Hacker Newsnew | past | comments | ask | show | jobs | submit | scdoshi's commentslogin


This follows a 5.7B investment by Facebook


Interesting set of videos by Jansen explaining systems used to create strandbeest: https://www.strandbeest.com/explains


And him on how you may help the strandbeests reproduce https://www.youtube.com/watch?v=U02qqB-2nbs&feature=youtu.be...


S2, linked from the README[1], does the same hierarchical subdivisions in squares, with each square having a string ID.

Squares divide into sub-squares exactly, but hexagon's don't sub-divide into hexagon's neatly.[2]

Can someone explain the advantages of hexagons over squares in this use case?

[1]https://code.google.com/archive/p/s2-geometry-library/ [2]https://www.illustrativemathematics.org/content-standards/ta...


Gosper island shapes do subdivide into gosper island shapes, if you want something tricky: https://en.wikipedia.org/wiki/Gosper_curve#Properties


See sibling post for links, but the short answer: you can use 12 pentagons, strategically interspersed. Like a (soccer) football.

edit: Oops, that only explains the tiling. For explanations of how the hierarchy descends, Dr. Sahr produced some GIFs that give a little bit of an overview of how the resolutions overlay: http://webpages.sou.edu/~sahrk/dgg/images/topogif/topogif.ht...


Thanks for that link.

Also found this which goes into some more detailed comparison with rectangular grids:

https://github.com/uber/h3/blob/master/docs/doxyfiles/usecas...


Boundaries between all 6 neighbors are equal. With S2, four are on the sides and top are equal and the corner neighbors have only a single point as their boundary.

This means you can better simulate dynamic systems where there is flow between cells.


Looks like a really good idea.

How do you plan on scaling without becoming part of the water mafia (or at least enabling them)?

After all, there are only so many independent suppliers with their own water source.

Would love to hear you thoughts on this.


That's a really good point. We've had to think long and hard about this one too. There's one interesting stat that is less known. Aboout 80% of India runs on groundwater, and in cities, we have estimated that about 50% of the people have their own private water source. BUT- less than 1% sell that water. Now imagine we've used devices to aggregate all demand (phase I). Phase II would be to introduce new supply into the market. Enabling anyone with a borewell to sell water to anyone else. The biggest cost of private water right now is diesel transportation costs. When we implement Phase II, we want to enable more micro entrepreneurs to sell water hyper locally, driving down water costs. We'll have to do this carefully, but that's the only way we see not becoming the mafia and doing our part to make clean water more and more accessible to people.


>>we have estimated that about 50% of the people have their own private water source. BUT- less than 1% sell that water.

I appreciate that you are trying to solve this important issue. I was wondering if individuals selling bore water is legal in India


> Enabling anyone with a borewell to sell water to anyone else.

What happens when the neighborhood loses its water sources due to overuse?


Yup, that's something we have to monitor carefully. That's why the backbone to our business is actually IoT devices that transmit data to monitor different aspects of water. We're starting with data around water withdrawals, and moving to using devices to monitor groundwater levels. We think we can collect enough data from our devices to model groundwater aquifer health, something that very few people are doing due to lack of data and overall costs. Once we can monitor aquifer health, we can make sure that we are not overdrawing, and recharging effectively.


Monitoring aquifer health seems like a good approach to start with. Great to see you have thought about the issues.

Water is an ethically challenging area to start a business in, but the status quo in India is pretty bad, so I hope you do well!


Thank you so much- we really appreciate it!


I thought of this too. Water is a tricky thing to deal in because it isn't like any other resource. When you sell water to someone, are you also depriving water to someone else?

This can easily turn into an ethical minefield if you aren't careful.


This is true- and I'll rephrase what you're saying. What are the environmental impacts of groundwater withdrawals? If someone is withdrawing, who else isn't getting water? These are exceedingly tricky issues, and the sad part is that right now, nobody knows the answer because there is no data to answer these important questions. The first step, which we are implementing from day 1, is to use IoT devices to monitor groundwater withrdawals from our sources. The next critical piece on our product roadmap is creating IoT devices to monitor groundwater levels and using algorithms to estimate groundwater aquifer health. Given that we will be aggregating demand and understand withdrawals, this second piece will be able to estimate our natural resource health in a city for the first time. This technology advancement is critical for not just India, but the world. Once we know which aquifers are depleting the fastest we can try to recharge and use a different aquifers. This is the only way we think the world can start managing groundwater and ensure everyone has enough clean water to drink.


