Linux Foundation 2021 annual report

Posted Dec 8, 2021 8:27 UTC (Wed) by tpo (subscriber, #25713)
In reply to: Linux Foundation 2021 annual report by raven667
Parent article: Linux Foundation 2021 annual report

> These salaries look fine to me, these people are well-compensated
> but this is not an obscene amount of money at all, this is what
> success should look like.

I do not think it matters that those people get paid "that much". The important question to me is whether so much $$$ corrupts those people. I.e. this simple question: "if you were not paid, would you have decided the same way?".

Many of those people have multiple hats on. One of those hats' duties might be "merge changes in the best interest of Linux". Another hat's duty might be "merge changes in the interest of $$$ corp". Do those different hats lead to concrete conflicts of interest? How are those conflicts of interest resolved?

It's not pretty for this situation to exist in the first place. However maybe that's real life: everyone needs to eat somehow.

Linux Foundation 2021 annual report

Posted Dec 8, 2021 15:26 UTC (Wed) by khim (subscriber, #9252) [Link] (18 responses)

> One of those hats' duties might be "merge changes in the best interest of Linux". Another hat's duty might be "merge changes in the interest of $$$ corp"

True. But that's because their compensation is too small, not because it's too big.

Linus gets more than, e.g., Greg Kroah-Hartman — because Greg works for Google and thus is Ok for him to get smaller compensation in Linux Foundation.

> It's not pretty for this situation to exist in the first place.

Oh, sure, it exists. Of course. But would it be good allocation of resources for Linux Foundation to raise Greg's compensation and ask Greg to leave Google to resolve that dilemma? IDK.

But if they would start paying less then this would make the problem more acute, not less acute.

Linux Foundation 2021 annual report

Posted Dec 8, 2021 16:41 UTC (Wed) by tpo (subscriber, #25713) [Link] (17 responses)

> But if they would start paying less then this would make the problem more acute, not less acute.

I do not believe that. Can you point to empirical evidence?

What those people earn is "more than enough". There have been studies that showed that increases in wage matter for your well-being up until about 100K and after that no improvement of well-being could be shown.

As a counter example to your hypothesis: my LWN account gets paid by HP. My laptop is an HP one (bought myself). It's a piece of shit: there are so many things that do not work as they should [1]. However, knowing that HP is sponsoring me, I hate to be say so and "hurt" HP: it's painful and makes me cringe -> HP *did* succeed corrupting *me* in this case. OTOH: If my laptop was by Lenovo and it was shit I wouldn't hesitate at all to say so. So here you go with an example that disproves at least the general applicability of your hypothesis.

(Sorry HP, thanks a lot for the LWN sponsorship, but your laptop sucks. To be honest I'd much rather you paid someone to fix all the things with your laptops that don't work under Linux than paying those sponsorships)

[1] https://wiki.debian.org/InstallingDebianOn/HP/Envy%2017-c...

Linux Foundation 2021 annual report

Posted Dec 8, 2021 16:53 UTC (Wed) by farnz (subscriber, #17727) [Link] (4 responses)

The studies I am aware of do not agree with your claim - there's a lower bound, below which happiness is not possible, and then increases in happiness are linearly proportional to the logarithm of your increase on the baseline.

For Manhattan, at the time of the study, the baseline was $85,000 per year. Then, you'd expect the happiness gain going from $90k ($5k excess above baseline) to $95k ($10k excess) is the same as going from $95k to $105k ($20k excess), and to get the same gain again, you'd need a $20k rise to $125k. And to get the same gain (3 doublings of excess) as you got going from $90k to $95k when you're on $125k, you'd need to go to $285k.

Thus, the RoI to you on pay increases goes down as you're paid more.

Linux Foundation 2021 annual report

Posted Dec 8, 2021 17:31 UTC (Wed) by tpo (subscriber, #25713) [Link] (3 responses)

"The best-known theory on this topic is that money can buy happiness, but only up to a point. This comes from a study by two Nobel Laureates, Daniel Kahneman and Angus Deaton (2010) , which found that emotional wellbeing rises with income. However, it rises logarithmically. That is, as an individual's income increases, their wellbeing increases at a slower and slower rate. And after income surpasses about $75,000 per year, Kahneman and Deaton's data suggests, wellbeing stops increasing altogether."

https://www.givingwhatwecan.org/post/2021/06/can-money-bu...

I guess this is what I remembered. However if there are other studies that show a "consistently logarithmic increase" then I have no problem with that.

Linux Foundation 2021 annual report

Posted Dec 8, 2021 18:21 UTC (Wed) by farnz (subscriber, #17727) [Link] (2 responses)

So, couple of things:

If you go back to Kahneman and Deaton's study, they did not state that wellbeing stops increasing above $75k - rather, they stated that with the data they had, there were no signs of an increase in happiness above $75k, but noted that their data set did not include enough people earning above $75k to draw conclusions from. It's the media that extrapolated from this to "wellbeing stops increasing", when the researchers said "not enough data".

