How Covid-19 spreads: simulations and visualisations

This is a list of visualisations of how aerosols and droplets spread. While not all are related to Covid-19, they are relevant and worth watching. Happy to add more if anyone finds them. 

The Dogs That Haven’t Barked (Or We Just Haven’t Heard Them Yet)

I’ve been wondering about the countries we haven’t heard from yet on Covid numbers, or which have just flown under the radar. What I call the dogs that haven’t barked. In short, who’s on the other side of the curve, and who isn’t?

The most interesting data I could find was at EndCoronavirus.org, a website put together by Yaneer Bar-Yam from the New England Complex Systems Institute. They have collected together data from 72 countries and plotted graphs of their curve flattening efforts. They’ve categorised the countries into three. Those which

  • have already beaten COVID-19 (where the curve is back to where it started)
  • are nearly there (where it’s half way down the slope)
  • need to take action (where it’s still near the peak
https://www.endcoronavirus.org/countries

That’s interesting enough, but I wanted to get a sense of what that means. First off, I wanted to get a sense of how many people that is — how many people are in countries that have beaten COVID (I actually it’s a little early to say they have, I’m using Bar-Yam’s terminology), and how many people are living in countries which still need to take action?

So I did a bit of downloading and spreadsheeting. This is what we have:

where populations sit on the curve

So according to that data, more than half of the populations of the countries measured haven’t really started to tackle Covid-19. Gulp. There are some big countries in there:

Eight of the 10 most populous countries don’t score well

Now, you could quibble about which countries are at which stage, and some countries have clearly done more than others, but the curve doesn’t really lie. At least, it’s as good a yardstick as any.

So my next question was: how many people are we leaving out? Turns out quite a lot. The countries covered had a population of 5.9 billion, which leaves about 1.9 billion people unaccounted for. The website says that the “set of countries is certainly not an exhaustive list, but we do highlight the countries which we find to be interesting or important in some way.” It doesn’t say why it found them important and interesting. Possibly the data was clearer for those included. Hopefully they’ll add more countries once the data is clearer.

So what does the pie chart of people look like if we account for the missing two billion? We get this:

The missing quarter

The population in countries that haven’t started yet goes down to 42%, which is still a big chunk of humanity, while the ‘good guys’ account for about 23%. A quarter of the world’s population doesn’t appear at all.

So what happens if we put all this onto a map? Where are all these people?

The gray of Africa and Central Asia

It makes for grim viewing. Several things jump out:

  • The countries doing well are flattered by the presence of China. And Vietnam, Thailand and South Korea. In population terms the largest country outside APAC to get a star is Greece, with 10.4 million people. It does not look good.
  • The gray areas are a big concern. Very little data from Africa and Central Asia was ‘interesting or important’ enough to include in the study thus far. As we’ve read in the media, Africa is unprepared for Covid-19, where two countries account for nearly half of all tests carried out in Africa so far, according to data collected by Reuters.
  • The other gray area is Central Asia. A piece in the Atlantic Council’s New Atlanticist this week said that the countries of the region were each ploughing their own furrow, with varied, sometimes suspicious, results: “Tajikistan and Turkmenistan had been clinging to a fantastical claim of zero cases for weeks, despite the recent spike in mysterious deaths attributed to pneumonia. Tajikistan finally reported its first fifteen confirmed cases on April 30, after weeks of speculations and warning from international experts.”

I don’t know what all this says, exactly, but it sheds a little light on the dogs that haven’t barked. Covid-19 is a pandemic, which means it infects people, not countries. Yes, we can seal borders, and keep people from traveling, but at the end of the day the disease will only be conquered when every country has flattened every curve — or, unlikely, never let it rise up in the first place. None of us is going to be able to travel as freely as we once did, or see supply chains and shops creak back into life again, until all the people in these countries are through the same tunnel we’re currently in. That could mean a lot more pain, heartbreak, and death.

