The mask we need

What do we believe now?

We think that coronavirus is moderately contagious and is spread primarily from person to person rather than by surfaces to person. We think the best way to get COVID-19 is to join an indoor dance and singing session and that outdoor spread is rare. We think indoor masks are valuable and outdoor masks are primarily social gestures. We think cloth masks work primarily by reducing spread from someone with early COVID-19 and minimal symptoms. As of today there's a suggestion that children get mild COVID-19 infections but don't spread them well.

We believe social distancing reduces spread but has a terrible economic cost that falls primarily on non-college workers and small business owners.

We believe effective therapies will emerge gradually over the next 4-18 months and effective vaccines over the next 6-18 months.

So what could we do now that would reduce infection, possibly suppress disease, and allow the economy to reopen?

We should test and trace of course, but given the state of American government and American media (Fox, Murdoch, etc) that's unlikely to be enough.

So what else could we do that doesn't require new technology or new innovation?

We could give every adult a better indoor mask. A mask that gives bidirectional protection, that protects against both infection and transmission. Give it first to 70+, then 60+, then every adult.

What are the features of this mask?

It's reusable of course. Washable with a filter module that's easy to replace. It comes with a UV light sterilizer than can hold several masks. It's a high air flow mask; you can wear it to your indoor CrossFit speakeasy and get your deadlifts done. It's not medical grade, but it's a hell of a lot better than a surgical mask.

If you're going to wear a damned indoor mask, it should work.

This doesn't require new science. It doesn't require new technology. It doesn't require closing the economy. At $100 a unit we could give every American adult this mask and a family UV sterilization unit for 30 billion dollars.

30 billion dollars. That's nothing. Jeff Bezos could do it from his change pocket.

Want to restart the economy?

Make this mask.

Wearing a cloth mask outdoors is like wearing a helmet in your car

I wrote this first on Twitter:

Outdoor masking is the equivalent of wearing a helmet in your car. Indoor masking is the equivalent of wearing a seatbelt in your car... 

 ... Formula 1 drivers wear helmets in their cars. Makes sense for them. For rest of us net gain is just background risk noise... 

... It took decades of struggle to get Americans to use seatbelts. Even now some don’t. Despite overwhelming value... 

... If you get hung up on wearing helmets in cars people will think you are nuts and ignore the seatbelts.

The best science I've seen on outdoor communication is the Chinese tracing analysis. We aren't going to see much more science -- experts consider the risk too low to be worth researching given all we don't know about indoor transmission (including transit).

There are two valid objections I know of to this stance:

1. Outdoor masking is of low value but it helps set social expectations that make indoor masking acceptable.
2. If you don't wear a helmet in your car the risk is on you, if you don't wear a mask outside the risk is on me.

To which I would say - True. But ...

1. We would never have gotten seatbelts in cars (high value) if we'd insisted that helmets were equally important (much lower value). If we don't have science we have nothing against the forces of stupidity.
2. Yeah, that does suck. Happily the risk to you is extremely low. As a matter of politeness we should give anyone wearing an outdoor mask a 10 foot space. It's a signal of strong personal concern.

Outdoor masking has a cost beyond damaging expert credibility. It's very uncomfortable to exercise wearing a cloth mask. The physical and mental health benefits of exercise dwarf the non-existent value of the outdoor cloth mask.

Indoor masking is where we should be putting our energy. We should be developing N95 equivalent reusable masks for at risk persons to wear indoors in place of the cloth masks most of us wear.

True story. My father, who was a geek before his time, specced seatbelts on his 1950s company car (to the chagrin of his boss no doubt). They came as 4 point restraints. When he showed up in the car for a date his guest refused to enter the car. Anyone with seatbelts in the car must drive like a maniac.

PS. Regarding those "outdoor plume" studies --- viral reproduction does not scale with respiration. That is, if you breathe 3 times as much you don't exhale 3 times as many viruses.

The year ahead

Massachusetts is going to try the Wuhan/South Korea path. Or something like it.

What's the alternative?

1. Shut down and open up -- trying to stay under healthcare capacity.
2. Buy time to find meds that work a bit, better care approaches for outpatient, hospital, ICU. There are some that look promising now (not HCQ).
3. Buy time to build up manufacturing, supply chains, new jobs in COVID-19 care and management, serology that actually works.
4. In a year or two we have some vaccines that work like those developed for animal COVID.
5. COVID becomes a second yearly severe flu, worse than the Swine flu. On top of the traditional flu.
6. We have fewer people over 80.
7. Many countries will stop Americans from visiting.

In Minnesota we'll let Georgia make the mistakes. If Massachusetts succeeds we'll try that.

