Hacker News: a99c43f2d565504

Recommend a web article to speech tool

a99c43f2d565504 — Thu, 18 Dec 2025 16:15:50 +0000

I'd like to give a URL to an article to some app and have it read the article out loud to me. I tried the Android apps of ChatGPT and Gemini already but neither of them could do it. ChatGPT provided me with a text input element in the speech mode but said that it can't read the article out loud: It could only summarize it. Gemini wasn't even aware of its own interface: It kept asking for the URL in a chat, which doesn't exist in the speech mode where we were communicating. It then started speculating about how websites typically work which has notning to do with our interaction.

Is there some app that could do what I ask for: I give a URL, and the app reads the article out loud, preferably in the background i.e. in such a way that I can turn off the screen of my Android phone while it keeps reading.

Note also that the app should be smart enough to realize what is the interesting content on the web page, i.e. not read out loud any irrelevant headers and sidebars and footers and menus etc.

Comments URL: https://news.ycombinator.com/item?id=46314611

Points: 2

# Comments: 1

New comment by a99c43f2d565504 in "Linux CVEs, more than you ever wanted to know"

a99c43f2d565504 — Tue, 09 Dec 2025 23:56:55 +0000

Right. Also, when it comes to the other aspects of TLS, such as preventing middlemen from making sense of what information flows between you and the server, what exactly is the threat in this case? I mean, it's a public blog post, which you only ask to read and so you are served.

New comment by a99c43f2d565504 in "Paper2video: Automatic video generation from scientific papers"

a99c43f2d565504 — Sun, 12 Oct 2025 11:22:25 +0000

Besides just porn or nudity, maybe we could also add violence into the arsenal of engagement. For example, maybe the viewer could use a virtual sword or shotgun on some key concepts in the presentation to initiate a tangent going on a deep dive on the concept, and then come back to the presentation once done with the rabbit hole.

New comment by a99c43f2d565504 in "Shai-Hulud malware attack: Tinycolor and over 40 NPM packages compromised"

a99c43f2d565504 — Tue, 16 Sep 2025 16:03:06 +0000

> Does anyone know of other major projects written in as strict a style as esbuild?

As in any random major project with focus on not having dependencies? SQLite comes to mind.

New comment by a99c43f2d565504 in "Models of European metro stations"

a99c43f2d565504 — Sun, 14 Sep 2025 09:39:46 +0000

It's also not an exhaustive list anyway. At least Helsinki, Finland is missing. I think Finland is unambigiously Europe.

New comment by a99c43f2d565504 in "Microsoft is officially sending employees back to the office"

a99c43f2d565504 — Tue, 09 Sep 2025 21:43:43 +0000

Maybe some positions are or feel worthy only when performed in physically social context. Jobs dealing with human problems have this tendency more often than those dealing more with non-human problems.

New comment by a99c43f2d565504 in "Claude now has access to a server-side container environment"

a99c43f2d565504 — Tue, 09 Sep 2025 21:37:51 +0000

I don't know what the question implied but generally speaking it is a known effect that making AI do the things you used to do accumulates cognitive debt which seems intuitively true too. Physical exercise is a good analogy perhaps.

New comment by a99c43f2d565504 in "Neovim Pack"

a99c43f2d565504 — Thu, 04 Sep 2025 14:22:17 +0000

Offtopic, and no one asked, but I'm going to advertise anyway: Helix is a good alternative to (neo)vim for anyone who wants a terminal editor with vimotions(ish) but doesn't like configuring or scripting. That was the selling point that made me stay with it: The out-of-the-box experience was close enough to what I was hoping to end at but gave up with neovim due to the hassle required. It is opinionated of course, but the default behavior and appearance of Helix felt much more appropriate than that of neovim.

New comment by a99c43f2d565504 in "The issue of anti-cheat on Linux (2024)"

a99c43f2d565504 — Sat, 23 Aug 2025 11:46:50 +0000

Many commenters have pointed out the frustration that even in these kernel level anti cheat system using games there are still cheaters and therefore the sacrifice of submitting to such privacy and security hole is close to worthless.

When it comes to the problem of cheating in games, I think the only solution is to bind the gamer's identity to a real life identity that is not trivial to change. That way the cheater only needs to be caught once.

New comment by a99c43f2d565504 in "F-Droid build servers can't build modern Android apps due to outdated CPUs"

a99c43f2d565504 — Wed, 13 Aug 2025 05:58:36 +0000

It's not the target that is now requiring new instructions, but one of the components in the build tools.

