax income per year (which you can round up to €5k pre-tax). Companies don't like when you point out how little the perk is actually worth.
That scenario is true for many. The core idea is somewhat like "what amount of salary will make you think of moving to a new job". Also the employer gets your details only when you approve them (one by one), so this might work in this scenario too.
Seems like a relatively basic CRUD site.
Considering the SSL issues which apparently weren't tested properly and the immediately published feedback that shows tons of private data from people, I really, really hope that the site won't be hacked in the next month or two. Seems like a prime target for folks looking to commit identity theft :-/
+ I believe it makes my writing better, even if only via the additional practice.
+ Journaling has been, at times, a sanity check. Writing about some craziness in my life provides a way to stop ruminating over it.
+ Journaling is a great way to explore ideas that I have.
Over the past three months or so. I've focused more on journaling as a writing and exploration tool and much less as a safety valve over observations of meanness or stupidity in the world or some sadness I might experience. Those things are easy to write passionately about, however I'm generating enough ideas I feel passionate about that I don't need to tap that resource.
On what day of class do they prove that "Microeconomics" is a Science? Would you care to demonstrate that?
I'm just asking because you're making an appeal to something in that argument, and I'm curious if you're appealing to a serious scientific field of study, or something passing for one.
Karl Popper said falsifiability is an essential feature of any scientific theory, so it's not weird to ask for a falsifiable statement from "Microeconomics" so that we can tell whether it's a scientific theory.
Mind you, I don't disagree with your conclusion that prevailing wages go down when labor supply is increased. What I am asking is whether "Microeconomics" is a science.