Posts by Dan 55 15423 publicly visible posts • joined 13 Jun 2009
Lack of data integrity checks, no way to encrypt contents, the usual CSV madness with fields with dates or spaces in them. To start with.
As a way of setting up a testing reporting method between Serco and PHE in record time with minimum contract penalties for the government to pay then yeah, it works. Wonderful. That doesn't mean it should be used any more than strictly necessary.
And whoever signed off on that contract should he taken out back and shot pour encourager les autres.
If I take my phone to the recycling centre, I don't have any say in what they do with it. In fact, if it's good enough to sell on and they do that then they've done their job, they've recycled it spending no energy breaking it down and generating no waste.
Apple expect to be able to tell people what to do with devices that have left their control. Someone ought to break it to Tim that they're not on his spreadsheet any more.
Most end up going to the third world to get recycled, and they don't have the Apple Approved iGlue Heater.
My mistake, so you were discussing something different (discrimination) where I assumed your meaning to be racism. My mistake.
From the Cambridge dictionary:
treating a person or particular group of people differently, especially in a worse way from the way in which you treat other people, because of their skin colour, sex, sexuality, etc.:
It is indeed your mistake.
Discrimination is the norm that people are different. Racism is to believe in superiority due to difference.
So, er, racism is a kind of discrimination, then? Yes. Are we there yet?
If it is so easy then go ahead and show all these boffins who have tried and failed how to do it. Remember this must work with varying quality of camera with highly varying surroundings. how good are these mask trained AI's with colour? If its good enough maybe half the face needs cutting off during training.
They've tried and failed to do it because they've shoved in a dataset which doesn't have enough examples, not checked the results enough, and put it into production.
The other possibility is that nobody does the exam because of covid restrictions and so screw the lot? Or is it best for the majority to test with this technology and those it cant work with have to do it a different way? Covid wont wait for more AI testing, its here. And the education system needs solutions, even imperfect ones.
That penalises people because they are black. Well done you. You do understand where this leads, don't you? Actually you're probably blissfully unaware, and I'm not willing to argue about it for the rest of the day. As always, the George Bernard Shaw quote applies.
I did not substitute, as the word "racism" is so loaded and it's difficult to have a rational debate about it, I tried to explain in a way that shows that as discrimination does not require intent (because that argument is generally accepted) neither does discrimination concerning race (aka racism) require intent. It's known as "casual racism", look it up.
As someone else pointed out, he had a beard. I wonder how it would cope with some of the fashionable moustaches.
Facial recognition algorithms are now being trained with masks to recognise people while wearing masks. This itself should give you a hint that solving problems with facial recognition algorithms and race is not impossible and the problem is not down to race itself, so we can't just shrug our shoulders and say "oh, it's because they're black, this is impossible to solve and they should just put up with their life being shittier because of it" or "he has a beard" or whatever. The algorithm must be trained with more images of black people's faces, more images of bearded faces, and more black people's faces with beards so that it reflects real life.
After many, many examples of algorithms having difficulty with black faces, perhaps it's time to start thinking that, maybe, algorithms have a harder time distinguishing black faces because it is an objectively harder problem?
No, it's time to start choosing better datasets.
And, here's the interesting question: what happens if it turns out that these algorithms will always be worse at distinguishing black faces? You know, like humans are.
Fuck me, we have to start from the very basics:
Imagine if you knocked up a website without accessibility which stopped blind or disabled people from using it. This is discrimination. You might never have thought to do add accessibility to the website but it was your job to do so and you didn't do it.
Now imagine you're training an algorithm and choose a dataset which wasn't large or varied enough for it to do the job required, and produces suboptimal or wrong results for non-white people. It was your job to ensure this didn't happen but you didn't. This is also at the very least discrimination.
Light is obviously not racist, you bleeding fool.
Samsung washing machine: you have to take the whole damn thing apart to replace the waste nozzle. It goes in the back, takes a route round the most inaccessible parts of the machine, and instead of being clipped to whatever it's connected to like all other tubes in the machine, it's glued to its destination.
As the other poster said: Don't. Ever. Buy. Samsung.
If it gave you the worst picture then it wasn't RGB through the SCART cable, it was component or S-Video.
Many TVs don't have SCART since France dropped it from being mandatory in 2015. You may rejoice, until you decide to plug something in which needs an RGB connection (game console, retro computer) then you may see things in a different light.
YPbPr needs a converter from RGB, cheap Amazon SCART RGB->HDMI boxes are laggy, OSSC is good but not cheap. What the console/computer uses internally is RGB, what the TV uses internally is RGB, anything other than simple RGB connection will give a worse picture or lag or both, with the exception of OSSC.
You could always install alternative app stores on Android, however now they're enforcing Google payments for the Play Store apps. Anyone who wanted to be on an alternative app store already was. Anyone on Google Play not using Google payments has just been screwed.
Been done but I'm not sure if it's still available:
Optimus Maximus Keyboard review (Cherry ML) (Chyrosran22, 11 minutes).
Fat enter is part of the ANSI keyboard layout. You also have one key less and no Alt-Gr key, just a copy of the Alt key from the other side.
The UK's keyboard layout is the ISO layout which is also used in many other countries. The thin enter allows one more key and the Alt-Gr key is useful as a selector for a third symbol on the key or as a compose key.
So you're really hating on most of the world's keyboards, except the Netherlands for some reason.
Facebook fact checking is too little too late, for every single fake video, and Zuckerborg knows it:
We fact-checked a fake viral video with a Trump supporter (CNN, 4 minutes)
When the CNN interviewer shows his interviewee proof that a video he watched was fake (that WaPo article he showed him would never even have shown up in the guy's Facebook feed), he says won't change the opinion he formed while watching that fake video. Repeated perhaps millions of times. That is the problem we are facing.
The law was badly specified anyway, the customer can scan the QR code in any app they want (doesn't have to be the NHS app) and in any case it says the business must request that the customer scan the QR code but it doesn't say the customer has to comply.
It seems designed to get moderation policies pared down to the minimum as otherwise if an example is found of a social network not upholding their policy then they could be sued for it. It also seems to make it harder for social networks to remove posted URLs which point to other social networks or websites.
In other words, would Gab have any problem with this? No, because they don't have any moderation anyway. Would Twitter? Certainly.
I think Google and Apple are working on Covid apps automatically generated from a relatively simple configuration file, in the hope that England will have one ready by the time the pandemic is over.
"I said I want a cut of the money to the US government because we made the deal possible."
This means:
Trump said Trump wants a cut of the money to the US government because Trump & friends made the deal possible.
And it's been in the news lately that his campaign has been hemorrhaging money.
Hopefully that makes it a bit clearer what he wants out of this.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
