I’m working on an open-source social bookmarking site in Elixir that is API compatible with delicious/pinboard. Its named linkhut and it’s currently able to import your bookmarks from pinboard and browser exports.
Two things that I think are very important in a bookmarking app (and why I‘ve been working on making one of my own on and off for a while): open source and offering an API for other tools to build upon. The current API on linkhut aims to be bug for bug compatible with pinboard while being more OAuth-y.
I’m still working on a snapshotting feature similar to pinboard’s, once that’s ready I think I’ll call 1.0 complete. Perhaps I’ll do a Show HN then.
Interesting! Got an account, but I wonder how you/we are going to pay for it. Even with just a few million bookmarks you're going to need a non-trivial DB setup, and (according to some napkin maths on my Pinboard account) ~2.5 MB of storage per bookmark for the snapshotting feature.
It's weird that this is at least the third iteration of "bookmarks on the web" that I've used. My trajectory has basically been del.icio.us > delicious > pinboard > linkhut (still kicking the tires on the latter though, staying on Pinboard for now but hope it doesn't stop working).
Tried it as an experiment (I'm planning on staying with Pinboard, at least as long as it keeps working), but it keeps failing, at least when trying to import as HTML.
Ok, I'll amend my statement to read: "if viewing a list of basic links that match a few basic filters requires client-side JS you're doing it wrong" and I stand by it.
If you'd like to employ JS to progressively-enhance your user experience, great. If you are architecturally requiring JS for basic functionality? stop.
If you were to have a static site (note: that was the original question), and want to perform a dynamic tasks (like filtering) - how else would you do it besides using JS?
I've been a pinboard user for ~6 years and I won't be renewing my subscription.
After a few years of using it I signed up for an archival account (which costs extra $39 per year). My credit card was charged, but I wasn't able to archive any page; the option to do so was never there. I emailed support (I guess it goes directly to Maciej) and NEVER got an answer, despite being a paid customer and following up several times. I ended up having to do a charge-back on my credit card. Very unprofessional.
On the flip side, that made me look into self hosting and now I happily run a linkding [1] instance on my NAS. I never really cared about the social aspect of it.
In the old, good days of del.icio.us social aspect was really robust. People were putting lots of stuff online. It was easy for me to find an academic or author I liked and dig through her bookmarks to my delight and betterment. It was still just the old good internet with real people online, sharing really interesting stuff. I would also post lots of stuff and we used sharing features with friends a lot (you could actually add #for:username).
I would be happy to have access to some delicious dumps, I know people made them before delicious went down.
People were starting to crawl it. Surely it was huge, they were probably crawling bits they were interested in. I am absolutely positive there were people trying to crawl a lot. It was a panic of kind, like with geocities. Still, would love to see what was preserved.
I doubt they got very far. Most of the pages cut off very early since humans never went more than a page or two in.
Maybe the new version wasn’t as good but when I left it wasn’t easy. Any at scale scraping was easy to detect since the site was so resource impoverished that it would have a performance impact.
The archiving happens automatically; you shouldn't need to do anything except request an archive backup[1]. Unfortunately those requests are not honoured, and haven't been for a long time (my last successful backup was in July 2020).
Yup. I have an archival account and none of my archived articles work. Some of them show an error, I hit the button to recrawl, but that just results in the error reappearing after a while. Others don’t show an error, but when I want to view the archived version, it says it didn’t find it. So I have a bunch of links in there, which already link-rotted, and weren’t successfully archived by Pinboard.
I’m considering moving to just using the SavePage (or whatever it was called) browser extension with a text file for the metadata (url, tags, titles, notes) on a NAS.
> I’m considering moving to just using the SavePage (or whatever it was called) browser extension with a text file for the metadata (url, tags, titles, notes) on a NAS.
Try Zotero. (I tend not to use its notes and tags features, however—even though it has them. I just defer to Hypothesis for that.)
Ah my memory is blurry in the specifics. I think the issue was then that the archived link never showed up. Definitely it didn't work. I checked an rechecked the FAQ to make sure that in fact it wasn't working for me, and it wasn't that I didn't know how to use it.