Aren't the drying bore wells and depleting ground water levels a testimony to that? http://bangaloremirror.indiatimes.com/bangalore/others/water...


Many (if not most) companies do stock refreshes annually, don't they?


Yes, but after the initial grant has vested entirely your opportunities for growth are limited.

Here's an example - say I'm hired with 1000 units of stock over four years. Each year I receive a refresher grant of 250 units of stock, again vesting over four years. Once I reach Year 4, my stock compensation stabilizes, assuming my refreshers remain the same (I'm am obviously grossly simplifying - you could be promoted and get larger grants, the company could do badly and your dollar value be significantly reduced).

But what you can do, and what the parent is alluding to, is jump ship around year 3-4 and get the company you're going to match your current amount of outstanding stock, meaning your initial grant becomes much higher, and your refreshers do too. At least, that's the idea. It's worked for me.

Additionally, although personally I did very well on my refresher grants I know other people at my employer did not. So that's another factor.


Normally refresher grants are not based on your starting grant, but what you might receive if you were hired for the same position today.


Are you able to put any actual dollar or % amounts against this?


It's meaningless alone. It'll need to be higher at roles with a lower base salary or can be lower at roles with a high base salary. Total annual compensation is what you should pay attention to.


There's refreshes and then there's the speed of the vesting schedule.

Some places offer a bunch to start, but very little after that, which means people tend to leave once they have vested all their options/RSUs.

After the first year, sometimes your vesting schedule will be monthly, quarterly, twice a year, or once a year. As to why they don't allow everyone to be monthly - it's probably because they get to claw back a fair amount of the equity if people leave.


> As to why they don't allow everyone to be monthly - it's probably because they get to claw back a fair amount of the equity if people leave.

I may be being a little cynical, but if you vest annually or bi-annually you can get a pretty good idea of when your employees are going to leave and use that to plan your crunch periods accordingly. People are rather unlikely to quit the month before the annual vesting date.


I think if companies could actually predict crunch times I'd agree with you, but scheduling is not an easy science.


To double down on the cynicism, you also need to be sure your employees value your options for this to be relevant.


My memory is fuzzy but the one place I got options that were worth anything, it would have taken me about 8 years to double the stock grant I was given when we were acquired. So I had this big block of initial options of which hundreds of shares vested every quarter, and then dozens of shares of the subsequent grants came due at a slightly different time.

To me, the first rule of rewarding someone is to make sure the recipient did all of the work before the reward. If you make them work afterward then it sours the desserts.

I would end up holding onto those little blocks until the next batch of original stocks vested because I had to push a bunch of buttons like a monkey (in a UI designed by monkeys) for a couple hundred bucks. Which just reminded me that I could be making more money and a bigger impact working just about anywhere else.

And then they cancelled the only project that any of us thought had any strategic value to the company. So my direct manager quit. Then someone else quit. Then I gave about 3 weeks notice, but was the 5th one out the door because 2 other people resigned in the week after I gave mine. In all about 8 people left in something like 12 weeks.

Because we all quit right after the end of the fiscal year, we were all owed a bonus from the previous one. When we got the bonus check (mine was a little over 2 month's pay, but the majority of the others got better reviews so got bigger payouts) we went out for beers. Someone, probably me, asked if it was worth it to stay for the last four+ months when things started going sideways. Only one person said yes, out of eight.


> To me, the first rule of rewarding someone is to make sure the recipient did all of the work before the reward. If you make them work afterward then it sours the desserts.

For executives and directors, I disagree completely. The shareholders are paying them to make and execute decisions that will grow the company (and share price) over a long period of time. The share-based compensation plan should reflect (and help align) that they are being paid to make long-term decisions.

Those are inherently "promise/award before the hard work is done" type of grants as I see it. (Why would I pay you more for a decision you've already made or work you already did, unless we're just talking about a modest spot bonus for a job well done?)


We aren't discussing keeping directors and executives happy. We're discussing directors and executives making stupid decisions and thinking the people doing the work won't care.

Also, stock options are worth less than nothing several times over if you are not in a position to make decisions. First, they're issued at market rate, so since they're options and not a stock grant, after fees you'd lose money selling them. Except you can't sell them because they are pieces of paper for a year.

