• 2 Posts
  • 34 Comments
Joined 3 months ago
cake
Cake day: April 26th, 2025

help-circle




  • I used enpass for years and was a happy user. one day it prompted me for some re-authentication bullshit security theater. although in that instant it was an easy task, took me all of 10 seconds, it demonstrated a scary amount of power they had as I couldn’t bypass it and access my data. from that point on, its days were numbered.

    the second issue is the export functionality that was seriously lacking and I had to resort to 3rd party converter tools to convert it to keepassXC; no way that flew by their QC, it had to be intentional.



  • you’re fine using lineageOS with microG and utilising it for cloud messaging, i.e. notifications. the actual content of the notfication doesn’t go through google (or apple), a push message just signals the telegram client there’s activity. then the telegram client wakes up, fetches the message from telegram servers, constructs the notification in-app and then displays it.

    google doesn’t have access to the contents of it, but harvests lotsa metadata that microG (as opposed to full-featured play store services) somewhat ameliorates.

    having said that, you should make every effort to ditch telegram as well, for a buncha reasons.



  • I’m like a test-bed for a) my business customers and b) friends and family. also, “wasting” time thusly is vastly better than my previous “hobby”, namely buying new and exciting shit.

    my customers benefit from me knowing how exactly (and why!) I should implement e.g. an unbound instance on-premise. or an in-house prosody communication platform. or the “dev team” (buncha dudes poking at wordpress) getting a slew of used elitebooks with linux for the price of one new windows-with-ai yoga the spec initially called for.

    f&f benefit from my early adoption by way of trickle-down tech. no way is anyone of them going to selfhost all this crap, but they get sprinkles of benefits in the form of “get this phone with that OS with those apps” and they’re dramatically better off. you don’t need the new ideapad ryzen that’s “on sale” (isn’t), have this 10-year old macbook I fixed and installed linux on - off you go. you don’t need the new phone that’s “free” with an exorbitantly priced plan, have the cheapest plan with this Redmi/Poco phone I swapped the battery on and installed LineageOS.

    as to practical considerations, any and all interactions with the likes of FAANG are and should be adversarial from the get-go, they are out to hurt you by any means necessary. them fucks lost the benefit of doubt ages ago so you not letting them have a millimeter of grasp in your domicile should be your primary task. as their gains are cumulative in nature, every battle won is significant and you’d do well to remind yourself constantly of that.


  • kodi and its derivatives are not something you should be using. it’s shit software on so many levels and we should burn it in the deepest volcanos we got.

    try one of these:

    1. run lineageOS TV (konstakang images) on it and install regular ATV apps for the services mentioned. so, like googletv except there’s no spying and ads and shit.

    2. create a normal linux box that has a DLNA sink e.g. using macast. there’s no remote control, you use your android/iOS device to send it stuff, like movie from Jellyfin or a youtube video, and it plays it back and allows some control (pause, play, rew/ff, etc)

    3. dedicated Jellyfin box; same as 2) but boots right into jellyfin client. it can be run in TV mode where it reacts to only up/down/left/right/enter/back, via gamepad or remote controller. if yours isn’t recognised, you can emulate it with InputRemapper.

    not familiar with how twitch does stuff.

    you also have the option of installing a normal raspi distro and then using a wireless keyboard and mouse/touchpad to run it, but I am of the opinion that once the device gets placed by the TV, it loses all keyboard and mouse privileges and should only be operated via the TV’s remote.


  • we had the centralised model (big corpos hosting all of the stuff) because our devices were shit and internet access was rare and precious. nowadays, with ever-present internet, when my $50 pocket computer has 8 cores and 8 GB RAM, the laptop many times that, let alone the desktop, we should be moving to Pied Piper’s vision of a decentralised internet and dedicate all of our resources to that goal.

    I’ve been a part of the fediverse some while now and I admit, I didn’t understand it fully. I operated under the premise that whoever put this thing together and then spent their time and energy promoting it has thought this through and then seeing more and more people jumping on, I took it as validation of that idea.

    a few years down the road, I have a better understanding, and I don’t really like it. it’s wasteful and disorganized and I don’t see a way where some order out of this chaos emerges.