--
Actually... I just checked my emails. My account never upgraded after payment (it was still standard) and the check-mark didn't appear next to my links. And of course there was no way for me to see the archived page or request an archive backup.
Was a Pinboard user since year 1 was my favorite online service for years,but the site went to trash after the creator became obsessed with politics in 2016 and moved his focus to tweeting and poli-larping rather than his customers.
This is true, even though I personally think his political thinking and rhetoric is brilliant and valuable. It doesn’t excuse him of taking care of the people who take care of him. Spend a few hours on your business, Maciej!
As a power user (21k bookmarks, 16k tags) I'm sorry to see Pinboard dying. I don't want any more features, but the killer feature introduced a few years, ago, archiving, is long dead:
> You asked to download your archive on 2022-10-04 17:36:42.
> You'll get email once the download link is ready.
Yeah, nah. I've requested this many times since it stopped working, and that was probably more than a year ago now.
Fortunately XML/HTML/JSON backups still work, so moving to another service should be easy.
Follow-up: Weirdly, it seems the backend for the archive feature is working fine - it's reporting what looks like a correct number of bookmarks, and says the backup is currently 51 GB, up from 30 GB when I last downloaded in July 2020. I guess there's an automation step missing (including throttling!), and then it could just run itself.
I've seen people (on here?) suggest that Maciej is on a long period of leave from Pinboard, and will definitely return to breathe fresh life into the product. But I don't think I've seen an official statement from him to that effect, not on the blog anyway https://blog.pinboard.in/
I'm a paying Pinboard customer (switched from a grandfathered-in one-time payment to an annual archival subscription). I've loved Pinboard for a long time, and really want to continue to love it.
It's felt to me for a long time that Maciej is burnt out and Pinboard is on a slow death spiral. It's one thing to leave good-enough alone and not develop new features, but it's another to not fix the things that are broken and let the product continue to gather rust - especially when you've moved your customers from one-time payments to subscriptions.
It's really too bad - it seemed like there could have been an enormous amount of inertia after the rather poetic del.icio.us acquisition, and the release of the archival feature was promising.
I'm still hopeful though, and that's why so far I continue to pay my $40/year.
I'm a Pinboard user and am happy with the bookmarklet which in my experience works flawlessly.
But... I (almost) never lookup any of my bookmarks. Googling is faster -- and safer, since it's possible I'm wrong and didn't actually bookmark what I'm looking for.
Bookmarking satisfies the mind; it's a bit like hoarding. Maybe the best bookmarking service would be a mock interface that doesn't actually do anything.
Ever since I found the 'highlight or hide search engine results' extension I have pretty much never browsed raindrop directly for technical stuff. I do still keep it around for the webpage highlights and tagging my twitter likes, though.
I agree as I've been in the same position but if you don't tag these properly then they are almost useless. Heavy bookmarker nowadays, 10s of thousands and I search it before I google because HIGH chances are it's something I've encountered before, even just a small tag like {math} helps greatly, even better, {math}, {vectors}, {gamedev}. I even dump research sessions there, eg. if I spend time on a gnarly bug or im grokking a new topic like ml and at the end I have 30 tabs open I save these as a group and label it as such "android opengl mapbox rendering issue", "math for ml" etc. One of the most annoying things is finding the perfect link and you know you did but can't find it again when you need to. Also god forbid you end up on a stackexchange site or forum and find yourself as the one answering (has caught me many times) [1].
This is happening to me right now! There's a bug in Reaper (a DAW) about handling gamepads. I found out that the bug was submitted first in... 2009, got no response from the devs and was not fixed. It does feel lonely sometimes.
Pinboard's performance was ultimately what led me to abandon it in the end.
I really like Firefox's built-in bookmark manager. It supports both hierarchy and tags and, if you export your Pinboard bookmarks to html, the file will include the tags which Firefox will preserve just fine when it imports everything.
