Category Archives: Smart Phones

Privacy: Why you should stop using Google Chrome browser

I use Firefox (mostly) or Brave. I use Google Chrome only to check GMail (no longer my primary email) occasionally to avoid logging in to Google services while online with the other browsers. I have done everything I can to minimize use of Google services.

As well as collecting your data, Chrome also gives Google a huge amount of control over how the web works

Source: It’s time to ditch Chrome | WIRED UK

Every time Google was caught making a privacy mistake, their errors were on the side of scooping up way too much. Google’s street view cars, driving all roads in the world, were scooping up all Wi-Fi traffic including unencrypted transmissions. When caught, Google said it was due to a software error. Which is unbelievable as it meant they were collecting terabytes of excess data and pretended no one had noticed this.

Google, contrary to their original stated purposes, is evil.

Public health confirms field is based on belief systems

Who needs data when you’ve got beliefs?

Though they don’t have hard numbers because of security safeguards, they believe plenty of lives have been saved.

Source: Wash. state’s COVID contact tracing app ‘WA Notify’ showing its worth | KATU

There is still no evidence that phone-based Bluetooth tracking has accomplished much. But there is a belief in some quarters; while others have said the tech found little that would not have been caught with other methods.

The end of smart phone based contact tracing apps

As explained in detail several times last year, smart phone-based, Bluetooth-based contact tracing apps were never likely to be useful.

Effectively with this announcement today, Oregon has abandoned it’s smart phone contact tracing apps, joining half the US states that chose to not pursue these apps.

Oregon pausing exposure notification app

This week, OHA decided to pause the ongoing planning of the Exposure Notification (EN) application project rollout for Oregon to focus on vaccinations and other priority efforts to end the COVID-19 pandemic.

Over the past six months, OHA has benefited from discussions with local public health departments and other partners, which highlighted the benefits and costs of any early notification app, including the intensive efforts state and local health officials would need to undertake to promote the app and address likely gaps in its adoption across Oregon’s diverse communities, as well as the added contact tracing demands full adoption would place on county public health staff.

OHA appreciates the feedback agency staff heard from our partners working in local communities. State health officials reached the decision after consultation with Gov. Kate Brown’s office.

OHA Public Health Director Rachael Banks said: “Approximately two dozen states have chosen not to deploy smartphone-based apps at this time and instead to rely on other tools to stop the spread of COVID-19. Oregon is focused on building trust with people in communities across the state to get all Oregonians vaccinated and sustain the other COVID-19 prevention practices, such as wearing a mask, staying physically distant and limiting the size and frequency of indoor social get-togethers. These strategies have prevented more than 4,000 COVID-19 deaths in our state. We’ll continue to prioritize these approaches because they remain our best bet to end the pandemic.”

Source: Oregon reports 33 more COVID-19 related deaths, 544 new cases – KTVZ

Google introduces mask filter for Android photos

April 1 – Google has updated its Android OS default camera app with a new “mask filter”. The apps uses AI techniques to automatically identify face masks and then filters them out of the photo, using AI to automatically generate the missing face area.

The update is available on all Android phones as of April 1.

Reviewers say the mask filter works as effectively as masks work in real life, which is to say, not that good. Google suggests applying the mask filter twice for more effective mask removal.

Boondoggle: Oregon’s smart phone contact tracing app failure

Smart phone contact tracing apps are a technology whose time never arrived:

Months after agency officials said they were actively “assessing the results” of a trial conducted at Oregon State University, state officials now say they have no final documentation about efforts to evaluate the project.

The decision by state officials not to formally assess the technology helps explain why the project has been delayed for months, leaving Oregon as one of just four states along or west of the Continental Divide that has failed to adopt the technology.

Source: Oregon can’t produce written evaluation of long-delayed COVID-19 exposure app – oregonlive.com

Oregon is incapable of doing IT projects anyway. Cover Oregon was the $450 million failed “health exchange” that never enrolled anyone before they pulled the plug.

One year later: Smart phone, Bluetooth-based contact tracing apps remain AWOL

One year ago, the tech sector jumped in with a plan to develop smart phone based contact detection apps. These would use Bluetooth to estimate potential contacts with someone later testing positive for Covid-19.

I predicted at the time (see past posts) that this technology was not likely to be successful for many reasons.

Here we are, one year later, and we are only now developing testing criteria for these apps.

The Science and Technology Directorate (S&T) of the Department of Homeland Security (DHS) awarded $959,305 to the University of Washington’s Applied Physics Laboratory this week to create testing criteria for a COVID-19 digital contact tracing app. The award was granted … Read More »

Source: DHS taps University of Washington to create criteria for COVID-19 digital contact tracing app testing – Homeland Preparedness News

The media gave much attention to alleged privacy issues but that was never the problem – the problem was the nature of the technology and the high false negative and false positive rates and that huge numbers of people would need to use it for it to deliver mediocre results.

One year later, these apps are mostly non-existent and public health and has found very few potential cases through the technology. In spite of being a brain injured idiot, my analysis was correct.

Update: As of March 2021, essentially no one is using contact tracing apps in Canada either. In Alberta, the app has found 0.02% (that is 1% divided by 50!) of the positive Covid-19 cases. And some of those cases might have been determined by other means eventually anyway. The score card: 0.02% by contact tracing app and 99.98% by other methods.

Smart phone contact tracing apps mostly ignored, unused

In reality, adoption of contract tracing apps by citizens was largely sporadic and unenthusiastic. A trio of researchers in Australia decided to explore why contract tracing apps weren’t more widely adopted. Their results, published on 23 December in IEEE Software, emphasize the importance of social factors such as trust and transparency.

Source: Why Aren’t COVID Tracing Apps More Widely Used? – IEEE Spectrum

Big reasons cited by the study:

  • Lack of trust in tech
  • Lack of trust in public health
  • Lack of trust in government

There are a host of problems with the Bluetooth-based tech. Some use network-side tracking (most countries) but some like the US and the UK would use Bluetooth due to legal and privacy issues on the network side.

I wrote a lot about these apps in the past. I have yet to see evidence that Bluetooth-based radio signal strength measurements are anything but error prone when used in the real world. Plus, they cannot detect contacts “across time” – that is, person with Covid gets off bus, you board and sit in their seat. The BT systems cannot detect this surface and airspace contact. One study found a 40+% false positive rate, others found high false negatives, for example. Others found decreased battery life of their smart phones, and many privacy questions, particularly about network-side tracking but I identified privacy issues even with anonymous Bluetooth methods.