How can a programmer help bleg?

I appreciate your frequent high quality work on covid 19.  I’m assuming you’re more plugged into various endeavors to fight covid 19 than I am, so I wanted to ask you for some suggestions.

For someone who has strong technical and analytical skills, but little domain-specific knowledge about pandemics, virology, or any of the other manufacturing challenges the world faces right now, what are some of the projects someone could most easily start or contribute to that would have the highest impact?

The answer could be a project that one person could start on their own, or other projects which already exist but are shorthanded of technical aides, even if they don’t have the specialized knowledge to lead a project on their own.

I ask because I’m someone who programs and models a lot, but is struggling with finding the best way to contribute something of substance.  I could go off and make my own dashboard for one thing or another, but I’d be weary that whatever I did might be jousting at windmills if I didn’t have the right direction.

That is an email from Jonathan Bechtel.

Comments

Mentioned earlier,

https://sniffoutcovid.org/

It doesn't matter what you do because the plan is to replace you with a cheap labor H1B.

Open standard protocol for videoconferencing

I feel there is already plenty of this

Can he drive truck?
If not, can he help load/unload one?
Is he getting takeout often and tipping heavily?
Has he checked the local Amazon warehouse for job openings?
Has he been in contact with those people who have relied on him for periodic but steady work, such as his dog sitter, baby sitter, house cleaner, tree trimmer, yard worker, the dry cleaners, gas station attendent, coffee shop barista, Safeway checkout lady?
Those are the people being killed by the corona virus, right now, and I can just about guarantee none of them give much of a fuck about actually coming down with the virus.

Your last sentence gives away the answer. According to economics, in what's sometimes known as the "sunken costs fallacy", the very fact that " I can just about guarantee none of them give much of a fuck about actually coming down with the virus" means that these delivery people will do their thankless task at minimum wage, and love it. This also means that the people hiring them, the 1% like myself, won't care about giving them a raise. Happens in every field. Supply and demand, and 'self-selection'

Works as long as rules of the game are enforced.

When pitchforks are out on the streets, it is a different story (especially at the country side!)

Ray Lopez, yes, but they might not be able to get work even if they are financially desperate enough to do it despite the risks. Many jurisdictions have banned non-essential house repairs done by tradespeople -- which I assume would include services like cleaning and yard work, unless perhaps the resident has a disability and is getting it done for that reason (in which case it might be done through an agency which gives the worker an iota more job security).

Also John J, although I sympathise hugely with the situation of these workers, it seems a bit rich to say "[t]hose are the people being killed by the corona virus, right now" when the US is approaching 6,000 actual COVID-19 deaths.

You're correct. Economically, financially killed, seeing their employers shutter their shops, knowing their jobs arent coming back, that (unlike the clueless academic economists) the "economic pause" or "economic hiatus" doesnt apply to them (or to us untenured folks also).

>the US is approaching 6,000 actual COVID-19 deaths.

Yes. Only 74,000 more and we can call it "a typical flu season."

The 2017-2018 flu season wasn't typical.

https://www.cnn.com/2018/09/26/health/flu-deaths-2017--2018-cdc-bn/index.html

To everyone who claims this is just a typical flu, how do you explain the overflowing hospitals and the fact that ice rinks are being used as morgues? That doesn't typically happen.

They're not being 'killed'. Unemployment benefits are a thing in all 50 states. And there are places hiring (Amazon, Walmart, grocery stores). Also, food is cheap and nobody's being foreclosed or evicted during the crisis.

I've been following this project
http://corona.help

Develop a toilet paper locator

i made a shared google spreadsheet for local suburbs to crowdsource store stock. only "innovation" was a script that shows the time/date of the most recent observation for each entry. publicized it on local facebook groups, got decent amount of interest, but actual usage never really took off and quickly fizzled. i was hoping to talk to store managers to see if they would keep their entries up to date, but haven't tried yet, expecting they're busy. if it worked, it would help a lot in minimizing everyone's exposure. it would be cool to integrate with a map interface and searching, etc, but i think the main problem is generating usage momentum. i think it's reasonable to expect worsening and potentially severe shortages for many months, so some reliable method to coordinate this info, especially with vendor support, would be a very valuable contribution. i'm interested in collaborating...

+1, through a fiber or even fiber-optic, micro-chip implanted in the roll, which is then routed to an ARC-GIS platform.

What about a mobile app to access / learn about new programs in the CARES Act. You'd select business or individual and input your zipcode. If business, it'd explain how the Paycheck Protection Program works and show participating banks in the area. If individual, it'd explain how the new programs work and provide links to the apply to the state's UI department / the IRS's free file portal.