I don't use Firefox as my day to day browser but all of my bookmarks live in there and I can search them from there much faster than any website. Those bookmarks live in a sqlite file which is easy to browse (good luck reverse engineering its schema though) so if anyone fancied knocking up a command line tool to search it I'm sure it wouldn't be difficult.
I realise this is a comparison article with a funny intro but I honestly don't understand that when you have a _list of links_ that you need to share with students you chose a third party website and then act shocked pikachu when that third party website has a hick-up.
There are hundreds of fool-save options that make it so you don't have to depend on a third party, here are some: You could email the links to students. List them on the personal website that this is written on. Have a reader that lists them. Co-host them with the rest of the assignment.
Man, I hope Pinboard stays up awhile longer. Genuinely one of my favorite simple, reliable internet tools. I don't need new features, I just need it to continue to exist.
Definitely, pinboard still does exactly what it's supposed to - store my bookmarks. I've got 17780 bookmarks[1] at the moment. I didn't try the auto archiving, but I happily switched to the new basic level payment model when he asked. It's been a stable part of my life for more than a decade.
It feels weird for you to address one positive comment here but not any of the folks saying they’re running into persistent issues with both their accounts and with attempting to get a response from support.
I’m really surprised the author hates on Raindrop so much because my experience has been nothing but good.
If anyone is building one of these bookmark apps, one of the killer features for me is the ability to highlight and annotate a copy (in case it disappears) of the page, so I know why I even cared about it in the first place.
> Raindrop displays by very-poorly-implemented “relevance” by default at all times (that is, not just in response to a search, where this might make sense if the implementation weren’t so awful), and this default cannot be changed, so I have to click twice every time I load a Raindrop page to switch to reverse-chronological.
When I open Raindrop it always remembers last sort (chronological), so I never have to click sorting button ever again.
Was surprised here too. You can display bookmarks in 4 ways (list, cards, moodboard, headlines) and choose up to 6 data fields per bookmark (cover photo, title, description, highlights, tags, info) and sort in 7 ways (date asc desc, name asc desc, url asc desc and manual).
That’s 168 ways to display bookmarks, and every collection can have a different sort.
The site and apps sync and remember these display options flawlessly.
I paid for Pinboard for life very early on, and remember getting an email a while back encouraging everyone to pay for a subcription instead so that development could continue.
I wonder if that didn't work out somehow, and revenue wasn't enough to encourage development/maintenance of the service. I've not logged in for years though, so the site's not been on my radar until today.
Raindrop is excellent value for the price, highly recommend.
Author makes some good points, but I still argue Raindrop is a superior product depending on your perspective as a bookmark tool user. Pinboard feels dead/abandonware unfortunately.
I paid for Raindrop for years, really loved it. But it got extremely buggy (forced me to relogin constantly). I'd open Raindrop, it would completely reset my auth info, and then close it because it's not worth the hassle of logging in yet again just to save a bookmark. And when they clear out login, they clear out everything (including the email). Their iOS share panel was an absolute disaster.
Their UX was great, but it's a nice case study that if you don't focus on the seemingly little but important things (like making Auth seamless and bulletproof), everything else is meaningless.
Dropbox is the same way. Opening a document never chooses the right account (corp vs personal), constantly clears SSO session, and I usually give up 95% of the time instead of using the app.
Hmm. I'm currently using Raindrop and haven't ever been logged out. It's been pretty solid for the past year or so I've used it on (Android app and Firefox extension on MacOS).
Pinboard has been rock solid for me? This is one of those HN topics like the phone numbers in Signal or the error handling in Go (two other perennially doomed projects) that just won't ever die. It's too bad that the whole thread here is a referendum on the health of Pinboard's development practices, rather than a discussion of the article, which is pretty solid and interesting.
Pinboard has extremely long page load times for tags that are above a certain volume, especially when it hasn't cached the results on the server: example shown takes 30 seconds to view by a tag with 3000 bookmarks: https://i.imgur.com/U8OeN86.png
This is why I stopped using Pinboard, otherwise I'd probably still be a paying user.
I do. I haven't requested a backup but I did check some pages and most were archiving as expected.
