Kavin is a bot. Recently flooded the piracy community with the exact same images.

I have one big frustration with that: Your voice input has to be understood PERFECTLY by TTS.

If you have a “To Do” list, and speak “Add cooking to my To Do list”, it will do it! But if the TTS system understood:

• Todo
• To-do
• to do
• ToDo
• To-Do
• …

The system will say it couldn’t find that list. Same for the names of your lights, asking for the time,… and you have very little control over this.

HA Voice Assistant either needs to find a PERFECT match, or you need to be running a full-blown LLM as the backend, which honestly works even worse in many ways.

They recently added the option to use LLM as fallback only, but for most people’s hardware, that means that a big chunk of requests take a suuuuuuuper long time to get a response.

I do not understand why there’s no option to just use the most similar command upon an imperfect matching, through something like the Levenshtein Distance.

What the fuuuuuck

Corporate-owned/non-FOSS software is just mindrot at this point, in the sense that the vast populous apparently cannot fathom that there’s another way than trusting big tech.

Take a look at Kavita for selfhosting bools!

Kagi lenses “focus” the search. So normal web search definitely can contain fediverse results, but with the lens switched on, you ONLY get fediverse results.

Named mine after “objects” from Iain M. Banks’ Culture Novels.

Currently I have:

• gsv
• hub
• excession
• drone

Nice and short, and map roughly to the “power level” of the hardware, so to speak.

And my Yubikeys are named after Special Circumstances agents 😄

Hmm, it’s a bit cheaper here (I think - it’s been a while!), but yeah.

Electricity is expensive here, I think the server setup draws 40€/month, but that is for the entire setup of course, not just pirating-related stuff; plus ~9€/month for the two usenet backbones, and a couple bucks for trackers.

My GF is VERY up to date on this (unfortunately? 😆)

That’s mostly true, although you usually (in my case at least) I am aware of all shows I have available on Jellyfin, and it’s only ones I like.

For discovering new shows to download, things like Jellyseerr actually do give recommendations… No idea how good they are though.

But frankly, Netflix used to recommend a lot of things that sounded interesting on the surface-level, and then turned out to be utter shit. Probably not an entirely bad thing to be lacking recommendations :D

For Non-English ones in my native language. There isn’t a lot of them. AFAICT the one I mostly use is free for a handful of requests/day, but generously lifts that limit in exchange for a “donation” 😄

(It’s only around 20/year)

Edit: and just to be clear, that one Tracker took us from “basically nothing is available in our language” to “literally everything is”, so it’s money well spent.

For a very long time, I was one of the people who kep saying:

“I used to pirate until Netflix came along; now I pirate because of the fragmentation of services; should a good service become available at a reasonable price again, I will be happy to switch back.”

But at some point, that stopped being true. More precisely, my *arr-Stack + Jellyfin setup become so stable, I do no longer really think about it, while also getting better quality content, and often faster than I would due to global licensing shennanigans.

Another factor also is that at some point, we crossed the “enough content to mindlessly scroll until we find something to watch” barrier, which my GF actually kinda missed from Netflix.

The crazy thing though, is that we pay actual money for this: hardware cost; electricity; access to usenet trackers and two usenet backbones. All in all, I do not think it’s cheaper than getting Netflix+Prime+Disney.

It’s just better. And we will not be switching back, ever.

I’m sorry, but have you never had actual Cheddar?

skill issue

Is it? If it wasn’t printed on the bottom, would you really be able to guess Ctrl+X, Y, Enter any easier than colon, q, Enter?

No

Nano is easier to get into, but far more limited.

No. I am not saying that to put man and machine in two boxes. I am saying that because it is a huge difference, and yes, a practical one.

An LLM can talk about a topic for however long you wish, but it does not know what it is talking about, it has no understanding or concept of the topic. And that shines through the instance you hit a spot where training data was lacking and it starts hallucinating. LLMs have “read” an unimaginable amount of texts on computer science, and yet as soon as I ask something that is niche, it spouts bullshit. Not it’s fault, it’s not lying; it’s just doing what it always does, putting statistically likely token after statistically liken token, only in this case, the training data was insufficient.