I'm not much of a programmer, but even if you can't significantly automate applications to these new programs, a one-stop-shop application with location-specific information might be really useful for the long tail of people who don't know what the heck is available.

Many state and local governments, as well as small businesses, could likely use some pro-bono programming help right now, too.

I posted a similar inquiry a couple of weeks ago on LinkedIn and Slate Star Codex and got connected to a startup I am now consulting for that makes a remote health monitoring system and needed scaling help. I also got pointed to the helpwithcovid.com clearinghouse site which has a bunch of other helpful looking projects, many of which look like they could use software engineers.

I don't see a website at helpwithcovad.com. Did you get the address wrong?

I believe it's https://helpwithcovid.com/ -- note the spelling of covid.

You can sort by project type, for instance software: https://helpwithcovid.com/projects?skills=Software has almost 300 projects that can be assisted.

A thermometer that plugs into your USB port.

It's important that you know if your usb is overheating.

Nice pun,
But, fever is serious business, it can cause swelling in the brain:
https://hotair.com/archives/john-s-2/2020/04/02/small-percentage-coronavirus-patients-neurological-symptoms/

Some Coronavirus Patients Display Neurological Symptoms

I just wanted to say thank you to Mr Bechtel for at least trying. Hope you find a match.

Yeah and if he's in any way related to Mr. Bechtel of Bechtel Construction, he's not in programming for the money.

Bonus trivia: Mr. Bechtel of Bechtel Construction was quite possibly killed by the Soviets on Stalin's order. Bechtel, who was staunchly anti-communist, anti-union, hated by the far left in the USA, accepted a proposal from Stalin to tour the USSR, and died under murky circumstances there at a time when the USSR had an active policy of assassinating dissenters and critics both home and abroad. BTW, the modern Russians (and the Chinese) are masters of assassination by stealth. Both groups have access to a rare Himalayan flower extract that, when inhaled or injected into a target, gives an immediate heart attack, with no trace chemicals that can be traced in an autopsy. Kind of like Covid-19 (almost certainly a chimeric virus, see my previous posts), but different.

Ray- you are an odd duck. I’m not complaining, nor should you change, but you are one strange character regardless of the forum you’re contributing too.

BTW - I was waiting for the punishment Sumner was going to bring down on you for bringing up his past comments. Looks like he calmed down. You really twirked his undies last week.

Thanks Slappy McFee. I'm an odd duck and that's fine with me, it got me into the 1% (that and I inherited a boat load of cash a few years ago from my childless uncle, bless him, who hid said cash in his house and I found it, or part of it, since the hired help got a good portion). I'm also now on Team Sumner as I accept there may be something to both money illusion (and sticky prices) and monetarism: have you seen how the US stock market has not crashed since QE-infinity was announced on March 12? If --and only if--the Fed is manipulating the stock market then monetarism works (people confusing real and nominal prices). I'm not sure however the Fed is buying stocks and stock futures, but if so, I'm now officially in the money is not-neutral camp!

That is quite the change for you. Are you ok with that? You have been firmly in money is neutral camp for as long as I’ve seen you post.

When the facts change... I'm willing to give short-term money neutrality a chance...for a short while (stay tuned to the soap opera at Sumner's site). Actually I am a believer in long-term money neutrality. You will note historically people like bullion coins to keep up with rising GDP (for unit of account purposes), which the archaeological data supports, which shows in the long run increasing the money supply helps the real economy, it's almost a necessity.

Write an app for stores to manage the number of people in their store. Something super simple, the equivalent of "taking a number" at the deli counter.

Is there a free app that allows someone who lives alone to 1) be checked that they're ok and 2) ask for (shopping) assistance. For all the ICU patients there are going to be a lot more who are just sick. For the frail, that could be a big problem. So, write it.

Just stay home, study, become a better programer on the other side. This train has left the station.

App for crowd-sourcing long lines at grocery stores. People could report when they are standing in line, either to enter or to checkout, so people could avoid crowded stores.

For more ambitious, develop sharing app for sharing items like toilet paper, masks, etc. where anti-price gouging laws or practices are creating shortages. Since people will be sharing use of, rather than selling ownership of, items that should exempt participants from anti-price gouging laws. (Just in case it doesn't, since you're doing this on a volunteer basis and not interested in making money yourself, release the software anonymously and make it run in a distributed fashion so that no central authority, such as yourself, can be held legally responsible.) Include Uber-like "surge pricing" algorithm to ensure that those that need use of masks and toilet paper are able to obtain it. Given the shortage of masks, if you are able to do this, you will be saving countless lives. (Actually, they may be countable, but that would be a project for economists rather than software developers.)