That said, I sent in a question / request months ago and never received a reply. I imagine they're busy but it was a little disconcerting being a long-term user and pre-paid subscriber for the next 5? years to not even get a 'thanks for the message' reply.
I was asked a question. I answered it. I don't know what you think is going on here, but I think it's probably simpler than you think it is.
This is my thing about Pinboard drama (and Go drama, and Signal drama) on HN. It's made to seem like a gripping psychodrama. But there's less going on than you think. I'm telling you: I use Pinboard every day, it's been rock solid for me, I don't use archiving all that often, but I clicked around and all my archivey things seem to be working. That's all I've got for you.
I meant your original post. You went out of your way to either gaslight or mock others who were sharing their troubles. Why?
Edit: I see you've edited your comment to answer my question. Well, all I can say is that's an interesting choice from someone whose bill rate is so high.
This is a deeply weird thread, but, I mean, I keep saying these threads are weird, so what do I expect? I'll keep going just to illustrate my point.
I don't doubt anyone who says they're having problems with Pinboard. Stop paying for it. Get your money back. It's a product, not a religion. Read the thread, though, and notice how many people are confused about that. You just asked me, a Pinboard customer, to go fix Pinboard for people having problems with it. Uh, no? I'm a user, not a parishioner.
I apologize if my replies seem weird. I'm genuinely confused because I've seen you be such a thoughtful and empathetic internet person.
I'm not asking you to fix it for them (That'd be crazy!) But if you're going to engage, perhaps you could engage beyond "wfm mic drop"? I'm not saying you need to get involved at all. But you chose to!
As one of many people on HN with the day-to-day responsibility of monitoring and troubleshooting a complicated Internet thingy that springs leaks and throws rods at inconvenient times, I am here to inform you that "works for me" is often a pretty valuable piece of information.
Now that our good intentions are worked out, it's probably silly for us to keep going with this thread. Somebody on another thread just said "Boiler Room is an incredible movie" and I need all my energy for that thread now.
I built a little hybrid bookmarking/notetaking app - https://www.showboard.ca/
It's not built for bulk usage like you're describing, but makes it easy enough to scrape, modify, share and search.
Let me know if this is of interest (it has a python backend ;)
Is there an alternative that focuses on local storage, which easily syncs between my devices via Dropbox etc.? Zotero comes close, but it has a weird full-text search, and the UI is not as nice as Raindrop's.
The other issue is to retroactively get local snapshots of these pages when the bookmark collection already consists of several thousand items.
Each bookmark is saved as a markdown file so syncing is straightforward. I mainly use it with the command line interface but you can also use a UI running in the browser.
ArchiveBox might be a better fit for your use case. Maybe a bit of overkill, but better to have more and not need it versus not having everything you need, right?
Bort does just that, uses browser's IndexedDB and syncs via Dropbox. But note though, that it's not similar to Pinboard etc. in the sense that it's not designed for a huge archive of rarely accessed bookmarks. Works well for something smaller, like what people commonly have as browser bookmarks.
By syncing between devices do you mean that the centralized server serves as the single source of truth, which is accessible to all your devices? Or is there storage locally on each of your devices too that is synced with this centralized server as well?
It’s way too complicated for me with a too steep learning curve, but as I understand it it’s the gold standard of “information management” and its active users adore it.
IndexedDB is probably what's meant, Local Storage (if taken literally) is a simpler and perhaps too limited a solution. There are some options that make IndexedDB easier to use, I've been using PouchDB, which works well. If syncing to Dropbox, you'd then dump the database into a JSON string/file and use the Dropbox SDKs to sync that. If syncing to a database, you can do something more efficient.
Raindrop released a major update with highlighting features two months after this blog post was written, so the author's concern about the lack of recent updates is likely due to that major feature being under development at the time.
I'm a heavy Raindrop user and have been extremely happy with the service. I don't really use the social features, though.
When Pinboard's price suddenly doubled I suddenly stopped paying. Wasn't a heavy user so the price hike wasn't relevant to me. I wish consumption-based pricing was more of a thing.
