Would like to know how the tech community deals with bookmarks and links. They might just be read-later entries that you want to visit later on. Or could the links you have been collecting be a valuable resource? How would you share them?
1. Every article I will read, I'll drop straight into Pocket, to highlight/read. I will never rea the article outright - this is important, because some clickbait can find me in a "vulnerable" state. After this time has passed, I'd delete it out of my Pocket without reading if I determine it's not worth it. I delete about 40% of the stuff I intended to read.
In pocket, I would highlight heavily (I have pocket premium)
2. All my highlights, notes from the books I read, my Tweets, articles that I write and newsletters I send end up in my Evernote.
This gives me the single source of truth for the knowledge I am gathering. I can instantly search for a figure from an article I've read 3 years ago. I organize it by topic and treat it as a "Commonplace Book".
As such, this is not for organizing "Links", because I dont care about links. I care about quickly finding the piece of information I've seen on the web, and -of course - proper attribution.
I’ve mostly given up, since web developers seem to have forgotten that URIs should be stable. If there’s a 50% chance it’ll be gone when I go to look for it, it’s not worth the effort to me.
This is like reason #17 why I wish we had a hypertext system (like we used to have) that was separate from the application deployment system (that’s been given to us).
I capture a lot of bookmarks into my org-mode and I don't spend much time deliberately curating. Once in few weeks I go through captured bookmarks and quickly assign a priority to each one (just a matter of pressing hotkey). Then I sort everything by priority, typically ending up with 10-15 higher priority ones that I would tag, put into my reading queue, refile etc.
Rest of it isn't curated and serves as my personal search engine [0]. Often instead of searching in google I'd first search in my emacs and find some relevant information in my knowledge base.
In addition, I'm working on promnesia [1], a browser addon that integrates links in my org-mode files with my browser. E.g.
- when I visit some blog, it would show me that I've got few blog posts from that blog bookmarked (along with my private notes and annotations), which typically means that the blog is worth exploring more
- when I visit someone's twitter profile it might prompt that I've retweeted/favorited some of that person's tweets
I usually send bookmarks to my separate email account. I add excerpts and quotes from the bookmark that I find interesting, also helps with searching.
Every few weeks I conduct housekeeping on my email. This moves the good bookmarks over to my personal git wiki or deleted. This ensures that every bookmark I have gets visited at least twice. This also allows me to not overthink before adding a bookmark without ending up with incomprehensible huge lists of bookmarks that I never end up re-visiting.
I've learned over the years to keep this system simple. Understand and realize that whatever bookmark system you use will eventually go away. If its good enough, they will try to monetize, if not they will be forced to close shop. Either way it gets killed, at least for me.
Another side note: not all bookmarks are equal. Some are just references you want handy when you want to point someone to them, usually in an online argument where response time matters. Some links have golden content on the other side. You want to read, re-read, internalize and remember (spaced repetition) this. Some are work related. Some are aspirational stuff you want to get to in the future. So knowing what and why I am bookmarking something was key to keeping it real.
I use Firefox Developer Edition as my primary browser and Firefox Sync to store and sync bookmarks across devices.
I have made 95%(2000+) of my bookmarks public on GitHub[0]. I have categorized them and host them on a public repository(110+ forks, ~1800 stars) and push every week. I have also made the entire repo available as a GitBook with search and export as PDF functionality.
Search (especially incremental) is very important, there's no way I'm going to read through someone else's 2000+ bookmarks, whereas searching for topics that interest me is feasible.
Wondering why you publish both gitbook and mdbook? I was planning to use mdbook to release my bookmarks to public; gitbook development has stalled as far as I understand.
I am learning Rust at the moment. I used to use that repo to try new things out. For example, to play with GitHub's API(the github-gitlab-stars.md file was created using a bunch of Python scripts, requests library), or learning branching and pipelines(Currently I am using Travis CI to build mdbook from another branch), etc. Initially, when I didn't know about GitBook or Static Site Generators, I made a Django App with search functionality and stuff out of those links.
I honestly stopped bookmarking for real when firefox introduced the awesome bar. Just typing any fragment I recall into the address bar made them pretty much obsolete for most of my use cases.
Every time I'm forced to use chrome for some stuff I am surprised how unusable the browser is without this feature. I'm sure you can fix that with some extension or other but I don't see why it wouldn't be a default.
I've tried dozens of services through the years and have always come away disappointed. I literally just want to store a URL and the page title — just text will do. I don't want an overwhelming visual list of my bookmarks. I don't want the page content scraped and stored. I don't need to share my bookmarks (other than syncing across devices). I would like some easy way to flag and clean up old 404-ing bookmarks. Simple categorisation or tagging. And a simple one-click add-to-inbox plugin that'd work on all my browsers including on iOS. And I'd like a good-looking UI — not an ugly mess like Pinboard. And of course there should be a solid export interface so that users aren't locked into some proprietary data structure.
