hotpathdev

Prior to smart phones, programming on phones was much more inaccessible due to hardware fragmentation. Mobile Apps are popular and APIs are available everywhere. Some deployed apps include implementations of BASIC, Python and other languages which you can use to develop right on the phone. The problem is, typing on a smart phone is terrible, and it's worse for programming. You need a keyboard and by then you might as well use a laptop.

00. pop "conspiracy theory"

10. "that's a conspiracy theory, that's a conspiracy theory"

20. mainstream media begins to ease their audiences into the new narrative

30. "we knew that all along, there's nothing to see here!"

40. goto 00

Now let's look back at the list provided: -Scientists -lawyers -engineers -artists -doctors -nurses

And compare it to programmers and writers. The non-programmer/non-writer world think programmers "just play on the computer" all day. So I think GP is spot on, non-programmers don't get how interrupting flow is catastrophic to what we do.

The most exciting thing I got from the discussion is this:

https://github.com/feizc/Visual-LLaMA

Someone claims to be working on a modified LLaMA model to understand images. He also appears to be working on a large number other projects, so it could be pure vapor.

[1] WebGL finger print is device specific and persistent. [2] Font finger print is device specific and persistent. [3] TLS finger print is device specific and persistent. [4] DNS is routed into USA by default. Incidentally, there are frequent dropped requests using this browser [5].

These are just a few that I spotted. Let's proceed with the discussion as though the above issues were not present.

After looking at the issue tracker, this project wants each Mullvad Browser user to look the same, per OS [6]. Blending into a crowd on the surface seems like a good idea, assuming the crowd was large enough, but that "per OS" detail is a big gotcha.

I personally don't see why a source-modified browser shouldn't be able to achieve perfect uniformity. It's especially suspicious to me that the Tor project never achieved it, despite having had multiple years of developer effort dedicated to this goal, and backed by funding. IMO, browsers should never have been flooded with so many uncontrolled privacy breaking features in the first place.

Modification of the browser is discouraged for any reason, including enhancing privacy features [6]. Now read that again, and this time assume hostile intent.

I mentioned in a different comment that the alternative to uniform blending is randomness. Some of the fingerprints in the browser are already randomized. Plausible randomness is far superior to trying to build up a large enough crowd and simultaneously solving the uniformity issues. The entire javascript engine should be ripped apart and reassembled so that all privacy invading features can only function for client-side specific tasks but cannot speak with the networking and storage features.

[1] https://browserleaks.com/webgl [2] https://browserleaks.com/fonts [3] https://browserleaks.com/ssl [4] DNS Leak test: https://browserleaks.com/ip [5] https://github.com/mullvad/mullvad-browser/issues/23 [6] https://github.com/mullvad/mullvad-browser/issues/1

As I suspected, this browser, just like the Tor browser, does NOT protect you from basic browser leaks. There are gaping issues and after looking at the github issues, the maintainers are certain that their current strategy is effective. This makes me continue to question this project and Mullvad.

I'm not advocating for this browser specifically, only encouraging more people to take pro-privacy and online safety measures.

It's pointless to say the problem is the employer, or the hacker who released the data, or the programmer who relied on bad algorithms, or the admin who didn't secure the data. One way or another, this data will get leaked, the old hashing and encryption techniques will be broken and there will be people searching through it all. Forget about the government, at least they are beholden to law and maintaining the appearance of adhering to it. Substitute employer for neighbor, girlfriend or internet stalker and you have equally valid scenarios which are even more disturbing in my mind.

The issue is that depending on the situation, and on the patient in question, the intervention will shift from decision, to persuasion to mandatory.

For example, if you have a positive test for cancer, they will persuade you to follow it up. They will apply very emotionally manipulative techniques.

Another example, if the patient does not have the ability to decide for themselves, the doctor's opinion will hold a very heavy weight and can lead into a legal issue.

If you believe that science and medicine in 2023 is offering complete understanding and solutions to all or even many problems, then you will naturally feel like these two examples should be handled that way. However, the daily "revelations" of "new medical research" which contradict past results and understandings should be enough to cause doubt in that high level of confidence.

Half the posts in this thread bring back the old SlashDot adage: The plural of anecdote is not data.

I find the application of this logic to be entirely one-sided. Positive anecdotes are data, negative anecdotes are wives tales.

Simplifying medical information into data points is a dangerous practice which by now should be revealing itself as entirely fraudulent. Anecdotes provide context which doctors and science love to dismiss. I'll take a handful of anecdotes and appreciate all the context they provide, meanwhile the context-free "science" will go back and forth never getting anywhere nearer to better health outcomes.

With all this downtime and reduced interface, are pro users expecting a refund on this months service ?

I may have missed the detail, but it also expects the pytorch conversion rather than original LLaMa model.

I didn't say it prevents tracking, I was offering a litmus test for a VPN to the question of red flags. If it doesn't pass the litmus test, preventing tracking is the least of your concerns.

If humans still exist, those things will still hold value. I can't say for sure humans will survive in the same quantities once all things settle, but let's imagine we are in a future where deep learning has matured and netflix, steam, facebook, etc are all now just "prompt terminals" that output your request.

Then, with infinite variations, there will be infinite duds. People will value the ones that aren't duds so there will be a market for curators. And there will be a niche where people will want to see the exact same output together and talk about it.

