Skip Navigation

InitialsDiceBearhttps://github.com/dicebear/dicebearhttps://creativecommons.org/publicdomain/zero/1.0/„Initials” (https://github.com/dicebear/dicebear) by „DiceBear”, licensed under „CC0 1.0” (https://creativecommons.org/publicdomain/zero/1.0/)BA
Posts
5
Comments
1,074
Joined
2 yr. ago

  • You don’t need to wonder, Apple has said as much that their AI is built on LLMs, just like everybody else. While hallucinations are still a major unsolved problem, that doesn’t mean they aren’t able to be reduced in frequency and severity. A ChatGPT like chatbot is going to hallucinate because you’re asking it to give extremely open ended responses to literally any query. The more data you feed it in the prompt, and the more you constrain its output, the less likely it is to hallucinate. It’ll likely be extremely rare that using the grammar check or rephrasing tools in Apple AI will be affected by hallucinations for that reason. Siri is more comparable to ChatGPT with regards to open ended questions, but it’s likely that they will integrate LLMs primarily for transforming inputs and outputs rather than the whole process. For example, the LLM could be prompted to call a function based on the user’s query. Then, that function finds a reliable result, either using existing APIs for real time information like weather, or using another LLM with a search engine. The output from this truth-finding process is then fed back into an LLM to generate the final output. The role of the LLM is heavily constrained at every step of the way, which is known to minimize hallucinations.

    You arguing that this is an unsolvable problem is defeatist and not helpful to actually mitigating the real issue.

  • It says:

    “Congress shall make no law respecting an establishment of religion, or prohibiting the free exercise thereof; or abridging the freedom of speech, or of the press; or the right of the people peaceably to assemble, and to petition the Government for a redress of grievances.”

    It’s definitely not as clear as one might expect out of a modern legal document. That being said, the precedent here is very straightforward, and any ruling in favor of the law would be a huge shift in how the separation of church and state is applied.

  • Most streaming services don’t do comment sections. That’s mostly a YouTube and TikTok thing where the sites depend on user submitted content and also function as a social network. Despite this, people talk about shows on those services, just elsewhere.

  • On embedded devices, how often are you parsing input that came from notepad (or any other text editor)? If your device has a UI or a web server, you’re likely already using something that handles various encoding and line endings. If you’re reading data you included at build time, consider a validator/sanitizer script that can run in your build environment where it can have easy access to off the shelf libraries.

    On a side note- as a software engineer who primarily works on things running in a general purpose OS but does occasionally have to make small programs that can function on embedded devices (albeit still usually with an OS, think routers and iot), I’m glad that the Rust community takes no-std development seriously. Large swaths of the rust ecosystem is available even in embedded environments.

  • Different OSes using different line endings is such a long standing and well known problem that I would only describe the bugs that come as a result as bad programming. Not even lazy programming, a lazy programmer uses a library that abstracts away these differences.

  • The part you’re missing is that while C++ does have newer safer ways of doing memory management, all the old ways are still present, in wide use, and are easier. Basically, C++ makes it easy to do the wrong thing and hard to do the right thing, and most codebases are built around the wrong things. It’s often easier to just rewrite it in rust than it is to refactor an existing code base, so if you’re going to expend that effort why not do it in a language that has stronger safety guarantees, a better dependency and build management system, and a growing community?

  • Google’s the main backer but it’s technically an open standard. Apple is implementing support in iOS 18 too, reluctantly. Google also has non-standard extensions on top of it like e2e encryption. I wish they would standardize it but that’s a pretty important feature and maybe it’s ok that they brought it out before it could become standard.

  • As a person who has been managing Linux servers for about a decade now, trust me that a few hours or days of learning docker now will save you weeks if not months in the future. Docker makes managing servers and dealing with updates trivial and predictable. Setting everything up in docker compose makes it easy to recover if something fails, it’s it’s self documenting because you can quickly see exactly how your applications are configured and running.

  • Yep, I have Kagi set as my default, but if I’m searching for places nearby or current events, I usually end up back on google. But 90+% of my search queries stay on Kagi which is impressive enough. In the past I’ve tried DuckDuckGo and always switched back in part because the results quality was very bad.

  • While Kagi does have a very specific audience in mind, I’m not convinced specialization vs generalization is a reason for a difference. Google also tailors ads for the audience, using information they have collected on you, people similar to you, as well as general time of day, current events, etc. It’s more that google wants you on their site for longer, Kagi wants you to find your result and move on, since the more you search the more it costs them.