WYGIWYG
I run a pi-hole on a pi 3 and another in a container in docker. Something rarely goes wrong with both and I have a script that sync them.
I replaced their google with searxng, but in the end, they needed ads for their free to play games, so I had to turn it off for them.
First: Every product we build must give people agency in how it works. Privacy, data use, and AI must be clear and understandable. Controls must be simple. AI should always be a choice — something people can easily turn off. People should know why a feature works the way it does and what value they get from it.
That’s a good idea to put first. Of course, like do no evil, priorities change, so we’ll need to keep a close eye on this.
Second: our business model must align with trust. We will grow through transparent monetization that people recognize and value.
Transparent is good, but if he things he’s going to add value to monetization, he’s smoking crack. There’s nothing we want from a browser that’s not already provided by a plugin.
Third: Firefox will grow from a browser into a broader ecosystem of trusted software. Firefox will remain our anchor. It will evolve into a modern AI browser and support a portfolio of new and trusted software additions.
Nobody wants that. We already had all we wanted from them in trusted software.
I pulled down some random ass pop song, and it ended up being some guy noodling on a keyboard playing a half assed version of the cantina song from Star Wars. I set it to my phone’s ringtone. No idea when that mp3 was long, it was a LONG time ago
I’m up to 45TB of actual used storage. I just want another tape analog. I want inexpensive, slow, long-term storage I can move off-site easily. This paying double to keep disks around and then moving them in boxes is just bad, and online storage is stupid expensive at those sizes.
Was running on Backblaze for years until they screwed around with my client enough that I can’t backup my NAS reliably. I’m not a company, I’m not going to pay the cost of my disks every year to store the content of my disks.
I’ve been considering for a few years standing up a 2u box in colocation.
Shit’s getting extra shitty everywhere. Its like a couple of countries went a little fash and every politician across the globe went FUCK YEAH, LET US IN ON THAT SHIT!!!
wow, sign me up for a couple of dozen terabytes of that!
I also remember people burning pitts on scotch tape, then rolling it up and reading it in 3d :)
Even if the cats are out of the bag, i’d like to remove as many privacy leaks as possible.
this is really no different.
For some, yeah. But with the way things are running in politics, we’re 6 inches away from Sony demanding to see our viewing logs to make sure we’re not pirating.
Today’s data collection is tomorrow’s felony conviction.
Hmm… What’s eating Windows’ lunch? Oh “Unknown” and “Other”. Cool.
If the data is coming from statcounter, it’s kind of vague where they’re pulling it.
It’s strange to see IOS and not Android, maybe because it’s fractured into so many sources
It could also be anti ad/tracker stuff.
They’re in order in this highlighted column:
I read somewhere Samsung does it too. Not surprised at all if they all do it, along wiht the streaming sticks.
prints article out
places it on an overflowing, ancient pile of documents of promising, science proved data storage methods that haven’t made it to public use yet
It’s been a problem for a long time before AI.
I have a Dell with UHD+Nvidia, took me a while to get Prime working to switch video cards. Even on UHD, it could do basic Steam games and Minecraft if you didn’t have high expectations.
Jesus, I need to read that source code.
they can’t manage to sell a single dollar worth of their product.
Ohh don’t worry, that’s not how this works :)
We’re still in the venture capital stage. The companies are circle-jerking, paying each other off with venture funds and stock splits. They don’t need to be making money at this point because they’re already getting everything they ask for.
Those $50-$200 packages from all the big companies are just there to get people used to the idea. They’re making all their money on selling each other useless support chatbots and horrible phone systems claiming they can reduce their staff by half. Well, they could always reduce their staff by half, customers have had to deal with shitty wait times for years.
You’ll pay for AI by the prices of your software rising. Those costs are absorbed and passed on to you as micro-transactions inside your actual subscriptions and payments.
Once they managed to get the AI intertwined in every system out there, they’re free to collude as a market and raise prices slowly. AI will be the cost of software inflation and hardware shortages that make anyone with a datacenter or enterprise hardware manufacturing capacity very, very rich.
It could even be that in the end, this isn’t a bubble, it’s just a grift and it never pops, but because so expensive that your average person can barely eat if they expect to use software tools for their work.
It COULD help the average person, but we’ll always fuck it up before it gets to that point.
You could build an app that teaches. Pick the curriculum, pick the tests, pick the training material for the users, and use the LLM to intermediate between your courseware and the end users.
LLM’s are generally very good at explaining specific questions they have a lot of training on, and they’re pretty good at dumbing it down when necessary.
Imagine an open-source, free college course where everyone gets as much time as they need and aren’t embarrased to ask whatever questions come to their minds in the middle of the lesson. Imagine more advanced students in a class not being held back because some slower students didn’t understand a reading assignment. It wouldn’t be hard to out teach an average community college class.
But free college that doesn’t need a shit ton of tax money? Who profits off that? we can’t possibly make that.
How about a code tool that doesn’t try to write your code for you, but watches over what you’re doing and points out possible problems, what if you strapped it on a compiler and got warnings that you have dangerous vectors left open or note where buffer overflows aren’t checked?
Reading medical images is a pretty decisive win. The machine going back behind the doctor and pointing out what it sees based on the history of thousands of patient images is not bad. Worst case the doctors get a little less good at doing it unassisted, but the machines don’t get tired and usually don’t have bad days.
The problem is capitalism. You can’t have anything good for free because it’s worth money. And we’ve put ALL the money into the tech and investors will DEMAND returns.
damn good idea, good on you!
Nobody was going to stay on 3.11
NT really lacked drivers and was super expensive.
We used 95, skipped 98 until se came out. We did a fair amount of Win2k, XP was pretty rough until service pack 2. When Serviceback 3 hit, we eradicated everything else on the desktop.
A lot of the industry artists are at the very least using AI to screw around with concept art for references. The kind of stuff where they used to use google to search. One of my friends fed a service a fairly raw hand-sketched drawing, told it how to finish it off, then asked it to put it in different poses at different angles, then used that to hand-make the character into 3D.
There are, of course, many artists who wouldn’t touch any of it with a 10-foot pole.