taosx

How did they manage to decrypt the messages, I thought it's very expensive to decrypt a correctly encrypted piece of data?

"About a billion messages were intercepted, about half of which had been decrypted by April 2021—further avenues of inquiry were expected to open as decryption progressed"

Not sure if it's because of starting as a web/application developer and but I wish I was introduced earlier to:

- in-process databases (rocksdb, sqlite)

- FoundationDB

- C/C++ and low level programming in general (I wished I learned those instead of js when I was younger)

- State Machines, Actor Model (Orleans Net), Event Sourcing

- Bittorrent for other things than pirating (it looks like it's dying)

- Arduino and similar

- Seastar

- Arrow (ecosystem)

I'm currently subscribed to 3 vod platforms. What I've noticed is that pirated sites load faster, require no login, quality is either stable or blocks (it doesn't lower itself), subs in many languages and of course are full of ads, but a good adblocker can handle that.

Edit: Just looked around and I think it's more complex than that. I'm usually either on a laptop or a desktop, so I appreciate sites that focus on pc first, have low friction, are simple with no animations, and give me control (I'll tell you if I want to see a trailer or not).

Tried it but honestly I like Iosevka too much. I love to split my code into-multiple side-by-side windows and Iosevka is perfect.

You can consider supporting smaller companies that provide VPS or bare-metal servers. Whenever I explored cloud services for hosting an app for a client, I felt like the costs were excessive, the choices are bit harder if you want to support multiple locations but doable.

Most adblocker lists I have have permanently blocked this domain.

Talking about credentials I'm wondering how much they spend on auth0 with their current MAU.

Exactly, but I still haven't made my mind if it's okay to do this. I dislike being tracked and I take some measures against that, would it be beneficial to more people by doing this?

NYTimes feels like a SSG when browsing (chrome) from Europe after initial payload but that's as an unauthenticated user with ublock. The sad part is that I can't read most of the articles due to the paywall.

Why we care about obscenity in pseudo-random ids and url?

Shift acts like god-mode so you can look up with shift and w and just fly.

I won't comment on the drama but I don't see a second worst thing. I feel like Rust makes the costs clear and you should* not use it in situations where a higher-level garbage collected language can work.

Thank you to everyone involved in this. I'm wondering, is Node.js fiddling with v8 in order to focus more on the resulting code performance than the initial startup.

Planning to learn C/C++ in order to investigate that and if there any gains there.

Thank you!

In cases where you care about performance, yes. Newer webassembly frameworks (rust from what i remember) are batching operations to the DOM for that exact reason.

I don't get what's wrong with his post about reflection, why would someone be uncomfortable with a tool?

https://soasis.org/posts/a-mirror-for-rust-a-plan-for-generi... - Is this the post ?

I'm a bit confused on what exactly helm was, would it be acceptable to compare it (functionally) to having an email and calendar server on a raspberry pi with a private vpn and a hosted ip address?

I use it a lot for small cli apps and during dev, it's so nice to be able to run typescript files with bun instead of tsx or ts-node.

> • or putting the words in the order of whichever language has been dominate at the time.

It's something along the lines of the third case, but there are also cases where a word by word translation is going on when I'm trying to express concepts I've learned in a language into the other.

I'm bilingual and a lot of times if I speak/write fast without post-processing my thoughts I get the order of the words mixed up to the other language. And yes, words are ordered differently from language to language (ex. french / english).

I'm not saying that systems are not in place, I'm just saying that it's too hard for the average consumer.

- After you've requested deletion the company has 30 days to "respond" but they can extend that with two additional months. - They can go to extreme lengths to verify the identity of the user which they have the right to do so and if you don't respond to the confirmation they are not obligated to delete anything.

I've encountered both practices in the past (and I've only made 3 requests, ever).

The dream would have been an automated way to do it, something like a government service where each company would have to publish metadata about captured user data and once you request deletion through the service the company would receive an event, a webhook call...

GDPR doesn't work as described and companies have found many ways to make the process difficult. There is a clause (last case scenario) where the company can say that the data is critical to the system and can't delete it.

I find interesting that most of pages take more than ~400ms to load on a new machine and 1G connection in Europe. Except the one at intercity-vpn.de one. I think after a point it's worth looking into serving from the edge than cutting size.

Some reasons that I can think of:

- Heat generation (harder to cool)

- Trade offs don't justify the pain

Yes, the data type matters a lot and it can be quite challenging, especially when dealing with media files such as videos due to existing compression as media codecs are very efficient.

In your case, with the database dump being a text file, I expect the compression results to be quite impressive. However, it's also crucial to consider the similarity of the data within your database. Although I haven't worked with compression in a while, I've experienced notable success in the past by increasing the dictionary size.

Take a look at for compression algos benchmarks: https://www.mattmahoney.net/dc/text.html

Congratz to LangChain. Wondering what's the dev workflow when doing prompt engineering using langchain/haystack.

- Is there a way to view the final log?

- Why there isn't a simple way how abstractions will look in the final output?

- Can we change the output of those abstractions (tools, agents..etc)?

Unfortunately, I see no plans about documentation whatsoever.

Private healthcare in US seems to be criminally greedy.

Classifying healthcare spending as "anti-poverty" spending may lead to an inaccurate assessment of the effectiveness of these programs. Including healthcare spending in the evaluation of anti-poverty initiatives, it could potentially mask the true impact of more targeted interventions, such as food stamps, housing assistance, and cash transfers. This could, in turn, hinder the development of more effective policies and programs aimed at reducing poverty.

Most subscription services are too expensive for the average user. They're great for B2B but as a user I rarely get enough value out of them to subscribe, with some exceptions.

That's the path I took and I don't really mind using github actions but deployment is just a pain compared to other apps I built in compiled languages. (really excited to try deno in the future where I have a single bundle)

I really enjoy fly.io so far and I got the impression that the primitives they have can take me really far, the only thing I'm missing is better documentation and more examples but I'm sure they will improve as usage grows. One thing that I can't wait to try is the machines api which has the ability to scale to zero.

Currently I have a directus app with sqlite and litestream on it but will probably move away from that to something custom as I don't have a good story regarding migration and ci/cd.