Wow! That was an hour of sheer panic and adrenaline! While in the process of attempting to locate, unplug at both ends, disentangle, and finally extract a power cord from my insane spaghetti of cables, I knocked out power to the same portion of my equipment that I did last time, doing the same thing, LOL. Clearly, this is a rather weak point in my physical infrastructure. But I knew that already. But have I done anything about it yet? Clearly not. LOL. Whose to blame about that? Okay, I should do something about it. Maybe. Sometimes, or perhaps more than sometimes, we so often don't learn, and we willingly and continually do things or don't do things even though we know they are wrong. Then bvad things happen, and then we ignore it some more, and it nags at us, and we ignore it some more. Wellp, this time, things did not power up as expected. The major component that got knocked out was the entire network plant. Upon reboot, the SFP+ fiber modules did not return to service as usual. The active module was layer 3 unreachable, even though the port showed as up. The secondary (backup) module was online, but of course, it did not have any fiber connected to it. I switched the fiber over, but the connection did not come back up. Then, some how, both modules became unresponsive, and this time, both ports went down too. Oh boy, this was not good. I rebooted the switch, and only the backup fiber module came layer3 online, but the Internet did not. Fortunately, the 5G connection that I canceled, and then they gave me two free months of was still up and running, and automatically powering the network, as per routing policy. So I pulled the latest firmware for the fiber module, and ran the upgrade process. This went off without a hitch. However, though the upgrader said it was rebooting, layer3 never became responsible, and The fiber link did not come back up. I rebooted the switch once again, and this time, both SFP+ ports became active, and both fiber modules were pingable on the management VLAN once again! Wahoo! Then, what's more, the Internet came up on the backup module. I did a quick swap of the fiber, and the connection is now back up on the primary fiber module, with the backup once again functional and on hot standby, for situations, I guess, just like this. Wow! Seems that the switch had a hayday with its SFP+ section. 10G-base-T worked perfectly the whole time. Very odd indeed. I really thought there for about an hour that I had initially fried one, and then both fiber modules, a $300 value. Wellp, turns out I got lucky, and nothing is fried. Just wow!
Hey, I have someone in my life who uses an iPhone and would benefit from a little bit of internet privacy help. Is there anything I can recommend them?
"Stop using iPhone/iOS" is not an option for them. Thanks in advance!
ssd.eff.org/module/how-to-get-… is pretty good. It's more "make your iOS device a bit more private" rather than generic "internet privacy", but there's good advice in there.
I made a video ranting about the algorithmic internet and ads, then started yelling about using Firefox and how forks are killing FOSS projects. I think I'm losing my mind and this won't be popular at all 🫠
If using forks killed Firefox as "relevant" architecture – using Edge/Opera would kill Chrome's "relevance". I can't see how that is the case. Using a gecko based fork won't show up as blink, but maybe I got you wrong.
I think, leaving Firefox sends a strong and clear message. Especially when forks get more attention. Same as growing numbers would. (btw Thunderbird has no notable forks – maybe there is something to learn from you 😉)
Staying with Firefox enables it to ignore criticism, keep default Google search (and bragging about privacy), add AI features, … in the end I value the freedom to vote with my feet, even in Firefox's case.
I'm on your side on most topics you cover – but come on, be a bit nicer to forks. 😍
Jujutsu is starting to grow on me in ways I can't fully articulate yet.
Essentially my workflow now is: create an empty commit describing some change. Do all my immediate work in a second empty commit on top of that. If the work fits within the larger change I described in the previous commit, squash the changes, then I'm back to an empty commit again and the change just accumulates below. If I realize "Oh crap, this thing I just did has nothing to do with the larger change but it's still good to have," no problem. Just describe it, commit, then I can easily rebase the previous squash on top of the change I just made. So then I have main -> unrelated nice-to-have change -> squashed commit with larger change -> empty commit for new work. It sounds confusing, and it still kind of is, but it feels like peeling back a level of burocracy I didn't quite realize existed until it was gone. I can even jj prev a couple times, rewind the worktree past the squash commit, and keep working on the new problem there if I want. Sure it's all possible in git, but how much magic do you need to know to actually want to use that on a daily basis? Willingly?
Speaking of Git, when it's time to not break Git users' brains (and my own brain still, if I'm being completely honest) I update my bookmarks (named commits, as far as I can tell) and push them to Git as branches. Then I just keep on working however I like or need to, and filter it out to upstream as needed. The fact that my working directory is tracked automatically unless I explicitly gitignore a thing worried me at first, but the key is making that squash commit what I want to filter out upstream. So all those design notes, sketches, test scripts and such that were useful to have and automatically version but probably aren't great to upstream? Just rm them in the squash commit and they're taken care of. I still technically have them on that branch if I need them again later.
No LFS support, no submodules, and no meaningful hooks are kind of rough, though. Hooks I can deal without except for at the Git boundary, but those other two...