So thinking of this by population size, and looking more deeply at the countries that either haven’t seen the curve appear, or aren’t reporting it for one reason or another, is one way of knowing how long we’re all going to be fighting this war.

Bike Fencing

Some interesting stuff going on in Singapore’s world of bike sharing.

They’re approaching the problem of errant bike-parking by regulating the companies via a licensing regime, which will begin later this year, according to Today.

From what I can make of it, operators must
– be licensed, or face a S$10,000 fine and/or six months in jail
– be responsible for the parking of bikes within designated parking locations, or lose their licence or find their fleet size reduced

Users will also be watched, under a geo fencing scheme that will require them to scan a QR code at the designated parking locations before ending their trip. Failure to do so will mean they’ll be charged continuously — I guess meaning the meter will keep running (not sure how this would work with the flat monthly rates all three operators are currently offering).

Readers have already pointed out potential flaws:
– what happens if there’s no space at the designated area?
– what happens if someone moves the bike after the user has scanned their code?

And Today pointed out in a piece that there need to be more designated areas to make this work. It’s fine picking up and parking a bike at a subway station or a bus stop, but what about when you’ve pedaled back to your home?

Singapore, as ever, is taking a positive but cautious approach to the sharing economy. I quite agree that companies are so far not incentivised to distribute their bikes with consideration, or to monitor them after they’ve been deployed. So something has to change. But also the usefulness of these bikes is going to decline rapidly if users aren’t able to leave the bikes within a few meters of their home for fear of draining their digital wallets.

More importantly, Singapore needs to consider what more it can do to encourage bike usage — by rapidly expanding its bike paths, by offering guidance to users about how and where they can use the bikes, and generally rewarding their use. As China has found, the more these bikes are used, the more other people feel comfortable using them and the quicker a social code of conduct emerges about their usage.

 

Investigators – New Kids on the Blockchain

Here’s a Reuters piece I wrote on a hitherto uncovered area of blockchain potential — helping law enforcement and others collaborate and collect evidence better, among other things. 

For security agencies, blockchain goes from suspect to potential solution

By Jeremy Wagstaff, Byron Kaye

(Reuters) – Police and security agencies have so far only taken an interest in blockchain – the distributed ledger technology behind cryptocurrencies like bitcoin – for tracking criminals hiding illegal money from banks.

But that’s changing as some civilian, police and military agencies see blockchain as a potential solution to problems they have wrestled with for years: how to secure data, but also be able to share it in a way that lets the owner keep control.

Australia, for example, has recently hired HoustonKemp, a Singapore-based consultancy, to build a blockchain-based system to record intelligence created by investigators and others, and improve the way important information is shared.

“They’ve been trying for years to come up with a centralized platform, but people are reluctant to share information,” said Adrian Kemp, who runs the consultancy, which was awarded a A$1 million ($757,500) grant by AUSTRAC, Australia’s financial intelligence agency, and the Australian Criminal Intelligence Commission.

Blockchain’s appeal for data sharing is threefold.

Its ledger, or database, is not controlled by any single party and is spread across multiple computers, making it hard to break. Once entered, any information cannot be altered or tampered with. And, by using so-called smart contracts, the owner of information can easily tweak who has access to what.

It’s a sign of how far blockchain technology has come within a decade since the publication of a pseudonymous paper describing bitcoin and the blockchain ledger that would record transactions in it.

Bitcoin has since become the preferred currency not only of libertarians and speculators, but also of criminal hackers. The bitcoin price is volatile, and hit record peaks late last month.

Governments are already exploring ways to store some data, such as land records, contracts and assets, in blockchains, and the financial industry, too, has experimented with blockchain technologies to streamline transactions and back-office systems, though with limited success.

SECURING SHARED DATA

The closest most law enforcement agencies have come to the blockchain has been working with start-up firms to analyze it for evidence of criminal deals.

But in the past year or so that attitude has begun to change.