Sentinel surveillance for COVID-19: nasal swabs of teachers, service workers, gym coaches and healthcare workers

Sentinel surveillance in pandemic control is typically based on identifying health care delivery sites that get intensive monitoring.

I have been wondering about how we'd do surveillance when we move away from our current stay-at-home condition. Assuming we use self-administered nasal swabs rather than obnoxious nasopharyngeal swaps we could distribute volumes of mail-in test kits (goal of 24h turnaround) to school teachers, healthcare workers, gym coaches, bartenders, and so on. Volunteers would swab weekly, mail in kit with their bar code on them.

We'd need a few million kits per week to do this.

Happily, this is discussed in the April 7 paper by McClellan, Gottlie, et al. I'd heard of the paper, I just needed to read it.

So I don't need to worry ... it's taken care of.

It's just the flu

When contrarians compare COVID-19 to influenza, they invariably mean to minimize it's significance. From what we know now this seems absurd, but of course it's not so simple.  The 1918 pandemic, after all, was just the flu -- and we don't think that's the worst influenza can be.

So how does COVID-19 compare to the spectrum of influenza? Wikipedia has an article on the CDC pandemic severity index that ranks various influenza pandemics. The 1918 pandemic was Category 5 - a case fatality rate (CFR) of 2.0% or higher. The worst influenza in my life was the Hong Kong flu with a CFR below 0.5%. It is said to have killed a million people worldwide (out of 3 billion).

The COVID-10 CFR seems to fit that range. We think its CFR is somewhere between 0.7% (based on presumed cases) and 1.5% (based on excess mortality). So by CFR it is arguably "just the flu".

What about if we look at the other half of the equation - the Basic reproduction number (R0 how contagious a disease is)? Wikipedia is again helpful; influenza ranges from 0.9 to 2.8, the early estimates for COVID-19 the 1.4 to 3.9. So COVID-19 fits the influenza model there as well, as long as we include monster events that cause historic devastation.

We can also look at who dies, and the disability of those who survive. Some influenza takes the young, some take the middle-aged, most take the old. COVID-19 seems to go for the middle-aged and old, so again flu like. As to disability, I haven't seen any reports on post-influenza disability.  I wonder if persistent lung damage will be one way that COVID-19 is not flu like. We don't know yet.

So, yeah, COVID-19 mostly fits within the spectrum of influenza, as long as we include pandemics that hit every 100 years or so. It's "just the flu," in the same sense that WW I was "just a war".

Hydroxychloroquine, COVID-19, and Lupus

Researchers are taking seriously the use of hydroxychloroquine for COVID-19 therapy:

Both chloroquine and hydroxychloroquine inhibit SARS-CoV-2 in vitro, although hydroxychloroquine appears to have more potent antiviral activity [75].

I think saw a post somewhere that claimed it interferes with viral replication inside infected lung tissue but I can't find it now.

That's obviously great if it works out.

If it does work out though, it might be worth looking again at old (and still current) ideas that rheumatic disorders that response to hydroxychloroquine (esp. SLE, RA) are infectious in origin. Maybe an RNA virus ...

Exercise and sanity in the time of you-know-what

Fellow exercise addicts -- let's examine our options...

1. Trail, road, gravel biking. Extreme aerosol diffusion. Antiviral ultraviolet radiation built in. Social distance built in. Definite good. Buy your gravel bike now before they're all gone, but any road bike will do. Start commuting by bike.
2. Mountain biking. Same, but different bike.
3. Running. Hard to be less social. Trail runs nicer. For St Paul MN -- Battle Creek!
4. Garage Gym. Get that car out of there. Who cares about cars anyway? Bench, bar, weights, programming from your local CrossFit franchise. Get some friends together at a distance (byob).
5. Your local CrossFit ... class sizes will be smallish. Open the big doors and spread outside where the air flow is amazing. Dress for the occasion. Learn to love open air deadlift and those $!$^@ runs and rows and such.
6. Golf. Chase (Tim) wants this. Lots of social distance! Don't share clubs :-).
7. Lake swimming. Cold in MN March, but eventually ...
8. Inline Skating! Time for a comeback. Get those blades out of the attic. Airflow, etc. Join the Facebook Minnesota Inline Skate Club Group.
9. Hiking. Slow trail running for the win.
10. Rock climbing outdoors. Sweat and blood do not transmit. Maybe don't spit on the rope?
11. Paddling! Canoe, Kayak -- air flow, social distancing, perfect.
12. Fishing -- not much exercise, but good for sanity.
13. Tennis -- don't spit on the ball.