The shame of this G7 meeting is that the Canadian ministers give zero f*cks about Canadian sovereignty, as it applies to Canadians and their online information, financial data, etc. EVERYTHING, in one way or another, ends up going through AWS, or some other offshore BS. ctvnews.ca/politics/article/mo… #cdnpoli #cdntech #FAIL #clowncomputing
Artificial intelligence is likely to take up much of the agenda as industry, digital and technology ministers from the world’s most powerful Western countries meet in Montreal this week.
A quick note that I'm shuttering my patreon account at the end of January. Everyone who has supported me on patreon has been awesome, but I'm simplifying things and will just stick with YouTube memberships. youtube.com/user/liamerven/joi…
I'm Liam, a totally blind guy originally from chicago, and now living in Oklahoma.
I create videos to educate, inform, and entertain. I play a mix of both console, an daudio-based games.
Seriously. if every fourth post on the local timeline is your automated radio station saying what song it is playing, you are doing Mastodon wrong. Quiet public is your friend.
First of all, what is quiet public? Second of all, sounds like someone needs to tone down their web hook. I know how to do the web hook thing in azuracast but I don't have it pop up every time the darn song changes. Just if the station goes offline, comes back, or I connect or disconnect the encoder. It's a private server with automation because I feel like it
@evilcookies98 Once upon a time, I had an azuracast webhook that posted to Mastodon *every* time the song changed, but I told it to post unlisted to avoid the account spam. I wanted to see how many posts I could rack up on a single Mastodon account.
@Bri I did that on discord and was constantly crying in a corner because I run automation. That reminds me if any of you guys ever need hosting for something like that I can point you to somebody good, and with good prices.
When the dynamic island on iPhone is telling me I have 10 percent battery remaining (and often preventing me from opening the app switcher at the same time for some reason), it has a VoiceOver action named "Collapse." This action doesn't actually seem to do anything.
Is there a consistent way, other than waiting, to tell the island that I get it and would now like it to get out of my face?
Each month we make a financial donation to an #opensource project we highly value or depend upon.
This month our featured project is @thunderbird , a superb and rejuvenated cross platform Email & Calendar client, distributed under #MPL2. We have used it ourselves for most of the past 22 years and would recommend it to our customers.
“Transgender for everybody”, but unironically. As in, spend some time deconstructing your gender and then actively choose what gender roles and expressions you actually do and don’t want to perform. Deconstruct your sexuality too. Deconstruct your religion. Deconstruct anything you were indoctrinated into so you can be authentically you
El hambre que las máquinas no conocen Hubo un tiempo en que pensar era un acto tan visceral como sangrar. Hubo un tiempo en que la duda dolía en las tripas, en que resolver un problema implicaba el…
We're joined by Zipline cofounder / CTO, Keenan Wyrobek. Zipline is on a mission to build the world’s first logistics system that serves all people equally via their fleet of autonomous drones that started in Africa delivering medical supplies and can now deliver packages (up to 8 lbs) directly to your door. They've solved a lot of gnarly technical and regulatory challenges along the way. We go deep with Keenan. We hope you'l...
We're joined by Zipline cofounder / CTO, Keenan Wyrobek. Zipline is on a mission to build the world’s first logistics system that serves all people equally via their fleet of autonomous drones that started in Africa delivering medical supplies and ca…
I've said it before and I'll say it again: someone really needs to ban me from using Native Instruments Damage in pretty much all of my music. It's just not okay. But I love it so much.
@mcourcel Yes indeed; thanks for asking. The piece I'm just about finished isn't up yet, though I'll likely drop it as a single soon. We ended up releasing two albums on streaming services - Spotify, Apple Music, YouTube, etc. - under the artist name "Electronight Alchemy". The latest album is called "Toward hope". The first one was "Love Loss lament".
@Scott @jscholes I mean, it's already inaccessible unless you're using Komplete Kontrol. But also, this is almost too real a possibility to be funny. :)
@Scott I don't *actually* give a shit, but I do find it funny that younger me probably would have been appalled by the idea of doing anything remotely formulaic or repetitive. Now I just do what floats my boat. :)
Apart from the fact that Damage is already from Heaviocity and not NI, there is no reason not to use it. In fact, its probably one of the most commonly used percussion libraries out there, for a good reason, so please, keep using it ;). I personally really like Symphonic Destruction as well. Heavyocity in general is just so fun to use!
From January 20th, 2026, #books released through #Kindle Direct Publishing (KDP) without #DRM will be downloadable as #EPUB or #PDF for users who have purchased them. It will also be possible for no-DRM status to be toggled for existing #ebooks.
It remains to be seen how many #authors and publishers will take advantage of this, whether the files will be watermarked, and if the feature will be available for non-KDP titles.
kfx is sometimes a pain to dedrm but otherwise it's pretty easy to dedrm Kindle books if you plug your device into your computer and use Calibre with the DeDRM plugin
i have sub-divided my list of follows & followers into several Circles. one of these is for all my followed #newsbots. i have found to my chagrin that a basic design feature of #Friendica is working against me here. atm, if i read a particular newsbot's post whilst in my "Bot" circle, then decide to quote-boost / quote-share it, with some pertinent commentary by me, it's a complete waste of time... the default Friendica permissions / visibility feature makes my resultant post "Locked", ie visible only to members of this circle, ie, a heap of bots. no actual human will ever see it.
