Live data from Hacker News

5G and Beyond for Contact Tracing

zenodo.org

81–90 of 95 posts

Re: 5G and Beyond for Contact Tracing

#81
post #52

I can't trust the authors when they start with a false premise: "The COVID-19 pandemic has suddenly raised the need for technological solutions capable to trace contacts of people and provide location-based analytics". No, it has not created any such "need". "It's for yours and _everyone's_ health" is the most common premise used to reduce privacy and increase control over people. No!

Yes. There is a demonstrable lack of need. Contact tracing has not worked and will not work for covid. As new variants emerge that spread faster, this will become even more true. The best measures against covid appear to be old fashioned common sense practices. Pay attention to yourself and if you’re sick stay home. Don’t cough, sneeze or breath on each other unnecessarily. Stand a bit further apart. Be outdoors more…

> Contact tracing has not worked and will not work for covid.

As an Australian, I can say with certainty contract trace does work for COVID. It is one of the tools Australia uses to keep it's cases near 0.

There are caveats of course. Contract tracing's big brother is the lock down, where everybody is forced to isolate. It's the big brother because all contact tracing does is limit the lock down to those who have been exposed. They are found by tracing the contacts of existing infections.

Contact tracing achilles heel it is labour intensive. It works beautifully, right up until the number of new infections overwhelm the labour force available to do the contact tracing. So in Australia contact trace is always paired with an initial lock down, to give the contact tracers a head start. Delta in particular with its R0 of 7 can easily overwhelm the man power before within a week if you don't stop it dead. (This has now happened in our most libertarian state, forcing said libertarian state into the longest and invasive lock downs we've seen in Australia.)

In Australia contact tracing was done with no technological help whatsoever at the start. It was all via interviews. Now we have QR codes everywhere, which you are by law required to scan when directed to do so by the Chief Medical Officer, which is invariable while there is local transmission. This is done as a force multiplier of course - it both raises the number of infections needed to overwhelm the system, and speeds up shutting down the outbreak. Time is money, and in this case the money is measured literally in billions.

The sad and bizarre outcome is those QR codes are very invasive - we had a much better solution available. If you visit places that must have QR codes (which broadly is any business), you by law must take a smart phone with the contact tracing app installed. You have to enter all your details into the contact tracing app. When you scan the QR code it uploads the exact time and place to government servers. The uploaded information has already been leaked beyond it's intended use.

In contrast the original bluetooth tracing app (and probably this 5G solution) was designed to never leak your position unless you become a close contact of a person contacted by the contact tracers. They were far more private and resistant to abuse. Yet Apple and Google effectively made them impossible to use, due to erstwhile "privacy considerations". And in doing so, forced the adoption of solutions where, as far as I can tell, privacy was never a design consideration.

Re: 5G and Beyond for Contact Tracing

#82
post #23

Earlier quoted context omitted.

Ah, the no true scotsman fallacy. If only we had destroyed our health and economy harder , it would have worked!

That's not what a no true Scotsman fallacy is, GP literally claimed that other countries did in fact take the necessary precautions and did better than us, which is would be a true Scotsman.

It's exactly what it is, look:

"Lockdowns work!" "Lockdowns didn't work in the West." "Those were no true lockdowns!"

This is essentially the summary of the parent comment.

Re: 5G and Beyond for Contact Tracing

#83
post #16

Earlier quoted context omitted.

You have to be kidding. This strategy of lockdown, wear a mask, contact trace and wait for vaccines clearly isn't working. Are you suggesting that it would work if the government could track our locations better?

Not saying we should follow their lead, but China managed to do exactly that with stricter lockdowns, masks and contact tracing without regard for privacy. After the initial wave China was covid free, people could travel freely and the economy was booming.

Nobody can look at the official covid numbers reported from China trended over time and believe they are true.

Re: 5G and Beyond for Contact Tracing

#84
post #82

Earlier quoted context omitted.

That's not what a no true Scotsman fallacy is, GP literally claimed that other countries did in fact take the necessary precautions and did better than us, which is would be a true Scotsman.

It's exactly what it is, look: "Lockdowns work!" "Lockdowns didn't work in the West." "Those were no true lockdowns !" This is essentially the summary of the parent comment.

Except the lockdowns in china were, according to GP, true lockdowns. Thus, there are true lockdowns. The no true Scotsman fallacy isn't just any time you clarify yourself.

Re: 5G and Beyond for Contact Tracing

#85
post #52

I can't trust the authors when they start with a false premise: "The COVID-19 pandemic has suddenly raised the need for technological solutions capable to trace contacts of people and provide location-based analytics". No, it has not created any such "need". "It's for yours and _everyone's_ health" is the most common premise used to reduce privacy and increase control over people. No!

Yes. There is a demonstrable lack of need. Contact tracing has not worked and will not work for covid. As new variants emerge that spread faster, this will become even more true. The best measures against covid appear to be old fashioned common sense practices. Pay attention to yourself and if you’re sick stay home. Don’t cough, sneeze or breath on each other unnecessarily. Stand a bit further apart. Be outdoors more…

> As new variants emerge that spread faster, this will become even more true.