The United States Air Force (USAF) has funded research into how blockchain could ensure its data isn’t changed. In May, the Defence Advanced Research Projects Agency (DARPA) awarded a grant to ITAMCO, the company behind an encrypted chat program to make a secure messaging service based on the blockchain.

Amendments to a recent U.S. Senate defense bill require the government to report back on “the potential offensive and defensive cyber applications of blockchain technology and other distributed database technologies” and how foreign governments, extremists and criminals might be using them.

Britain, too, is exploring several uses of the blockchain, say consultants and companies working for several departments.

Cambridge Consultants, a UK-based consultancy, said it had worked with the Defence Science and Technology Laboratory, a UK Ministry of Defence (MoD) agency, on using a blockchain to improve the trustworthiness of a network of sensors on, for example, security cameras.

The UK’s justice ministry is looking at proving that evidence – video, emails, documents – hasn’t been tampered with by registering it all on a blockchain, according to a blog post on its website.

Marcus Ralphs, a former soldier and now CEO of ByzGen Ltd, which makes blockchains for the security sector, said he was working on projects with the MoD using blockchain to track the status and level of individuals’ security clearance. Other work included helping the Foreign and Commonwealth Office (FCO) improve the way work permits are issued and records stored.

“PASSING THE BUCK”

These are early days.

Kemp says there’s no guarantee his project will be deployed more widely. And some who have worked with AUSTRAC are skeptical, saying such projects have more to do with agencies turning to the private sector because they’re running low on resources and ideas.

“The government is just looking to pass the buck on to private industry,” said Simon Smith, a cyber private investigator who has worked on cases involving AUSTRAC.

Many police forces and armies aren’t ready for the technological and mental leap necessary.

The Police Foundation, a UK think-tank focusing on policing and crime, is pushing British police to explore the blockchain, but its director, Rick Muir, said “we are still at the stage of ‘what is blockchain?’.”

Neil Barnas, a USAF major who last year wrote a thesis on the potential of blockchain in defense, said U.S. military and security agencies were slowly waking up.

The problem, he says, is that military minds are more inclined towards centralized systems than the decentralized ones that blockchain’s distributed ledger embraces.

That said, blockchain’s association with the criminal underworld has not dented its appeal to those who see its potential, said ByzGen’s Ralphs.

“The negative narrative around it has not at all watered down or diluted interest of the people we’ve been engaging with,” he said.

($1 = 1.3201 Australian dollars)

The Ugliness of Short Term Hacks on the Road to Wireless

Here’s a Kickstarter project to solve the problem of no audio jack which illustrates just how thorny it is: iLDOCK – charge and listen to iPhone 7 at the same time by ildockgear — Kickstarter

ILDOCK lets you use your wired headphone while charging your iPhone 7. You can also add storage via SD, TF and USB ports with Plus. 

NewImage

The problem, as some have highlighted in the comments, is that Apple rarely grants MFi status to accessories where the lightning cable doesn’t plug directly into an Apple device. In this case, one does but the external one doesn’t. Most manufacturers get around this by making the external one a microusb. I don’t mind that, in fact it helps me, but some folk aren’t crazy about it. 

There are other issues too, of course: if you have lightning headphones and want to charge, this isn’t going to help you. 

I know I’ve written before that the future is wireless, but this project, worthy though it is, merely illustrates how ugly the interim is going to look like. 

(Via Kickstarter)

Singapore’s M1 aims narrowband deployment at the sea

Singapore telco M1 is getting Nokia to install an NB-IoT network atop its 4G one, interestingly with an eye not just to land but to sea. 

NB-IoT stands for Narrowband Internet of Things, and is the GSM world’s answer to narrowband technologies such as LoRa and Sigifox that threaten to take away a chunk of their business when the Internet of things does eventually take off. Why use expensive modems and services when you’re just trying to connect devices which want to tell you whether they’re on or off, full or empty, fixed or broken?  

