Fundamental questions

[Syndesi]

Hi,

thank you very much for the repository and road map in the first place, I really appreciate this effort :D

I would like to ask a few fundamental questions, which are likely helpful to keep the resulting roadmaps on a clear path:

• What is the difference between Graph Engineer and Graph Scientist?
• What is the common overlap between Graph Engineer and Graph Scientist? Related: The overlap between both professions could be documented as a "general graph roadmap", such that you can avoid duplicated content in both specializations.
• Where should the Graph X roadmap start? I.e. should we keep basic and mid-level computer science topics like Vim and APIs? I personally would advice against them, as they are not themself graph topics. However you could add a "related computer science topics" page, where such things are collected.

Graph theory suggestions:

• The listed topics are a great start, however adding some more context to them could be useful:
  • Link to a page which further explains the topic.
  • Maybe a minimal example, if possible.
  • A very short description what it does / how you can use it?
• A specific block for graph algorithm classes could be very useful (community detection, shortest path finding, ...).

Graph language, graph modelling and graph database suggestions:

• Most graph technology can be put into one of the following two categories: Property graphs vs triple stores. As the distinction is on a quite low level, it affects most higher level concepts as well. Explaining the difference between both and their history, strengths and weaknesses could help beginners to better understand the graph technology landscape :)
• Databases can furthermore be identified by the following categories:
  • Graph native storage vs multi modal storage: Graph native storage is usually faster, while the second category usually supports alternative query languages as well.
  • Open source vs closed source, or free access vs paid access.
  • Support of query languages (Cypher, Open Cypher, GQL, Sparql, Gremlin, SQL, custom, ...). Most databases support multiple languages.
  • Does the database support horizontal scaling? (Most don't).
• Query languages are often inspired by each other, or directly evolve from an older standard. Highlighting this tree / timeline can help with understanding smaller differences between databases.
• Query languages are often standardized. Links to these documents should be included, although marked as advanced and optional content. E.g. for Open Cypher: https://opencypher.org/ , https://s3.amazonaws.com/artifacts.opencypher.org/M23/railroad/Cypher.html .
• Data modelling: A few techniques like normalization apply to most databases, others to specific technologies. Every major variant (property graphs, triple stores) should be represented.
• The list of graph databases is unlikely to be ever fully complete or correct, so listing a few major ones including their websites, documentations and forums is likely the best idea. To also represent other ones, you could link to https://db-engines.com/en/ranking/graph+dbms .

[denironyx]

Hi @Syndesi ,

I appreciate you initiating this issue and providing insightful questions:

Differentiation between Graph Engineer and Graph Scientist: Graph Engineers focus on designing and implementing graph data structures, while Graph Scientists delve into data analysis, employing graph analytics and data science techniques. Their expertise lies in deriving insights and predictions from graph data, often with less emphasis on data storage or management.

Common Overlap between Graph Engineer and Graph Scientist: Currently, the common ground lies in understanding network fundamentals and graph concepts. However, their backgrounds may vary, with Graph Engineers often coming from data engineering or software development backgrounds, while Graph Scientists may have experience in data science or analysis.

Starting Point for the Graph X Roadmap: Personally, I see value in addressing foundational knowledge, although I acknowledge that this approach might seem too broad. Including additional resources for beginners could offer a more accessible entry point.

Graph Theory Suggestions: The suggestions for graph algorithms seem geared towards the Graph Scientist roadmap, focusing on sophisticated data analysis techniques. However, suggestions about examples, descriptions and linked pages make sense. I am starting to think we might end up creating a markdown page.

Graph Language, Modelling, and Database Suggestions: Your suggestions regarding graph technology are valid, especially considering the continuous emergence of new tools. I propose reviewing these aspects annually, possibly in conjunction with the Year of the Graph project.

In conclusion, We welcome pull requests if you're inclined to contribute further. Your feedback has been invaluable, and I appreciate the thoroughness of your review.

Thank you once again for your insights.