Can you imagine facebook ? "Facebook, please give me a friend that <prompt here>"

My hot-take is that people trying Arch derivatives as their first linux has and will continue to hinder linux desktop uptake.

That being said I believe the Steam Deck OS is an arch derivative. I will allow this one to be the exception to the rule.

With the progression of the technology, you may find there are not so many "in-between" steps as you expect in your current list. It will quickly go from the current stage "employee being more productive" to "steam, I want to play a game where <insert prompt here>". No need to generate art assets or have anything resembling a game engine, just pure interactive "imagination".

Very accurate. The CIA is actively involved in the interference of their 'allies' governments.

Japan didn't break from US in this news article, it received permission to do so, likely so that it can remain in a defensible position against China.

Excellent use of ad-hominem, incorrect use of whataboutism and to top it off, a classic strawman.

Long-term services are great targets for governments.

If you were to looking for some trust in a VPN, you would want them to offer locations in privacy friendly countries, and highlighting them as such. That would potentially funnel more used to those servers which would be beneficial. You would also want the VPN to ensure the servers in those countries are run by companies based in that country, and not be head-quartered in some other country.

Bingo.

Check all the browser leak tests too, they are important and different tests.

Browser fingerprinting is exactly that. And the browser leaks are an even more concerning issue that must be confirmed. Websites want to know who you are or at least that you're not a bot. As a pro-privacy user, you don't want websites to know either of those things. That's low-hanging fruit that a few simple browser tweaks can help with.

Yes either you want everyone to look the same, or you want every page request to be totally random.

This isn't useful to 'dark web' types. This is at best useful for 'mom and pop' who heard about 'china tiktok' on the news.

Simply download the Tor browser and evaluate its performance on one of the many browser fingerprint [1][2] and browser leak [3][4] web services. The last time I checked, it didn't pass every test.

[1] https://www.amiunique.org/fp [2] https://coveryourtracks.eff.org/ [3] https://browserleaks.com/ [4] https://www.dnsleaktest.com/

The last time I tried the Tor browser, it did not sufficiently handle browser finger prints. I don't have high expectations out of this project either, but at least they offer a firefox extension. I'd have to dig into it to determine how effective it is, but as it stands there are other firefox extensions that already do an excellent job.

[GNU/]Linux.

I should be able to stop there, but I understand that if you never really gave Linux a shot, or you tried one distro once and gave up when the first thing broke, then you didn't get a chance to learn anything about Linux at all. If that's you, try Linux Mint next time. It's a good beginner distro, you won't break it easily, you will be able to google solutions easily and after a few months you will have learned enough to try other distros. It's important to start with a stable beginner distro to have the "things that work on linux desktop" benchmark.

Here's another thing you need to know about Linux: The look and feel of your Linux desktop is not locked-in. All those reviews you read about a Linux distro that talked about the look or feel, are actually just misleading. The look and feel are distro-independent for the most part.

Linux desktops are as customizable as you want them to be [1], or as little as you want them to be. GNOME and KDE are not the only game in town, XFCE is also a solid out-of-box minimalist experience which is very much like old-windows. If you want to boost productivity, tiling managers are waiting for you. You can also install all of them at once and switch at the login screen. Even the login screen is something you can customise or swap out.

[1] https://teddit.net/r/unixporn

The issue tracker and pull requests are being hit with very funny suggestions. Many people suspect this is an April Fools joke. It's possible this entire repo was generated by a LLM to appear plausible.

I especially like the suggestions to rewrite the algorithm in Rust [1] and this pull request which simplifies the algorithm to a single c file [2].

[1] https://github.com/twitter/the-algorithm/issues?q=is%3Aissue... [2] https://github.com/twitter/the-algorithm/pull/712

The simple truth is that a single talented, unencumbered (ie, as repo owner you're less encumbered by PR etiquette and feature direction), motivated developer in open source software, especially from what I've seen in the game-engine/framework space where multiple disciplines come into play, is far more effective than a large group of contributors will ever be as far as pushing large new features is concerned.

When it comes to picking a smaller open source game engine, it becomes essential to evaluate the hero developers involved: how committed are they to the project? is their effort sustainable ?

I think Bevy will work out long-term, because they have structured as a sustainable foundation with solid project management, however the lack of a hero dev that who can contribute to all the sub-crates [1] within the bevy ecosystem also means you have odd quirks in Bevy that will remain longstanding on the issue tracker, like for example setting the texture wrapping mode.

For that reason, Fyrox is worth serious consideration if you're constraining to the rust ecosystem of game engines/frameworks. Just one developer, but he's got the old-school experience and dedication, and the framework looks/feels much more familiar to a C++ game developer. It's not the sophisticated beast that Bevy is, but that's a good thing to the kind of person who would even consider these game engines/frameworks in the first place.

What really saddens me is that these hero game devs are still choosing C++ instead of rust. There are so many great flash-in-the-pan game engine/frameworks in C++, but they will never take off for a number of reasons.

When you look at how game devs approach development in C++, you come to realise that Rust is the language we wanted all along.

[1] https://github.com/bevyengine/bevy/tree/main/crates

Ever since I saw your pokemon fine tuning, I've been looking forward to trying it myself ! I find your results to be convincing, as a long time pokemon fan !

I made an account to say thank you for sharing your fine tuning code for others to use !