Techgoondu reports: “While that network caters to heavy users who stream videos or songs on the go, a separate network that M1 is setting up at the same time is aimed at the smart cars, sensors and even wearables.

They said pricing will likely vary with each solution or package, with some companies saving costs from deploying large amounts of connected sensors. However, others that require the bandwidth, say, to deliver surveillance videos over the air, would likely stick with existing 4G networks.

And while many NB-IoT devices are still on the drawing board – standards for the network were only finalised in June – M1 executives were upbeat about jumping on the bandwagon early.

Alex Tan, the telco’s chief innovation officer, said the technology would open up new business opportunities in the years ahead.”

A press release from M1 says it’s working with the ports authority — Singapore is one of the biggest ports in the world — to  “explore the deployment of a network of offshore sensors to augment the situational awareness of our port waters,” according to Andrew Tan, Chief Executive of the Maritime and Port Authority, MPA.

This follows Sigfox’s deployment in the city state last month. It also pips to the post rival Singtel who have been talking since February about running a trial of NB-IoT with Ericsson.  (Update: “Our preparation to trial NB-IoT is well underway. We are working with our vendors and industry partners to conduct lab trials in December, with a view to launch an NB-IoT network by mid-2017.”)

Here’s my earlier piece on LoRa

Pocket-sized Smartphone Breathalyzer

Further to my piece on smell sensing tech, it seems that breathalyzers, which use gas sensors like this one: Alcohol Gas Sensor, are getting smaller. This one attaches to a smartphone, fits in a pocket and costs $35. (Via Interesting Engineering)

That’s not the cheapest one out there — this BACtrack Ultra-Portable Personal Keychain Breathalyzer probably is — but I think it’s probably the cheapest that connects to a phone. 

Some more links on the matter: 

Drinkmate | Specifications

 Blood alcohol content – Wikipedia, the free encyclopedia

 

Apple Takes on Evernote?

Apple’s update to OSX allows users to import Evernote notes into Notes (if you see what I mean) painlessly and effectively: Import your notes and files to the Notes app.

As far as I know, this is the first time an app with some heft has included this capability — there are third party tools for OneNote, but no native functions. 

To me, this is the first serious challenge to Evernote, since why would you bother with Evernote if you’re an iOS and OSX user? 

There are limitations, I suspect. I can’t find any way to add tags and it seems the tags preserved in an enex/xml file are lost on import. That’s a showstopper for me. And of course some of the deeper features of Evernote aren’t there — saved searches and what have you. And if you use Android and/or Windows this is not going to help you.

But I suspect the bigger thing for most heavy users will be a sigh of relief that a player like Apple sees it worthwhile to add this feature. For many users there’s been growing disquiet as to just how  long ‘Ever’ means for the company, and the ramifications for their vast Evernote collections. 

Uber, $70 bln company, doesn’t seem to test some of its code

Screenshot 2015 12 21 07 40 15

Growing pains, I guess, but this should not be what big disruptive companies look like. I noticed that Uber’s web app offers filters to create lists of historical data — your rides — via criteria like which credit card you used, the city you took the ride in, the month etc. Great for expenses. Except it doesn’t work. The filters simply don’t work.

Uber have confirmed it and said they’re working on it. (It’s still not working.) But for a feature like this, wouldn’t you have done even basic testing, like, well, to see that it worked? 

 

Making 3-D objects disappear

I’m a big fan of invisibility cloaks, partly cos they’re cool, and partly because I think the principles behind them could end up in a lot of things. Here’s another step forward: Making 3-D objects disappear:

“Making 3-D objects disappear: Ultrathin invisibility cloak created Date: September 17, 2015 Source: DOE/Lawrence Berkeley National Laboratory Summary: Researchers have devised an ultra-thin invisibility ‘skin’ cloak that can conform to the shape of an object and conceal it from detection with visible light. Although this cloak is only microscopic in size, the principles behind the technology should enable it to be scaled-up to conceal macroscopic items as well.”

(Via ScienceDaily.)