But it does not understand or know that either; it just keeps talking. I go “that is absolutely not right, remember that <…> is <…,>” and whether or not what I said was true, it will go "Yes, you are right! I see now, <continues to hallucinate> ".

There’s no ghost in the machine. Just fancy text prediction.

I’m a programmer as well. When ChatGPT & Co initially came out, I was pretty excited tbh and attempted to integrate it into my workflow, which kinda worked-ish? But was also a lot of me being amazed by the novelty, and forgiving of the shortcomings.

Did not take me long to phase them out again though. (And no, it’s not the models I used; I have tried again now and then with the new, supposedly perfect-for-programming models, same results). The only edgecase where they are generally useful (to me at least) are simple tasks that I have some general knowledge of (to double theck the LM’s work) but not have any interest in learning anything further than I already know. Which does occur here and there, but rarely.

For everything else programming-related, it’s flat out shit.I do not beleive they are a time saver for even moderately difficult programs. Bu the time you’ve run around in enough circles, explaining “now, this does not do what you say it does”, “that’s the same wring answer you gave me two responses ago”, “you have hallucinated that function”, and found out the framework in use dropped that general structure in version 5, you may as well do it yourself, and actually learn how to do it at the same time.

For work, I eventually found that it took me longer to describe the business logic (and do the above dance) than to just… do the work. I also have more confidence in the code, and understand it completely.

In terms of programming aids, a linter, formatter and LSP are, IMHO, a million times more useful than any LM.

LMs give the appearance of understanding, but as soon as you try to use them for anything that you actually are knowledgable in, the facade crumbles.

Even for repetitive tasks, you have to do a lot of manual checking to ensure they did not start hallucinating half way through.

Thanks lol. The thing is, they’re getting away with this. Forcing Freestyle to open their protocol would instantly solve all of those problems, as you could use good hardware with good software, since you’d no longer be locked in.

The app is genuinely better in many ways. It at least allows you to set two thresholds for low/high each, and it does repeat them after a while. At least as far as I remember, it is also better with the “jittering re-triggers alarms” thing. Back when I was using it, you could not mute alarms in advance, but apparently that does work now for up to 6 hours, good on them. BUt this is my rant, so I am ignoring that :D There’s still “nice to have” features missing, for example “if you are trending up, don’t repeat a low glucose warning” and vice versa, but my initial impression was that this might actually work. (The Libre 3 had gotten so bad that I had disabled all alarms permanently for multiple months, so when the G7 came out I immediately switched, but at that point xdrip+ did not support it yet.)

Except - you cannot disable the “very low” alarm, and you can set its threshold no lower than 55 mg/dl. Big deal, I mean, it’s important to be warned, right? Except in many situations, I know more than the app. Imagine this: You are woken at night by your low glucose alarm, and are trending down. These sensors lag by about 15 minutes compared to blood measurements, so even if you drink some juice right now, and KNOW that that will be enough, the “very low” alarm will DEFINITELY still go off. You know, eventually. Maybe in 15 minutes, maybe in 20 or 25. But you will be woken again, and all you can do is try to stay awake until then, or turn your phone off completely (why not just mute the app, you ask? Well, I answer, because the Dexcom G7 app will STOP SHOWING YOU YOUR GLUCOSE LEVELS if you do that, warning you instead that you must let it warn you. It’s great! /s). Of course, turning your phone off means you won’t get any more alarms until you switch it back on in the morning.

Before I end this rant, I need to mention a couple more things. The first is how patronizing Dexcom is towards its customers. Not being able to disable an alarm is one example, but my “favorite” example is this: I had been using their app for a while, and went on holidays abroad. During that trip, my sensor expired (I knew in advance and brought spares). Imagine by AMAZEMENT when I discovered that the app would not let me start a new sensor because my GPS location was not IN MY COUNTRY OF RESIDENCE.

Luckily at this point I discovered that xdrip+ had just released experimental support for the G7, and switched immediately and without problems.

In a similar vein, the app is actually only supported on a handful of devices from the biggest phone companies. There’s a community project from at least the days of the G5 which removes this restriction, and the app does actually run perfectly on basically any phone, you just aren’t allowed to, I guess.

