• 0 Posts
  • 62 Comments
Joined 2 years ago
cake
Cake day: June 19th, 2023

help-circle





  • Interesting, I hadn’t seen news about that Apple feature before… There seems to be a lot more press around Recall, which in turn amps up the amount of consumer attention and backlash.

    That said (and I wouldn’t want Apple’s “semantic search” even if I had an Apple device), I’d still trust Apple more to manage the dataset securely compared to Microsoft. The Apple ecosystem is far more strictly controlled, whereas in Windows it’s more of a free-for-all (most people just used XP as an administrator, the UAC could be easily disabled on Windows Vista and 7, etc.). Especially with Microsoft’s move to put advertising in Windows 11 and complete lack of security measures in the initial version of Recall, it is very hard to trust Microsoft in this regard.


  • Currently testing out Shiori. I have tried both it and Readeck on my Raspberry Pi 3B running 32-bit Raspbian ‘Bullseye’. From an installation and performance perspective, Shiori is easier: 32-bit binaries are available and its web interface is a bit more responsive. If you are not running a 64-bit version of the Raspberry Pi OS, you have to build Readeck yourself (which I successfully did).

    The main attraction of Shiori over Readeck is that it has both a readable and archive view. The main downside for me is it lacks the read/unread feature of Readeck and Pocket (although that’s apparently something planned for future and in the meantime, it’s possible to workaround using tags).

    Both have a browser extension to enable capturing pages that contain Javascript and do not gracefully degrade. Both also have an export-to-ebook function that may suit your e-reader.

    Bottom line: if you have an RPi4 or later, then Readeck is probably want you want. If you’re running an older Pi, though, Shiori may be a better option.










  • There’s already talk-to-your-dog/cat products such as FluentPet. Probably the biggest issue with cats in particular is that their “vocabulary” is quite limited (usually less than a dozen distinct “meows”), but some of the FluentPet users (examples on Youtube such as BilliSpeaks) seem to suggest basic reasoning. A full-blown language is beyond them, but they do seem capable of understanding more concepts than we give them credit for.





  • Australis13@fedia.iotoTechnology@lemmy.worldaight... i'm out..
    link
    fedilink
    arrow-up
    44
    arrow-down
    1
    ·
    3 months ago

    The irony is that, according to the article, it already does. What is changing is that the LLM will be able to use more of that data:

    OpenAI is rolling out a new update to ChatGPT’s memory that allows the bot to access the contents of all of your previous chats. The idea is that by pulling from your past conversations, ChatGPT will be able to offer more relevant results to your questions, queries, and overall discussions.

    ChatGPT’s memory feature is a little over a year old at this point, but its function has been much more limited than the update OpenAI is rolling out today… Previously, the bot stored those data points in a bank of “saved memories.” You could access this memory bank at any time and see what the bot had stored based on your conversations… However, it wasn’t perfect, and couldn’t naturally pull from past conversations, as a feature like “memory” might imply.