shortwavesurfer

cross-posted from: https://slrpnk.net/post/28482551

I'm looking into installing a door lock w/ key pad at home for two use cases:

1. I'm out of town and need to allow someone to enter my home, in an emergency or for any reason.
2. Nice to have - "oh shit, did I lock the door" - ability to lock the door remotely from my phone, would also solve use case #1 by unlocking remotely.

If there are no privacy respecting / self hosted apps for remote control (use case #2), then a "dumb" electronic lock w/ key pad that enables me to set a PIN that I can give to a friend or neighbor in a pinch and then reset the PIN after I get home, that would be good enough. If no such keypad/electronic locks exist, then my backup plan is to just make a few copies of my key for trusted friends & family and/or hide a key, but I'd like to explore the keypad route.

Currently, I have an automation that turns on my air conditioner for 37 minutes, every 3 hours. However, during that time, I cannot do anything like update, HAOS, etc. Because it breaks the timer, and then I have to manually shut it off.

I'm doing my best to continue using Firefox, but it's becoming incredibly annoying, and I'm seriously at my wit's end and just about to give up on it permanently.

I've noticed during video calls it causes my computer to completely freeze up and require a reboot sometimes.

And on my phone, sometimes when I'm reading, it will just make my screen completely stop responding to me and I have to force reboot my phone in order to fix it.

I know that Gecko is the only other web engine option to fight Chromium's Blink, but I'm seriously just about fed up enough with Firefox to say fuck it.

Servo or ladybird may be the only real hope left.

I have a Seeed SenseCAP T1000-E running fw 2.7.3 and with boosted gain turned off my node drained to 5% in 3 days and 6 hours. With boosted gain on though i have only gotten 2 days 15 hours to 7%. I know they are not the same, but i doubt i would get 16 more hours out of it.

So I just noticed that my local router has been up for 170 days. It has the public-private key cryptography introduced in version 2.5 because it has the lock icon. But since it's been up for 170 days, I know for a fact that it does not have version 2.6 at all, and there was two key issues fixed in 2.6.3(ish and 2.6.11.

It comes in 5,000 mAh and 10,000 mAh variants and will use 95% of the battery in power bank mode and reserve the bottom 5% to run the mesh node. I think you can run the mesh node and use all 100% of the battery if you wish, but the power bank mode can only use 95% of it. I think it's based on the NRF52 and SX1262 radio.

Edit: Sorry for the duplicate post. Please feel free to delete it, because I can't figure out how to.

• Channel Name: ShortFast
• Channel Key: AQ==
• Channel Uplink: ON
• Channel Downlink: ON
• Device Role: CLIENT
• LoRa Hops: 7
• LoRa Preset: ShortFast
• MQTT Topic: msh/US

The Channel name and key are auto set by choosing the ShortFast LoRa preset. Once there are enough people we can start state topics.

If you dont use the default LongFast meshtastic is really empty. I switched my node to MediumSlow, since it's like 86% faster than LongFast. I knew there would not be any in my area, so I subscribed to the msh/US MQTT topic and theres still nobody else. I'm going to leave it like that for a while and see if I hear anything during any kind of band openings or catch any other MQTT nodes trying out that topic. And yes, both uplink and downlink are enabled. I do know that there are some nodes in the Bay Area on msh/US/mqtt-bayme-sh, but I am not near their.

So I have been using Meshtastic for a bit over a month now, and I've been hearing about MeshCore, which sounds interesting, although it's much smaller. Tonight, I decided to give it a test, and while there's nobody around me who I can test with, my initial impressions of the firmware is that it's early, but good. With that said, from everything I can tell, the applications to control it from Android and iOS do not appear to be open source. And I absolutely refuse to become interested in something that is not open source. So that's a big no-no for me. If I am wrong, I would definitely be willing to try it again. But if I am right, the only way I would consider using it is if somebody else writes an app that is open source.

So I live in a small city of around 50,000 people and we have a router that's 200 feet up on a tower with a 5.8 DBI antenna.

There's a guy 17.5 miles away who wants to get into the mesh and his node is on a 30-foot flagpole and also has a 5.8 DBI antenna.

There is no major elevation change between the two nodes and according to a distance calculator, the line of sight between both antennas should be about 24 miles, which would cover the distance with no problem.

With that said, the nodes are not connecting together. And I'm wondering if that's because of the 5.8 DBI antenna gain on both sides, or if there's something else I might be missing.

Edit: On a side note, I live 3.7 miles from the router, and it has trouble hearing me, but I do not have trouble hearing it with my T1000E. And I'm also wondering if that's because of the antenna gain on the router side.

I used i2p from InviZible Pro (F-Droid).

I was trying to connect my Monero wallet to a Monero RPC Damon that somebody I know runs and while it did connect the absolute best speed I could ever achieve through it was 45KiB/s. I changed no settings at all and just used the defaults. Turned it on and had 33 client tunnels.

Tor usually gets me ~400KiB/s to the hs, but i thought i2p would be faster.

Edit: it used 2 hops as default and i left it that way.