Things governments should do to help morale:

1. Free fishing licenses!
2. Waive state park fees.

COVID-03 and COVID-19: influenza co-infection and multiple strains

I remember COVID-03 (Coronavirus disease 2003, known then as SARS) caused by Novel Coronavirus 1 (SARS-CoV-1). It was frightening and puzzling, especially in Toronto Canada (from 11/2003, emphases mine):

The entire SARS story puzzles the heck out of me. Why did so many nurses die, even in locations that should have had strong infection control? Why did the disease seem so contagious in some places, and not at all contagious in others? Did the virus attenuate? Was the epidemiologic behavior due to an unidentified cofactor infection that was common in some places and not in others? (eg. a second virus was needed to develop full fledged SARS).  

I can't believe that the infection control measures were so effective. The disease was loose in China for months. Why did it not spread in India at all?

A year later I wondered if there were multiple strains circulating, all mutually immunogenic, some more toxic than others. (There may be multiple strains of SARS-CoV-2 as well.) I wondered if that suggested a pandemic management strategy - a kind of "backburning"...

Create a contagious synthetic pathogen that's relatively benign, but induces immunity to the major pathogen -- and spread it actively. I say not entirely novel, because this is how Polio was suppressed. The oral vaccine was an active contagious pathogen that was excreted in stool. It immunized a vast number of persons -- but some became sick, disabled, or dead. When Polio was less of a threat we switched to a non-pathogenic inoculation. The difference is the successful Polio strategy was probably unintentional (I suspect some people understood even in the 1950s), but in the future we'd be deliberately exposing an entire population to an immunogenic pathogen that would almost certainly harm many people.

Now we are enjoying COVID-19, the bigger, uglier, brother. Again there's tremendous variability from place to place and time to time. Again India seems unbothered. Again young healthcare workers are vulnerable. Again I wonder if some of the sickest patients have multiple viral inflections or more aggressive strains. Perhaps as our seasonal flu finally fades so will the worst of COVID-19.

I hope this time we'll understand it better.

India suggests COVID-19 will behave like a bad influenza pandemic

I've been watching India on the Hopkins CSSE map for weeks. Even as case reports appear to the east and the west, India remains quiet.

India, population about 1.4 billion. India, not known for Singapore class public health. India, with life expectancy of 69 years (China is 76y). India, connected to the world.

What are the chances that India has 3 cases of COVID-19?

One in a hundred? One in a thousand? Lower than that I think. It must be everywhere in India (including in Trump's mass party).

If COVID-19 had a 2% mortality rate even India would notice. If it has a non-Wuhan China rate of 0.7% (where denominator is limited by testing criteria which is in turn limited by test cost), India might not notice.

So somewhere between 0.2% to 0.7%. Bad enough to justify a smarter and better response than the Trump administration is capable of providing, but not 1918 flu pandemic levels. In an ideal world it would be a wake up call for the US to do what it should have done years ago. In an ideal world, of course, Donald Trump would never have gotten within a thousand miles of the presidency. The only way we'll get better is if he's defeated.

Someone is hacking at my Vanguard account and Vanguard can't stop them locking me out

So this has been happening.

Every few days for the past few weeks I have received an email from Vanguard like this:

Of course it's not me. Someone (some bot most like) is running passwords against my Vanguard user name. When they fail I'm locked out.

It's not supposed to work this way. This was a common problem in the 1990s, but then security teams learned to use timeouts to reduce the risk of password attacks. The chance that anyone will guess my quite long and random unique password is infinitesimally low.

I don't know the motivation. It might be harassment or it might be someone locking out the password so they can then do a social engineering attack. Given Vanguard's approach to lockout security I think there's a good chance they'll succeed.

I've written Vanguard about the problem but the representative tells me there's nothing they can do. Their security is working as it should.

I've gone through their password reset several times. It's the usual - last 4, birthdate, name of first boss, then text a code. The usual poor quality reset process that's been routinely broken. (Of course the answers to my secret questions are also unique strings unrelated to the question.)

Since Vanguard can't fix the lockout problem I'll have to try changing my username to a random string. That will take a phone call with Vanguard and a bit of hassle but I really don't have a choice.

Although the account rep didn't know this, there's an option to restrict logon to only recognized computers. This is a bad long term solution, but I've enabled it for now.

There's no relationship between the wealth of a corporation and the quality of their security.

Update 3/1/2020: Vanguard responded:

Our Fraud Team has reviewed your profile and the incidents you described.

They have determined that your account was locked multiple times by another client with a similar user name. Fraud has recommend you re-register for account access to change your user name to avoid this situation  going forward.

