That's a valid point, that it could potentially be more expensive than it's worth. In other words, it is possible to spend too much on trying to make your employees happier.

I wasn't trying to argue that the optimal amount to spend on employee turnover is infinite, but I do think most companies dedicate less than the rational/optimal amount of effort on it.

I totally got your point. I am playing Devil's Advocate to those who are recommending going to the other extreme and spending enormous amount of money for employee retention, including a significant increase in per employee salary.

I agree completely with this. I was writing about turnover of high performing employees, which is crazy expensive. Turning over low performers is valuable and important for companies to do.

Exactly. In this article I was focused on regrettable turnover (when you wish the employee didn't leave), as opposed to non-regrettable turnover, which is very healthy.

In my experience, almost every time someone leaves someone regrets it and someone doesn't. The manager who supervised the person will usually say it is for the best. I have worked with great managers who clean house and terrible managers who drive people away, or treat them as scapegoats. In either case it may be identified as healthy turnover.

This is a great point. I've seen companies that are great and companies that are horrible about substantially increasing the compensation of their top performers. It makes a huge difference and should be an easy decision to make if companies take the time to think about it.

I agree with you -- I wrote this with regrettable turnover in mind. Parting ways with a low performer is definitely a valuable thing to do!

I see now - we're talking only regrettable employee losses. Thanks.

Great point, we're definitely aware of one more tool issue, so we have to make sure we're sufficiently valuable and lightweight to use.

It's not meant to be task management, but rather a shared repository for longer term goals with period status updates. Most companies today keep these goals in google docs and they don't get much interaction as a result, so we're hoping to address that.

I definitely see the use for it and if you're able to solve it with Lattice, that'd be something. It's been a pain point of mine in prior companies where nobody knew higher-level goals. Good luck with it!

I love that these guys think of 4 years as short term and 20 years as long term. Young companies often set up this dichotomy as a couple months vs. a couple years, which just isn't enough. Long term thinking is a big potential advantage startups have over public incumbents that have quarterly expectations, but it's usually not taken advantage of. Not just on founders obviously, whole startup ecosystem fuels this.

>> "Long term thinking is a big potential advantage startups have over public incumbents that have quarterly expectations"

I completely disagree. If a public company has a poor quarter their stock price might fall a bit and shareholders might get pissed off. If a startup has a bad quarter that could be enough to put it out of business. Startups can't ignore the long-term but naturally all new businesses live day-to-day, especially when relying on venture capital and not a reliable revenue stream.

I thought about this more and I think you're right. Some companies do become very short term focused when they go public, but when you not longer have to worry about short term survival you can plan much farther out in the future.

>Long term thinking is a big potential advantage startups have

This seems unlikely, given the financial insecurity associated with startups.

One example of long term thinking would be military procurement and similar. It takes a decade at least to develop a new weapons platform, which is then deployed for several more decades. One example that comes to mind, and that might appeal to HN readers, is the fact that the technology in the space shuttle was mostly developed in the 1970's and only decommissioned circa 2011.

> "I love that these guys think of 4 years as short term and 20 years as long term. Young companies often set up this dichotomy as a couple months vs. a couple years, which just isn't enough."

You can think 20 years ahead, you can have your guesses, your vision, maybe also a cheesy vaporware "concept video" to prop up your PR.

None of this speaks about how your company truly operates. Because a modern tech company can't really execute 20 years ahead. 20 years is an eternity in technology. In 20 years the world will be nothing like what you thought it would be, and everything you thought would be cool to build up to would already be obsolete.

I've been thinking about Google X's "moonshot" efforts, can those efforts succeed? Their self-driving cars. Not impossible, but they move too slow. They're thinking like a giant corporation now, where getting up in the morning and putting their pants on takes 4 years.

Big carmakers have been playing with electric cars for decades. They were thinking "long-term". But it took Tesla to wake them up, and show them that their long-term vision is possible now.

And that's how it's with Google as well. Far more likely a hungry, angry, scrappy startup will eat their lunch. Someone who's not happy to imagine what it'd be like 20 years forward.

When you don't have a proven business model and a few billion dollars in the bank, you don't have time to think 20 years ahead. You want to have the future now, and you'll work your heart out to deliver it.

Mostly true for startups, and nearly all other businesses.

At the same time, knowing that the type of speech recognition they wanted to build wouldn't be possible for a long time, the Bakers thought, and then executed in decades. Contract work, government work, whatever it took to build a bridge to the future they envisioned.

And I can't see any definite reason to count Google because they are big. Being big didn't stop them from grabbing the majority of the smartphone market, or from teeing up so that most of the next billion people who will have a smartphone as their computer will have an android phone. One can make the case that they sometimes move too quickly--is there any reason why Glass couldn't have come out after Android Wear?

Microsoft sold no computers themselves while Windows held 95% of the personal computer market.

Android is under Google's control. They purchased it, evolved it, put the resources into making it into the monopoly mobile OS that it is today with over 80% of the global market. It was Google's backing and strategy that made Android's market share dominance possible.

So yes, Google has grabbed the majority of smartphone share. Everyone understands who controls Android, even mighty Samsung has absolutely no confusion over this point.

If a market is to be anything, it must be people buying and selling stuff. Unlike MS and Windows, nobody buys Android, and Google certainly doesn't sell it.

Google built Android but does not control it. What Google controls is:

1. The Android marks, like the green robot

2. The closed-source Google services: Play, Maps, etc.

Those are Google's levers. Any smartphone vendor who wants to use the above must agree to other terms, like making Google the default search engine.

In Russia or China, Android phones without Google services are very common. Google can hardly be said to control those markets.

Only half of the 'android' phones use GMS which is licensed from Google. The others are based on AOSP, and Google has no control over those.

Since Android as a whole has around 80% market share globally, Google controls at best 40%'of it.

Samsung is a GMS licensee, and is certainly beholden to Google for their version of Android, but this in no way supports the claim that Google has grabbed the majority market share. They simply have not.

It serves Google for people to think they have more control over the market than they actually do, but it's bad for everyone else.

Also, the comparison of Android to Windows breaks down when you look at smartphone profit share. Unlike Windows, OEMs don't pay to license AOSP or GMS. While Google monetizes GMS through mobile advertising in their apps, the majority of the profits in the industry are from hardware sales/carrier subsidies (unlike the PC business which is mostly a commodity business). That is to say that most of the profits on "Android" are captured by companies like Samsung, HTC, and Amazon.

Is Android not to smartphones as Windows is to PCs? Your point might be an interesting one, but you haven't explained it sufficiently.

Non-CS people doing CS will give a worse experience in a single specific case, but will improve the overall customer experience in the long run.

I'm not arguing that companies should get rid of departments, just that they should take CS seriously. I think having all members of the team talk to customers is the best way to give the whole company an orientation toward customer service. 37signals, Stripe, and others do this too, not just Amazon.

CIL is a public company in the ran-by-a-government sense, so it's a little different. BTU is the largest private sector coal company.

CIL is a publicly listed company though govt is the biggest shareholder. You are right about the private part.

I think you should be able to compare the value of any two companies, no matter how different their distribution of possible outcomes or how differently they fit into an investor's portfolio.