Follow Slashdot blog updates by subscribing to our blog RSS feed

 


Forgot your password?
Close
typodupeerror
×
Medicine Linux

Linux Foundation Creates Its Own Versions of Apple/Google Coronavirus Tracing Apps (businessinsider.com) 50

Posted by EditorDavid from the additional-apps dept.
"The Linux Foundation has formed a new group to provide public health authorities with free technology for tracking the spread of the coronavirus and future epidemics," writes Business Insider. Launched in July, the group has already released two apps "that notify users if they've been in contact with someone who has tested positive with COVID-19." Since these apps are open source, people can contribute code and customize them, allowing regions with similar needs to collaborate, general manager at Linux Foundation Public Health, Dan Kohn, told Business Insider. Developers that want to build an app off these projects can access or download the source code.

These apps take advantage of technology launched by Apple and Google, which can be integrated into any app, that uses Bluetooth on people's smartphones to track who a user has been in close proximity with, without identifying the specific people. If anyone tests positive for COVID-19 and uploads that information to a database run by a local public health authority, any user who has been in close contact with that person will get a notification through their app saying they may have been exposed — again, without identifying who has COVID-19. If someone knows that they may have been exposed, they can either self-quarantine or get tested.

"Essentially we think exposure notification could have a big impact on reducing the overall rate of exposure," Kohn said. An Oxford University study in April said that if about 60% of the population used a contact tracing app, it could grind the diseases spread to a halt. Researchers on the team also found that digital contact tracing can cut down spread even at much lower levels of usage.
This discussion has been archived. No new comments can be posted.

Linux Foundation Creates Its Own Versions of Apple/Google Coronavirus Tracing Apps More

Linux Foundation Creates Its Own Versions of Apple/Google Coronavirus Tracing Apps

Comments Filter:

  • Comment removed (Score:3)

    by account_deleted ( 4530225 ) on Sunday November 01, 2020 @11:49AM (#60671820)
    Comment removed based on user account deletion
    • it would've required the two parties to work together on messaging and science and, well, that's pretty much a no-go at this stage [thehill.com].

      I think the best we can hope for at this point is to get somebody in charge who takes the pandemic [bloomberg.com] seriously [cnn.com] and do what the epidemiologists suggested years ago [pbs.org].

      Trouble is we're still going to get a mountain of push back from anti-vaxxers and now anti-maskers.... Both of which will mean it's difficult if not impossible to stamp out COVID. By all accounts immunity only las

      • Re: (Score:1)

        by Anonymous Coward

        By all accounts immunity only lasts 3-4 months

        LOL, no. You read clickbait headlines like like this, [marketwatch.com] then come back here and blab fear-mongering nonsense without actually reading the article or the study it references.

        There are many [nejm.org] studies [cell.com] suggesting immunity lasts longer than 3-4 months, some say far longer.

        All you're doing is spreading misinformation, probably due to your irrational fears about the virus.

        • The conclusion of that study you linked to was âoe we demonstrate that antibody production persists for at least 3 monthsâ. Sure letâ(TM)s all be hopeful that immunity lasts years even though it doesnâ(TM)t for Influenza. But we donâ(TM)t know. And we donâ(TM)t know if we will ever have a highly effective vaccine against this virus, so letâ(TM)s not shoot down people for being concerned about catching it and the possible ill health or death that may result.

    • Re: (Score:2)

      by shanen ( 462549 )

      Good point and I should have mentioned trust in my longer suggestion below. However I was trying to stay in contention for Second Post (after the flaming troll) and I failed.

      So the way it works is that quick and stupid wins the race. Three sentences might place as #2. Five paragraphs finishes fifth, and that was without reading the preview, just posting as it stood.

    • Certainly mistrust of government, Google, and Apple plays into this.

      The published *design* of the app, the spec, is a brilliant piece of work to preserve privacy while being able to tell if you've spent time near someone who turned out to be infected. Really pretty impressive - it doesn't allow anyone to know where you've been or who you've been around, while letting you know if you're at risk. From a cryptography and privacy academic standpoint, it's really nice work.

      I can certainly understand why people w

  • What I really wanted to know... (Score:5, Informative)

    by bento ( 19178 ) on Sunday November 01, 2020 @11:58AM (#60671848) Homepage

    So this article contained a linked to a BI article which mentioned a Linux Foundation initiative which had been running since July. The things I was really interested in were link to the actual initiative and the code for their projects.

