DVassallo

Author here. Docker is the only one in that list I’m on the fence on. However, I feel that the EC2 AMI can be the equivalent of the container image, and Docker would only be adding another layer and another OS to deal with. Sure, the AMI is not as portable as a docker image, but all in all I prefer working directly on the EC2 VM just for the sake of reducing layers.

There are ways to address that: One of them is to provide a downloadable copy of the web app. Another is to offer a browser extension to verify integrity. Here's an example from Cloudflare (https://blog.cloudflare.com/e2e-integrity/).

But even without code integrity checks, client-side encryption of user data should still radically improve user data privacy compared to the status quo of storing it in clear in a database owned by the app developer.

It looks like it can also happen if you simply open an HTML email with Apple Mail: https://twitter.com/funjon/status/1148464952480374784

The jump in 2016 was because I got an offer from Oracle and Amazon matched it. I actually lost a bit from stock depreciation that year because most of my RSUs bested a week after Trump’s election.

For 2017 and 2018, yes stock appreciation is the reason for the growth over 2016.

I’m not sure what you mean, but there is no proof of safety for GMOs. No evidence of harm is not evidence of no harm. The only adequate test is the test of time, and testing on the global population is highly irresponsible.

“Natural selection” is the process that makes something “natural”. If it hasn’t passed the test of time, it’s not natural. Which part of my list doesn’t overlap with yours? You consider Impossible Foods natural?

Eating meat-fed cows is not natural. Eating trans fats is not natural. Eating GM foods is not natural. The first two proved to have catastrophic harm. Why are we confident GM foods are safe?

We’re treating no evidence of harm as evidence of no harm. And we’re doing it on a global scale.

The same was said for meat-fed cows and trans fats. Turned out their consumption had fatal consequences.

You think that the risk of consequential harm (say, death within 10 years) is the same for Brussels sprouts and GM foods? Haven’t we learned anything from mad cow and trans fats?

Edit: The risk is reduced significantly after we’ve eaten something for a few generations. The problem is if we test something new on the global population before a local tribe tries it on themselves for a few generations.

Every GM product is different.

They did it in local tribes. If the experiment proved harmful, the impact was local. In this case, the impact could be global. Big difference.

Yes: meat, fruit, vegetables, nuts... Quite a lot.

It doesn’t have to be 50000 years. For example: I trust that drinking cow milk won’t kill me even though it hasn’t been common for long. Can’t say the same for GM food.

Broccoli have been eaten for over 2000 years. If they killed us, we would have known.

The burden of the proof is on the GMOs. We have plenty of evidence that eating food that we’re not used to eat can cause catastrophic harm. If this was a local tribe experimenting with GMOs, then it’s fine. But if we expose virtually every person to that downside, the risks are considerable.

With “natural” I mean natural for humans to eat them. Those foods can be assumed safe because, by definition, they have been eaten by humans for hundreds of thousands of years. Can you say the same thing for GM food?

You really don’t see the relationship?

The potential consequence of eating unnatural food that did not pass the test of time is catastrophic. Do this on a global scale, and the consequences could be global.

Trans fats are another example.

The potential downside of GM food is human extinction.

In the 80s and 90s the beef industry meddled with nature and started feeding animal meat to cows. The result was a disease that could have killed us all if we had eaten contaminated beef in the previous 10 years. If the genetic conditions that make people susceptible to vCJD were more common or universal, we might not be here today [1].

Grass-fed beef can be assumed safe because it has been eaten by humans for 200,000 years. Animal-fed beef turned out to be very dangerous very quickly. People who think GM food is as safe as food we’ve been eating for hundreds of thousands of years are highly delusional.

[1] https://en.wikipedia.org/wiki/Variant_Creutzfeldt–Jakob_dise...

The incentives are quite soft. You get a “good job” email from your boss and maybe your boss’s boss. But nothing else. (8 years at Amazon, and 5 granted parents — I’m embarrassed to admit!)

They were going to be useful for my immigration visa renewal though. My L1 renewal got stuck with a request for evidence, and the lawyers thought that patents are excellent evidence of outstanding ability to the eyes of the gov. Eventually I got my green card before my L1 evidence got processed, so I didn’t get to see if that would have helped.

I worked at AWS for 8 years. Overall it’s a good place to work, and I’d highly recommend it to junior developers just starting out. I left 4 months ago, but it wasn’t Amazon’s fault. I was treated well, was paid very well, and I liked the work. But I was always going to be working on someone else’s terms there, and I realized that wasn’t the ideal career path for me. I like working for myself better, but I don’t regret my time at Amazon.

——

Edit: Here’s my comp progression at Amazon for those curious:

2010: €50K - Joined Amazon in Dublin, Ireland as an SDE-1 (entry level) in the AWS CloudWatch team.

2011: €75K - Still at Amazon, same team.

2012: $120K - Moved to Seattle with the same team. Promoted to SDE-2.

2013: $150K - Still at Amazon, same team.

2014: $185K - Promoted to SDE-3 (senior level), same team.

2015: $230K - Still at Amazon, same team.

2016: $390K - Still at Amazon, same team.

2017: $470K - Still at Amazon, same team.

2018: $511K - Still at Amazon, same team.

2019: Left Amazon last February. You can read more about why here: https://news.ycombinator.com/item?id=19135399

All figures are gross income as shown in my W2. My last 3 year-end (Dec) paystubs: http://imgur.com/a/EgIVQln

Author here. Just my housing and insurance costs are $86K/year. That’s before food, utilities, etc. My health insurance alone (the cheapest plan from ACA for a family of 4 is $1.5/mo with a $14K deductible).

Is it an option to let employees find their own health insurance through ACA (or privately) and just pay them more to make up for it?

Yes I have a decent amount of savings. About 5 years of runway if my income remains $0. I wrote in detail about my financial preparation here: https://danielvassallo.com/from-employee-to-bootstrapper/

True, but for all intents and purposes, it's basically public domain with some attribution clauses and litigation indemnity:

"Grant of Copyright License. Subject to the terms and conditions of this License, each Contributor hereby grants to You a perpetual, worldwide, non-exclusive, no-charge, royalty-free, irrevocable copyright license to reproduce, prepare Derivative Works of, publicly display, publicly perform, sublicense, and distribute the Work and such Derivative Works in Source or Object form."

https://github.com/apache/kafka/blob/trunk/LICENSE#L67-L72

Permissive licenses are basically public domain with some attribution clauses and litigation indemnity.

Yes I agree. They just need to find a good business model for self-hosting though.

Sure is. I didn't mean a product that Amazon can't do, but other factors that Amazon would likely be unwilling to do or find very hard to do. For example, offering high-touch support, or having a great UI... Or being an open source champion! Lots of things that Amazon is likely incapable of doing in the near future.

lol. True, but not much to differentiate on that unfortunately, unless they're able to point out something fundamentally wrong with the Amazon platform.

I estimate the odds that AWS Aurora will offer "massive scale SQL without manual shading" in the next 3 years to be >95%. Then why would anyone want to to use Cockroach? They need to differentiate on something else. Ideally something that Amazon is unwilling to do, or would find very hard to do.

Stealing an open source project would be like saying stealing quick sort. You can't steal something if it was given to the public domain.

This is not the best way to compete with Amazon. The believe that their feature set is enough of a differentiator is probably delusional. CockroachDB does "massive scale SQL without manual shading". BSL won't hinder Amazon offering those features too.

If CockroachDB remained OSS, there would have been a chance of Amazon offering Elastic CockroachDB eventually, and at least make the Cockroach brand more popular. But now it will likely be an Aurora feature.

If you want to compete with Amazon, do something Amazon cannot do.