New comment by a99c43f2d565504 in "Rust running on every GPU"

a99c43f2d565504 — Sat, 26 Jul 2025 14:08:10 +0000

As far as I understand, there was a similar mess with CPUs some 50 years ago: All computers were different and there was no such thing as portable code. Then problem solvers came up with abstractions like the C programming language, allowing developers to write more or less the same code for different platforms. I suppose GPUs are slowly going through a similar process now that they're useful in many more domains than just graphics. I'm just spitballing.

New comment by a99c43f2d565504 in "Plain Vanilla Web"

a99c43f2d565504 — Sun, 11 May 2025 20:04:51 +0000

Are you implying there exists a better way of making UIs that just hasn't taken over the world for some reason?

I also don't like the state of the web.

New comment by a99c43f2d565504 in "Show HN: My AI Native Resume"

a99c43f2d565504 — Mon, 05 May 2025 11:33:02 +0000

Chat bots require a special API I suppose, but an intelligent agent would just learn to use the existing way for programs communicating with other programs over a network. Unfortunately the I in LLM stands for intelligence.

New comment by a99c43f2d565504 in "Finland's National Allergy Program Successfully Reduces Allergic Diseases"

a99c43f2d565504 — Sat, 15 Mar 2025 11:25:32 +0000

Torille

New comment by a99c43f2d565504 in "EU to mobilize 200B Euros to invest in AI"

a99c43f2d565504 — Wed, 12 Feb 2025 12:46:20 +0000

I have no experience in this whatsoever but I feel like it's good to have some transparent bureoucracy even if it's slightly inconvenient if the alternative is to toss around billions in an opaque and corrupt manner like the way bookkeeping (or lack thereof) in global scale climate funds works. Note that I don't have experience in the latter either nor sources to give, yet this is the impression I have. Feel free to correct.

Where Are the Intelligent Robots?

a99c43f2d565504 — Thu, 02 Jan 2025 19:07:28 +0000

I recently watched a YouTube video where someone reacted to Sam Altman's statements. One comment he made left me wondering if I've missed something in the current "AI summer." Altman suggested that "robots will do all current jobs on our behalf, and we humans must therefore find new things to do."

If I understood correctly, it seems like OpenAI's leader is implying that large language models (LLMs) have something significant to contribute to robotics. This strikes me as curious because, as far as I know, these are two very different problem domains. Only one (LLMs) has experienced rapid advancements recently, while the other (robotics) hasn’t.

Robotics in the physical world demands fast and precise responses. A robot walking to a grocery store and picking up items, for example, needs quick and accurate decision-making to avoid falling, breaking things, or picking up the wrong items. LLMs, on the other hand, provide slow, approximate answers that often resemble truth but can include hallucinations. This works well for paraphrasing non-exact information but seems ill-suited to the precision robotics requires.

So, my questions are:

1. Have LLMs contributed to advancements in robotics? 2. Is there any reason to believe they will?

From my understanding, the current paradigm of LLMs relies on scaling compute power for diminishing returns, which is the opposite of what useful robotics needs: efficient and accurate computation.

Comments URL: https://news.ycombinator.com/item?id=42577600

Points: 2

# Comments: 3

New comment by a99c43f2d565504 in "Asterinas: OS kernel written in Rust and providing Linux-compatible ABI"

a99c43f2d565504 — Wed, 16 Oct 2024 11:14:34 +0000

Zig kernel when

New comment by a99c43f2d565504 in "I Am Tired of AI"

a99c43f2d565504 — Fri, 27 Sep 2024 10:29:08 +0000

Perhaps "trust" was a bit misplaced here, but I think we can all agree on the idea: Before LLMs, there was intelligence behind text, and now there's not. The I in LLM stands for intelligence, as written in one blog. Maybe the text never was true, but at least it made sense given some agenda. And like pointed out by others, the usual text style and vocabulary signs that could have been used to identify expertise or agenda are gone.

New comment by a99c43f2d565504 in "Exploiting CI / CD Pipelines for fun and profit"

a99c43f2d565504 — Mon, 09 Sep 2024 00:16:29 +0000

To not have to remember that one can not use a .dockerignore at all and instead explicitly pick the files and directories from the build context that need to be in the image.

Where are we at with automating commercial software development?

a99c43f2d565504 — Thu, 08 Feb 2024 19:28:29 +0000

I know I can discuss matters and get code snippets from ChatGPT but is it yet viable to integrate some AI-whatever to an IDE and some project management system and just observe the tool generate a draft pull request from some ticket in the project management system that I as the human developer responsible for the task can then review? My assumption is that there is no such seamless solution yet and I'm not sure whether will ever be but I also feel like I want to keep myself up to date about whether there has been advancements worth noting. If you know of such working solutions that you are satisfied with, please share the exact toolchain: which IDE, which AI-thingy and which project management thingy?

Comments URL: https://news.ycombinator.com/item?id=39306403

Points: 1

# Comments: 1