    • Another thoughtful but brief response that didn't win FP. Congratulations on the possible third place show? However it mostly reminds me of another aspect I probably should have mentioned in my longer comment (lost somewhere below).

      In terms of a solution approach, I'm seeing the Linux Foundation as mostly acting as a funding mechanism here. But as a possible donor, I would NOT donate to help fund this project, whereas I might donate to help fund the simpler solution approach I described below. How about you

      • Re: (Score:1)

        by bento ( 19178 )

        I'm not sure I'd be willing to donate to such a thing, but this is the kind of thing I'd want to see developed as part of government/private/public research initiatives and would be perfectly willing to accept my tax dollars being spent on it. Any approaches including the simpler one that you suggested down thread.

        I don't see contact tracing as useless... though I think the ship sailed on it being useful for the current outbreak back in January. We needed to be quarantining travelers, or at minimum doing co

        • Re: (Score:2)

          by shanen ( 462549 )

          I feel like you're raising a new issue, and I think the only "solution" there would be some kind of mandate to basically force people to use the contact-tracing app. Maybe a carrot instead of a stick, but I think it would have to be a pretty expensive carrot.

          The exposure-counting app is motivated from the other side, appealing to each user's personal self interest. Maybe I'm too selfish and narrow-minded? I just don't trust other people to warn me about their diseases, but I trust my own smartphone to warn

          • Re: (Score:1)

            by bento ( 19178 )

            I don't see any reason that it couldn't remain opt-in similar to how it's being used now. But I think you're right that you'd need to incentivize turning it on. In a situation like a pandemic though you could make turning it on grant you access to some additional stimulus money or something like that and people could choose. Not sure where you'd get the most bang for the buck... but the census tried to incentivize me to fill out one of their other surveys by mailing a $2 bill along with the survey (which di

            • Re: (Score:2)

              by shanen ( 462549 )

              You raise a lot of interesting points in your long comment, and they deserve more substantive responses than I have time for right now. Sorry, but I'm going to try to limit my replies.

              On the contact-tracing app versus the exposure-counting app, I posted several other comments in this discussion. Here I will just clarify that I was focusing on the simple Bluetooth approach, which handles the distance question moderately well. It obviously works better if more people to turn on their Bluetooth, but I don't kn

    • Other big names [wikipedia.org]

      Including M$ and Facebook.

      Just asking if there's something behind.

  • Inability to learn from mistakes is insane (Score:3)

    by shanen ( 462549 ) on Sunday November 01, 2020 @12:01PM (#60671860) Homepage Journal

    Near as I can tell, this app is basically emulating an approach that has already been tested. Can anybody point at the differences in this version that will prevent it from also flopping? Looks to me like an insane attempt to do the same thing hoping for different results. Actually worse than that, since they are doing the same thing, but more feebly (in terms of resources).

    Which comes first? The problems or the solutions?

    I'll be traditional and go with the main problem: The incentive structure is wrong. You are supposed to go to a lot of hassle to set it up, then it drains your battery, and finally it fails to detect people infected with Covid-19 who haven't gone to the trouble of setting it up. Even if you're using it, the most likely outcome is going to be false alarms as the app cries wolf when you weren't infected (especially if the person with Covid-19 was just wearing a mask).

    My currently favored solution approach: A simpler exposure counting app that lets me adjust my behavior to reduce my risk. Just count all the smartphones that come close to me, when and where, and help me adjust my schedule and movements to increase my social distance. Add little incentives and rewards for taking fewer risks. Maybe some rich companies with big cash reserves will even feel like buying favorable publicity by funding small lotteries for the people who keep their exposure low or who change their behaviors from risky to less risky.

    Cue the incoherent naysayers of Slashdot 2020. Or is it conceivable you have a better idea? I think there are probably plenty of better ideas, but it's barely conceivable one will appear in such an environment as Slashdot has become.

    • Re: (Score:2)

      by shanen ( 462549 )

      What? I can't find the typo? There must be one considering my haste. Should I apologize preemptively?

      Main thing I missed was to note that the exposure counting app should include a white list for "safe" people. For example, your family. Hopefully your family members aren't infected, and you spend lots of time with them that doesn't need to be considered as affecting your risk. Maybe a shake-together registration option for exclusions?

      (But this now reminds me of the bike tracking app that as far as I know st

      • The point about the Linux app and other apps is: it is anonymous.
        If you can/have to white list ppl: it is no longer anonymous.

        Sorry, your idea is just bullshit.

        To change my behaviour, I do not need an app. I have eyes, and know how to use them.

