Aqueous

I come to that conclusion as well. Literally millions of people died of this thing you called ‘minor.’ But those deaths don’t register because they happened out of view, in an overwhelmed hospital or out in the street or at home when capacity ran out. You don’t think about them because you didn’t see them. All you can see is the inconvenience to yourself.

Let’s flip this conversation on its head. why is it that people are surprised that a failure to make a basic decision to protect both oneself and others from a potentially lethal illness is met with sharp criticism and scorn, as it should be?

Please tell me the way in which a virus that swept the globe and killed nearly 6 million people (and those are just the deaths we counted) in 2 years is ‘minor in the scheme of things.’

Um….not sure if you were joking but even compared to Amazon, in no way, shape or form can Walmart be considered a small retailer.

Ah interesting - I vaguely remember hearing about this, but since I hadn't heard about it in so long I assumed it died.

I don't know what the uptake on this product is, but this specific product might be a case of "Solving a problem no business is having." For instance I think it'd be hard to justify purchasing this service for an enterprise without a compelling use-case. But I certainly think that Facebook has the resources make something that solves a problem that people really do have.

On the other hand, Slack seems to solve a problem pretty well that most businesses have.

Finally, I don't think VR as it currently exists solves a problem any business (or person) truly has beyond sheer entertainment. It remains "cool tech" that has yet to find a solid, non-gaming use case for the average person. This may change but this is why I currently think Meta is in trouble and I'm puzzled as to why Zuckerberg would go all-in on VR.

It's a platform for managing a kind of appointment, but it doesn't have an appointments table. The appointments data is combined with a different table, and there was no way to disentangle them easily because the entire system was built around that object model. Case study in failure to normalize.

The hard part is done, but basically we had to switch the engine while the car was running so to speak. In order to switch this out you need to start writing the correct data shape, then switch everything over to reading that shape. When people work strict 9 to 5s this will take you forever especially when managing a large amount of volume, which requires you to be extremely risk-averse and slow.

I think the correct thing to do would have been to pivot away from user data and ad targeting as business model and start offering premium, paid services. For instance Facebook could have very easily been Slack for businesses, as well as an enterprise internal social network. Another thing would have been to start offering appliances for private hosting of individual Facebook instances. Finally Facebook could have also started competing in cloud and leveraging its massive infrastructure for other companies to use. I might be wrong but it seems like these could have replaced Facebook’s current revenue model with stable, reliable recurring revenue while also resolving its longstanding PR problems.

As someone who is currently over a year into rewriting a massive system with a fundamental design error by the original designer I can assure you that failing to plan your data model up front can have huge costs not just for you but for anyone who picks up your code in the future, and can hamstring a system so that it is impossible to extend or evolve.

often poor code is a symptom of poor data structure design. having the wrong data structure can make your code heinously complex. conversely designing the data structure correctly can make your code simple and clear, because it’s obvious how to manipulate the structure to your code.

i also see a lot of code where people effectively throw 30 responsibilities into the same class. Really difficult to update. strictly enforcing single responsibility solves a lot of problems.

The vaccine results are clear. You are much, much less likely to have a severe health outcome if you are vaccinated, even with omicron. The vast majority of the people hospitalized with omicron and overwhelming hospitals right now are unvaccinated people. For people vaccinated and boosted, omicron is significantly milder.

Sorry to break it to you, but you are a selfish person who won't even do the bare minimum to prevent spreading the illness to other at-risk people.

Yes. My biggest irritation with the show is that literally the entire point of the books, and the reason to read them at all, is the central, intriguing idea of psychohistory and the Seldon plan. These ideas are front and center in the books and literally inform every scene but in the show they are almost an afterthought. The result is that the show feels nothing like the books, at all. Frustrating. Someone will have to remake this show again and hopefully do better at capturing the ideas and feeling of the books.

I would say it indicates junior tech skills as well. If the value of the linter isn’t self evident to him he is a junior.

Alternate title: I recklessly ignored health guidelines around a pandemic that has killed 5.3 million people in 2 years.

ever tried to scale monolithic development beyond 3 teams? the point of micro services is simple: to make technical infrastructure resemble the human organization it is designed to support. that’s it.

Can someone remind me what the purpose of conventional finance is again?

no, no, no, no.

if you deliver the product only by taking shortcuts that impede other engineers you do not get credit for ‘delivering’ the product.

having been the next engineer to come along, this attitude drives me up a wall.

