[Smile] What's something small that really makes you smile?

Forums › Off topic › [Smile] What's something small that really makes you smile?

1 | ... | 268 | 269 | 270 | 271 | 272 | 273 | 274 | 275 | 276 | 277 | 278 | ... | 428

11294 posts

Uber Geek

Trusted

Lifetime subscriber

#3046309 6-Mar-2023 13:59

rb99:
Bournemouth up next. Yes, the team placed 20th in a league of 20. Be worried. And I am a fan of theirs.

Hey, at least it's not Bradford, they would have come 21st or 22nd :-).

Wise

Send money globally for less with Wise - one free transfer up to NZ$900 (affiliate link).

floydbloke

3494 posts

Uber Geek

ID Verified

#3046913 7-Mar-2023 17:01

Operating a 15 tonne digger. If you're ever down Invercargill way I highly recommend Dig This.

Roses are red, that much is true, but violets are purple, not ****ing blue!

Tinkerisk

4157 posts

Uber Geek

#3047556 8-Mar-2023 18:13

OMV 6 on a fast ARM CPU 🙂

- NET: FTTH, OPNsense, 10G backbone, GWN APs, ipPBX
- SRV: 12 RU HA server cluster, 0.1 PB storage on premise
- IoT: thread, zigbee, tasmota, BidCoS, LoRa, WX suite, IR
- 3D: two 3D printers, 3D scanner, CNC router, laser cutter

neb

11294 posts

Uber Geek

Trusted

Lifetime subscriber

#3047561 8-Mar-2023 18:28

The current fireworks display in the Tasman. Pity there's no webcams out that way.

(For people reading this later, don't know how long it'll last but it's been going for an hour or two now).

matisyahu

1615 posts

Uber Geek

Trusted

#3047616 8-Mar-2023 19:16

On a really cold and crappy night, coming home to a hot bowl of soup and fresh bread with garlic butter - being able to cuddle up on the sofa watching a good movie or binge watch a season.

"When the people are being beaten with a stick, they are not much happier if it is called 'the People's Stick'"

neb

11294 posts

Uber Geek

Trusted

Lifetime subscriber

#3047620 8-Mar-2023 19:32

matisyahu:
On a really cold and crappy night, coming home to a hot bowl of soup and fresh bread with garlic butter - being able to cuddle up on the sofa watching a good movie or binge watch a season.

It was a dark and stormy night; the rain fell in torrents—except at occasional intervals, when it was checked by a violent gust of wind which swept up the streets (for it is in London that our scene lies), rattling along the housetops, and fiercely agitating the scanty flame of the lamps that struggled against the darkness. Luckily, I had a hot bowl of soup and fresh bread with garlic butter waiting for me, and Netflix ready to go.

Geektastic

17927 posts

Uber Geek

Trusted

Lifetime subscriber

#3047631 8-Mar-2023 19:56

Rob Brydon live in Christchurch last night.

Very funny.

neb

11294 posts

Uber Geek

Trusted

Lifetime subscriber

#3047771 9-Mar-2023 00:34

This comment on a tech forum, after some guy was a complete ---hole and kept telling one of the regulars that he didn't know what he was talking about:

BTW, XXXX is a Siglent dealer. He has considerable first hand experience with these scopes and a direct line to Siglent to get any hard questions answered. Being flippant with him might not be your best approach.

neb

11294 posts

Uber Geek

Trusted

Lifetime subscriber

#3048140 9-Mar-2023 16:18

The wonder of (snort) "AI". Using an extremely brittle DNN to make a security camera "smart" may look good in the demo but when it's trained to distinguish humans vs. cars, pretty standard for a security camera, then getting repeatedly warned of a human crossing the trigger zone when the human has four legs and a tail shows just how dumb the smart tech really is.

Ed209 may have been a Hollywood cutscene but it'd probably happen with today's tech.

For those who don't know about how this works, neural networks can only match what they've been trained on. Train a DNN on cars and people and it'll classify anything it meets, which high certainty, as either a car or human, whether it is or not. And the only way to extend the recognition to something new is through catastrophic forgetting, drop the model and retrain a new one on { cars, humans, parrots }, after which the model will class anything it sees, with high certainty, as either a human, a car, or a parrot.

Tinkerisk

4157 posts

Uber Geek

#3048259 9-Mar-2023 18:06

@neb

Yes, I used to think that too - I just saw yesterday how AI sorted out 70% of the roughly 100 applicants per day in 3 minutes based on the PDF application documents (cover letter, CV and references in a single file), which are 70% no longer even read by a human in the HR department. The system should not learn at all, but only check.

neb

11294 posts

Uber Geek

Trusted

Lifetime subscriber

#3048260 9-Mar-2023 18:15

Tinkerisk:
Yes, I used to think that too - I just saw yesterday how AI sorted out 70% of the roughly 100 applicants per day in 3 minutes based on the PDF application documents, which are no longer even read by a human in the HR department.

Luckily HR applications aren't a target for adversarial learning otherwise the DNN would be approving, say, convicted serial killers as the perfect job applicants.

Come to think of it, just as CAPTCHAs created a huge underground market for CAPTCHA-defeaters, I wonder if there's enough money in job applications for someone to automate defeating it. The work isn't in the aversarial stuff for which there's off-the-shelf tech but simply configuring it to target whatever DNN the HR is using.

Tinkerisk

4157 posts

Uber Geek

#3048261 9-Mar-2023 18:18

Even small companies can buy it as a central service of a specialized enterprise, don‘t underestimate it. The remaining 30% WILL be read by humans and they can check, if applicants are/were criminals… How stupid AI is also depends on how stupidly it is used.

neb

11294 posts

Uber Geek

Trusted

Lifetime subscriber

#3048262 9-Mar-2023 18:25

What I meant was that once it becomes a big enough target to be worth attacking, and if there's a single centralised enterprise running things it will be, then there'll be a market for defeating it, so every candidate no matter how bad they really are will appear perfectly qualified and it's back to humans checking every one again.

Having a secret knock to get in only works if the bad guys don't know what the secret knock is.

Tinkerisk

4157 posts

Uber Geek

#3048668 10-Mar-2023 14:54

There must be a small, pop-up coffee roastery in the neighbourhood - since recently it smells like freshly roasted coffee on the balcony every weekday’s morning!

neb

11294 posts

Uber Geek

Trusted

Lifetime subscriber

#3049754 13-Mar-2023 19:20

Good to see our dams are going above and beyond in holding all the water that's come down:

1 | ... | 268 | 269 | 270 | 271 | 272 | 273 | 274 | 275 | 276 | 277 | 278 | ... | 428