What is really annoying is that there are a lot of really good data modeling applications, they are just in research areas. Generative AI is absolutely a waste of resources, but a ton of money and energy is spent on that instead of on the applications that are actually bearing fruit.
You can set it yourself but it then verifies your address is real using Bing maps and the database is really lacking. If it doesn’t find an entry it won’t let you enter it. I am told this will be moving to Azure maps soon which I hope is better.
Anyway we are leveraging manual network entries tofind phones at our locations using the WAP bssid or, for ethernet, LLDP but the latter isn’t working. I can show LLDP coming in on a pcap but Teams doesn’t see it - another ticket for Microsoft.
I am in the middle of trying to get e911 functional for Teams direct route calls, based of lis data, my Teams can’t correctly determine the state I am in, much less my current address. It took multiple tickets to get our corporate headquarters to show up correctly instead of an address a half-mile away.
I forsee getting a lot of tickets from this feature.
The gate here is really cool, I remember from my optical classes all the different ways to encode bits on a photon over fiber, I am curious which properties are more and less suitable for this application.
If I want to know when I’m going to die, I’ll ask an actuary like we did in the old days.
Because they want to use antiporn laws to restrict books and other media with LGBTQ content.
When the 8 bit quants hit, you could probably lease a 128GB system on runpod.
Yeah, I mean the AI being shoveled at us by techbros. Actual ML stuff is currently and will continue to be useful for all sorts on not-sexy but vital research and production tasks. I do task automation for my job and I use things like transcription models and OCR, my company uses smart sorting using rapid image recognition and other really cool uses for computers to do things that humans are bad at. It’s things like LLMs that just aren’t there - yet. I have seen very early research on AI that is trained to actually understand language and learns by context, it’s years away, but eventually we might see AI that really can do what the current AI companies are claiming.
Back in the 90s in college I took a Technology course, which discussed how technology has historically developed, why some things are adopted and other seemingly good ideas don’t make it.
One of the things that is required for a technology to succeed is public acceptance. That is why AI is doomed.
deleted by creator
ChromeOS is Linux.
“We are sorry you noticed, we didn’t think anyone would read all that.” -Adobe, probably
Gmail is probably the hardest one to kick. I’m fine with paying for an email service if it’s functional and doesn’t siphon my personal data, but finding a quality trustworthy provider and then migrating 20 years of data to it seems so overwhelming.
They say that, but when Ken Paxton subpoenas them they will say they have no choice. It would be better to use an app that doesn’t store this data server side at all.