Alternative viewpoints for Categorization Mindmap to find matching patterns more easily

[Michadelic]

This idea comes from breakout #3 of the ISC Fall Summit 2020, documenting it here so that it does not get lost:

Offer additional visual mindmaps like the pattern categorization mindmap but with different viewpoints:

• Phases/Adoption - this is covered by the extisting InnerSource Program Mind Map
• Roles/Stakeholders - a mindmap listing patterns grouped by specifc roles or stakeholders
• Challenges/Categories - a mindmap listing patterns based on challenge categories (technical, organizational, culture, ...)

Side-Note: The mindmap cannot be browsed online currently, a deployment to gh-pages or some post-processing would be great so that it can be browsed and linked directly.

Kind Regards,
Michael

[fwan2000]

Love the idea of having different viewpoints to categorize the patterns! One more angle could be program oriented vs project oriented. OSPO/ISPO might be more interested in program level patterns. Personnels involved in doing an InnerSource project might be looking for patterns at project level.

[lenucksi]

Another axis:

• Likelyhood of applicability of a pattern by organizational size

[spier]

This is a great idea to make the patterns more easily discoverable.

To make the maintenance of such alternative viewpoints more manageable, we should probably look into automating it, as already raised in #162

An additional thought:
In the context of the gitbook I have also been exploring if one could generate a very simple index of important terms, and which pattern they appear in. Much like the index at the end of some books. Visually very different from a mindmap, but with the same goal of improving discoverability.

[fwan2000]

What do you think of adding labels as another section in the patterns to help discoverability? I think labels could help to generate the index or mindmap to get the same category of patterns together.

…

On Sun, Feb 14, 2021 at 6:33 AM Sebastian Spier ***@***.***> wrote: This is a great idea to make the patterns more easily discoverable. To make the maintenance of such alternative viewpoints more manageable, we should probably look into automating it, as already raised in #162 <#162> *An additional thought:* In the context of the gitbook I have also been exploring if one could generate a very simple index of important terms, and which pattern they appear in. Much like the index at the end of some books. Visually very different from a mindmap, but with the same goal of improving discoverability. — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#212 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AO55Y2KLTSLJZBQXZM22SM3S66YHLANCNFSM4ROT5W5Q> .

[NewMexicoKid]

Labels make a lot of sense; however, if it is completely free form, we might end up with many labels that belong to what should be the same label/category. If we introduce a labels section, perhaps we could have a common central labels document where people should go to find labels or add new labels (to minimize unnecessary proliferation). —Tim

[spier]

Adding labels to patterns could lead to interesting new features for the readers.

If we were to create an index page based on all patterns then that could also serve as the "central labels document" that the authors/writers can then use to use the labels in a more consistent manner.

Also we can make it part of the review process to check if labels have been used consistently, so the Trusted Committers can help with that.

Speaking Implementation for a moment:

• The labels could be placed in a ## Labels section (not would be most similar to our current template structuree
• We could experiment with using Jekyll Frontmatter as for those we would probably find ready-made parsers that we can use to extract that data.

What's the smallest experiment to try any of these ideas out?

• We can try labels as a way to generate the mindmap => Auto-Generate the Mindmap on .md-source changes #162 (This experiment doesn't deliver new value for the reader immediately, as we already have the mindmap. However technically this is interesting because we already have the labels for these patterns.)
• We can add some other viewpoints with new labels (see the various ideas in the thread above)