Many of the popular models make dubious assumptions, and don't do their best outside of places with high density and areas can be seen as a single point. See, for instance, what the health data.org does in Missouri: A state where over half the population has been hunkered down for weeks, because there really are stay at home orders on the three largest population centers. Instead, the model claims that the state will peak Mid May. I also bet that pretending that Illinois is a single dot, instead of a large city and a vast rural area, also leads to errors.

It's not difficult to customize and enhance the models by disaggregating geography, and get local officials to be interested in said outputs.

Those who are working might consider tripling their charitable contributions to low overhead charities.

A new blog post on that might be useful.

I have done some limited search without much luck for best ways to increase charitable giving locally, given that my employment and income are unlikely to be affected by COVID. I wish someone would create an easy to use tool to redirect funds provided by upcoming stimulus spending to people who need them more.

Obviously there are the standard good giving options, but I would like to give money that I will receive because of this crisis to people who need it because of this crisis.

Overall I am surprised at how little work in this area I have seen.

This pandemic is terrible.

But you’re still better off sending money to the people buying malaria nets in Africa. As bad as Corona-chan is, malaria in rural Africa is much much worse.

I agree that is a worthy cause and have donated to it before. In this instance I would specifically like to direct relief money I will receive from the government due to this crisis to people affected by this crisis more than I am. I am thinking mainly of those who have lost/will lose their jobs.

An unemployed American earning $1200 a week in benefits and free healthcare is infinitely less deserving than an African family who needs malaria nets

Both/and... my point was philanthropy is something we can all do.

There's a group called Helpful Engineering based in Slack (think something between a message board and whatsapp), with tons of projects going on, some of them software based or needing software help. A really great group that I encourage everybody to join.
https://www.helpfulengineering.org/
And here's a link to join the slack, though it will expire in a day. If anybody wants another, let me know.
https://join.slack.com/t/helpfulengineering/shared_invite/zt-d6ytoynb-_tJjyC7oEd4fT8L0I8ug8Q

One way to find a cure for Covid-19 is to grep ("grep is a command-line utility for searching plain-text data sets for lines that match a regular expression") the literature, including patents.

For example, a quick search of Google patents shows 200 patents (issued or pending) listing Wuhan and SARS as keywords. Likely some of the people doing SARS research have pointed to, in their patent applications, possible cures like vaccines. Wuhan researchers, Zhengli-Li Shi and Xing-Yi Ge, Key Laboratory of Special Pathogens and Biosafety, Wuhan Institute of Virology, Wuhan, China, who developed SARS-CoV in 2015 as a chimeric virus (man made virus), and published the genetic sequence of SARS-CoV-2 this year (after all, they probably designed it), would be persons of interest who might have a vaccine. Indeed SARS itself had a vaccine made, and shelved, after the SARS crisis went away. Testing this vaccine in vitro and then in vivo, in a mass produced way, would likely give a viable vaccine in a year or three. Probably a dozen teams are doing this now, independently of each other.

Reminds me of my senior project as a science undergrad. The professor said: "reinvent the wheel if you want, or, search the literature". Everybody but me did the former. I did the latter (I was working as a part-time patent searcher for spare money, so I knew the ropes). I went to the US Patent Office (at that time it was not automated, they actually had printed patents in wooden shoe boxes), found a solution to the problem (after interviewing a patent examiner who was an expert in the field), and got an A grade for my project. The rest of the class reinvented the wheel with mixed results (and complained to the professor about my effort, but he pointed out I was within the rules). The "Not Invented Here" syndrome where each scientific or engineering lab tries to do everything on their own, Robinson Crusoe style (published 1719, the first English novel), is alive and well I've found over the years.

Quite possibly the solution to Covid-19 is already lying in some dusty shoe box somewhere. Many a solution to various scientific questions was found in a museum or in the literature (e.g., recall in genetics Mendel's -- but not Mengele's--experiments)

Developing an app from scratch is not something you can just crank out in a week or two. Especially not a solo programmer working at home.

My suggestion is to do something local. Stand up a web page that provides information your local community could use. A local case tracker, a site for finding local stores that are open, whatever.

Or, look for larger projects that are seeking programmers and join them. it will also look good on your resume if you can show that you spent your covid-19 downtime working on something to help others while keeping your skills sharp.