      • It's cute that you think that your family is invulnerable. Anyhow, measuring risk requires precisely taking into account how many people you interacted with. Interacting with 100 different people every day is risky; interacting only with your wife and 3 kids is safer (and the app can tell that they are the same people each day). The app can hopefully include it in the risk factors, even without you telling it explicitly that it's your family.

        • Re: (Score:2)

          by shanen ( 462549 )

          I certainly do not think that my family is invulnerable, but I'll dismiss your rudeness as just your attempt to fit into the atmosphere of Slashdot 2020. In fact I am quite worried about my wife's inability or unwillingness to adjust her habits to reduce her risks in accord with The New Normal. In that specific context, at least the exposure-counting app might help her shift her shopping times to when there are fewer people in the stores.

          Your reply does indicate that you have some understanding of what the

          • Yes, I think that what you perceive as rudeness is just the current style of communicating on Slashdot. Sharp, brash comments often get the point across better and are encouraged by the upvote system. No offence intended anyway. My point is that those 4 contacts are not "background noise". They are part of your level of risk, and simplifying them out does not change it. If you display "20 contacts" or "16 contacts*" with a footnote "*: plus 4 family member", that does not change your actual probability of

            • Re: (Score:2)

              by shanen ( 462549 )

              I would say that they went too far into the outer space of fancy algorithms without considering the real world and the motivations of the people who were supposed to use the contract-tracing app. Or maybe they were projecting from their own concerns about the possible abuses of the personal information?

              Perhaps my speculations about the perverse incentives are incorrect. (Maybe I should have used "perverse incentives" in the Subject back at the start of the thread?) Anyway, at this point I think it's pretty

              • I, for one, have strong concerns about the possible abuses of personal information, and about the effectiveness of contact tracing. Influential experts like Bruce Schneier have voiced them publicly: https://www.schneier.com/blog/... [schneier.com] , https://www.schneier.com/blog/... [schneier.com] .

                For me the answers are "yes" and "no, but even if I wanted it doesn't work on my phone, which has a non-conventional OS". Where I live, it seems that a serious bottleneck in controlling the infection is the number of tests that they can perfo

                • Re: (Score:2)

                  by shanen ( 462549 )

                  Not sure if I need to clarify, but the exposure-counting app is not saying anything about the number of possible infected people you might have been close to. However, it could link to a database for the level of infection in your area, and in that case it would give you a meaningful estimate of your risk. I see it as a standalone data source that helps me reduce my exposure to other people in a time when that is advantageous.

                  In Japan one of the most effective countermeasures was just telling people how man

  • And why does the app ask to get permissions for my contacts, my photos and my emails?
    Something is fishy here!!

  • The Italian official apps, created by a private company on government input, are also open source, actually free software under AGPL 3.0: here is the main repository [github.com]

    Good part of the documentation comes also in English

  • So much for switching to a linux phone as a means to avoid mandatory network tracking.

    King George remains safe.

    • Re: (Score:2)

      by PPH ( 736903 )

      avoid mandatory network tracking

      Cheapo flip phone master race reporting in.

  • I wonder whether anybody else has used that source yet to roll their own.

    It also does not store detailed tracking data on a server but uses an approach that works mostly on the phone.

    It's on github: https://github.com/corona-warn... [github.com]

    • The "store it on your phone" app is also how the New Zealand (NZ) Covid tracer app works.

      Given that NZers do largely trust their government to act sanely with private data, you might be wondering why NZ took this approach. At a guess it has something to do with NZ being in the "five eyes" intelligence group ...

      How effectively the NZ app works is another question, since it requires people to scan in QR codes at places they visit, rather than automatically recording data via blue tooth.

    • I wonder whether anybody else has used that source yet to roll their own.

      It also does not store detailed tracking data on a server but uses an approach that works mostly on the phone.

      It's on github: https://github.com/corona-warn... [github.com]

      The Portuguese app is also open-source. The GitHub repository is this [https://github.com/stayawayinesctec] [github.com].

  • What irks me about the Android versions of these apps is that Google decided that Bluetooth Low Energy (BLE) can be used for location estimation, therefore Android requires that the location system of the OS needs to be turned on. Yes, that part that used to be called "GPS". So although the Covid19 apps that I know about don't use the location data, all other running apps on the phone can use it.
    This includes some Google services that I can't take away the right to read location data from, because they're

Slashdot Top Deals

It is easier to write an incorrect program than understand a correct one.

Close