I thought opposite. Contact tracing should work well for strongly spreading (without much contacting) variant because it just trace nearby user rather than "contact".

Re: 5G and Beyond for Contact Tracing

#86

Earlier quoted context omitted.

It's more practical at this point to start weaning off carrying a phone at all times. Not necessarily going off grid into the woods, just not carrying it unless you'll need it for something specific that day. I'm not sure more tech is going to solve this issue, we're just digging the hole deeper.

The special thing about mm wave antenna arrays is that you don't need to have a phone on you to be identified by the unique fingerprint of your gait. There's enough information density in that spectrum to resolve your identity accurately even if you're not carrying a phone. And there are papers demonstrating how to do just that. https://scholar.google.com/scholar?hl=en&as_sdt=0%2C5&as_vis...

Just noting that some of those articles are about mm wave devices explicitly designed as radars. But if you dig through you'll find articles geared towards "mm wave antenna arrays", which is a basically a keyword for 5G radio hardware.

Re: 5G and Beyond for Contact Tracing

#87
post #79

Earlier quoted context omitted.

Plenty of things in the world work like that. Half assing something often has worse side effects than whole-assing it. If we had 4 weeks of strict lockdowns and then returned to normal we would have both less covid, a better economy, and better mental health.

That's just wishful thinking. The virus is so highly contagious that as soon as the lockdown ended, the exponential spread would start up again. So forget it, most rational people aren't willing to tolerate more restrictions. At this point we need to focus on vaccinating as many people as possible and otherwise just accept the risk.

I'm talking about the established reality of what actually happened, not speculation.

Re: 5G and Beyond for Contact Tracing

#88
Longtime HN lurker here, finally something I can chime in on, as I was involved with the development of the Covid19 tracing app for my country.

> limited deployment – any technology requires a sufficiently high number of users;

Sure, yes, that's true for anything.

> security and privacy concerns – many weaknesses and privacy leakages are documented

I am yet to hear of any such cases, when it comes to tracing apps based on the google/apple tracing API.

> lack of reliability – BLE-based apps are prone to errors due to the usage of signal-strength measure- ments to infer contacts;

We did extensive testing and the only issues we found was that BLE couldn't tell distance very accurately, but still, we could, with high enough confidence, tell that two phones were within 5m of each other or so. There was also the issue of BLE signal going through walls, which meant you might get a false positive with a neighbor, by placing your phone next to a wall, but that's really an edge case.

> data governance – there exists a double standard in governing users’ location data, with fine-grained tracking of device location available only to private companies such as Google or Apple

Plain false. Apps based on the API do not have users location data, only proximity data and even that is only stored on the users device and never transmitted elsewhere. GApple covid19 tracing api is based on the DP-3T tracing protocol, the (very human readable) whitepaper of which can be read here: https://github.com/DP-3T/documents/blob/master/DP3T%20White%...

There was a lot of disinformation regarding contact tracing apps, both from tin-foil hat wearers, incompetent (or maybe corrupt?) state officials as well as competing, centralized (and actually privacy-breaching) solutions. I can still see a lot of this going around.

Re: 5G and Beyond for Contact Tracing

#89
post #85

Earlier quoted context omitted.

Yes. There is a demonstrable lack of need. Contact tracing has not worked and will not work for covid. As new variants emerge that spread faster, this will become even more true. The best measures against covid appear to be old fashioned common sense practices. Pay attention to yourself and if you’re sick stay home. Don’t cough, sneeze or breath on each other unnecessarily. Stand a bit further apart. Be outdoors more…

> As new variants emerge that spread faster, this will become even more true. I thought opposite. Contact tracing should work well for strongly spreading (without much contacting) variant because it just trace nearby user rather than "contact".

faster spread will lower the signal-to-noise ratio and put too much restrictions on scociety. uk already had to dial back it's contact tracing sesitivity.

Re: 5G and Beyond for Contact Tracing

#90

Longtime HN lurker here, finally something I can chime in on, as I was involved with the development of the Covid19 tracing app for my country. > limited deployment – any technology requires a sufficiently high number of users; Sure, yes, that's true for anything. > security and privacy concerns – many weaknesses and privacy leakages are documented I am yet to hear of any such cases, when it comes to tracing apps bas…

> I am yet to hear of any such cases, when it comes to tracing apps based on the google/apple tracing API.

data leaks in the api itself will come to light, rest assured. but as you said, the problems with everyting besides the vendor-api have already caused a lot of privacy havoc.

> ou might get a false positive with a neighbor, by placing your phone next to a wall, but that's really an edge case.

in a literal sense; yes :D but let me assure you that ppl putting their phone near a wall is absolutely not uncommon.

> Plain false. Apps based on the API do not have users location data

again you seem to willfully misconstrue the allegation. let me tell you that just because the tracing app itself does not have precise location data, the problem of creating yet another record on your phone, attesting your behaviour, is not going away.

i take it you are not (yet?) critical of privacy implications of your $dayjob, but try to respect the concerns of ppl that are unilaterly affected by them.

Post reply on HN