In other words, not a malevolent hacker, just someone who is not very good with credential management (maybe a bit further ahead on the dementia curve than I am). Based on my username it's probably a distant relative (it's a County Leitrim Ireland name, small cohort). Vanguard should be using time delay management of password attacks, instead they're locking me out. The re-register option is a real nuisance.

For now I've configured Vanguard to only allow access from my Mac (presumably a cookie). Maybe after a few weeks of getting a different error message my confused relative will figure out they're using the wrong damned username. Then I can try returning to standard access.

Update 3/13/2020: Locked out again, so the restricted access trick didn't help. I'll undo that. I really hate to have to change my username just because Vanguard can't implement 10 yo security technology.

Update 3/28/2020: Finally logged back in again doing the usual reset. Except now I discover the "restrict logon" is implemented by a cookie -- and I cleared my Safari cookies a week or two ago. So even with the reset I can't log in. It didn't work to stop my nemesis, but it sure stopped me.

I had a chance to review Vanguard's troubleshooting pages and looks like they haven't been updated for 5-10 years. So now I have to phone them some time during their limited service hours.

Update 11/7/2021: About 6 months ago I finally quit Intuit's Quicken software. After I did that I didn't have any more Vanguard lockouts. Despite my disabling Intuit's online account feature I think they were polling and storing my Vanguard financial records. They weren't logging in successfully, but they did lockout my account.

Apple can beat Google Maps -- by investing in bike route maps

Google Maps seems unbeatable. Every time Apple does an upgrade Google does three. It seems Apple can't win.

But Google has weaknesses. Google maps are increasingly hard to read, particularly in sunlight. Google has no options for scenic routes; even when I choose an alternate route for the pleasure of the trip Google aggressively reroutes me to the fastest option. Apple maps at least have a "no highway" trip option.

These are small weaknesses though. Apple still gets big things wrong even with their latest revisions. Apple hasn't learned much from Google's Local Guides program. My Local Guide score lets me relocate a business in seconds -- something that's made me quite popular with CrossFit gyms and medical clinics that have moved (sometimes they've suffered wrong location listings for months).

Most of all Google has bicycle routes and Apple doesn't. That gap means I can't consider Apple Maps for everyday use. Bike routes are a map moat and Apple hasn't tried to cross it.

But ... Google's bike map moat is silting over. They aren't updating them any more. Google once accepted bike route suggestions from Local Guides -- but now they direct us to treat omissions as road errors and even those are ignored. For example, here's Google's current map of bicycle trails around Hastings Minnesota:

That map makes it seem there's no route from the urban core to Hastings. In fact there's a lovely trail from Hastings to the blue dot on the left, then a brief gravel road, then a trail to St Paul and thus Minneapolis.

Google's neglect is Apple's opportunity. This is an area where Apple could actually beat Google Maps. I think they'd like that.

And, of course, if Apple did make a move maybe Google would accept some improvements ...

Bodies are weird, episode 26

My right wrist hurt the other day. A sudden sharp annoying pain. Maybe a tendon, maybe my arthritis acting up.

So I did my usual amateur self-therapy. I avoided the sharp ouch, but I moved with weights and resistance through a proximal path that was sometimes achy but not ouchy. I had lots of opportunities to load the wrist with weights, I am obliged to do CrossFit six times a week [3].

After about 4 or 5 days of this I noticed the wrist was pretty good. No more sharp pains.

It seemed … familiar. Eventually I remembered it happened before, back in Oct 2015, a bit before my formal Dec 2015 arthritis diagnosis. Resistance work was the fix then too.

This isn’t what we were taught in the medieval medical school of my youth. We were taught to rest sore joints, not to put them under painless load. We weren’t taught that running might make knee cartilage better.

Bodies are weird. Back in 2015 my knees were quite sore. I figured my CrossFit days were numbered; I even tried underwater hockey.

But then the knees got better. I continued my back squats and lunges and all the CrossFit rest. Maybe it was the exercise, maybe it was the hydroxychloroquine my atypical rheumatologist prescribed [2] maybe it was both.

Over the next four years I sometimes had knee and wrist effusions, sometimes not. Lots of things came and went. My hands got beat up, but they didn't bother me much.

Then this past summer came around. I felt weaker. My back was fragile in late July. I developed “pseudo-claudication” (look it up). I lost a bet with my daughter when I missed my birthday Bar Muscle Up. I figured age had caught up.

But then it turned out I had the pseudo-claudication was pseudo-pseudo. Probably a protruding disc. It got 80% better after 6-8 weeks of modified exercise and 100% better after 8-10 weeks. (Discs do that — it’s even in the textbooks.) I hit new lifetime best lifts in clean & jerk and back and front squat. Equaled some others. Got even closer to that elusive bar muscle up.