There is a time and a place for a denormalized schema. If you are trying to search a large data-set a denormalized schema in a single table with some good indexing is much preferable to joining across 30 tables.

Whether you choose normalized or denormalized depends very much on a) what you need to do now b) what you may need to do in the future. Both are considerations.

I thought this was partially explained because the AI aliens were advanced but extremely limited in number i.e. there were only a few hundred left. so they need humans and human resources to make up for their limitations in numbers.

‘Managing Data Is a Nightmare Keeping data in-sync across microservices can be challenging.’ Leaving aside the debate over whether micro services vs monolith is appropriate for your org, this shows me you are thinking about the problem wrong. You shouldn’t have to keep data ‘in sync’ because your domains should be defined in such a way that it doesn’t have to be. In a loosely coupled, event driven system each domain only needs to know information relevant to the business functionality it manages - and in event sourced model even within the domain you have eventual consistency.

In a micro services environment the concept of the ‘global system” state and how the system behaves end-to-end shouldn’t even be something it’s that useful to have to think about and you should start thinking in more local terms about the business functionality you care about.

Congratulations. How long ago did you start looking? Did it take all 18 months to find a new position or did you find some as soon as you started looking?

Though this can also backfire. I observed a huge architecture error in the central data structure / schema of our system. Everyone agreed that it was a huge problem so I got put in charge of fixing it. Of course when you build an entire system around a bad schema, it gets baked through every inch of that system. Over a year later and it’s mostly fixed, but my career has stalled and I took the major hit to fix it.

Like on every other major OS the publisher of those applications. And I mean they should be distributing the applications as well. Not the OS publishers.

No, I’m suggesting that the authors of the applications are responsible for patching and distributing their own application like on every other major OS.

But should distributing and ensuring the security of software that users install on their own systems even be the responsibility of operating system maintainers?

Have to say this because I've seen this attitude become pathological: You don't have to be "killing it" all the time. There's no rush. Time isn't running out.

Finish school. Slow down. Think about what you want to do.

$21 / user / month - so if you have 100 engineers that's $2100 a month or $25k a year.

Still should be doable for most businesses that size but licensing costs can blow up when you start to have a lot of seats. An annoying thing about the company I work for is that they have a limited number of licenses for things like IDEs, so they ration them. And so I'll boot up an IDE for a language I work in less - like say, PyCharm - and it will stop working because my license got taken away and given to someone else. I'll have to request another one be given back to get working again, which is pretty annoying when I'm trying to get something done. I work mostly with Docker / Kubernetes so if I'm in a situation where my core tools are being constantly taken away, I'll be pretty miffed.

I agree that Docker has every right to charge big companies for this software. Just wanted to point out that the costs can be more than you'd expect.

The excuses Coinbase uses are ludicrous. As far as I can tell, deposits on Coinbase are not directly on-chain. And transfers out aren’t blockchain-only transactions either - they are mediated by Coinbase. So the wallet you see on Coinbase is a not a blockchain wallet - it's a Coinbase wallet, effectively a bank account.

Blockchain transactions are irreversible, but that has nothing to do with why they won't reverse these thefts. They don’t reverse these transactions because it would cost money to insure people’s deposits. Not because blockchain transactions are irreversible.

I'd be curious to hear a lawyer's take on this - whether something that is effectively a bank can get away with not being called a bank in order to avoid insuring deposits.

I will say, in this case it is not what happened. The previous system launched successfully with an incorrect schema. The platform grew and when it came time to add more features, some of them were either impossible to do because of the incorrect schema, or they would have been so complex to implement that the system would have fallen apart. Needless to say, the schema needed to be corrected, but the wrong schema was already baked in all across the platform.

So it's not like the main developer handed something off that was unfinished. They handed something off that was finished but wrong. And because of that we had to replace the engine while in flight, so to speak.

Precisely. If I were to have neglected to point out the issue, or refused to lead the fix, I would probably have been promoted by now. Instead I chose the hard path and got punished for it. However, the team will benefit far into the future because they will no longer be fighting a schema that was just wrong. The only cost they had to pay was my career velocity.

Haha, yep. I just think it's fundamentally unjust for the developers doing the (easiest) green field development to get promoted while making significant architecture mistakes, then punishing later developers for pointing out those mistakes and getting stuck with fixing them. Until someone makes noise, this is just going to keep happening.