And I know this will stick in the craw of a lot of developers, but as a non executive, any control you believe you have over the stock price is either tiny, illusory, or something the executives will take away if they notice. This is their game and to them you are a pawn, no matter how brilliant and indispensable your team and boss think you are. If you're affecting the stock price then it'll be up or out for you. You'll either be made a manager or treated as an unwanted variable.


In that particular line I'm talking more about execution of incentives, then painting options with that brush. Have you ever won a contest or award at work and then they can't seem to deliver it? The check or the tickets never shows up and you have to keep working for something that was already promised you? It sours the whole experience.


Some companies do small refreshes to almost everyone.

Others do performance based refreshes, so job hoppers often think they’re never getting more stock, because they had their eye on the next gig or their side hustle or whatever.


From what I know the refreshes are typically pretty small.


I've never seen it myself. My understanding is that it's pretty rare.


This might be just me, but the most important thing to me when I get advice is:

- who am I getting it from

- what do I know about their life/experience

- under what circumstances

Because that let's me adjust how much importance to put on it. Getting advice is generally not the hard part, figuring out which advice to follow and which to discard is.


Totally agree.

We help people build external social support networks as well.

In addition, as is true with all good coaches, most conversations circle around introspection and talking through analysis rather than imposing our own advice. We still give plenty of advice, but we also are aware enough of the nuances that exist and we-don't-know-what-we-don't-know.


I might be missing something, but seems to me like the products are still recycled, but as raw materials, not as full devices.

This makes sense, given data is almost certainly bound to be present on phones etc.

Maybe there's a more nuanced way to do this, but this doesn't seem all that bad.

OTOH, having these phones go through a refurb program and selling them probably involves a fair amount or resource use as well.


Forcing recyclers to shred perfectly fine phone components makes sense because data is bound to be present 'etc'.

My "Hacker" "News" moment of the day.


Who said they are "perfectly fine"?

Also, it's not up to you, the costumer decided to deliver the device for recycling and not for scrap.


Reusing is more efficient than recycling can ever be.

The resources used to refurb and resell are a tiny fraction of those required to go from raw material (recycled or mined) to a new phone.

As far as data goes, soft tools will stop anyone but the NSA. If you're worried about the NSA you don't toss your phone in a collection box.

Yes, you are missing something.


> Maybe there's a more nuanced way to do this, but this doesn't seem all that bad.

Like removing the flash storage first, maybe...?


With all the human labor that'd take, you'd pay more for refurb than for new.


I find it highly unlikely that human labor is necessarily required to remove the backplate of an iPhone, remove flash storage and recycle the rest. But let me know if I'm wrong - the entire example is extremely contrived in any case.


It'd be a hell of an expensive robot, and probably not all that reliable - more complex than "Liam", what with the masked hot-air rework requirement, and even that was just an ad that never actually got built. And that assumes you only need the one kind of robot, which is a bit optimistic given that sizes, shapes, internal configurations, and assembly methods tend to differ among iPhone versions. You might need several expensive robots. (And have you seen what it takes to tear down an iPhone? It is not a simple process! Special tools are required, and I don't just mean for getting the unicorn screws out.)

In any case, you'll pay much more for a refurb that's helping cover cost of the considerable labor involved in that kind of rework, whether it's expensively done by humans or done by expensive robots. With that kind of production cost, and especially with a huge upfront automation investment to amortize, it's going to be a struggle to keep the prices of refurbs from exceeding those of new product, and at that point it's no longer economical. The whole point of buying refurb is to get the same product for less money, or more product for the same money, as you'd get buying new. If you can't sell refurb at that kind of price point without taking a loss, you may as well not bother.


If only there were a way to encrypt the flash storage and then throw away the key.. Alas, one can dream.


This has been a fun blast from the past. viaCycle was in YC (S12). And yes, we shut down in 2013 around 3 years in.


Hey, I would love to hear more about your experience, would you be free for a quick chat ?


The article describes a hypothetical scenario.

It's clear if you read carefully and pay attention to the dates, but this is the kind of thing that seeds fake news and forwards. Disappointed that shock value and clickbaity titles are being used by a rights advocate group.


> It’s January 2018 and after enjoying a meal with her family

I did not realize that either until I read the comments. That is some over-dramatic click-baity bullshit.


Yes, it's fiction, but it seems to me to be a reasonable extrapolation from legislation that is moving through the UK parliament right now.


>this is the kind of thing that seeds fake news and forwards

One could argue that careless readers are "that kind of thing", more so than anything else.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: