_lex

> it's possible that she's the only woman on the panel because she thinks this way

Note I said "it's possible". I did that because I can't prove it. I don't think anybody can prove either possibility.

>why do you assume anything is racist in the first place instead of thinking it is just normal human behaviour?

I'm not originally from the US, though I live here now. While you may look at me and see a black man, I grew up in a country where everyone was black, and therefore nobody was. As a result, I'm probably the least indoctrinated into any movement or society based on identity. The idea that I might identify with racial issues faced by American blacks was one that was forced onto me by reality.

Imagine you're a black person who's never seen racism before. You'd only start assuming that the problems you face are related to race when you are repeatedly treated worse than your cohort, and when you're treated so badly that there is no plausible narrative that you can construct that will lead to another explanation. For example: Employers are shocked (and uncomfortable) when a black face shows up with your resume. Your job offers are lower and less frequent, although you're at the top of your field. People find issues with you and your work that they just can't explain, but accept and praise your work when submitted under other names. Senior leaders think you're a janitor when you lead a department. Patients refuse to be seen by you, even though you are indisputably the best in your speciality. I can go on (note these events have not all happened to me, but each has happened to someone in my inner circle). After a while, even someone who didn't know that racism existed would discover it. You're lucky you don't have to.

Movements like feminism and BLM must sometimes be wrong about specific events - there's no way to really prove what's in someone's heart. And sometimes bad things happen to women and black people - things that would have happened independent of gender and race. But when there's a distinct, recurring pattern, it's really hard to pretend like nothing's wrong.

Separately, you should know that racism and sexism are actually normal (meaning default) human behavior. People form groups based on similarities and exclude those who are dissimilar. These groups may include employment, or leadership at companies. Thats why this is so hard.

I think it's worth considering that we're seeing some survivorship bias here - it's possible that she's the only woman on the panel because she thinks this way. In other words, it's possible that other women who would have seen this event as being rooted in sexism may eventually be a part of so many similar events during their career that they effectively burn out and switch to more female friendly fields.

This has been my experience working alongside women in software engineering - the ones who pay attention to how they might be treated differently leave.

It's also one of the hardest things about being black in tech - you must ignore many possible slights and assume that nobody is purposefully racist. You must construct alternative narratives to explain things that you see day in and day out, otherwise it's easy to get swallowed by doubt and lose your ability to contribute effectively. I call it assuming positive intent. It's how you survive.

I particularly like the discussion of LOIs in https://stripe.com/atlas/guide#transactions-and-agreements. Should be very useful to new companies, and is rarely discussed.

This should have been a series of posts, compiled into a guide. It should be easy to cut this up into digestible pieces, but it's sort of a massive throw up right now.

(edit) Ah - now i see- it is a guide. But it's a single page layout, with navigation on the right. So I guess it just violated a few of my assumptions. I think I also assumed it was a blog post since the h1 title doesn't call it a guide (though the html title does).

It sounds like there's a space for a company that simply validates these issues and supports opensource software, for customers like DOD. I'd expect that such a company could charge each customer quite a bit, and that each customer will want pretty much the same verification of the same libraries, with additional work only needed as new stuff gets requested. Thoughts?

This analysis is very low quality. Engineers move to the bay area not for salaries, but for equity and RSUs. It's not unusual to make your salary again in RSUs, which are usually not that risky. Equity is a mixed ball, however. But these two factors are the unique element in the Bay area, and why you should move there. If you are an engineer in the bay area without loads of equity or RSUs, you're getting ripped off.

Blacks are disproportionally policed, leading to higher likelyhoods of being accused of a crime. Blacks are convicted more often than white people arrested for the same crime. When black people are convicted of a crime, they are more likely to be sentenced to incarceration compared to whites convicted of the same crime. Blacks generally get harsher sentences when compared to whites who've commited the same crime. https://www.americanprogress.org/issues/race/news/2012/03/13... http://www.huffingtonpost.com/kim-farbota/black-crime-rates-...

This is pretty bad article because it focuses on # of funded startups, not # of attempted startups. What if 99.999% of startups attempted were attempted by solo founders? It would make you expect that the # of funded startups should also tilt towards solo founders - if all other odds were even, you'd expect 99.999% of funded startups to be solo founders, and it would not mean much.

OAuth is, like much of authentication and authorization that has been well marketed, very technically flawed. It makes a big show of not trusting the recipient application with credentials, but anyone who's actually interested in stealing creds still can.

Plus it's way more complicated and has way more failure scenarios than simple password auth.

The only saving grace that I saw was that a service no longer has to store users' passwords to other systems, for persistent interaction with their data. I think this is really why people bother using it.

Yeah, this is exactly what my dev teams need right now. I reached out to you just now with my @agco email address on your "schedule a demo" button. Some advice - MAU is not connected to value delivered.

[edit] How much is premium support?

Well, that is the end of Theranos. They should just return what money is left to their investors at this point. This explains why they could never get the tech right...

Yeah, something that most people don't know is that extrinsic motivation dissolves intrinsic motivation.

see http://www.spring.org.uk/2009/10/how-rewards-can-backfire-an...

This means that fitbit will leave you as either a fitbit addict (if you stick with it), or as someone who's less healthy than you would have been otherwise (if you don't, because now you'll exercise less on your own).

I'd hope he's passionate about his business. I think you may have thought that I meant passion project as in "small, unimportant thing he's passionate about". I meant "thing he's passionate about". I think YC is probably sama's passion project.

Seems like a particularly poor article. The title is an unsubstantiated claim based on the author's passion project, rather than any sort of real analysis. Ok, so banks do loss-leaders and cross sell. SO does your supermarket, payroll provider, and many, many other businesses. Yes, that means that the things they cross sell will be more expensive because they have to subsidize other parts of the business - but without that entry point to the customer, they'd be paying google and all sort of other marketing companies to deliver customers for each segment of the business - and they'd have to overcharge to fund that marketing as well.

Or if the argument is that he plans to charge fairly for each bit of the ecosystem, he's about to learn some hard lessons in strategy.

To the author: I hope everything works out for you - but do some research first; you're underestimating your competition.

When companies complain of shortages, they generally mean "shortage at the price I would like to pay". The talent is there, they just need to be willing to pay the _true_ market clearing price, and to advertise as such. Having trouble hiring an iOS lead? Advertise the job with a $200k salary, and you'll see many more applicants.

Granted, there's going to be related costs of sifting through many unqualified applicants; but that's part of the cost of doing business.

Finally, I explained above why businesses don't want to train employees. It really is that simple. Th e math is basic and as inescapable as gravity. (note that planes do fly, aka, you can defy it by devising a clever contraption).

It only seems that way because as a company you're amortizing the investment in education over 2 years, when as an individual the employee can amortize that investment over the effective life of the technology or how long he thinks the knowledge will benefit him, which should be way, way longer (since he can switch companies and sell himself as a battle-tested expert). So the $5000 premium does include a payoff for investing in education, but the payoff is spread out among many more years.

This is a fundamental economic problem in training employees in transferrable skills. It's the reason why companies rarely pay for employees' graduate degrees (even though it can be extremely tax efficient for them to do so), and in the case that they do, they include term requirements (e.g. you have to work for us for 4 years afterwards, otherwise repay a prorated portion).

There are strong economic disincentives to training employees in transferable skills: the second you do, their market value increases, which means that if you don't pay them more, they will be recruited by other companies willing to do so. So you wind up paying for both their education and an increased salary because you educated them. It's cheaper to just hire someone who's already educated.

For example, consider the employer who decides to pay 10,000 to train their employees when doing so will cause their employees' market salary to increase by 5000/year. Assuming the employee will stay with him for 2 years if he does the salary bump, he's effectively paying the employee $5000 (salary bump) + 10,000/2 (training) = $10,000/year extra, when he could just go to the market and hire someone who already has the skill for $5,000/year extra.

Going to the market may also be faster if the skill is both complex & modularized (so it doesn't require deep insight into the current code base). You'll also benefit from the experience of the person who's done it before and gained battle scars.

So instead of promoting a jr engineer and training them, it's often just cheaper and more efficient to hire a senior guy who's done it before.

