What's the draw for an engineer to work at Google these days? Seems like you'd be a small toad in a gigantic pond where your ripples make little impact. Kind of like working at Oracle or IBM or Microsoft or <Big Software Company>. Except there are those dumb bikes all around the campus. Yet still there seems to be this perception there is something 'special' about working at Google that sets it apart from any of the other places, isn't there? The emperor has no clothes? It's not like you're going to get any wealthier from options there than you would at any other relatively same sized company.
Here are the advantages of working at Google (from my perspective.)
- Excellent infrastructure. If you want to run a job on 2000 machines, write 10 lines of boilerplate, press go, and you're done.
- Lots of people smarter/better-educated/more-talented than me. (Though maybe this is common. :)
- Lots of resources. For a batch job, you can spend 10k machine-hours doing something frivolous without getting approval from anyone, since there's usually enough idle capacity somewhere. For instance, rendering the largest nebulabrot that has ever been rendered. http://www.danvk.org/wp/2007-04-06/nebulabrot/
Or finding the highest possible scoring boggle board. http://www.danvk.org/wp/2009-02-19/sky-high-boggle-scores-wi...
This is also great for prototyping something new that requires crunching a lot of data to create. I made a rhyming dictionary that was based entirely on datamining lyrics pages. It's also super convenient that we have a copy of the web on disk to play with.
- Google's brand behind any product you launch. Things might fade into obscurity, but they at least don't start out that way.
Disadvantages of working at Google:
- It's hard to fail fast because everything has to preemptively scale. It's probably easier to scale here than at most other companies, but still much harder than spit+duct-tape on a machine in your apartment.
- Google-wide goals/approvals can eat into the time to focus on product-specific things. There is some big-company overhead, though it's kept to a minimum.
- Established products are somewhat ossified, whether that's due to refinement or over-fitting is up to debate (and debated.) Thankfully though Google is remarkably not risk averse about making major changes to major products if you have the data to back it.
There are things that are easier to achieve in a big company than in a small one. Working on products used by almost everyone means that even small incremental improvements will have a larger effect than the whole product of an average startup has. Even a smaller Google property is probably on a much larger scale than the successful main product of a startup. There's better infrastructure and more machines to throw at problems than elsewhere, making it easier to solve really hard problems.
I'm sure that at some point Google will go the way of any large company and become an organization of middle managers shuffling paper around with nothing getting done, the rot hasn't set in yet. There's a big cultural difference to a standard big corp. (No, I don't mean the food or any other shiny baubles).
And while nobody is going to get fabulously rich from Google stock options/grants any more, I have absolutely no complaints about my total compensation.
Google still has great perks. If you just need a job and having the best cafeteria outweighs working with dinks, it's the best choice next to Facebook, and the only choice of giant dot-conglomorates if you live in NYC.
They're having the same growing pains as Microsoft had in the mid- to late-90s. Everyone prior to Windows 95 had a LOT of options sitting in the wings waiting to be vested or for a galactically large liquidity event (like, say, the revolution of the Desktop PC), and everyone post windows 2000 were left with the big company sitting on hoards of cash and a stock price that will never go above their options value but maybe 2-5 bucks max.
Google has all of its earnings built into the stock price for more than a decade, so anyone they hire (unless it's a chief muckety-muck) is going to be even-steven in stock. So, Google has to differentiate on culture, salary, and benefits, which they can obviously do.
laszlo bock always says that according to some very quantitative surveys they do, compensation is not the primary reason why googlers leave (impact is).
Whatever variation may exist, carbon dating has been independently verified via other dating means such as coral ring formations and dendrochronology. Unless whatever is having an effect on carbon dating also affects these other, unrelated, forms of dating in the exact same way we can expect the variation to be small and not significant enough to upend everything we know. The evidence that our dating mechanisms work is undeniable - tectonic movement, linguistic evidence of language dispersion, physiological changes in species and their geographical distribution, etc etc all happen at predictable rates and all match up with what we know about our dating methods.
Really, creationists should be better suited applying their ultra-skepticism at the much larger gaps in evidence and analysis found in their own ideas. The thing about people who value science is that whenever data presents a challenge we have to account for it, unlike the idiot creationists who only embrace data when it agrees with their already pre-conceived ideas.
There's nothing to be terrified of. Just live your life and do your best. If you're going to live your entire life with fear about retirement or debt, then by the time you reach retirement age you'll have lived most of your life in fear. Which is worse? Working during retirement age or living the vast majority of your life in fear? If things get really really bad you can always put a hole in your head or jump off a bridge. So don't worry about it. Just work hard, enjoy your life, and be happy if you want to be.
