User:AmilOda/Blog

Status report for CASE study [CASE - Praxisprojekt (HS16)]

This page only reflects the status for the given dates. For a more up to date status on my current progress please visit my notes page.

16.09.2016

The official first week of work. Although I already unofficially started working in August. I had some difficulties understanding the structure of Wikidata, but after consulting the notes of Beat Estermann I think I have a basic grasp of it now, though a deeper understanding will surely come as I progress with the project.

I have not yet cleaned up the data from the Swiss PCP inventory, as I am not sure yet on how to tackle the large amount of data. I’m not sure it is possible within Excel, so I might have to resort to another program. I need to ask my Auftraggeber or Expert on how best to do this.

So far I started with the mapping of the data fields (this has basically been done already, but I did again for myself in order to better understand the data I'm dealing with).

Started with Listeria Tool in order to find out which datasets from the Swiss PCP Inventory are already in Wikidata. After first few tries I managed to get some results, though I still have to interpret and order them correctly. Weirdly enough I cannot seem to be able to display the PCP reference number (P381) yet.

As soon as I manage to have the Listeria display properly I will start ingesting 2-3 data sets automatically as a test.

30.09.2016

After discussing with Auftraggeber and Expert I was advised to first clean up the datasets and translate the most common words before uploading.

However I was not able to work on this because of work commitments.

14.10.2016

I started setting up my CASE document with a structure and began writing the introduction.

We had a workshop this week about appropriate sources and literature for our paper which made me realize that I don't have any appropriate literature yet. I will need to work on this matter ASAP.

28.10.2016

The last two weeks I was not able to do any significant work for my CASE. However starting November I will be reducing my workload for my employer from 70% to 50%, which will give me the much needed time to work on my CASE and catch up.

11.11.2016

On advice from my supervisor I created a Kanban board with Trello in order to retain an overview on the open tasks.

I extended the dataset (KGS-Inventar, A-Objekte, Juni 2016) with columns for labels and descriptions in German French, Italian and English. Then I continued with translation work and clean up.

The coordinates are currently using the Swiss coordinate system (Q349628)   , but should be converted to ‘coordinate location (P625) ’. I found an online tool to convert coordinates one at a time, but ideally there’s a tool to bulk convert. So far I found one here, but I need to be active in the community in order to download the file. I reached out so that I can access the file, but maybe I will need to create the excel myself using this formula.

We also received an SQL dump of the WLM data. One table (monuments_ch_(de)) was missing in the dump, but I still tried to convert it to csv-file. I was not able to do the convertion, so I enlisted help from a developer colleague for this weekend.

27.11.2016

I managed to export the MySQL to different csv files (one needs to run a local MySQL server on the machine, it is not enough to just install the workbench). Further I then created separate lists based on A or B category. Then using OpenRefine I managed to merge the A_monuments_ch_(de) list with a-kgsinventar-2016-20160607 list. Using kgs-nr as the unique identifier I copied over the Lat and Lon, so that there will be no need to use a formula to convert it. For the missing datas the formula will still be used.

With the A_monuments_ch_(fr) and A_monuments_ch_(it) lists I was constantly getting "Error: Cannot retrieve field from null", even though the Lat columns in those files had data in them. I will need to do some "debugging" here.

Now I will start experimenting with the QuickStatements Tool.

09.12.2016

Using instructions from page 12 of this document I was able to convert some coordinates. Unfortunately my results don't seem to be correct. Using y=700000 and x=100000 I wasn't able to get the same numbers. The final numbers for λ and φ were not correct. But the second-to-last numbers of λ' and φ' are.

First steps with the QuickStatements Tool. Managed to Update an existing entry (Government Building (Q2137740)   ):

Q2137740	Lde	"Regierungsgebäude"
LAST	Lfr	"Siège du gouvernement cantonal"
LAST	Len	"Government building"
LAST	Dde	"Gebäude in Aarau (Schweiz)"
LAST	Dfr	"Bâtiment à Argovie (Suisse)"
LAST	Den	"Building in Aarau (Switzerland)"
LAST	P969	"Regierungsplatz"
LAST	P281	"5000"

Also experimented on adding new items (residential house Huber (Q27986591)   ):

CREATE
LAST	Lde	"Wohnhaus Huber"
LAST	Lfr	""
LAST	Len	""
LAST	Lit	""
LAST	Dde	"Gebäude in Riehen (Schweiz)"
LAST	Dfr	""
LAST	Den	""
LAST	Dit	""
LAST	P17	Q39
LAST	P131	Q12172
LAST	P18	""
LAST	P373	""
LAST	P625	"47.573637, 7.648167"
LAST	P1435	Q8274529
LAST	P969	"Hackbergstrasse 29"
LAST	P31	Q41176
LAST	P276	"Riehen"
LAST	P281	"4125"
LAST	P381	"1904"