Make our presentation file to public - what is the best way?

[hfu]

The issue

We would like to publish our presentation file, such as the presentation file for #214 (https://docs.google.com/presentation/d/1OwuZeywXDsbAmpsuYT3m70hp4psQag7O8nfHJZNzvrs/edit?usp=sharing), in an organized way.

How we address the issue

I appreciate advice from @albertkun from Documentation perspective.

I do not think this is urgent. Let's think about it when we have time.

[albertkun]

@hfu Thank you for asking me about this issue!

Overall recommendation

A) Automation! 😄

In specific, if we are going to host the presentations on the Smart Maps website, I would suggest creating the markdown file for the documentation through the use of GitHub Actions. I currently do something similar for this icon repository with a simple bash script in the GitHub Actions workflow:

https://github.com/LACMTA/metro-iconography/blob/main/.github/workflows/update-svgs.yml

Digging deeper, I have two suggestions, depending on the type of presentation you wish to host:

Online presentations

Examples

• Google Slides
• hack.md
• canva

In this case, I would recommend either a CSV file, Google Sheet, or hosted table in a database that can be read to generate the markdown content. The columns in the file should contain metadata like, author, presentation date, etc., and this can be used to automatically create the markdown file. The flow could look like this:

flowchart TB

A[`Google Sheets` or `CSV` file] -->|Add info to `table`|B(Round)
B[ `cron job` or `github actions`] -->C{Was `table` changed?}
C --> |Yes| D[Update `index.md`]
C --> |Yes| E[Create new `.md` file]
C --> |No| F[Do nothing]

Loading

Offline

Examples

• *.pptx
• *.pdf

In this case, I would either recommend hosting the file in a folder presentations in a git-lfs enabled repository that can then trigger a workflow to generate the content based on the file name. Metadata for these files (author, meeting type, etc.) would be harder to add for these files.

The flow would look like this:

flowchart TB

A[Upload file to `presentations` folder] -->B
B[ `cron job` or `github actions] --> C{Did `folder` change?}
C -->|`Yes`| D[Update **Presentations** `index.md`]
C --> |`Yes`| E[Create new `.md` file]
C --> |`No`| F[Do nothing]

Loading

B) Alternative to automation

Manually updating the presentations and the index.md file on the documentation website! This would mean going to each link and adding the content there! Could be an interim solution as it would be the quickest.

Feel free to let me know what you think would be the best going forward!

• Resources
  • https://github.com/marketplace/actions/generate-update-markdown-content

[hfu]

@albertkun Thank you! It looks to me Automation of Online Presentations would be the best option!

[albertkun]

Ok, that sounds good!

Do you have a preference on whether we use a static csv file, database, or Google Sheet?

Static CSV

➕ Easy to access, but method for getting the data is simple
➖ Harder to input data for non-technical

Google Sheet

➕ Easy to input data
➖ Dependent on Google's API access (which could change in the future)

Database

➕ Potentially useful to setup larger database infrastructure (can build a front end to input data)
➖ More time, harder to setup because need to find a host

[hfu]

I definitely like static. Can we use a static CSV file?

[albertkun]

@hfu yes we can! we should probably decide what fields we would like to use. Here is some suggestions:

Column	Data Type	Description	Example
Name	String	Who gave the presentation	Albert Kochaphum
Date	Date	When the presentation was given	2023-08-20
Link	String	Link to the presentation	https://hackmd.io/@MXxudfFtSvSvSonMo4L3qQ/SyT0kFwqn#/1
Topic	String	Keywords for the presentation	documentation, UN Smartmaps

[hfu]

I like the simplicity of your suggestion.

Discussion question

1. regarding name how should we handle a presentation done by two or more persons? e.g. the last UN Smart Maps Quarterly presentation was both developed and delivered by yuiseki and hfu.
2. maybe we can discuss whether the name should be of file owner or the actual presenter.
3. The result of the discussion may depend on how we should handle the name. If we can handle name as just annotation or metadata, we can keep name as flexible as possible, like free text annotation. If we use name to sort or categorize decks, there should be a standard.
4. My suggestion is that we can keep the meaning of name as flexible, and we use only date to sort the decks. We may use topic as tags.

FYI: example from speakerdeck

[albertkun]

Good point about the use of the name field! If we do want to keep authors, we can have an additional field for presentation title!

Regarding multiple authors, we can include as many names as long as it is in quotes like the following:

"Hidenori Fujimura and Yuiseki Matsumura", 2023-08-09, UN Smart Maps Quarterly Update, link

[hfu]

Kaz Sato-san from Shizuoka prefecture kindly offered me that he can provide his presentation material for #219. He said he can provide the material when we are ready.

[albertkun]

@hfu Thank you for letting me know! I am gradually working on the presentation automation (of course, it is a bit slow since I am still traveling around Tokyo..) However, here is the link to presentations.csv where you can add the link to Kaz's presentation:

https://github.com/UNopenGIS/smartmaps/blob/main/presentations.csv

As a reminder, the format is:

Column	Data Type	Description	Example
Presentation title	String	Name of the presentation or meeting	UN Smart Maps Pacific 5 Documentation Update
Name	String	Who gave the presentation	Albert Kochaphum
Date	Date	When the presentation was given	2023-08-20
Link	String	Link to the presentation	https://hackmd.io/@MXxudfFtSvSvSonMo4L3qQ/SyT0kFwqn#/1
Topic	String	Keywords for the presentation	documentation, UN Smartmaps

or in the csv file:

"presentation title",date,speaker,link,keywords

[hfu]

#274