Two interesting things: 1) this predicts difficulty in controlling contractors and getting them to stick to schedules. It explains why taxis suck, homejoy struggled, and showcases that uber circumvented the entire issue of driver reliability by only doing on-demand rides. It also predicts that scheduled contractors will attempt to steal clients from the scheduling company (direct measurable profit impacting skill - you can't own them). Perhaps the on-demand service strategy that works requires either the use of employees or zero-scheduling/real-time request resolution.

2) As an employee, you must figure out how to measure your profit impact and consistently represent that metric as fact - socialize the idea that you have profit impact. You also want to figure out exactly how what you do and how you do it can increase or decrease profits: being the person who's able to both represent a meaningful metric and show how you've affected that metric, and how that metric should have affected the bottom line will make you the most valuable. If what you do does not impact profitability, then do something else.

This is incomplete, but very, very good economic analysis. In the end his points are simple:

0. Employees are willing to trade "freedom" for job stability and increased average returns. Contracters will more easily defect for sporadic higher paying opportunities, which makes them less reliable.

1. Employers can overpay employees while ensuring that they know that they are overpaid, to 'domesticate' their employees and improve reliability of outcomes.

2. Decades ago, employers could also promise long-term job security to achieve similar effects, however rapidly changing markets mean their future existence (and ability to grant stability) is in question.

3. Instead, employees have fixated on cross-company careers, optimizing for general employability , which means that they must still conform to the requirements of a domesticated employeeship.

4. However, when employees have significant, direct, measurable profit impact skill, and can make or break the company, they can not be owned.

He also does an economic analysis of cursing, while explaining that maximizing employability tends to strip away the parts of people that can make them great.

One of the most basic rules of doing extraordinarily well in business - Commoditize your complements. Google's hope here is to drive the cost of android developers down, while driving the number of android devs up.

He may not keep it, but he can control it, and it will make him significantly more powerful.

Exactly true, however he's donating STOCK not money. So he's avoiding the taxation on the stock gains that he would have seen, which could easily have been upwards of 50%. Then, yes, he can be employed by and can expense costs to his foundation - so that money that he's not being taxed on is still within his control.

However, the bigger deal is that he can use that foundation to wield significant power and to distort public policy to his benefit.

Depending on how he does this and the basis of the equity, this move may be an amazing tax shield, worth WAY more than it seems.

{edit} - I actually think this is exactly the strategy being applied by him and his team of tax strategists - He's using this "donation" as a tax shield and keeping the money in a foundation, which he can use to help "the underprivileged" as he pleases, instead of just paying an amazing amount of tax like Zuck did.

I gotta say, this is a great move for founders + companies: Commoditizing your components will lead to downward price pressure on engineering wages. Also, it's superficially great for employees too: apply to a common app and get prequalified for N companies rather than 1. But I doubt any of those companies will ACTUALLY change their hiring funnels, which means that this is just an extra, gratuitous step in the process. Perhaps there will be the benefit of being introduced to many hiring funnels at once, but that could have been achieved by an "are you hiring" email, or by checking out those companies' hiring websites.

This shows me how myopic and out of touch these guys are - I already use the biggest loser on xbox kinect, and it's replaced both a gym and a trainer for me.

This should not be an app - it should be a service, billed at $12/year (or $10 for a 3 month pass), with fresh content/languages added regularly. This would turn your company into something potentially investable, and would give you a steadily increasing stream of income. You'd also be able to increase the price over time as you add more and more value to your offering.

This should also be a web app, as many people have pointed out to you. As awesome as mobile is, web is still easier to share and convert users on, and while everyone has a web machine, not everyone has a tablet.

I might be drinking crazy juice, but I see everything EXCEPT a business model. How does this company make money? How do you attract customers? How do you reach scale? What you've written down is a bunch of ways to spend money - not a business. Most businesses do spend money, and some may even benefit from some of the ideas presented, but please don't call it a business model.

I think you need to define your own language if it's better. 'Deck' is clearly better than 'Dockerfile'.

Feel free to innovate - you're a startup, and it's what we love about you.

CEOs are employees, and stocks are considered passive income. Granted, he might still have to be on the "board", but he could get pretty close to passive income.

To spell this out: Most successful businesses eventually hire someone to run them w/o much intervention from the investors/founders. He can do the same.

Sounds like you took the best parts of digital ocean and are trying to push it as a platform with docker baked in. I like. It seems like you're also trying to simplify using docker. I like even more.