[18-24.3] Monitoring COVID19 effects

Let's set up a monitoring of the societal response to the Covid-19 epidemic and the actions of the authorities! We are counting on your support. If you would like to contribute to making data and analysis on the impact of the Covid-19 epidemic visible, we would be happy to hear from you. We will support you with data, advice and action. This is a continuously developing hackathon initiated by Statistik Kt. Zürich and the Swiss open data community.

Lasst uns ein Monitoring der gesellschaftlichen Reaktion auf die Covid-19 Epidemie und die Massnahmen der Behörden aufbauen! Wir rechnen auf viel Unterstützung aus der Community. Falls ihr mithelfen wollt, dass Daten und Analysen zu den Auswirkungen der Covid-19 Epidemie sichtbar werden, freuen wir uns über eure Beteiligung. Wir unterstützen gerne mit Daten, Rat und Tat. Dies ist ein sich fortlaufend entwickelnder Hackathon initiiert von Statistik Kt. Zürich und der schweizerische open data community.

Join in: https://db.schoolofdata.ch/event/7 Twitter: #covid19mon Follow: @statistik_zh / @opendatazh / @opendatazurich / @opendatabs / @opendataswiss / @opendatach

3 Like

Seven challenges are set up and a couple of them are being actively pursued already this morning: Visualisierungen zu Fallzahlen in den Kantonen, Data on hospital infrastructure, Auswirkungen auf Luftqualität, Wirtschaftliche Aktivität im Grundversorgungsbereich, Konsumentenstimmung, Daten zur Netzauslastung, Frequenzen im ÖV, MIV und LV (Velo), and Publibike API. The data sources are being collected and discussed on our chat server and social media. The hack is on!

1 Like

We have started a project to collect and standardize data sources around covid-19 and make them accessible together with „no-setup“ computational environments. The project is in its infancy and available here: https://renkulab.io/projects/covid-19/covid-19-public-data

We are setting up data ingestion pipelines for various sources including the data compiled by OpenDataZH. The first dataset is the JHU data and the simple viz is available via the notebook here: https://renkulab.io/projects/covid-19/covid-19-public-data/files/blob/runs/Dashboard.run.ipynb

The project runs on our Renkulab platform where anyone can make an account - you can fork the covid-19 project and do your own thing, spin up an interactive environment for working with the data, or if you’d like to get involved directly we can add you to the group that is developing it. We will try to add the swiss data sources today!

1 Like

Can R scripts be integrated in the pipelines?

1 Like

What is the best way to set up a data set which can be used collaboratively, so that multiple people can edit and add data? I have started the hospital infrastructure challenge, but may experience with collaborative data collection is very limited. Any recommendations? I was thinking of Google docs, but maybe there is better way to do this. Needs to be „view only“ for public and „editable“ for registered and approved users. TIA

1 Like

Thanks for sharing your project @rokroskar ! Feel free to join our live chat and post a challenge if there’s something you could use help with. I’m a user of renkulab and will likewise be happy to join in or to support anyone here doing some data analysis.

That’s a great question, and you should post it in our Buffer Overflow if you want to have a wider discussion. I’d say Google Sheets are fine as a starting point. We’ve got some other suggestions in the What tools do we use for data wrangling? thread, which I’ll update now with a couple of suggestions.

1 Like

Welcome to the forum @bafe. Assuming your question is about Renku pipelines, I don’t see a renku-r module on GitHub like there is for Python, but the documentation mentions R support, and I’d be very surprised if nobody is using it for statistical exploration. @rokroskar is the right person to ask, and I’ll try to connect you two in the chat for an explanation.

1 Like

Given that it a manual work, to avoid duplication of effort do you want to start already sharing your dataset on google docs?

1 Like

yes, on it. Will upload a first partial dataset in 1-2 hours

1 Like

@bafe indeed, you can use R! We have images with R and python pre-installed - just let me know what you need and we’ll try to set it up. If you want to join the existing project that is trying to collect open data on covid-19 and you need R in the image, we can do that easily.

Just to follow up on my post above - we’ve added the ZH Open Data sources to the covid-19 public data project on renkulab.io - we will set up daily automatic updates in the next day or so.

1 Like

Some calls to action and acknowledgement of the importance of the work the Canton of Zürich has started in the Swiss press this morning. We are trying to not let it get to our heads, and are quietly continuing the work. Thank you all. Our colleagues at Opendata.ch and elsewhere are setting the stage for building on top of this and involving more of the community. Keep calm and datamine on!

Meanwhile up north, possibly the biggest health hackathon in history is about to begin…

Check out the fresh-off-the-press Open Data Institute newsletter for a short sweet round up of news from around the world.

A quick update of our progress on the global forums at Open Knowledge:

Things are coming to a wrap today in our first #covid19mon sprint. Here is the plan.

:tickets: This afternoon starting at 16:00 we will run a livestreamed video conference and go over the results. If you are working on a project, please check that your documentation is ready for prime time at the latest by then:

  • We will collect project descriptions from db.schoolofdata.ch and publish them in a blog post.
  • Please make sure there is a link to your visualization, dashboard, notebook, dataset, etc.
  • Your (nick)name should be listed in the project page (click JOIN…) and linked to your social media profile or web page if you wish to be credited.
  • We will ask 1 person from each topic to present their results live in the conference.
  • At this time we will also share notes about „What’s Next“ and how you will be rewarded for your participation.

This is not the end of hacking activities, the datasets will be continuously updated and worked on further. This is a milestone that we set 7 days ago, and we hope you can join us on the stream or on social media, to ask questions, debate the conclusions, and launch our datasets into higher level orbit :rocket:

The call will focus on acknowledging the results of everybody’s efforts. We also absolutely want to hear about your experience, what worked - and what did not. A form will be sent out tomorrow to collect and systematically go through ALL of your feedback, and respond to each of you here. THANKS in advance!

1 Like

:tv: Our closing presentation with representatives of each of the datasets that was being wrangled over the past seven days. You can watch it from the beginning, or skip to each of the demos in the links on the project pages at https://db.schoolofdata.ch/event/7

:tada: ATTN: we have prepared a short survey for participants and followers, that should only take you 5 minutes and will really help us to plan how we can further help you, your projects, and the open data and open source resources you worked with ----> feedback #covid19mon :writing_hand:

:hugs: Thank you so much for your contributions and engagement. „A gift for future epidemologists“, as one participant expressed our outputs. Today we pave the way for tackling the effects and supporting each other through this pandemic… And we are far from finished yet! We are making our datasets available to the world, while Switzerland gears up for big hackathons this weekends at #CodeVsCOVID19 and next week with #VersusVirus. Stay safe, stay home, keep at it :rocket:

Updates on all work that the team has continued to do over the following days:

In the meantime the crowd-boosted data source is now being used by many top-tier press publications including the national broadcasting service. Power to the people - when their government works hand in hand with them :handshake:

Relaying an important announcement for everyone using the #OGD ressources provided by @statistik_zh, partners & contributers to openZH/covid_19. From Thursday, 2020-04-09 10:00, case numbers will be delivered with an adapted data structure; see example CSV:

Changes include:

  1. newly added row new_hosp (= new hospitalisations since last date)
  2. row ncumul_hosp renamed to current_hosp (= number of hospitalisations on that date)
  3. row ncumul_icu renamed to current_icu (= number of ICU patients on that date)
  4. row ncumul_vent renamed to current_vent (= number of patients required to receive ventilation on that date)
  5. all row titles are lower case (_icu instead of ICU)

For background discussion, visit issues #473, #463, #456.