Let’s close this with a look at the hardware. Where Dexcom is moderately ahead of Freestyle in terms of software, they severely lack behind it in hardware. And I get it, this can be hard. They measure every 4 minutes (which is more than enough IMO), and the sensor is a good bit chunkier, yet it only lasts 10 days. Also their adhesive sucks. To this day, you need to put on the giant bandage-patch-thing they include in the box, and even then it’s easy to rip off. They’ve gone through at least 4 adhesive variations, and all of them suck in different ways. But again, I think this is forgivable; the Freestyle Libre 3 is a miracle in terms of hardware, so whatever.

What is NOT forgivable however is making your big-ass, shittily-sticking-to-skin sensor RUBBERIZED and have SHARP, DOWNWARDS POINTING OVERHANGS THAT WILL SNATCH ON EVERY PIECE OF FABRIC IT CAN GET ITS HANDS ON AND RIP OFF IMMEDIATELY.

Why?? Why do this, Dexcom? Why did no one go “actually, only the should-stick-to-skin side of the should-not-stick-to-anything-else thing should be sticky”???

OK, right. But as long as they send spares and make RMAs easy, it should not matter that - WRONG. THEY DON’T. Worse, actually, but for this I have to circle back to the Freestyle.

I mentioned earlier that I had never had a defective unit, and that is true. But once (once) in two years I accidentally ripped one from my arm. I was low on sensors at the time, so I called them. The call went through immediately, and the guy on the end was super apologetic, even though I was also super apologetic for ripping it off and being late with renewing my subscription, hence being low on sensors. He immediately told me they’d be sending two replacements, just in case, free of charge. They arrived the next day. (Actually, they always sent at least two spares with the quarterly delivery, too.)

Back to Dexcom. As mentioned, I started using it very shortly after the G7 came to market. That was a mistake. For the first NINE MONTHS, SIXTY PERCENT OF SENSORS I GOT FROM THEM WERE DEFECTIVE, with 80% of those being dead on arrival, and NOT including sensors that ripped off due to shit adhesive.

I have a day in my log from May 2022 where I put 5 sensors on in a row (they do prick pretty badly, btw), waiting for the warmup period to end for each of them just to read “Sensor defective”. The 6th one finally worked. By this time I had given up on putting the extra adhesive sticker on, since rubbing them off is also a pain because they shred instead of coming off in one piece. So of course, it snagged on my shirt and ripped off within an hour. I cried that day, and I will not apologize for it.

“Hold on, your log? Why do you have that?” I hear the attentive reader asking. Well, let me tell you. Dexcom G7 sensors last 10 days, and the year lasts 365 days. Every 90 days you get your quarterly delivery of 9 sensors, except for Q4, where it’s 10. There are NO spares. When you want to RMA a defective sensor, you have to

• find their pretty hidden support page online
• fill out a questionnaire for every single sensor, including three separate sensor ID numbers (which are either on the sensor applicator or its box, not both, so better keep both), all of your personal info EVEN THOUGH I ALREADY AM LOGGED IN AND YOU HAVE ALL MY INFO DEXCOM, rather leading questions speaking to exactly what YOU were doing wrong when the sensor failed, as well as the date you applied the sensor on and the date it failed.
• a couple of days later you will get a phone call (better hope you hear your phone ringing, they won’t be calling again, and calling them is a two-hour wait!), and a tired service rep will go through ALL of that info you spent half an hour inputting AGAIN. Then, they will calculate how many days were left on each sensor, sum all of them up, then divide by 10 days and send you not a single sensor more than required. 60% of those would inevitably fail again, and the cycle would begin anew. (Also, the first rep I ever had on the phone scolded me and told me too many RMAs would put me on their watchlist. Fuck you, Dexcom.)

…

Oh god this has gotten out of hand, I’m sorry to whoever read all of that.

TL;DR:

• Fuck whoever is in charge of Freestyle’s software team, fuck ALL of Dexcom, but especially the product designer who I’m sure will LOVE my idea for a rubberized dildo with snaggy edges.
• Use open source software with medical devices.
• Please mommy EU whip medical companies into compliance.