• 0 Posts
  • 19 Comments
Joined 2 years ago
cake
Cake day: June 23rd, 2024

help-circle

  • It fits into the whole philosophy. There are several posts ( Initial Kuketz, discussion on Kuketz critique, reminder/restart discussion, criticism on usage of OpenAI in /e/ and poor communication, same questions again with no or wrong answers) criticising /e/ for heavily ignoring privacy and security flaws and only one response post on this Duvals answer on OpenAI usage in which they clarify to see ‘emotional’ reactions and look for alternatives while still finding it acceptable and criticism is ‘FUD’ and ‘hurt of reputation’ instead of valid concern.

    Additionaly the points postet by Kuketz are not addressed since today. Updates are a bit faster but still with weeks delay and still not including several parts of security updates (instead it’s the bare minimum).

    I looked for several posts on social media and Duval always ignores the points and yells that all the people are only up to harass him. He also uses false arguments to convince (probably) himself of this ridiculous behaviour.

    I started using /e/ in summer 2022 and was positive and hopeful because of the idea (long updates, privacy in mind, degoogled). But over the years learning that nearly all internal community and external expert criticism was ignored or marked as irrelevant or harassment when it’s not, my opinion changed and I’m no longer willing to talk or write about /e/ diplomatically as it is inappropriate.



  • Had the same problem. Contacted the admin who said that I was blocked due to many requests. The problem: They block critical queries sending something like 0.0.0.0 or 127.0.0.1. So the problematic connection cannot be established. That is the intended behavior. But the problematic will keep retrying and therefore you get blocked. When having startet my companies notebook (I got blocked) due to many Windows requests. Don’t know why but the treshold is not high enough






  • OpenAI feature is called Murena Voice to Text, so users don’t know, it’s OpenAI. They use tracking ids in their own update process. They ignore critique.

    Graphene is currently only available for Google devices and therefore no alternative.

    Lineage contains connections to Google as well.

    Recommending iOS is ridicolous.


  • Undertaker@feddit.orgtoOpen Source@lemmy.ml/e/OS is not a secure OS
    link
    fedilink
    arrow-up
    18
    arrow-down
    1
    ·
    3 months ago

    OpenAI feature is called Murena Voice to Text, so users don’t know, it’s OpenAI. They use tracking ids in their own update process. They ignore critique.

    Graphene is currently only available for Google devices and therefore no alternative.

    Lineage contains connections to Google as well.

    Recommending iOS is ridicolous.



  • Undertaker@feddit.orgtoPrivacy@lemmy.mlGraphene vs /e/ os
    link
    fedilink
    arrow-up
    7
    arrow-down
    1
    ·
    edit-2
    3 months ago

    /e/OS is not Google free (several calls, integrations and so on are connecting to Google). It makes use of OpenAI as well, uses tracking ids for updates. It is far behind regarding updates and thus risking privacy due to lack of security. They ignore any sort or critique.

    Graphene: You have to buy a Google device. Even second hand is support as it increases the value of their devices (or stabilize) and you walk arround with their name.

    Advice: Have a look at Iode.

    Pick a device that is not meant to be used for many years as Graphene plans to support a non Google device in coorporation with an unknown manufacturer.

    (Written from a Fairphone using /e/)



  • It is not. It is better than bloated Google Android but not “great”.

    • They use tracking ids for updates
    • They are far behind with updates lowering security and thus risking privacy
    • They use testing keys which is ridicolous
    • They added OpenAI integration and called it “Murena Voice to Text” so that unfiliar users could unintentially leak sensitive data
    • They ignore valid critique and call it ‘personal harassment’ although nothing personal nor something like harassment happen there (for example in related /e/OS threads)




  • No, BFS is ment to explore all nodes abd edges systematicaly, while ant algorithms may never explore a whole bunsh of nodes. They do they search not systematically but by walking paths randomly at the beginning and spreading pheromons while doing so. Finding shorter paths will result in a higher pheromon level remaining and thus direct next ants to these shorter paths instead of longer ones.

    You wouldn’t know, what they will do before executing the program and they will propably find shorter paths if you use 1000 ants instead of 100.

    In my results I drew a graph the highlights the ant’s paths by thicknes. I drew such a graph after (for example) 10, 100 and 1000 ants. You could see, that the first ants took a path randomly and after 100 iterations, some of the shorter edges became a bit thicker than others. At the end, only the shortest edges where drawn with high thickness while the longest one where rarely used if at all