I'm certainly not the only one to remember all the survivalist BS from the 80ies. Maybe that'd be a good business to be in these days. "Survive in the woods with only your wits and a loin cloth!" "Bob and Doug's guide to finding beer after the apocalypse!" and so on...
> " If things get really really bad you can always put a hole in your head or jump off a bridge. So don't worry about it."
That philosophy might work if you're a single person.
That philosophy suddenly does not work so well if you have a spouse, children, and aging parents, perhaps sick or dependent families. Suddenly the whole "oh well, I can always just shoot myself" solution isn't so clean and practical.
yeah live your life at the fullest,,,everything eventually works out.
And if worse case scenario it doesn't...and you are broke at 55-60....go rob a bank. If you succeed you get some cash...if you fail you get free room, food and health care.
I swear I thought those words would get edited out by someone else at Cloudkick. I wrote that half jokingly and I guess it made into the actual blog post. :P
It has nothing to do with time travel, if you didn't care to read the article, why care to comment.
Also you can probably only know the future of different, possible similar timelines. Whatever changes you make in those other timelines will not affect your own. You can alter their future but not your own, otherwise you can alter causality. I don't know if altering causality makes any sense.
A perfect deterministic world simulator will imply the ability to see into the future, which would give rise to the "Dbz Paradox", even if that was not the point of the story.
It is not exactly a new revelation, if you include a model in the model, the problem potentially arise. What you typically do is to keep running the model with the previous answer, and hope the answer converge. It is a common technique for finding numeric solution to the more complex equation systems. I have done it many times. Of course, there is in general no guarantee that there is a solution, that the solution is unique, or that you will find it using this technique.
But if the simulated worlds are deterministic as implied then you could run the simulation forward into the future if your reality also followed the same determinism. Doing this would cause an interesting paradox where knowledge of the simulated future could change what you do in the real world. For example, simulate forward to the next lottery drawing to discover the numbers. Then play those numbers in the real world and win the lottery.
This all depends on the fact that the simulation and the real world are deterministic which, it would seem, is highly improbable.
In the context of this story, the simulation below you would also be looking into the future, and the simulation above would be looking at your future. Whatever "future" is seen has to be reflected in every simulation.
This doesn't change much, the problem irons its self out. I see no paradox. And I don't know what you see that makes determinism "seem improbable."
For the sake of argument, let's say that world zero (W0 for short) is at some arbitrary level. Worlds being simulated below that level would be -W1, -W2, etc. Worlds above would be W1, W2, etc. If everything is deterministic then worlds at all levels take the same path and arrive at the same outcome. But if W0 has control over the -W1 timeline, then W0 can view it's own future by moving the -W1 timeline forward. W0 now knows it's own future (lottery numbers for arguments sake). W0 can now use that knowledge to change it's future. The W0 future is now different from the W1 future which contradicts the ability for all worlds to be deterministic.
As I said, every simulation within the context of this story would be looking into the future via the lower simulation.
In this argument, every simulation would look into the future to see the lottery numbers, and would act out on this knowledge. There would be no "changing the future." Every simulation would see the lottery numbers, then see themselves go to the store and get a ticket with the numbers, and every simulation will win with those numbers. There was no other future. The future you see in the simulation is the future that will occur in your own simulation.
Calling this a paradox is like calling recursive mirrors a paradox.
The funny thing is that, given in the story that it is possible to simulate the universe, even the top level Timmy and Diane can't be certain that they are not in a simulation, only that they are at the top of that particular line of simulations involving them. Imagine that when they publish this paper and others try Diane's program out on their own quantum computer, they will also create infinite simulations, which all also simulate this account of Diane and Timmy creating simulations at a previous point. In each of those infinite simulations of others running Diane's program there will be an infinite number of simulated Dianes and Timmys experiencing themselves as top level in the occurrence described in this story (not seeing the black ball behind them). Basically if the universe can be simulated, we are probably all in a simulation.
They'd probably know when they didn't see the hole in space behind them.
They probably wouldn't have any qualms with turning the simulation off either. My guess is that all those infinite simulations were on borrowed time until the top layer got bored.
I don't think you read my comment carefully enough (could be my fault, I may not have been very clear). Just to clarify, unless nobody ever runs Diane's program ever again on their own quantum machine there are an infinite number of Dianes and Tommys not seeing the black hole behind them even though they are also simulations, and they would not know it one way or the other but odds are they could guess they are in a simulation and would more than likely be right.
