I’ve been looking into some kind of simple SSO to handle this. I’m tired of entering passwords (even if it’s all done by the password manager) a single authentication point with a single user would be great.
Keycloak and friend are way too complex. Ideally I would like to have something in my nginx reverse proxies that would handle authentication at that level and tell the final app what user is logged on in some safe way.
XMPP is way more open and interoperable than all the solutions available, it works like email any user can can talk to any other and doesn’t depend on a some proprietary / closed service centrally owned by anyone. That’s a good selling point.
XMPP doesn’t really force users to sign up with email address, it just happens that XMPP addresses use the same format, many public servers will give you an address like username@server.example.org that is never mapped to a real email address and only works for XMPP. The decision to actually ask people for their real addresses is up to who owns the server and won’t be directly exposed on the XMPP network.
It is as dead as we want. There's no reason to reinvent the wheel, probably the only thing that XMPP lacks is a bunch of money into a very good, cross-platform (but native) client like Telegram has that actually works 100% of the time and a bunch of large scale public servers to handle regular users who don't want to host their own. Also... easy registrations and setup on said client.
For a regular user and most privacy aware people, they just don't care if the protocol is Matrix, Signal or XMPP - they just want a good end user experience and a solid thing, that's what XMPP lacks today and it's all client side.
Bottom line is: XMPP as a protocol is great, lacks someone with vision and money to drive it into mass adoption.
No, Matrix isn’t the best in terms of privacy. It is a metadata disaster and most other platform are a lot more performant.
Matrix’s E2EE does not, however, encrypt everything. The following information is not encrypted: Message senders, Session/device IDs, Message timestamps, Room members (join/leave/invite events), Message edit events, Message reactions, Read receipts, Nicknames, Profile pictures
Matrix is developed by a for profit entity, a group of venture capitalists and having a spec doesn’t mean everything. The way Matrix is designed is to force into jumping through hoops and kind of draw all attention to Matrix itself instead of the end result.
XMPP is the true and the OG federated and truly open solution that is very extensible. XMPP is tested, reliable, secure and above all a truly open standard and decentralized it just lacks some investment in better mobile clients.
What most fail to see is that XMPP is the only solution that treats messaging and video like email: just provide an address and the servers and clients will cooperate with each other in order to maintain a conversation. Everything else is just an attempt at yet another vendor lock-in.
People need to get this through their heads, XMPP is the only solution for their problems.
It is good indeed they're actually decent people, however I'm sad they switched to a piece of shit of a software that fails to sync messages at any opportunity and delivers a poor user experience in most platforms by not being native / following platform specific UI guidelines.
I like Jabber and Monal a LOT and I hope they can replace the garbage that other IMs and Matrix is however the clients must be a bit better. Monal is a very good attempt, it is almost there but there are multiple pain points that won’t be acceptable for a regular user.
I expect them to be on the level of other modern chat apps, not something barely working, hard to setup for most people and prone to failure at any point.
I tried Monal once, and let's face it doesn't even look close to any other basic iOS app, poorly designed, sometimes crashes and getting audio/video to work is just painful because of some bad taste decisions they've made.
I like Jabber and Monal a LOT and I hope they can replace the garbage that other IMs and Matrix is however the clients must be a bit better. Monal is a very good attempt, it is almost there but there are multiple pain points that won't be acceptable for a regular user.
I don't live in a place with particularly high energy prices, but I've been reading about countless complaints about home labbers changing to new hardware because of that.
The thing is that, even if the power is cheap a Pi when you're comparing a PII to a RPi 4 or a modern machine with a "T" CPU if you've the money to spend on the new thing you'll way better, faster CPU, faster RAM, less noise, less power, more modern features, less software issues... at certain point it makes no sense to run that old hardware. Did you ever try to SSH into a Core Duo machine with a Ed25519 key? The CPU doesn't have the modern crypto extensions making the login unbearably slow, similar happens with other SSL stuff.
That's another good pick yeah. N350 stuff will also be interesting.
2020 has shown that the Pi Foundation sees itself as an industry-first product.
I think they never saw themselves as anything other than an industry-first product. The "hobbyist" market was just a way to develop, test and enter the mass market and gain critical mass in terms of FAB capacity and support / mind-share. IMHO their goal was always to go into the industry and disrupt some markets but you can't just get there without the scale. They just played an entire generation of hobbyist making them addicted to their product to grow it and test it for them.
Now that they're public and partially owned by Broadcom it will just get worse.
The Pi Foundation kind of held the SBC market hostage to their ecosystem because software support is important and all the shinny Python libraries people are used to are typically only fully compatible, stable and tested with the Pi GPIO. With the RP2040 chip they make it so software/library compatibility is no longer a barrier to other CPU makers to enter the market - Intel or even Rockchip and Mediatek SBCs can include the RP2040 and gain instant software compatibility with any software library made for the Pi GPIO. Note that right now when RK releases an SBC it take a while for libraries to catch up with the GPIO definitions and whatnot.
I'm sure they aware of this risk, however, there's a much bigger market opportunity there - the SPI / i2C / GPIO bridge market typically held by FTDI. When you want to make low level hardware communicate with computers, usually on USB ports, you'll need some kind of hardware to handle the low level SPI / i2C / GPIO signals and convert them into something the CPU and the OS can understand, this is where FTDI has a big market share, even the Arduino uses a chip made by them for that.
The RP2040 can do this exact same task - it is what it does on the RPI 5 after all - and that’s a very big market. Almost every peripheral we connect to our computers is using one of those bridges to connect low level hardware such as microcontrollers to the computer or to simply toggle LEDs. Broadcom is now an investor of the Pi Foundation and they do a lot of hardware that does require those kinds of bridges… maybe they were the ones pushing the Pi guys into this direction because business wise it makes sense - they can test the reliability of those chips on the SBC market and once they’re sure they perform as good as FTDI ones they can use them everywhere for a fraction of the cost.
Let's see who gains more from the RP2040, the Pi guys obliterating FTDI or Intel and others taking chunks of the SBC market.
There's already a couple of examples of what I'm saying here:
For 80$ I can get an 8th gen HP Mini with 16 GB of RAM + 256 GB M.2., case, power brick, all cables and have a much more stable and powerful system (second hand on eBay).
Frankly the power consumption of that thing x performance delivered will be just bad. Take for example this example, a more modern Pentium D vs a Pi:
If you don't have any kind of attachment to the machine, just trash it, get a Pi or a second hand 8th Gen i5 HP Mini for around 80$ and enjoy. If you do have an attachment to the machine you may as well run some OS from the same era. https://winworldpc.com/library/operating-systems
But then again I’ve always found the AltStore devs practices leaning toward shady
Yes they're shady, after all we're talking about the guy who made Delta only available in Europe through the AltStore in order to push it. They're also the guys who make banners saying "3rd party apps are here" showcasing UTM, qBitControl, PeopleDrop and iTorrent when none of those apps are available on their store - just delta and 2 other random utils.
Hmm this is actually interesting, passkeys would indeed make things simpler.