16 Feb 2024 Datahub Cloud Coworking and planning
16 Feb 2024 Datahub Cloud Coworking and planning
Present: Daniela, Yoana, Rufus
https://lifeitself.org/tao/meetings#types-of-meeting
Agenda
- Check in
- Building our agenda today
- Clearing up our doubts and questions we need to answer
- Editor Guide ✅2024-02-16 we worked on this extensively with results in shaping doc in datahub-next
- Sorting the front page etc ✅2024-02-16 rufus wrote up options.
- AOB
- Access https://github.com/datahubio ✅2024-02-16 daniela and yoana have acccess now
- How can Rufus do development without setting up an entire local environment - at least for rendering a project (can we render local content for testing somehow - so we don't even have to hit github??)
- Are we trying to hire another mid-level PM
Question about the product
Rufus (later): see https://github.com/datopian/product/blob/main/areas/pages.md and the hypothesis tree https://coggle.it/diagram/YguYTjL-yUb8Ivf0/t/datahub-pages-hypothesis-tree-a-product-a-data-marketplace
Key point here is that we have credible belief here that people could want this and we are going to try and rapidly adjust based on feedback
- We know that X amount of people have signed up for Flowershow
- We have a funnel
- We have built tools like this for 2 decades and people are still adopting them (adoption is inclining)
- We know X amount of people have published their datasets on Github
Daniela's questions
- What problem are we trying to solve? Not being able to publish my dataset elegantly and share with others doesn't seem like a problem to me but rather a solution masquerading as a problem.
- RP: Please say more about what is the actual problem you see if this is a "solution masquerading".
- Why do we think Datahub Cloud is worth investing into?
- What problem are we solving LONGER TERM?
- 🔑 (RP) I have a few ideas for where we can go e.g. data marketplace, (open) data community (like kaggle etc), data processing, enterprise etc. AND right now we are focused on this being useful to at least some people
- Are there enough people having this problem? Eg. is our solution a must have or a nice-to-have rather 🔑 (RP) we don't know until we offer it. So we should offer asap.
- Are we really going to produce a page that provides more value than a standard Github repo?
- What specific challenges do data professionals and enthusiasts face when trying to present and share their data (from Github?) using existing tools and platforms? 🔑 we know people like 538 produce their own mini catalogs. We know github doesn't quite cut it for us … And we will only find out by asking people and having an offer
- I can create diagrams in Github https://docs.github.com/en/get-started/writing-on-github/working-with-advanced-formatting/creating-diagrams and you can't create a line chart there but I can add it as an image or maybe another way of doing it. Or is there a way to have a line chart on Github repo?
- What specific challenges do data professionals and enthusiasts face when trying to present and share their data (from Github?) using existing tools and platforms? 🔑 we know people like 538 produce their own mini catalogs. We know github doesn't quite cut it for us … And we will only find out by asking people and having an offer
- We keep mentioning elegant and beautiful but what we have right now or what we are planning on having in v1 is rather standard/average atm. 🔑 ok 2 answers: a) we still soft launch with what we have, do an invite process and rapidly iterate from people's feedback b) you do the GM approach - but then you need money and buy-in. I'd opt for (a) which is what we are doing
- Eg. FlatUI table component is GREAT but it seems there is an issue with it and we'll prob need to
- What problem are we solving LONGER TERM?
Yoana questions
- How do we position DataHub Cloud within the 2024 trends to ensure it is perceived as a cutting-edge tool for data presentation and data storytelling? Who cares? Everyone goes on about AI etc and we have a hypothesis people want to share datasets better. We may be wrong and in that case we will find out but we aren't offering an AI
- Given the rise of AI and machine learning in driving innovative data interpretations, how can DataHub Cloud integrate these technologies to enhance data visualization and storytelling? Who knows at this point
Answers
- We are building a product ON TOP of Github and not such that will compete with Github
- We are using the gaps that Github currently has and providing solutions for these gaps
- Daniela: Are we for the v1 really? We dropped random data, we dropped the catalog, we dropped the nav. V1 will be around an elegant and beautiful page. Atm it is not very elegant and beautiful IMO. It is okay and average. (I felt a relief after putting this in text)
