After years of writing books, the Bubble User Manual, forum threads and blog posts, Iāve finally jumped into video content. Iām starting with the basics, and will dive into more advanced topics over time.
Check out the first video below, and subscribe on YouTube to stay in the loop!
This is separate from the work I do with Bubble, so Iāll put my personal spin on it ā check it out and let me know what you think!
However, I think the Bubble documentation is terrible.
Iām sorry to hear that! What do you think we should focus on to improve it?
search in both the bubble manual and the forum is oddly badā¦
for examply, i was looking for a way to count characters in a text field and a search for ācharacter countā in the manual came up empty. the forum only had a solution from 2018 or so using regex (yikes), so i asked on the forum and it turns out thereās an operator called ā:number of charactersā which -one would think- would/could/should have been suggested when searching for ācharacter countāā¦?
also, i asked once about knowledge management in the forum and the response was to go into google and enter the topic plus bubble.io. hmm.
would be useful to have a cross-referencing KMS mechanism with all the bubble resources (or at least the manual and the forum) - maybe a task for bubble AI ā¦?
I agree with you that Gitbook (the platform used for the docs) unfortunately doesnāt have great search capabilites. Itās fairly limited in terms of understanding context versus simply searching for keywords (like your example highlights, you wonāt find ācharacter countā unless we specifically say character count in the text (and preferably in the header) of the relevant section). Rectifying this one example is of course perfectly doable, but itās a challenge for content overall.
That being said, itās clear that AI and LLMās will gradually replace traditional search. As you probably know, LLMās are much better at understanding a given context, as opposed to simply searching through text for keywords.
Gitbook has its own AI feature which does help with some questions, but overall, in terms of search, Iāll have to ask for some patience while we work on implementing more robust AI/search features. It may seem on the surface that an LLM simply can be trained on the material we already have, but itās worth remembering that as Bubble hosts millions of apps that need to stay stable, performant and secure, itās important that we can confidently say that an AI agent answers questions correctly and doesnāt hallucinate or spread misinformation due to misinterpreted content.
I canāt share the specifics of our discussions at the moment, but based on Bubbleās official communications and the broader tech industry trends, itās evident that support and documentation are shifting toward AI. This shift will significantly enhance context-aware inquiries.
Indeed a new video (hopefully high quality ) is out today! We wonāt be diving into the editor just yetāthis time, Iāll be focusing on design theory. Iāll cover some key design terms and principles to kickstart the conversation about one of the most important aspects of app design: responsive design