As a product designer I'd love to sink my teeth into this problem but unfortunately it's a dev-heavy project and that's outside my comfort-zone.
For now I'm sticking with my fallback solution — dozens of 'link dump' notes in the Mac Notes app. It's searchable, lightweight, and flexible but a nightmare to manage. There must be something better out there.
> And I'd like a good-looking UI — not an ugly mess like Pinboard.
I’ve been using Pinboard for the last 10 years or so and I’m super happy with it. Love the speed, the text-heavy interface and the high information density. I love that it never changes. I use a separate app my phone (the website is unusable on mobile) and it works great. I don’t need anything more.
I would add that the API makes Pinboard even more powerful. Add to Pinboard then use the API to do whatever you like. This can be as simple as a CLI curl. If you understand the power of the Pinboard API you would appreciate it far more.
I've been messing around with raindrop.io recently, and it certainly fixes the "ugly" problem of Pinboard. It does some things you say you don't want - page content scraped and stored - but give it a look.
I used to hoard bookmarks to never look at them again. It wasn't adding any tangible value to my life, so I got rid of that habit.
What I'm doing now instead, is curating a set of links around some topics, and sharing them publicly. The most portable way of storing and presenting them is either an HTML page on one of my websites or a markdown file mirrored to Github. That way I can be sure that my data isn't locked into some proprietary format by an entity who might get Incredible Journey-ed for whatever reasons.
I have a system of subscribing to RSS feeds and following accounts on Twitter as my source of news and continuing education, starring/bookmarking/favoriting items that look interesting, then returning later to read them - a two-pass filter basically. I'll then remove items I've read if I don't plan to return to them later.
In order to view all these saved links as well as share with anyone interested, I created a website to republish them all on one page:
I use Emacs org mode capture and syncthing and org-protocol via Firefox.
On a page that I think may be useful I click a menu item in firefox which activates org-protocol sending the URL to an Emacs server, starting Emacs if needed. Emacs prompts me for some tags and then makes an entry in an org file which I sync to all my computers, home and work, via syncthing. I'll then either simply close the Emacs frame or if appropriate leave it open while I read through the page and add notes to the org entry. Later, if I have reason to revisit then I'll search on terms my brain is able to associate and that hits on the tags, text or page title that were stored.
Scripts and configuration are propagated to different computers using Git via a personal Gitea instance and vcsh/myrepo tools.
Not really. Long-form things that I want to read later, I put into Pocket [0]. Everything else I don't bookmark, I can just search for it later, or it's lost forever like tears in rain.
I built my own service and client as a hobby project to scrape and categorise articles for myself. I realised throughout the project though what a monumental pain scraping article content reliably is and have have spent less time on it lately, I still dogfood it however.
You can save links in the web app, via email, browser extension or pull automatically by integrating with a Medium blog or any RSS feed. Link collections get automatically turned into an email newsletter that can be personal, public or shared within a group.
We recently launched 'Picked for you' which suggests links you might find interesting. Also, we launch new features daily so feedback and requests are more than welcome!
The Alfred workflow makes capture and tagging pretty seemless.
Edit: just read the op properly, not after pinboard. I also would like a visual manager mainly for design refs. I don’t personally use it but Are.na is a cool platform for visual bookmarking of media alongside articles / sites.
I have written a service where I store link, timestamp and addendum where I can explain why I went to that link, and a counter for "if this link already exists in the database" then increment. I save links when I have too many tabs open so as to close them. This gives me a way to query on those links at a later point of time in mysql command line. It has worked well so far.
A couple of people have mentioned Evernote, which is a great product, and a great service, with a great plugin for Chromium-based browsers (even Edgium!). However, I love me some markdown, so I've started using Boostnote for my actual "thought" notes. I like it quite a bit, and I've found a few tools that help me "clip" articles from the web and turn them into markdown. So far, I prefer `clean-mark` (https://github.com/croqaz/clean-mark#readme), because it has more consistent conversion and better meta-data generation, but I've also used `hget` (https://github.com/bevacqua/hget), which, in my experience, is not as robust but sometimes catches things that `clean-mark` misses.
Let's just say I'm seriously dissatisfied by how I deal with them. After trying multiple variants I (once again) ended up basically just using Pocket (which I hate and promise myself to replace by Wallabag or such, but... yeah) for everything I'm going to read (removing permanently after it's done), and just bookmarking everything I want to keep between history cleanups in my address bar suggestions without sorting much.