It’s not like I’ve stopped aging. I look a few years older than my age. I feel pretty old. Everything could fall apart tomorrow. So I’m not expecting to carry on like this. I’m just saying bodies are weird and “osteoarthritis” / “idiopathic arthropathy” [2] is weird. We do not understand. We might as well keep moving.

- fn -

[1] The process likely began with some rogue antibodies before 2010 and a single acutely inflamed distal finger joint in 2012.

[2] The one study I’ve seen on HCQ and OA says it doesn’t work. OTOH, I think “osteoarthritis” should be renamed “idiopathic osteopathy” to underscore our ignorance of what’s likely many different conditions with similar appearances. My mother did relatively well on it FWIW — before she went full RA.

[3] I leave it as an exercise for the reader to imagine why a sane person would actually need to go 6 times a week, even foregoing my ice hockey. It’s not for (my) health or training!

The killer application for Apple's AR glasses will be driving

Sucks to get old. At 60 my night vision is probably half of what it was at 25. I drive slowly at night to reduce the risk of missing a pedestrian.

What I need are AR glasses that receive input from forward facing light sensitive sensors and that enhance what I see as I drive. Draw circles around pedestrians. Turn night into day. With the usual corrective lenses of course.

I’d pay a few thousand for something like that.

Seems quite doable.

Carbon sequestration

It’s the year 2,500. Civilization has long recovered from the chaos years following the Trump regime.

Alas, even that catastrophe only transiently slowed CO2 accumulation. It continued for decades near the 2020 height of approximately 4*10^13 kg (40 gT) of CO2 a year.

Happily in 2,500 post-AI solar powered nanotech can extract atmospheric carbon and produce diamond sheets and glues. It’s possible to build a diamond wall to hold back the Atlantic and restore the lost cities of Manhattan and Miami.

How much CO2 would that take out of the atmosphere?

Diamond has a density of 3.5 gm/m or 3,500 kg/m3. If the wall is 1000km long by 50m high by 10m wide it will consume about 1.75*10^12 kg of carbon.

So we only have to build 22 such walls to undo a single year of CO2 emissions.

Update 9/24/2019

I put the numbers into a Google Sheet. Since the wall only required a few weeks of carbon output I decided to try the diamond base of a floating city modeled as a disc with a radius of 50km and a depth of 200m. That did the job!

A single floating city absorbed 145 years of 2018 carbon production.

It would be easier, of course, to build something attached to a landmass, but where’s the romance in that?

Clark Goble 1967-2019

Clark and I had different opinions on the future of ISIS. This was years ago, when we conversed often on app.net/ADN. He under his own name, me, less boldly, using my John Gordon pseudonym. (Clark knew I was John Faughnan, we corresponded by email too.)

Clark was a slippery fellow in a debate so I suggested a wager based on the state of ISIS in a year or so. I forget what I wagered, but Clark offered to pay off in Amano artisan chocolate. His company’s chocolate that is.

Some time later, a bit prematurely I thought, Clark decided I was right and he shipped me a wonderful supply of some of the finest chocolate I’ve ever had. I tried to stretch out the supply but I’m bad with chocolate. I ate it too quickly.

I was in Clark’s debt after that. I thought I’d have a way to pay it back, but I didn’t expect that to be a donation to his posthumous gofundme account. He died 9/6/2019 of a ruptured cerebral aneurysm. He was 52yo, married, with five children.

My wife’s mother died the same way at the same age. I remember, from medical school days, a young woman dying suddenly the same way. It’s like being hit by lightning. Here one day, gone the next.

I never met Clark in person. We chatted for years on app.net, then continued on Twitter. We had some things in common. We both loved the outdoors - mountain biking and, in his case, mountaineering. We enjoyed science and especially physics — though he was the real thing and I am just a hobbyist. Both of us were happy fathers; Clark's youngest child was about a year old when he died. 

We were both born in Canada. We both liked travel, he wrote often of his mission work in Louisiana.

Politically and spiritually we were pretty different. Clark took his faith and Mormon theology seriously, I’m a functional atheist. Clark was a Western old school Republican, I’m an Obama-Democrat. Many of our discussions were on politics — he loved the ins and outs of democracy.

I miss Clark's voice. I’m sad I’ll never get to drop in for chocolate. I mourn for his family and all who loved him.

• David Clark Goble Obituary - Visitation & Funeral Information
• Remembering Clark Goble | Times & Seasons 
• Clark's Tech Blog | Trying to figure it all out a post at a time
• Clark and my last twitter conversation - Aug 27, 2019