Really interesting to see these sorts of comparisons both as someone who was previously a heavy user of Pinboard and really admired what Maciej was able to achieve with solid execution on a simple idea.
One thing that I found really lacking in Raindrop and other alternatives was the support for treating online commentary as a first class citizen, which ultimately led me to develop my own little service which has served me very well since 2020.[1]
I have made it available to others to use over a 2-year-long free beta test which will end this year, but ultimately I'm starting to think that this kind of service is just so personal (or maybe, that my own needs are so individual?) that it's hard for users to settle on a service in this area that leaves them feeling completely satisfied.
>Notado is a content-first approach to online bookmarking, where the paragraphs and sentences that make you want to bookmark something are treated as first-class citizens rather than pieces of additional metadata. With Notado, your bookmarks focus around selections of text, which are fully searchable and can be organised however makes sense to you with a powerful automatic tagging system.
With first-class paragraphs, how do you position yourself in relation to Obsidian and Roam?
Looking at that example, is it possible to add comments to the quotes? I am looking for a timeless service where I can share a bookmark with others and have them add their comments, preferably openly with ActivityPub. Your Kulli service seems to be a complement for that:
It would be nice to discover people with the same interests and their links about the same topic.
Do you have plans to extend kulli.sh to include Twitter and other social networks? The difficulty of ranking a huge amount of comments seems to be accessible with the history of good comments from notado.app.
Thanks for leaving such a thoughtful comment! I'll do my best to reply to specific parts:
> With first-class paragraphs, how do you position yourself in relation to Obsidian and Roam?
I don't use or see the utility of tools like this personally; they are in my mind a great example of Žižek's take on Lacanian "surplus enjoyment".
I imagine some people will nevertheless want to take things they save to Notado and export them to Obsidian, Roam, Notion etc., and I have built out incremental export APIs for people who have this particular use case.
> Looking at that example, is it possible to add comments to the quotes?
No, this is an anti-feature in my opinion for the same reason I mentioned above. People who want this would be better served using Notado as a unified content ingestion engine (comments, article highlights, kindle highlights etc.) and redirecting to the tool of their choice; anything from org-mode to Roam.
Commenting on content is such a difficult thing to get right and so many people have their own deeply ingrained ways and expectations around this that I believe it's best to delegate to the end user.
> I am looking for a timeless service where I can share a bookmark with others and have them add their comments, preferably openly with ActivityPub
I generally like having these conversations in real-time as much as possible, but I think there is a great idea somewhere in here. Let me chew on this for a bit.
> It would be nice to discover people with the same interests and their links about the same topic.
This is what Notado Feeds are intended for; I would love for it to catch on by people including links to their Feeds home in their bios on Mastodon, Twitter, Github etc. (this is what I do!)
I'm also looking at creating a hub to improve the discoverability of user-curated topic feeds, searching by tags, showing the most popular feeds by number of RSS hits etc.
> Do you have plans to extend kulli.sh to include Twitter and other social networks
Twitter was never really on the cards because it's such a developer-hostile platform to work and integrate with, and given its ongoing implosion this is even less likely to happen. I'm interested in somehow trying to include Mastodon (or specific Mastodon instances maybe?) in the future, but that's about all that's on my radar for the immediate future.
> The difficulty of ranking a huge amount of comments seems to be accessible with the history of good comments from notado.app.
This is a really cool idea; I would love to be able to even highlight or filter comments on a link from users whose comments I have previously saved on my Notado account on Kullish!
I feel you may be a bit more optimistic than I am about how many people will pay to use Notado once the free beta ends. ;)
It's not something I've thought extensively about. I imagine the worst case scenario (outside of having to comply with legal orders) would be having to disable public feeds for somebody, but I'm not too comfortable getting into specifics based on hypotheticals at this point. Hopefully I'll never have to cross that bridge, but if I do, I'll cross it when I come to it.
I always found bookmark apps to be missing the features I needed to I ended building my own open-source bookmarking tool back in 2015 and I still use it to this day. I'll share it here in case it's helpful to any of you.