I would like to use bookmarks more, assigning tags, using them in my planning, but Firefox bookmarks just suck and I end up spending too much time without any benefits. There were some external tools for that, but they aren't better than a plain text file because of poor browser integration.
Anything that can be annotated and put into context ends up somewhere in my *.md notes.
I hope this doesn't sound overly self-promotional, but I have a newsletter with around 1500 subscribers (and a few Patreon supporters) which I started a few years back to address a very personal issue of bookmarking/curation: I realised I was storing tons of links which I never visited and forgot about, but that certain links sort of stuck with me and I kept talking about them, most of which related to data.
So I started sending them out as a curated list of a handful of links per week, alongside a short explanation of why it mattered or a quote.
8 years later, I still sort of remember if I sent something, and I no longer use bookmarks.
apologies for site not being done yet and article not fully fleshed out. but the topic has come back many times now that i think i needed to share what i thought about it.
to answer your question: we need to share and distribute our bookmarks in GitHub. it works for developers because our search queries are keyword-based.
pocket, pinboard.in, etc. don't work because they feel like "todo" lists.
I use HN submissions as a better bookmark/log of what I've been reading that day (or previous day). Usually post things I might want to revisit later.
I use browser bookmark folders only for links for private stuff e.g. fun+memes, cooking, curated lists of <FOO>, ... Also I maintain markdown files for documenting changes/tweaks and modifications made to various machines on my network.
It is intended to replace the “new tab” page of the browser and works really well!
When I first “launched” it had a minimal set of features and scratched my own itch. I wanted to charge $12/year. However, I discovered that nobody even wanted to pay $1 per month!
I’ve since created designs for a new UX, which is what you see in the screenshots, and put up a landing page. All out of the startup playbook. However, nobody has expressed an interest in the product, so I haven’t opened up access for others.
I blog. I can't seem to get significant traction in terms of an audience, but I very often can find some link I want for some reason because I blogged about it. In cases where searching the internet or searching HN is failing me, it's often been a huge relief to turn to my own writing and unearth it in short order.
It's a primary reason I continue to run my big fat mouth on blogs the rest of the world largely seems to have a yawning disinterest in most of the time.
I do curate my bookmarks, in the sense that I'm renaming them (the website title is sometimes way too long and bloated, and I put tags on them to find them where I believe they might be relevant (ex: note-taking, selfhosted, e2ee)
If the website contains an important bit of info, I try to send the page to the Wayback Machine so that it's not lost to time when the website eventually dies, and I also copy that info in my notes (currently using Joplin)
What I’ve found is that it really doesn’t matter which tool you use. There’s many options that will allow you to manage your bookmarks, consider a simple txt file.
The challenging bit is providing a sound enough structure so every item finds a place almost organically. If it’s a simple list then it can take a while to find a specific entry and if it’s too categorised you’ll run into a similar problem. I like to use trees with symlinks.
I'm really careful with bookmarking now. I found a txt file backup of my del.icio.us account from 2004, I couldn't find a single link which still worked. There was some gold which I would have loved to be able to look at again.
If something is interesting enough to keep I clip in into Evernote and tag it, then I can be confident it will be there when I go looking for it at some undefined point in the future.
Yes, of course I do. I collect bookmarks in many places. Easy to access every day use bookmarks are in my Firefox bookmarks. Others are in my chrome browser which I use for different browsing and the one I want to keep forever I use https://yabs.io which allows me to search and tag bookmarks.
I used to do this for a few years: https://f5n.org/stack - but my interests have shifted a little and not too many people gave any feedback or indication that they read it.
These days I just save them to Pinboard, like I used to, and sometimes go back and find something I vaguely remember...
I use a self-hosted wallabag for long-form articles that I want to get back to or are part of a particular string of research and it ensures I can always read it even if it disappears.
Bookmark wise I mostly just use the standard bookmark manager. I have considered a self-hosted syncing mechanism but I only have a few hundred bookmarks, it is stuff I use.
I put every blog, site I want to follow on my freshrss instance(rss.ishanjain.me).
I use Reeder(on iOS) and FeedMe(on Android) to read those posts. When I like something, I add it to Pocket where it gets archived and stuff and I can revisit it later at any point.
I use Browser bookmarks to store interesting websites, Like tools, Stuff other than articles basically.
We built a tool for this as we had the same problem. A place you can collect, tag and share you links in a visual way. Right now it's only for articles, videos (YouTube) and films, but will expand into other domains. Have fun, let us know if I can help you navigate it. dean@discovereel.com www.discovereel.com
I started to use https://getmemex.com/ a few months ago. It's working nice for now (locally stored, screenshots, text search, include results in duckduckgo, ...)