    I thought it’s a sort of fail-over distribution of content. so if lemmy.bing is offline/gone, you can interact with lemmy.ding or lemmy.bong and access all data and post and comment and whatnot. not so, when ding is gone, it’s gone. its radiated content may be present on other instances, but still there’s a ton of issues that way.

    instead, I believe a decentralised and distributed system, with no single point of failure, no admins spending their hard earned cash on maintaining lemmy or mastodon instances or, god forbid, dedicated hardware in the vein of i2p or similar, should be the end goal.



  • it’s not “forever”. it’s however long they don’t have any ideas to the contrary.

    why it was implemented - so that executive #279 can show executive #114 that number go up. look how our engagement is rising! look at all them people downloading our app! when I took over from exec #317, number was this big, lookie now!

    same way google made their search worse, so you have to search multiple times, thus upping the engagment, page views, etc. and then exec X goes to exec Y and say “look there’s a huge rise in searches where my bonus at!”


  • you must’ve me confused with someone who does shit on your behest, go find out yourself.

    this is just for onlookers, as it’s obvious it’s weirdo’s shill: the term in the ToS is “all comms must be readable by all other clients” which an E2EE capable client would be in breach of and would be promptly kicked off telegram’s infra, as was mentioned by those same FOSS developers in lemmy threads regarding that subject. as for you, plonk.



  • appreciate the effort, but kinda went overboard with the deets :) I run several prosody XMPP servers so I’m familiar with the underlying tech. what you describe should be feasible with it as well, but there are constant issues with devices not being able to access history, so I was wondering if things were better on your end.

    so, based on this, I’ll spin up a snikker docker and try it out for a coupla weeks, see what’s what. many thanks.

    edit: turns out this snikket thing is conversations (standard XMPP client) and prosody (XMPP server) with different branding.


  • your argument boils down to “the fully functional and loaded gun is in this weirdo’s holster and he won’t use it”.

    the whole point is not relying on the benevolence of the weirdos out there and not letting them even be in the position to do any harm. encrypt my 1on1 comms and I don’t give a fuck what happens in the pedo/terror/carding/etc public groups. ample time to implement that in the past decade+ and be on par with practically every messenger out there. but he/they won’t implement it, they insist on all your shit being in the “cloud”, in plaintext, forever. there is no scenario where there’s not a malicious intent behind that.

    I’ve been using Telegram since the early days. it was phenomenal vs the crap of its day - magical, even. like many, I was enamored with the vision of durov the folksy hero battling the forces of evil (in a bozo nightmare) and bequeathing us this tech marvel.

    but I can’t trust it with anything any more. if weirdo can’t be trusted about some stuff, then he can’t be trusted with anything. enough for me, YMMV.



  • aside from the dogshit UX and the uber reliance on Evilcorp’s infra, having more than two devices (I know, shocker in this day and age!), the arduous migration process to a new device, the limited chat history (I think it’s 40ish days) and many more.

    same way Telegram adamantly refuses to implement E2EE, and not only that, it actively prevents 3rd party devs (a number of clients are FOSS) from implementing it on their own.

    both PJ Harvey and durov respond the same way when asked about any of them things - smokescreens, FUD, whataboutisms, etc.

    any of them things woulda been acceptable in 2015, here’s a PoC looking for funding, limited devs and resources; remember TextSecure and RedPhone? nowadays, they are nothing short of malicious.



  • although it’s interesting research, I think it’s a weak text if you’re even tangentially aware of Telegram’s bullshit narrative as it focuses on the wrong thing. the main point should be “this dude was caught lying on a number of occasions and throws out smoke-screens and FUD when confronted about super-simple stuff. therefore nothing that comes outta his mouth is to be trusted” which should’ve prompted a mass exodus from this bullshit platform a long time ago.

    the way more important issue is the collective action problem of dumping this crapware - leave it for what?

    I run XMPP and Matrix servers and use various clients, along with Signal. all of those things are fucking dogshit software, there isn’t one that can come close to Telegram’s UX. if you fell into a coma in 2014 and woke up now to Element of Fluffy or whatever, you’d think someone’s pulling your leg. this is what a decade of development looks like?!