picardo

The diameter of an atom is between 0.1 to 0.5 nanometers. Does it mean when the researchers build a 0.1nm scale we can't expect to have any more improvements in chip production?

Sure. Here is a story from 4 years ago. [1] Barclays says blockchain cut this process from 10 days to 4 hours.

[1] https://www.reuters.com/article/us-banks-barclays-blockchain...

It's not in the public eye. The blockchain is only accessible to trusted parties, like banks and merchant representatives.

No, we don't. That's why it's slow and expensive to trade across borders, especially if the counterparties have never done business before. It's an arduous process of proving creditworthiness through letters of credit. You often see document fraud where someone can tamper with a letter of credit, which is hard to detect.

Banking systems are complex. You can't build a single centralized system that works for everyone, let alone get everyone to trust each other on the system. The solution needs to be decentralized, trustworthy and provide built-in checks for integrity.

Permissioned blockchains can be useful in trade finance and international trade. Financial transactions across borders can be delayed to allow for clearance and settlement between several institutions, which requires multiple banking systems to verify the integrity of transaction and collect a lot of paper trail. There are several competing projects solving for that problem right now.

I wouldn't call that creepy, either. You were opted-in to a feature that makes your account more secure and it is better for you to have more security than less security. It's for your own good and protects you against real creepy actors, like hackers and data harvesters. It's not really lining Amazon's pockets to give you more security. They're opting you in by default for your own good. Some people find that kind of nudge paternalistic, so you can easily opt-out from your account page. It's not mandatory.

There is nothing creepy about 2-factor authentication.

You can borrow against it.

There is some truth to that. They are designed to serve a number of personas in an organization, so if you just need to do one thing or another, they will seem overly complex, but I think you could say the same thing about a cockpit or any interface for a complex system.

Snowpack is a bundler, not a build tool. It works with a variety of file formats, and it uses esbuild to compile JS/TS assets.

esbuild can be a good alternative to Webpack in some cases, but Webpack solves a lot of edge cases well. For example, just today I debugged an issue where a library had used `require` in an ES module instead of `import`, and that broke the build with esbuild. So I had to fork the library to fix that issue. Webpack handles that case without problems.

I can think of at least a few cases like that where Webpack works better, but esbuild is (hopefully) going to catch up. It'd be ideal if Webpack and other bundler tools ditched their homegrown build tools and replaced them with esbuild.

Just a throwaway line, written, no doubt, in a fit of high-mindedness.

Yep, that's probably what I'll end up doing.

hey, thanks. That sounds interesting, but this is a volunteer project. I'm donating my time, so I don't think I can justify a service like that right now. The datasets are around 200MB each, just over the 100MB limit of Github.

Thanks for making Datasette! I've been using it to build a project to make US ranked choice election results more accessible. It's very intuitive and easy to use. The main difficulty for me is how to get around the limitation that the sqlite files have to be colocated on the same server. These datasets can get pretty large, and I can't host them on Github, and since I can't put the datasette db on an S3 bucket, I've been exploring mounting AWS Elastic File System to the Docker container. Is there a better way?

It's not mentioned in the article but Amazon had disabled their affiliate link program for Goodreads ahead of this announcement, which cut off a major source of their revenue for them, and forced them to sell. They had no choice.

The strategic reason for Amazon is obvious. As someone else mentioned, Amazon doesn't want Goodreads data to be used to add value to their competitors' offerings.

Speaking as a developer who tried to build on top of Goodreads API, I also want to add that this was a long time coming. The API had been neglected for some time. And some of the most interesting datasets weren't even made available through the API.

This is exactly what I was looking for!

Wow. That's crazy. I can't imagine what they're thinking. The weird thing is I have gone on Linkedin and looked for who they hired after a role I applied for is closed, and it's usually journeyman programmers who don't have any domain experience. My suspicion is they're hiring from their own networks, and the job listings are just a show.

I'm a full stack dev with 16 years of experience. I've applied to various roles at Thorn nearly 30 times over the last 6 months. Never even got a rejection. I realize it's a small team and there are a lot of applicants, but I don't understand why I never hear back from you. Maybe my resumee is getting caught in spam filters or been blocked because I applied so many times?

At any rate, I like your mission and wish you the best!

I do this sometimes, but I don't email the people. It's a personal exercise. It was recommended to me by a therapist. While it does lift up my spirits and fill me gratitude and compassion, it'd make me feel worse if the person on the other side didn't respond for whatever reason... You've been lucky with your reply rate so far.

I like the idea, though. I'll give it a shot at sending them in the future.

I never used Absinthe, but if you're initializing an ORM in your resolver, loading the entire record into memory is unavoidable. How does Absinthe get around that? (Sounds like it generates the SQL?)

I haven't used Apollo Server lately. But the way you describe it doesn't address the core issue, which is the initialization of the intermediate objects in-memory. So just to give an example, if I wanted to query for the projects of listings of my company, I can write it this way in GQL:

   me { company { listings { projects { id name } } } }

This will initialize: a User, a Company, Listings and Projects of all listings.

I can also write this in SQL using a couple of joins and return an array. The memory consumption is trivial in comparison to the original request.

I'm not sure if this is mentioned in the thread, but one of the reason it takes so long for the requests to return is when GQL initializes the entire record in memory and then reduces it back to only the fields you wanted. This can be a big problem if you have a deeply nested data model, and potentially many results. The memory consumption can hit the roof. I find that the best approach in those cases is to create a one-off REST endpoint (or to create a field higher up the GQL hierarchy) and handroll the SQL query.

True, but these requests cannot be made simultaneously. The code is loaded only when the UI needs it.

I think what he is saying is since there is a network delay of 300ms for every request, it's better for him to make a single request for a large file than to have to make 10 requests for 200k files and wait 300ms * 10.

Is there a similar conversion tool for Flow types?

Is there an English translation?

Ah, I get it. Thanks!

I'm stuck on the first question. I'm not sure why this is wrong. Can anyone help me?

> select name, sum(weight) OVER (PARTITION BY name) as running_total_weight from cats order by name

http://www.windowfunctions.com/questions/1