# Open Data @ BFH CAS DA - Part II
Part II of the study notes ([part 1 here](./bfh-opendata-1.html)) for the Open Data lecture by Oleg Lavrovsky at the Berne University of Applied Sciences [CAS in Data Analysis](http://www.ti.bfh.ch/de/weiterbildung/weiterbildungsangebote/cas/datenanalyse/tabs/uebersicht.html). The course of study is designed for professionals interested in data projects, building experience in the analysis of data using desktop tools. The intent of this lecture is to present a practitioner perspective as well as some introductory background on open data, the open data movement, and several real-world projects - with details of the data involved, legal conditions and technical challenges.
We started with a quick recap of last week's lecture: how attention to the questions that arise in data usage lead to virtuous cycles of **data, information and knowledge**, and how the opening of data accelerates this cycle. We covered **definitions** of open data, including the licenses, guidelines and publication standards involved. With [R scripts](https://gist.github.com/loleg/aef3fd6aa91e2a65c80627bb0f29f49d), we searched for and accessed live open data to visualize it directly in R Studio using the CKAN package, as an exercise.
Our focus is on the origins of the open data movement in **Switzerland**, and we summarized opportunities and challenges exist here in regards to public data. For a good summary of progress in the area of Open Government Data, see [Yearbook of Swiss Administrative Sciences](https://ssas-yearbook.com/articles/10.5334/ssas.120/) 9(1), pp.66–79.
After discussing the role of the **Community** in validating use cases, our next goal was to explore open data ourselves in a **Hands-on** way, looking at what happens behind the scenes and trying out some open source tools on datasets we researched together in class.
We looked at some community projects, in particular discussing [Open Budgets](http://make.opendata.ch/wiki/project:open_budget) ([Open Data Camp Bern 2013](http://make.opendata.ch/wiki/event:2013-03)) and [Predict Delays](https://github.com/OpenDataDayZurich2016/ODDPredictDelays#synopsis) ([Open Data Day Zürich 2017](https://hack.opendata.ch/project/81)). We also watched a short clip of last year's [Open Food Data Hackday](https://twitter.com/OpendataCH/status/1060145640569798656).
The hallmarks of open development explained let us launch into a mini-hackathon inspired by the [make.opendata.ch](http://make.opendata.ch) events. We divided into 4 teams of 4-5 people each, and took up roles (Expert - Designer - Developer), brainstormed and researched open data sources, and tried to find or even (very) rapidly build prototypes with a ticking countdown clock. Two blocks of 45 minutes were spent on the exercise, which was followed by 5 minute pitches from each team, along with discussion and feedback focusing on unravelling the barriers which prevented teams from getting closer to the challenge they picked or using the data they wanted.
The topics our four teams picked were (the original top-voted A and B challenges were not responded to):
### C. How many snow days will there be in Bern in 2025?
The [NCCS reports](https://www.nccs.admin.ch/nccs/de/home/klimawandel-und-auswirkungen/schweizer-klimaszenarien.html) were compelling, but the available data could reportedly not be used to make a prognosis. However, [Meteoblue](https://www.meteoblue.com) ([CSV](https://www.meteoblue.com/de/wetter/archive/export/basel_schweiz_2661604?daterange=1985-01-17+to+2019-01-24¶ms=¶ms%5B%5D=677%3Bsfc¶ms%5B%5D=678%3Bsfc&utc_offset=1&aggregation=hourly&temperatureunit=CELSIUS&windspeedunit=KILOMETER_PER_HOUR)) allowed us to export data for the Basel region, and the team showed us a compelling visualization. The team also quickly assembled a document of their work, complete with code and screenshots.
### D. Does religion have an influence on life expectancy?
Although the data is available from the Federal Office of Statistics, the team had trouble merging it into a single data analysis. See
[Sprachen-Religionen](https://www.bfs.admin.ch/bfs/de/home/statistiken/bevoelkerung/sprachen-religionen/religionen.assetdetail.4242801.html) and [Lebenserwartung](https://www.bfs.admin.ch/bfs/de/home/statistiken/kataloge-datenbanken/tabellen.assetdetail.3524901.html).