[how] can i avoid this? how can i manually allow any/all posts / replies / boosts / shares that i do from this circle, be visible across all the fediverse?
when circles comprise actual human accounts, this baked-in Friendica privacy concept makes sense. it is however actively unhelpful, indeed plain obstructive, for the specific case of a bot circle.
When you are in a circle, then the default permission for the post is set to that circle. Means: You have to change the posting permissions before posting to avoid this behaviour.
@Michael 🇺🇦 @Droppie [opensoc] @Friendica Support that's easy enough to do from the web client from the permissions(?) tab. Is there a way to do so from other clients, like Racoon?
@Michael 🇺🇦 oh Michael, i feel so stupid! I did already look there before posting, but somehow entirely failed to notice i could manually change it. i've now done a test
this is an experimental quote-share from my #Friendica #newsbots #Circle, whose default Permissions disallow anyone outside this Circle being able to see or access it. i have now done a one-off flip of the Permissions, just for this individual post, to Public, so...
Hello? (Hello, hello, hello) Is there anybody in there? Just nod if you can hear me Is there anyone home? Come on (Come on, come on), now I hear you're feeling down Well, I can ease your pain And get you on your feet again Relax (Relax, relax, relax) I'll need some information first Just the basic facts Can you show me where it hurts?
I for one don't understand how the circles feature is supposed to work. I don't think it existed when I started using Friendica years ago, this was one of my goals for the holidays was to play with it more and update the documentation.
Circles existed even before I started to use Friendica. But then they had been called "groups". We renamed them some time ago to not get confused with the groups that had been called "forums" back then.
In the hours following the recent earthquake off the north-eastern coast of Japan, Google's #AI "displayed outdated information and wrong answers, including the wrong magnitude of the earthquake."
"False information must not be displayed -- even once -- in the field of disaster response, where lives are at stake"
@thunderbird I love your work, but pushing this tab into Thunderbird, causing Thunderbird to pop in front of my current meeting window while I’m presenting and have to then scramble to close it and get my notes visible again, is _utterly_ unacceptable.
We are so sorry for this focus-stealing bug, and we're actively looking into this. The appeal should open in a new tab, but that tab shouldn't ever steal focus and do things like disrupt your presentation. If you'd like, you can report on/follow the bug here: bugzilla.mozilla.org/show_bug.…
persons 2, 3 and 4: You *really* should use Wireguard. It's more secure, way faster, and easier to deploy. Pretty much everyone has dropped it in favor of Wireguard. Here are a bunch of valid reasons why you shouldn't use OpenVPN in favor of Wireguard, with resources explaining things in great detail in ways that are easy to understand.
Person 1: Yeah, cool. But I'm going to deploy OpenVPN anyway.
Sometime later, in a completely differenct place:
Person 1: I have just deployed a new self-hosted OpenVPN. Wow, this is really slow and kind of annoying to deal with.
Persons 5, 6 and 7: You should have used Wireguard.
There are also just way more resources and scripts and Wireguard is much easier to set up. I remember following a two-part guide to deploy OpenVPN and create shell scripts for certificate management. I still screwed up somewhere and never got it working, until someone made PiVPN.
Native WireGuard still doesn't offer an easy way for clients to auto-join a network without having to do per-client configuration on the server.
It's great for site-to-site deployments and such, replacing IpSec more than OpenVPN, but for traditional hub-and-spoke, you need non-standard solutions on top.
@klittle667 Well, yeah, but there are some very specific use-cases where native Wireguard is still preferred. I love Tailscale myself, but when you need every part of the system under your complete control (yes, I know about Headscale) then native Wireguard might still be a better fit.
Chatmail provides FOSS infrastructure for interoperable, secure, speedy and reliable end-to-end encrypted messaging. Check out clients as Arcane Chat, Bots or Delta Chat today!
Everything Trump does is a metaphor. He knows only how to destroy, not to build.
Destroy the East Wing with no new design or blueprint. Destroy US healthcare without having a replacement. Destroy the US economy without having a fix. Destroy FEMA with no plan for emergency response. Destroy environmental regulations without any new protections.
Me: how do I get this text off of this FreeBSD 4.11 VM ? It's too old for me to ssh to it, it doesn't have netcat, and I can't install any software very easily at the moment
This is a repository of apps to be used with your F-Droid client. Applications in this repository are official binaries built by the original application developers, taken from their resp. repositories (mostly Github, GitLab, Codeberg).
Borris
in reply to Adam MacLeod • • •Adam MacLeod
in reply to Borris • • •Borris
in reply to Adam MacLeod • • •