Here's a list of open source covid-19 projects you might check out.:

https://opensource.com/article/20/3/volunteer-covid19

Hack into all computer systems globally to steal and public all trade secrets and patent applications for everything that might help everyone to figure out how to produce drugs and vaccines and tests for any potential pandemic which would speed up producing what's needed for ending the current pandemic and treating COVID-19.

This would include all the documentation and machine instructions to build ventilators and the software to operate them.

If a ventilator built in China or Vietman looks identical to a Germany made Phillips ventilator will be approved for use by the FDA, right?

And Trump can legalize the theft "with his pen" just like Truman did by executive order 9604. (https://www.presidency.ucsb.edu/documents/executive-order-9604-providing-for-the-release-scientific-information-extension-and)

>I appreciate your frequent high quality work on covid 19.

Where is it posted?

Well played.

Just do your day job and donate money to people/groups who are doing effective work (generally, there will be organizations who can put the money to much better use than some spare time from a programmer). Some suggestions are here: https://forum.effectivealtruism.org/posts/wpaZRoLFJy8DynwQN/the-best-places-to-donate-for-covid-19
There are also people buying masks/etc and donating them to healthcare workers, etc.

A lot of work on digital apps and contact tracking for covid:

https://www.statnews.com/2020/04/02/coronavirus-spreads-too-fast-for-contact-tracing-digital-tools-could-help/

One thing you could work on would be proximity and density models for museums and public spaces.

So, for example, museums might want to close off or open up areas based on room density limits, or might want to adjust admissions based on projected density, or have access which is time limited so that people leave and not linger so more folks can come in.

A friend is working on a project for an art museum and I suggested this along with retaining the assistance of a infectious disease doctor to examine their premises and make recommendations.

Density, access point interaction with staff (digital payment rather than tickets), and movement of people deserve a little modeling and maybe some apps.

If Jonathan's talents run toward data visualization, automating the generation of what we're calling skyline tower charts will be helpful, particularly since it would help facilitate prioritizing appropriate resources and setting mitigation strategies for states and regions facing different rates of spreading infection. The data is available from the COVID Tracking Project via API, the charts themselves are simply modified stacked bar charts that are grouped together for close comparison, where what you want to present the data showing the same time periods and for a set percentage of population.

dschnr has built an dashboard to do something similar to what we're describing using available graphing tools, but which doesn't lend itself to the additional detail that's the skyline tower charts we've concepted would make possible. I haven't come across a pre-built graphic tool that could just plug in, so that's where help in programming development would be most needed.

Join one of the competitions on Kaggle. You can also try to find a teammate on Kaggle who may be able to add some content area knowledge to your technical skills.

An option is to subscribe to https://crowdfightcovid19.org
You get an email every day with requests from covid researchers and occasionally firms which lack a specific expertise. Sometimes they need programming work. Other times, legal advice. More often, some obscure (for me at least) biological or medical information.
(I have yet to find something I can do well enough to contribute, but an expert programmer had several options in the past days)

As someone with fairly deep technical skills (data engineering) I was in a similar situation. What I found was that finding a good technical volunteer opportunity was harder than finding a job. And at least with a paying job the offering company is constrained and has to say "no" to many good people, you'll rarely hear that from a charity project even if it's not a good fit. For what's likely to be a short-term position you also need to find an opportunity that doesn't have a high-upfront learning curve, i.e. they need to use the stack/tools that you're already familiar with. And they have to be implementing something worthwhile. Most projects I looked at would have little impact unless the potential users came in droves, most of those projects are going to have zero impact. Alas, I never found a worthwhile project that I could credibly assist. After more than 30 hours of searching, I gave up.

On the bright side, I came away in awe of how well the economy works at matching job opportunities (non-charity) with job seekers.

Check the Coronavirus Tech Handbook. Tons in there.

Also check CovidBase, another wonderful aggregator.

If you want to work on datasets for machine learning, you can check out https://github.com/coyotespike/covid19-images

I think now is the time to be contrarian and work on another problem that is being neglected because of COVID 19.

I don't think the IQ elasticity of COVID 19 is very high right now.

Honestly? Coronavirus needs to be fought with PPE, ventilators, and ten million volunteers.

Is there a food pantry that needs help making and delivering? Is there a group making masks, shields, or vents that needs more manual workers? Does the local grocery store need more people picking out groceries for delivery or pickup? Does your local Medical Reserve Corps need help calling people to organize test screenings, send test supplies around, etc.? And have you set up your computer to [email protected] to help them model the main Coronavirus proteins?

