wdencker

4 lata temudiscuss

Great list! Overlaps somewhat with my own (incidentally also titled “life advice”): https://trove.to/wes/trove/de40440b-ed21-4fdb-ba61-b5f83322b...

wdencker

4 lata temudiscuss

This is very much in the spirit of what we were trying to do with trove.to [1] — give people an easy way to curate & annotate lists of websites, and layer a social graph and endorsement system on top of those lists.

The problem we encountered is that the vast majority of people are not hyper-organized list makers — the 1% rule of the internet [2]. To create a "human curated search engine" with any utility, you need a massive amount of manually-categorized data — data which most people are simply not interested in generating. This is why no social bookmarking site (e.g. delicious, pinboard, etc.) has ever taken off to hundreds of millions of users.

I still think there's something exciting to be built here, but it will likely need to take a more "automated" approach as you suggested.

[1] https://trove.to/

[2] https://en.wikipedia.org/wiki/1%25_rule_(Internet_culture)

wdencker

5 lat temudiscuss

Hey HN! It's been 3 months since Trove's original Show HN [1], and since then we've released tons of new features and improvements including:

  - a fully-fledged WYSIWYG + Markdown editor
  - a Chrome extension [2] to take and save notes directly from the browser
  - image uploading
  - collaboration (can invite others to add to your trove)
  - feeds (to see new blocks from people you follow)
  - a new logo!

Trove can be used in so many different ways, e.g. for saving and annotating articles, organizing your recipes, sharing book recommendations, planning trips with a group, doing product research, creating a syllabus for self-teaching, or even just jotting down ideas. Use it long enough, and eventually you'll have a curated "knowledge library" for every topic you're interested in — see my profile [3] for instance.

Give it a try and let me know if you have any ideas or feedback!

[1] https://news.ycombinator.com/item?id=26582658

[2] https://chrome.google.com/webstore/detail/save-to-trove/fioh...

[3] https://trove.to/wes

wdencker

5 lat temudiscuss

Agreed. I’m working on one such tool: https://trove.to/

wdencker

5 lat temudiscuss

Interesting observation. I wonder if these are issues intrinsic to the sites themselves, or if instead there’s some fundamental law that states once a publishing platform reaches a critical mass of popularity, it ceases to yield high-quality content (on average).

I do wonder how Substack is thinking about this problem. They obviously have a different business model than Medium, but I think they are still susceptible to this issue.

wdencker

5 lat temudiscuss

Interesting... where can I see Lex Fridman's shelf?

It seems everyone and their mother is working on a knowledge management / list curation tool these days—well, including myself (I'm working on Trove [1, 2]). IMO, the key value that has yet to be unlocked (though we're trying to do this with Trove) is layering a community / social network on top of a tool like this—so that it's not just a single player utility.

Always happy to chat / share notes if you have any thoughts on this! (email in bio)

[1] https://trove.to/

[2] https://news.ycombinator.com/item?id=26582658

wdencker

5 lat temudiscuss

For an academic take on "how-to Linux", I'd recommend CMU's CS 15-213 course [1]. The systems class I took in college borrowed liberally from it, IIRC.

Also, I'm putting together a master list [2] of the best resources from this thread and the other one OP mentioned. Let me know if I'm missing anything!

[1] https://www.cs.cmu.edu/~213/schedule.html

[2] https://trove.to/wes/trove/learn-linux

wdencker

5 lat temudiscuss

There's a ton of different note taking methods, as you can tell by the variety of responses here. People will evangelize one method as being superior to another, but I think you just have to try them out and see what works best for you.

IMO, a prior question to understanding the information we consume is deciding what information to consume. There's an essentially infinite amount of content out there, but we have a finite amount of time to learn. The strategy we usually employ is random and mindless—we consume whatever happens to show up in our HN or Twitter feeds. But I think there's a real value in thoughtfully curating the knowledge we take in, which is why I built Trove [1,2].

I'm curious to hear, how does everyone else here optimize their information diet (if at all)?

[1] https://trove.to/

[2] https://news.ycombinator.com/item?id=26582658

wdencker

5 lat temudiscuss

Yeah, for your workflow here I don't think there is any better solution, especially because of the Kindle / Goodreads integration (which Amazon will never open up to other services).

As the Fred Wilson piece I linked above notes:

> I wonder if listmaking is really a vertical thing instead of a horizontal thing. That would suggest that there will be successes in verticals like food, travel, shopping, reading, film, music, etc but that each will be its own thing and not part of some meta listmaking community.

With Trove we're trying to build something closer to the "horizontal" thing, but I suspect power users of a particular vertical (e.g. "power readers" like yourself) will still flock to vertically focused and integrated solutions (like Goodreads for books).

wdencker

5 lat temudiscuss

While it's a very difficult problem to crack, I do think there's a way to serve both the bookmarking ("single player") and recommendation ("multiplayer") use cases with one product. Fred Wilson has a good article from 2015 on this dilemma [1].

My personal opinion is that Goodreads and other similar services have failed to do this because they don't make their user-generated content a first-class citizen. The focus is on rating and making lists of books, not their users' thoughts and annotations on those books (which would tie content creation and consumption into a much tighter feedback loop).

I'm working on Trove [2] because I think this is a problem worth solving — we recently did a Show HN [3] you may have seen. Would love to hear if anyone has more ideas on how to tackle this.

[1] https://avc.com/2015/11/lists-2/

[2] https://trove.to/

[3] https://news.ycombinator.com/item?id=26582658

wdencker

5 lat temudiscuss

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-...

wdencker

5 lat temudiscuss

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 :)

wdencker

5 lat temudiscuss

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

wdencker

5 lat temudiscuss

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.

[1] https://trove.to/wes/trove/the-ultimate-wfh-setup

wdencker

5 lat temudiscuss

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.

wdencker

5 lat temudiscuss

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?

wdencker

5 lat temudiscuss

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!

wdencker

5 lat temudiscuss

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.

wdencker

5 lat temudiscuss

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!

wdencker

5 lat temudiscuss

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.

Submissions

Show HN: Trove – Curate and collaborate on internet knowledge

Educational Youtubers

Show HN: Trove – Curate and publish a knowledge base on any topic

Substack and Crossroads for a Revolution

The ultimate WFH setup

Software that actually improves my life

Bootcamps and ISAs: Economics, Challenges, and Opportunities

comments