An alternative, one that I began using a year ago after losing confidence in Pinboard is the self-hosted Espial - https://github.com/jonschoning/espial.
The visual presentation is a near complete clone of Pinboard. It also provides a route for Pinboard import.
Weird, I totally had the opposite response on everyone one of these dimensions. Actually what became clear is that the author has a specific use case (sharing with students) that Raindrop doesn't fit well for (individual bookmarking), which is probably not the use case that 90% of the users are looking for.
Even though the main site isn’t really actively being developed anymore, the Pins for Pinboard iOS is a third-party app that is continuously getting updates and is very polished. If it weren’t for this very functional and pretty app I probably would be more annoyed.
I switched from Pinboard to Raindrop to AnyBox.
Pinboard hasn’t received much love for years and when I received an answer to my support request a year after sending the email, it was clear that I have to look somewhere else. Raindrop felt like an okay replacement, but never felt solid to me and introduced a bunch of “undefined” categories. Very inconvenient to deal with.
I am now using AnyBox [1], a macOS and iOS app that syncs using iCloud. The developer is super responsive and I could not be happier!
Another good alternative is https://www.are.na/ it is also tiny team but they have been around for a while and they pretty openly communicate their state/financials.
This one is amazing, never heard of it and thanks for sharing! Appreciate that you added lifetime subscription as well, but for now subscribed monthly :)
Great comparison. I am working on Savory which tries to keep it very simple and might be interesting to folks looking to jump off Pinboard: https://getsavory.co/. There is a chrome extension to save links and tags are a first-class citizen. There is no support for archiving or sharing yet, but it is something I want to add soon.
I used hypothes.is for highlighting, but now that tons of sites (like this one) block external javascript I needed something with a good safari plugin to get around that without too many hacks. Raindrop fit the bill. I’m a very happy customer, would happily pay more.
Can anyone tell me the purpose of bookmarks in 2022? I have only one hypothesis that it is for users who hate keyboard. Any frequently visited site pops up after Alt-D and first 2-3 letters of it's title or URL. Any infrequently visited site is easier to find with search engine after your bookmarks number is above 100.
A problem with amateur user generated content (which is what I mainly like) is that the uploaders are really shit at describing their post. Sure, describing a picture of a shady copse of trees as 'my favorite place to relax' will look good in an algo feed, but how am I supposed to find that later? Okay I will 'save' the post on reddit but it is saved exactly as it is to a chronological list. Same goes with your twitter likes or whatever.
Bookmarking services let you write your own title/blurb/tag so now it is /actually/ findable.
Ah, I see. This particular case I am handling by filing a geoposition in one of my Google Maps lists with a comment or maybe a link to the source. This way I also have a chance to visit it while nearby. This adds a small dependency on Google, but so does any bookmarking service, unless it is self-hosted.
I’ve been messing around with a little cloud service to do something like pinboard for personal use. I never thought about turning it into a commercial service because I figured there were already lots of great solutions out there. No I might reconsider it if there’s interest.
The author used a SASS bookmarking service when any number of general purpose options would have sufficed. Google doc, HTML file on a web-server, text file on a share somewhere...
The only time I bookmark something is when I find something interesting that I wasn't looking for OR some obscure page that's hard to get to in a SASS product.
It's too much work curating bookmarks, and for me the utility marginal at best. Often times when I go to look up a saved bookmark, the page has moved so I have to find it again anyway.
The flagship instance is: https://ln.ht
The source code is hosted here: https://sr.ht/~mlb/linkhut/
The documentation: https://docs.linkhut.org/introduction.html
Two things that I think are very important in a bookmarking app (and why I‘ve been working on making one of my own on and off for a while): open source and offering an API for other tools to build upon. The current API on linkhut aims to be bug for bug compatible with pinboard while being more OAuth-y.
I’m still working on a snapshotting feature similar to pinboard’s, once that’s ready I think I’ll call 1.0 complete. Perhaps I’ll do a Show HN then.