I know we engineers love to think in terms of multipliers - our whole job is to leverage our skills to have massively outsized impact on the world - but the fact of the matter is that most of us don't have the expertise to truly multiply here. Sometimes, delivering food to someone in need, or assembling face masks for medical workers, is enough multiplier.

As always, make money by working for a corporation, and donate to charity. The world doesn't need a million new apps with specific use cases that have no chance at adoption.

I sometimes wonder about the software quality of fold.it or [email protected] If they are open source, you might be able to help optimize them. A 1% increase in speed or decrease in memory usage might go a long way when multiplied by hundreds of thousands of computers.

Another place your efforts might be useful:
New York State COVID-19 Technology SWAT Team
https://www.ny.gov/programs/new-york-state-covid-19-technology-swat-team

Unlike the Helpful Engineering thing I posted above (which I still completely endorse) where you can largely pick your own projects to work on, this is (presumably) a more centralized program where you fill out a form and they will get back to you and tell you what to work on (if you choose to accept). But New York seems to be doing some real innovative things right now so I would definitely encourage you to read more about it and fill out their form. And the state backing might also help with some of the doubts above that no one programmer or one app have a chance at adoption or filling actual needs.

Depends how you want to help. You can help fight the virus. Or you could help people retain their sanity, physicality, and humanity while under global lockdown. If the later of interest, I have a project for your consideration.

The dance community like so much of the service industry is reeling from the pandemic. Dancers have lost the ability to perform and teach. Most have no safety net. I am part of one such international dance community called "Gaga." Their website: https://www.gagapeople.com/en

There are certified Gaga dance teachers all over the globe. They've taken their teaching online via zoom. Our participation has been phenomenal. We are getting 300-1000 students from all over the world for live zoom classes at the same time.

What we need is a global calendar.

So any Gaga teacher across the globe can add their classes. So users in any city is presented all the classes that are available to them. The primary challenge is time zone.

Teacher Use Case: Teacher registers. The system checks they are a certified Gaga teacher by checking the official Gaga website. https://www.gagapeople.com/en/teachers. The teacher creates a username and password. The teacher picks home city from dropdown menu, and the system figures out their time zone. The teacher adds their classes by entering day, time, zoom link, and crowdfund account e.g. gofundme.

Student use case: From a dropdown menu, the student selects their city. The time zone is set appropriately. The student can view available classes by day, week, or month. For each class, he or she is able to see who is teaching, click to see official bio on Gaga website, click to join zoom meeting, and click on crowd funding link.

This seems so simple, but my coding skills are a few decades out of date, and I have other interests now. I last coded in Fortran, Ada, and Pascal:)

Suggestions for implementing the above use cases with low-code, no-code approaches would be appreciated as well.

Thanks for your consideration.

Does Calendly not do what you need for this?

No it does not.

Make a website for hospital workers to anonymously report shortages of ppe and other gear.

Make the website look like the google home page, so that their manager can't tell what they're doing.

Scrape the DHCP as part of the submission, so that you can tell what city and hospital they're at.

Display the resulting data on a map of the USA.

Make sure you know something about privacy/security before doing it.

This is the time that we're glad not everyone "learned to code" other than worthless MSM editorialists. The 'useless' low wage laborers are the real heroes right now as are big business.

Hello Jonathan. For group testing (as described here: https://www.forbes.com/sites/kotlikoff/2020/03/29/group-testing-is-our-secret-weapon-against-coronavirus/#4a2778ae36a6) how about an algorithm to estimate the most efficient way to socially map and group people, for testing, based on their primary social circles--work, school, church, and maybe preferred grocery store? Citizens fill out a survey (with whatever privacy protections required). The algorithm groups them. If testing in groups is the most efficient way to test a population for disease, then efficient groups fed into group tests could be even more efficient. Unlike with random groups, where positive hits would probably find one-off infections, positive hits with this method would seem more likely to lead to mother lodes of infection.

+1 Good way to allocate resources initially. Exponential models, though, after a period of time, spread across many nodes, so it has to be fast acting response.

Bill, Ah, good point. Sort of like on LinkedIn. You sign up and your circle is small, but once you connect to a hundred people directly you seem to have half the planet as a third-ring connection.

But what if the social groupings represent repeat interactions instead of one-off groupings like a party or wedding? Value could be captured until the group's infection rate matches that of the general population. Flare-ups would also offer virgin territory, although the value in that scenario would be attenuated by herd immunity.

Comments for this post are closed