### E. In which hospital should I seek treatment?
This group was very interested in open data apps, and broke down the question into four contexts - the media and opinion, subjective treatment satisfaction, reviews of the corresponding treatment area, waiting times in the hospital emergency room. For all but the last question they found interesting leads at [opendata.swiss - health](https://opendata.swiss/en/dataset?q=health) ([Hospital Statistics](https://opendata.swiss/de/dataset/key-data-on-swiss-hospitals-2016), in particular, e.g. as used in [Compare Hospitals](http://make.opendata.ch/wiki/project:compare-hospitals) app) and the [Swiss National Association for Quality Development in Hospitals and Clinics - anq.ch](https://www.anq.ch/de/fachbereiche/akutsomatik/messergebnisse-akutsomatik/step3/measure/1/year/2017/page/2/search/Hirslanden/?no_cache=1). Finally, they found through the [opendata.swiss Apps](https://opendata.swiss/en/app/) catalog the [welches-spital.ch](https://welches-spital.ch/alle-bewertungen.php?hid=72) app, which was demoed to the class with interest.
### F. How ecological are electric cars?
This team started with a lot of initiative, and a clear hypothesis: the electric car has a lower energy consumption than a combustion car. Their initial interest was in the maintenance cost, and they compared various co-factors, such as purchase cost and depreciation, taxes / duties / subsidies, maintenance costs, energy / fuel during lifecycle as well as production and transport (total environmental impact), infrastructure (charging stations), environmental impacts (emissions, noise). They cited sources such as TCS (Swiss motor owners association), Frauenhofer Institut, ASTRA (federal department of roads), and the open data portal. They also researched press articles - [TagesAnzeiger](https://www.tagesanzeiger.ch/wirtschaft/standard/Studie-Elektroautos-sind-doppelt-so-umweltfreundlich/story/21623211), [ADAC.de](https://www.adac.de/der-adac/motorwelt/reportagen-berichte/auto-innovation/studie-oekobilanz-pkw-antriebe-2018/), [Energie-Experten.ch](https://www.energie-experten.ch/de/mobilitaet/detail/wie-stark-belastet-die-batterieherstellung-die-oekobilanz-von-elektroautos.html), [movi-mento.ch](https://movi-mento.ch/de/artikel/2016/wie-oekologisch-sind-elektroautos-wirklich.html), [EMPA](https://www.empa.ch/de/web/s604/batterien). That last one led them to an interesting data source ([ecoinvent](https://www.ecoinvent.org/database/database.html)) for further research.
Besides all this information digging, the team took the hackathon format very seriously, outlining the "building blocks of success" in their notes, which I will paraphrase here:
* Common understanding of the problem
* Formulate hypothesis
* Organization in the team
* Inform yourself about the topic
* Data retrieval
* Explore, clean, add more data to the data
* Analyze, explain / present
Very impressive, all these projects - everyone put a genuine effort into teamwork, data exploration, and the pitching. In my wrapping comments, I mentioned several improvement suggestions about communicating insights and documenting the journey.
One of the questions asked was about the methodology of hackathons, to which I replied with information about the new BFH research project that Opendata.ch is cooperating with: [#hack4socialgood](https://hackmd.io/s/HJPhKLz5M). Several people stayed back to chat some more about hackathons and civic tech, machine learning and career paths.
Many thanks once again to all of the students who enthusiastically took part in the module, to the BFH staff and my fellow teachers of the CAS. I’ll be following your progress and projects, and ready to answer any questions on the course forum, or via contact options at [datalets.ch](https://datalets.ch).
---
© [Oleg Lavrovsky](mailto:open@datalets.ch), January 2019
This work is licensed under a Creative Commons Attribution 4.0 International License.