Matthew Killingsworth had a longer term study than Kahneman and Deaton, which started in 2010, but did not publish results until 2021. Killingsworth had more high earners than Kahneman and Deaton, and due to the length of the study, Killingsworth had data on people who'd seen their pay rise considerably during the study period. This extra data agreed with Kahneman and Deaton up to $75k, but showed that as pay rose beyond that, wellbeing continues to increase, remaining proportional to the log of your income.

One other thing that comes out of Killingsworth's data is that there exists a floor level of pay where you are at risk of being made unhappy by random life events - coincidentally around $75k - because you cannot afford to throw money at the problem. Things like your home heating/cooling system breaking down fall into this category; if my home heating fails, I can afford to just pay 10x the normal price for an emergency replacement of the broken bits, where when I earned less, I couldn't.

Linux Foundation 2021 annual report

Posted Dec 8, 2021 19:41 UTC (Wed) by rahulsundaram (subscriber, #21946) [Link] (1 responses)

> One other thing that comes out of Killingsworth's data is that there exists a floor level of pay where you are at risk of being made unhappy by random life events

Yep. This is the key part of the study. Below a certain amount of income, it is impossible to be not worried about random life events. Every blip can be a tragedy. It doesn't say that beyond that income level, happiness doesn't increase. It is not unreasonable to say that beyond a point, happiness isn't likely to increase proportionally but I don't know that we have seen studies that explicitly show that.

Linux Foundation 2021 annual report

Posted Dec 9, 2021 13:12 UTC (Thu) by Wol (subscriber, #4433) [Link]

While it may vary for a lot of people (things like rent and mortgage levels have a big impact), this can clearly be identified as a line below which you do not have the ability to ride out those little things which knock you.

More to the point, its a line where you are wasting money buying things in small quantities, because you cannot afford to take advantage of those little discounts if you buy in bulk. If you can't afford a week's toilet rolls, you can't afford to save money by buying a month's supply for only twice the price ...

Breaking through that ceiling can be HARD...

Cheers,
Wol

Linux Foundation 2021 annual report

Posted Dec 8, 2021 17:01 UTC (Wed) by khim (subscriber, #9252) [Link] (10 responses)

> So here you go with an example that disproves at least the general applicability of your hypothesis.

Perhaps I misunderstood you example? Or you misunderstood what I was trying to say?

Because to me your story sounds like the best way to justify what I'm saying.

> What those people earn is "more than enough". There have been studies that showed that increases in wage matter for your well-being up until about 100K and after that no improvement of well-being could be shown.

How any of that relevant? Are you US citizen or have any US friends? $100K per year wouldn't even be enough to rent decent apartment in California. Not counting any other expenses. Just a decent place to live. Consequentially if you visit the linked in you would see that average pay for the senior software engineer would be $182212+$19282+$107383+$10300 = $319177. Once again: that's average. Not someone like Greg or Linus.

Consequentially few of these guys are getting money solely from Linux Foundation, they have other source of income (mamy? most? haven't investigated, actually). Greg KH, in particular, works for Google and I doubt he does that for free.

As your own story shows this makes him more susceptible to the *corruption* from Google side, isn't it?.

That, in turn, means, that if we want to stop that all of them should be moved to the Linus status who receives money solely from Linux Foundation and from noone else.

Linux Foundation 2021 annual report

Posted Dec 9, 2021 18:53 UTC (Thu) by anton (subscriber, #25547) [Link] (9 responses)

Are you US citizen or have any US friends? $100K per year wouldn't even be enough to rent decent apartment in California. Not counting any other expenses. Just a decent place to live.

Looking for the average American income, I find that the median annual wage in the USA is $34,248.45 (i.e., half the people in the USA earn less), and the average is $51,916.27. Concerning California, the average (not median) salary is shown as $62,586.

Linux Foundation 2021 annual report

Posted Dec 9, 2021 19:00 UTC (Thu) by khim (subscriber, #9252) [Link] (8 responses)

Just try to look on the price of apartments in San-Fancisco.

Yes, the fact that prices have grown to the point where large percent of people couldn't afford to live in San-Fancisco is a problem.

No, it's not the good enough reason to ask Linux Foundation members to live in cars.

Linux Foundation 2021 annual report

Posted Dec 9, 2021 19:19 UTC (Thu) by Wol (subscriber, #4433) [Link] (7 responses)

Look at London for the same problem here in England. My house on the edge of London (a SMALL house) is worth over TEN times my salary. And I'm earning maybe 150% of minimum wage (a typical wage for a Uni graduate). How on earth is your typical person struggling to earn enough to live going to afford a place to live?

I'm lucky, I bought my place ages ago, I paid a LOT less for it back then ...