Worth a try I would say, but I won't make a definitive recommendation for now.
Currently using Pocket to quickly bookmark important articles. Their simple tagging system allows us to organize links by topic. We have a "read later" tag for quick bookmarks and after reviewing content we will add a new topical tag. Sharing with peers is easy as well.
https://pinboard.in does exactly what you are asking for, additionally it keeps an archive of the pages (Similar to archive.org) and you can see public bookmarks of your "friends" on the service.
I use RSS to quell the internet firehose, specifically Feedbin (also great for Twitter lists).
I do a fair bit of reading in Feedbin. Anything I read and want to save, I tag and send to Pinboard. If I want to read it later, I mark it as such and also send it to Pinboard.
I use Inoreader's "Save to Inoreader" bookmarklet which lets you tag and save the link which can be accessed across devices. Their search works well enough on the saved links and they have a very generous free tier.
You could just bookmark everything, and use tags for specific categories.
Maybe add a 'ril' tag for every thing you want to read later, no tags for generic queue of things to look at, 'archive' for things you want to keep, etc.
I use checkvist.com to curate. I have both links and saved information, organized to multiple levels. I back up monthly. I have over 10,000 lines in my straight text backup. I share this list with friends and acquaintances.
I've used pocket for many years to collect articles and web pages that I want to read later, and kept using it almost by default when Mozilla took it over and integrated it into Firefox.
I use Joplin, and make my own awesome lists. Sometimes I'll just use their webclipper app to clip parts of pages for my own 'doc' system for things I want to remember.
The main reason behind my building Histre is the idea that we throw away a lot of the signal we generate while doing things online and this can be put to good use for ourselves.
Bookmark management is just a special case of knowledge management. What you really need is a knowledge management tool that is easy to use. You'll get a ton of other benefits too.
As it is right now, Histre aids the casual online research we all do (ie the explore -> filter -> decide loop). For example, it removes friction in taking notes on links you're looking at, with free-form tags that you don't have to create first and other such niceties that add up. And it easy to group notes into notebooks and share. In short, when you have to look at a bunch of links for something (decide on AirBnB, people to hire, material for your next blog post, etc) Histre makes your life easier. But this is just the starting point for what Histre intends to do.
IMHO the biggest problem with apps like Evernote, Notion etc is that it becomes digital hoarding, and not a knowledge base. And the knowledge base focused apps out there involve a lot of manual upkeep, which almost never happens, especially at work. Things start out okay and quickly fall into disrepair. I'm differentiating from the other note taking apps by automatically putting together a knowledge base (grouped by topic etc).
One idea I'm excited about and I'm working on right now is: Histre automatically fetches updates from the websites you visit, ranks the websites with things like lack of ads / referral links and ranks the new posts with your 'revealed preferences' of what you tend to actually read from the list of updates previously shown etc. Personally I think this is will be what replaces social news sites, which are too sensitive to people who bother to go upvote on /new and has to cater to the lowest common denominator.
Automatic Upkeep (WIP): Histre detects links/notes related to your existing notebooks and offers to update those notebooks with the new links and notes. This is similar to how Google Photos suggests new photos for your existing albums. This solves the upkeep problem. Currently people create knowledge bases with good intentions and it becomes stale and useless quite fast.
Exports: I'm working on org-mode exports, as I'm an Emacs user myself. Other export formats coming soon.
Integrations: There is Hacker News (import and optionally share your upvotes) and Telegram (take notes using the Telegram app) integrations now. I'm working on Twitter next (lists of Twitter profiles -> follow all / block all).
Commented this elsewhere but this is cool so I'll help exposure; "This is cool, I'd dig a $2-5/mo unlimited account for 1 person/team with the same unlimited settings." Personally I just want something like this for myself, I was really surprised I couldn't share a Pocket list/tag publicly when I tried. IE, I have tags on "how to contribute to X project" which has a ton of bookmarks I'd like to allow public contributors/confluence to read without an account.
1. Every article I will read, I'll drop straight into Pocket, to highlight/read. I will never rea the article outright - this is important, because some clickbait can find me in a "vulnerable" state. After this time has passed, I'd delete it out of my Pocket without reading if I determine it's not worth it. I delete about 40% of the stuff I intended to read. In pocket, I would highlight heavily (I have pocket premium)
2. All my highlights, notes from the books I read, my Tweets, articles that I write and newsletters I send end up in my Evernote. This gives me the single source of truth for the knowledge I am gathering. I can instantly search for a figure from an article I've read 3 years ago. I organize it by topic and treat it as a "Commonplace Book".
As such, this is not for organizing "Links", because I dont care about links. I care about quickly finding the piece of information I've seen on the web, and -of course - proper attribution.