101011

Nice writeup! It'd be helpful to mention that SNS allows you to setup a retry policy and also allows for you to configure a DLQ when retries are exhausted.

I think KaiOS is probably the closest thing that you can find. It will also have a browser, but trust me when I say that on most phones it's so slow and unusable that you won't be tempted to use it often.

I had a similar issue for myself, only my strongly desired application was Spotify (instead of WhatsApp). Ultimately I weaned myself off Spotify through a phone with KaiOS and ultimately I took the plunge on buying the Light Phone II.

It's a real bummer that there are so many limited solutions for things like this. FWIW, the contracted software company that worked on the Light Phone OS put together a blog post detailing how they used Android to build their phone here:

https://medium.com/sanctuary-computer-inc/building-lightos-w...

> All of that is to say: when we refer to the LightOS, we are referring to “our custom fork of Android 8.1 that embeds a platform-signed React Native app as the default launcher” (amongst other drivers and low-level customizations).

Is Spark what you're looking for? You can do all sorts of joins, groupings, and aggregations with parquet(s) acting as your source(s)

To be fair, I think their algorithm is way better once it gets a bead on your interests. Now, that definitely results in some shows never pop up in your 'feed,' but overall I'm in 2 modes of watching:

1. Scrolling into something that suits my interests 2. Navigating straight to something I want

If I'm in mode 1, then the feed works pretty well. If I'm in mode 2, then I can search straight away for the thing I'm looking for (and usually that search starts at the Google layer so I can be sure I'm going into the correct app in the first place!)

Which is funny because you always hear how fast crypto transactions are. What everybody seems to leave out is that fully settling on the blockchain is slow and costly.

Any surveys of India are going to amount to a small sample size across the entire population - India is massive.

Also, they're almost certainly undercounting the extremely poor, a very sizable group that is going to fall through the cracks no matter what standards you're trying to abide by.

There's a reason the title says "may" - there's an acknowledgement of room for error here.

The point is, India, in spite of their lack of adequate healthcare, are showing mortality rates well below other countries. I applaud the attempt to discern what the toll may truly be, flawed as it may be.

I'm sure if it was as simple as looking at public health data then they wouldn't have spent the money calling people in a survey.

I'm fine with people calling out error margins for research, but do you really have to say this is lousy clickbaity work?

I respectfully disagree. I have found great value out of database tests that did not require the overhead of end to end tests, which I've found can be brittle and slow - especially if your service depends in some way on another application or outside piece of infrastructure.

As an aside, anytime I hear "not ever" from somebody wrt design patterns, my doubt sensors jump to alert.

To be fair, orthogonal just means something that’s at a right angle to another line. I’ve heard people refer to something that is opposite, or an antonym to parallel (or in sync) as orthogonal.

Spoilers dude! I haven't started yet.

I wonder how much money on market research went into that.

It's almost like there's no clear answer here.

Just curious: do you notice a difference in the population over time, or is it one of those things that mostly just works while it's running?

I'm not convinced it is a bad thing. Good scale, relatively low latency, keeps your infrastructure costs down, and many of the other alerting options (looking at you Datadog and NewRelic) are crazy expensive.

Open to hearing other opinions though.

Gee, I'll bet you were a peach to teach.

Let's just fire all the teachers and see if the next batch of applicants do any better.

I'm a software engineer and I'll happily admit that I'm grossly overpaid for the contribution to society I provide.

I'm curious what you think teachers should be paid for trying to educate our future.

Oink...talk about the glory days. If it existed, it was there.

Worth calling out here that SemVer has been around for a decade now, so "recent" is relative

Don't do mobile?

Have you read the Body Keeps the Score? It felt pop in the sense that it was on the leading edge, but I felt the author made a best effort to put forth the available science

Don't most IDEs handle this already?

Interesting, according to this blog post: https://blog.youtube/news-and-events/better-protecting-kids-...

> YouTube now treats personal information from anyone watching children’s content on the platform as coming from a child, regardless of the age of the user. This means that on videos made for kids, we limit data collection and use, and as a result, we need to restrict or disable some product features. For example, we no longer serve personalized ads on this content or support features such as comments, live chat, notification bell, stories, save to playlist, and others.

So it's to keep kid behavior from being tracked. Kinda cool.

While I agree that making something illegal won't stop it from happening - in this case, banning a currency drastically curtails the functionality of it. That reduces the value of it, which in turn would reduce demand and supply.

You are very opinionated for what ultimately is a piece from a fiction writer.

Of course he's going to play on words and conflate emotions when waxing about the meaning of life.

I don't understand why so many have to be such a harsh critic. Yeesh.

More like, they're just wanting to make sure they're legally covered. There is no added benefit for the company in making that experience better.

Source: working on a product engineering team for a larger company. Why would we upheave our roadmap for something that doesn't help provide explicit value to our product offerings? We've got a list a mile long of improvements and new capabilities to deliver on, and no engineer likes working through legal hoops anyways.

The NYTimes just had a big scoop on Trump's tax returns.

Would it be considered a hack if somebody within the IRS gave those documents to the press? What if the source was a foreign entity working through an intermediary?

That is to say, what distinguishes a source vs a hack is a function of the press' investigation into the authenticity of the materials.

I'm not sure blocking the NYPost is the right call here, much as I hate to say it.

Very interesting - can you link your source? I wasn't able to find that quote myself, and wanted to read more.

Which points in particular are you disagreeing with from the article?

I've probably tried a dozen or so.

For MySQL, I haven't found anything that beats SequelPro. For Postgres, I haven't found anything that comes close to parity, but my favorite is Postico.

I know people that swear by IntelliJ for their db stuff, it just never hit home for me personally though.

Calling someone a "gaslighting c*" didn't seem angry to you?

I don't think the XKCD comic is wrong. It's a commentary on 'computerized voting.' Not a commentary on 'computerized voting with a paper trail,' which is what the writer of the article presents as a solution.