Cheers,
Wol

Linux Foundation 2021 annual report

Posted Dec 14, 2021 14:27 UTC (Tue) by rbtree (guest, #129790) [Link] (6 responses)

Eh, this is pretty typical for many parts of the former USSR (where I'm from). A decent (not lavish by any means) two-room apartment in most towns/cities would cost you around 10-15 years of what you'd make in that city working an average white collar job.

Living in Moscow and making $2k a month? Here's your crappy single room apartment for $170k, thank you very much. Good luck with saving every penny and paying for it in 8-10 years.

A relative of mine started living in Moscow when the was less than 30 years of age. She just recently paid out the mortgage for her apartment on the edge of the city. She's 50.

Living in the sticks (like I do; not Russia BTW) and making $600 a month? Get ready to shell out $70k for a similar apartment and live on macaroni and bread for the next 10-15 years.

(It's a very rough estimate, I did not spend more than a few minutes checking realty prices; but you get the idea.)

So it seems to me, a large chunk of the world lives like you do in London, sans all the niceties and opportunities it provides.

Linux Foundation 2021 annual report

Posted Dec 14, 2021 16:06 UTC (Tue) by Wol (subscriber, #4433) [Link] (5 responses)

I'm from the generation that didn't have easy loans. So when you bought a house, couldn't borrow more than 3x salary, and interest rates were 12-15%, getting a decent house wasn't too hard. Sure, you lived on bread and cheese for a year or two, but 10% inflation rapidly shrank your mortgage, your repayments stayed the same, and your pay went up.

(And the generation before me had inflation at 20-25% - if you managed to get a house in the 50's or 60's that was your mortgage paid inside a few years!!!)

Now prices are much higher, inflation is much lower, money is hard to borrow (what! cheap money is everywhere! Yes, for people who don't need it!), and today's youngsters just can't get that starter home. The rich are borrowing to buy up all the housing stock for silly money, and then charging "what the market will bear" so youngsters can't save to get on the ladder.

If you're lucky, you have a job where demand for housing is low, so it's cheap. If you're a youngster in a thriving town, sorry, rents are sky-high ... (My daughter was lucky - her 4-bed detached cost about 2/3 my little pad ...)

Cheers,
Wol

Linux Foundation 2021 annual report

Posted Dec 14, 2021 16:58 UTC (Tue) by farnz (subscriber, #17727) [Link] (4 responses)

Just for context on that "hard to borrow" line; the trap is that mortgage lenders can't take rents into account when looking at affordability. So if I'm paying £3,000/month in rent, but could buy with an £1,800/month mortgage, I can get caught by the stress-test in affordability that says "what if interest rates climb, and you have to pay £2,900/month in mortgage?". On my salary, £2,900/month may be more than the lender is willing to accept I can pay - and yet I'm paying more than that already.

Linux Foundation 2021 annual report

Posted Dec 14, 2021 20:36 UTC (Tue) by Wol (subscriber, #4433) [Link]

And of course, the sting in the tail is that that rule was introduced to stop lenders being reckless ... except that like most of these rules the regulator never thought it through ...

Cheers,
Wol

Linux Foundation 2021 annual report

Posted Dec 14, 2021 20:58 UTC (Tue) by nix (subscriber, #2304) [Link] (2 responses)

The other fun bit was that (until recently) they couldn't even take your *existing mortgage payments* into account when you were remortgaging. So you might be trying to remortgage from a mortgage on which you're paying £1500/month to one on which you're paying £1000/month and find all the mortgage providers (including the one you're currently using) saying "no, can't give it to you, it's unaffordable" without being allowed to consider that the mortgage you are *now* on is even *worse*.

Linux Foundation 2021 annual report

Posted Dec 14, 2021 21:01 UTC (Tue) by farnz (subscriber, #17727) [Link] (1 responses)

For bonus fun, the stress test does not apply when you're remortgaging with your existing provider and the monthly payment is the same or lower than it was before for the life of the mortgage deal. So last time I remortgaged, I could reduce the payment considerably via my current lender, but I would not have met affordability criteria for another lender.

As it happens, my current lender had the cheapest option for me, but had they not been as good, I'd have failed the stress test on another mortgage provider.

Linux Foundation 2021 annual report

Posted Dec 14, 2021 22:20 UTC (Tue) by Wol (subscriber, #4433) [Link]

And that has bankrupted a fair few people - ex Northern Rock mostly - who couldn't remortgage with their existing provider because they didn't offer new mortgages, and couldn't remortgage with any one else because they failed the stress test.

And those poor people couldn't afford the Standard Variable Rate but could have easily afforded a discount remortgage ...

Cheers,
Wol

Linux Foundation 2021 annual report

Posted Dec 9, 2021 13:57 UTC (Thu) by ballombe (subscriber, #9523) [Link]

FWIW, HP stopped sponsoring Debian LWN subscription several years ago...