Show HN: Trove – Curate and publish a knowledge base on any topic
trove.toCreator here, happy to answer any questions.
Trove is a project I’ve been working on for about 3 months now. The initial idea was borne out of a frustration my cofounder and I had while researching startup ideas. We would often come across excellent resources that related to one of the ideas we were investigating, but we found there was no great way to organize these resources, annotate them with our own thoughts, and share our annotations with one another. Text messages were unwieldy for long-form thoughts, Google Docs & Notion didn’t provide enough structure, and emails quickly became buried. In a world of abundant information, simply being able to sift through and organize it all is now the biggest challenge.
What we believe is missing is a “GitHub for knowledge” — a place where we can not only store our own knowledge, but also build from others’ ideas [1]. Trove has a long way to go as a product (right now, it’s a relatively simple React/Next.js app with a Django REST backend) — we would love to hear any feedback or ideas you might have!
i'm someone who keeps a ton of github knowledge bases:
- https://github.com/sw-yx/spark-joy/
- https://github.com/sw-yx/launch-cheatsheet/
- https://react-typescript-cheatsheet.netlify.app/
i should be your ideal user.
i was very excited at the idea, but tried out the product and its clearly not there yet. but keep going.
specifics:
- i was looking for search scoped to my trove
- collaborative editing, if it exists, is not obvious
this is what i was looking for and i stopped once i didnt see them exist.
good luck!
I recently created a web application for curating one big shared public knowledge base.[1] Its contribution system is conceptually based on GitHub's pull requests and StackOverflow reputation.
It's focused on tips for learning resources, so it seems ideal for your cheat sheets. Currently, creating lists of SDKs, tools, etc. is not very well supported, but we are working on it. And I already use it for these things, see for example my list of Flutter libraries.[2]
I guess it's also kind-of similar to Trove, but it is more focused on creating one curated catalog of knowledge rather than personal recommendation lists (although that is also possible now via personal study profiles, e.g., [3]).
[1] https://knowledgepicker.com
cool - yea i think some amount of personal perspective/curation is what makes this special to me. nice project!
Thanks for trying it out, and for this candid feedback! You're right, the product is still a ways off from becoming a true "GitHub for knowledge" — things like "forking" a trove, live collaborative editing, and more powerful search are some of the missing elements.
We'll keep working!
That launch cheatsheet link is going in my "Entrepreneurship Journal" Google Doc ;D
Are you familiar with https://kit.co/ (a service Patreon acquired and later dealt)?
There's also https://satchel.com/ in the B2B SaSS space.
Does anyone think there is a market for a wikipedia/wirecutter/consumerreports-like building tool, for communities to build recommendations by consensus?
I am familiar with both — as far as I can tell they're both great products in their own right, very focused on their particular niche (Kit for products, Satchel for B2B SaaS).
You can use Trove to create similar things; for instance I have a trove on my recommended WFH setup [1].
Long term, we believe a general-purpose tool may have the most potential. The ability to search for any recommendation, and get back a result that's been vetted (i.e. "starred") by my network would be game changing, IMO. Obviously, we've got a long way to go before we can reach that level of utility.
I would be interested in what is sort of the inverse of Medium's latest action. I very much disagree with Ev William's statement of "the idea of an imprimatur that establishes credibility or trust. Trust is more important than ever and well-established editorial brands still have meaning. But today, credibility and affinity are primarily built by people — individual voices — rather than brands."
What would make your product most valuable to me as both a consumer and a contributor would be a way to establish an aggregate reviews in a specific way, such as by editorial board or brand. A group of reviewers working together. Sort of like a subreddit. Not only being able to sort reviews by author or people I follow, but compare the rankings of the consumer reports team to the wirecutter team (examples of competing review brands.) I want to look at a masthead and trust their overall review quality standards, to find the reviewers I can trust.
A CMS to host reviews is cool. A way to collaborate and build complex team consensus is even better.
In march 2020 I started http://ontol.org/ - my version of "Github for wisdom". Wisdom - is from DIKW pyramid "Data-Information-Knowledge-Wisdom" (https://en.wikipedia.org/wiki/DIKW_pyramid)
Glad to see what I am not alone! World need our hearts, hands and minds!
Now Ontol in close beta stage and I testing hypothesis in tiny russian auditory. But after this discussion, I will speed up by 300% ))) Thanks!
(My dream is to meet you in YC batch)
It looks like it lets you organize and comment on URL's? Is it of any use if you have content that isn't already on the web and isn't neatly tagged by URL? E.g. your own photos.
