I’m a 10+ (cumulative) yr. experience dev. While I never used The GitHub Copilot specifically, I’ve been using LLMs (as well as AI image generators) on a daily basis, mostly for non-dev things, such as analyzing my human-written poetry in order to get insights for my own writing. And I already did the same for codes I wrote, asking for LLMs to “Analyze and comment” my code, for the sake of insights. There were moments when I asked it for code snippets, and almost every code snippet it generated was indeed working or just needing few fixes.

They’ve been becoming good at this, but not enough to really replace my own coding and analysis. Instead, they’re becoming really better for poetry (maybe because their training data is mostly books and poetry works) and sentiment analysis. I use many LLMs simultaneously in order to compare them:

• Free version of Google Gemini is becoming lazy (short answers, superficial analysis, problems with keeping context, drafts aren’t so diverse as they were before, among other problems)
• free version of ChatGPT is a bit better (can keep contexts, can issue detailed answers) but not enough (it does hallucinate sometimes: good for surrealist poetry but bad for code and other technical matters when precision and coherence matters)
• Claude is laughable hypersensitive and self-censoring to certain words independently of contexts (got a code or text that remotely mentions the word “explode” as in PHP’s explode function? “Sorry, can’t comment on texts alluding to dangerous practices such as involving explosives”, I mean, WHAT?!?!)
• Bing Copilot got web searching, but it has a context limit of 5 messages, so, only usable for quick and short things.
• Same about Bing Copilot goes for Perplexity
• Mixtral is very hallucination-prone (i.e. does not properly cohere)
• LLama has been the best of all (via DDG’s “AI Chat” feature), although it sometimes glitches (i.e. starts to output repeated strings ad æternum)

As you see, I tried almost all of them. In summary, while it’s good to have such tools, they should never replace human intelligence… Or, at least, they shouldn’t…

Problem is, dev companies generally focus on “efficiency” over “efficacy”, wishing the shortest deadlines while wishing some perfection. Very understandable demands, but humans are humans, not robots. We need our time to deliver, we need to cautiously walk through all the steps needed to finally deploy something (especially big things), or it’ll become XGH programming (Extreme Go Horse). And machines can’t do that so perfectly, yet. For now, LLM for development is XGH: really fast, but far from coherent about the big picture (be it a platform, a module, a website, etc).

Have a proper radio ham license. Buy a 40-meter transceiver and a software defined radio dongle. Convert your code into esoteric programming languages such as Whitespace and Brainf, then spell it. “Plus, plus, next, plus, dot, open bracket, next, …”. Transmit your spelling over 40-meter band, while a receiver across the continent is tuned to the frequency. Ask it to repeat and record the QSO. Set the SDR recorder to I/Q packets instead of demodulating AM. Publish it as an audiobook.

null

Raytheon, Epirus and other manufacturers already have “human microwave heaters” and they’re mobile, except that they’re called “Active Denial System”.

A mix between the last one and the previous. The key error should be set in order to indicate the presence of an error, while status and code represents, respectively, the error numerical code and the error type, while the error message is supposed to be concatenated to an human-friendly error text.

They can’t without the given permission from the browser to do so. While they can indeed track the mouse, when they try to access mobile motion sensors (I’m considering a CAPTCHA inside a webpage being accessed through a mobile browser such as Firefox mobile or Chrome for Android), they need to use an HTML5 API that, in turn, will ask the user for permission, something like “This site wants to use sensor motion data. Allow or block?”

Nowadays there are some really annoying CAPTCHAs out there, such as:

• “Click over the figures that are upwards/downwards” and various rotated bears
• “Rotate the figure until it matches the given orientation” and a finger pointing to some random direction, as well as rotation buttons that don’t work the way you would mathematically expect them to work
• “Select all the images with a bicycle until there are none left” and the images take centuries to fade away after you click them
• “Select all the squares containing a bus” and there are squares with the very corner of the bus that make you wonder if they are considered as part of “squares containing A bus”
• “Fit the puzzle piece”, although this is the least annoying one

In summary, the CAPTCHAs seemingly are becoming less of a “prove you’re not a robot” and more of an forced IQ test. I can see the day when CAPTCHAs will ask you to write down a Laplacian transform for the solution f(x) to the differential equation governing the motion of a mass considering the resistance of air and aerodynamics, or write down a detailed solution to the P versus NP problem.

A lot. Not the whole thread, but the sub-thread that was started by my comment. I’m both a writer and a reader. I often use my comments to embrace different perspectives or ideas, bringing new possibilities to the table and contributing to the overall discussion, and these comments I made tend to get replies that, in turn, will be replied. This creates a good environment of dialogue, when many people contribute with different perspectives. And I like to read and ponder every one of them.

I replaced the part of the link you gave and indeed it potentially notifies about every reply on the thread, except that it doesn’t work with specific comments (for example, https://thelemmy.club/post/16151677/13197569), only with the entire post. I’m not used to use RSS Readers, but it’s worth to know it’s a possibility for being notified, even if it works only for the entire thread. Thanks!

It’s a fairly common thing on onion websites, especially those who offer real-time interaction (e.g. some onion web-chats), they use this Transfer-Encoding: chunked method for fetching messages and content because JS is often discouraged and sometimes automatically blocked by onion-enabled navigators while surfing DW. HTML Forms with submit buttons are also used for this kind of interaction.