Not yet, though this is a direction we've been thinking about. We know a fully-fledged "knowledge base" may not only consist of annotated URLs — it may also contain the author's own thoughts / content (whether that be text, audio, images, or video).
Ultimately, we think there's a need for a multimedia publishing platform that's more lightweight than something like Substack or Medium—everyone has great knowledge to share, but few want to start their own newsletter.
How is the infrastructure that you run this with? I would love some advice for future startups and side hobbies. What are your costs and plans to scale? I'm especially interested in the plans of new startups who might not go all out with hosting and servers due to budgeting and what not in the beginning.
Everything is run through 2 Digital Ocean droplets (one for the Next.js instance, and one for the Django REST server). I'm not a sysadmin by any means (I just followed a guide[1] online), but it seems to be working well enough.
All in (including the managed database) it's $65 / mo, which seems reasonable to me. (Could be even less, but I scaled up the droplets for this launch).
[1] https://www.digitalocean.com/community/tutorials/how-to-set-...
The concept is great. One major thing stopping me from trying this out though is ease of putting something I see to within the CMS (which I am feeling like conceptualizing as?). A browser extension would go a long way towards helping this.
A separate point since I see you're interested in YC and similar startup culture, I know this was made as a deep dive, pretty detailed explanation of using Trove, but some small details can be skipped over because when you've got the benefit of both visual and auditory media, people can jump to answers far quicker than you can say them. e.g. explaining the concept of your top Troves. It's readily understandable without explanation (maybe even without a mention as people see it in the hotspot of the page as you scroll!) and at least personally, I prefer to see things and interpret them in videos as if I was scrolling through the page myself.
Thanks for your feedback here!
On your first point, we are in fact working on an extension to help make trove creation easier. I also think we can make the editor experience a lot better, especially for less technical users who don't know Markdown.
On your second point, you're totally right that the demo could be a lot more compact. Honestly, I know I could have made it a lot more polished / rehearsed — but since we're changing the product a lot at this stage, I figured I'll be making another one soon enough :)
This is a very cool idea, and I was actually working on something similar a few years ago (a "GitHub" for design documents, mainly aimed at Product Managers). My enthusiasm fizzled because of extreme competition in that sector, but I really like what you guys built.
Thanks for your comment! Very cool. As an ex-PM the use case makes a lot of sense to me—Figma makes things a lot easier but still so much of the Design-Product back and forth happens over email, DM, and Google Docs. I do wonder if there is still something missing.
I enjoyed putting thought into curating Troves that weren't just visual references. Would love to see more expression tools for collections beyond just adding annotation.
Great to hear! I would love to hear what you think these new expression tools might be. Just spitballing here, perhaps you could record audio on a block instead of writing? Or are you thinking even more outside the box than that?
This is great. A feature request: screenshots. I've been on the hunt for a way to collect UI snippets and ideas and this is near perfect.
Dig the UI, name, and logo, too. Nice work :)
Thanks for trying it out! The ability to upload your own screenshots / images is on our roadmap. For now, you can still include images hosted elsewhere on the web through standard Markdown syntax.
For example, one trove that does this: https://trove.to/wes/trove/interesting-wikipedia-articles
A lot of internal knowledge I work with can't be moved to SaaS.
Does anyone know of any open source solutions for similar KB user experience (quality as the first order primitive). COTs are fine... but me and everyone else are already on Sharepoint and that's not going away -- so any solution needs to integrate; at least on the surface... understandably there are indexes/embeddings that need to exist for KB retrieval.
Really cool idea— have you considered adding a chrome extension? It'd be nice to save resources as I browse the web.
Thanks for your comment! In fact, we're working on a Chrome extension now. Beyond just allowing users to save things as they browse, we think there's some other interesting applications—for instance, if I came across a page that someone I follow had in their troves, I could then see their notes / annotations directly in the browser.
Open to suggestions if you have more thoughts on this!
Awesome product! I really connect with the social & knowledge sharing aspect of the platform that enables a more meaningful way of navigating the web. Are there any plans for expansion into user generated content to expand on the knowledge hub vision?
Thanks for trying it out!
In some sense, we believe we are already a platform oriented around user-generated content—the act of curating and annotating something is itself a medium of expression. But yes, ultimately we would like to allow users to create blocks that are not just URLs (e.g. free-form text, images, audio, and video).
For now, our focus is on nailing the use case of sharing and curating great resources, since we believe they are the critical component of any great knowledge base.