Ask HN: Why did delicious.com fail?

In general, despite great promise and utility (IMHO), curated bookmarking of this sort never really took off. It's always been around, and still is. Why? some guesses:

Not that many people are data-oriented enough to want to use such a service. Same reason RSS is relatively unpopular.

Delicious and related services never generated the sort of social proof / payoff that drove other services. E.g. desire for friends / followers / karma were essential to FB / twitter / HN (since we are here!). Delicious was high on the personal utility, low on the social reward.

I'm not sure this is that related to Delicious' functionality, but might just be a cultural / community function, that never really developed in this case. It's not tech that makes HN awesome, it's people. But given that Digg / Slashdot faded, this may not be as popular a category. Reddit is huge now... but it's more of a forum vs. link sharing.

And these services are so easy to create, who would pay for them? In the absences of hype / popularity, not much of a business there. I switched to raindrop, which is great. But others were close.

It's very sad to me. I think that the ability to socially curate the web, and to vote up/down on such resources, is actually really valuable to the long-term health of the internet.

Maybe we're just not ready yet...

* major redesigns (the current Hulu redesign is about as bad)

* competitors, there were a lot of good bookmarking services. I think I went with Evernote

The wikipedia page has a lot of what went wrong https://en.wikipedia.org/wiki/Delicious_(website)

Now it is finally owned by Pinboard. But most of the users who abandoned del.icio.us had already switched to Pinboard. Now all is well.

Again if I remember correctly there was a decent API for del.ico.us but this was not a priority for delicious.com.

The site was premised on the idea of "folksonomy" ie that structure would emerge from the accumulated links. This hypothesis didn't bear out.

IMHO that's also the reason why Facebook was winning above all the other Social Networks, it was "real" and not anonymous (or pseudonymous). Virtually all popular social but (kind of) anonymous services of the past didn't survive or at least not make it to today's mainstream: forums, MySpace, ICQ... Reddit is kind of popular but almost exclusively used by techies or people considering themselves nerdy.

I wonder if one could make a modern Delicious, with mandatory Facebook connect or something. Not sure though if that would still work out of privacy concerns. Actually one might consider Pinterest something quite close to it.

Now Chrome does this feature pretty well native I don't really need delicious anymore, i never used it much for its social aspect.

I lead the team on the re-write that brought Delicious into the single page app (SPA) world. Most people wondered why we built the app from scratch again. The python codebase that the Yahoo! team managed, I am not sure if they started their version of the app that way or if the inherited it, but we concluded that it would be quicker to re-write the app as opposed to dissecting the exiting application codebase. This is the primary reason why stacks were canned, regardless of what others might have said. Many times rewrites require reducing features in order to build a new foundation. This is sometimes a good thing, and sometimes not.

A rewrite also gave us the opportunity to build the app as an SPA. While we made an SPA, React, Redux, Webpack, Gulp, etc. did not exist, and we definitely ended up spending more time than expected creating inferior toolsets due to the lack of options at the time.

Another major challenge we faced was that the database was sharded in such a way that if your username was "zebra" the backend would hit MySQL server 1, then 2,3,4,5,6,7,8,9, and finally 10 before returning results. If your username was "apple" you'd get a response quite quickly. I am not sure if we ever completed the migration into new shards/replicas.

There were also some random old API endpoints where we had to double encode periods in the url or the API would throw an exception. Obviously there was no documentation on for our treasure hunt to find which API endpoints had this major issue.

We didn't get any notice at AVOS before the team was moved off of Delicious to work on other projects. Perhaps we took too much time building the rewrite, or perhaps we realized there was no business model worth the time.

AVOS would later make a unilateral decision to move all projects to Angular, which was not a framework I believed would benefit my career development. I left AVOS when StubmleUpon called me and said, "hey if you can make Delicious a SPA, do you think you can make our app an SPA?"

Boy, I sure know how to pick 'em...

P.S. I will say that the Delicious team at AVOS was a very talented team and I am thankful for each day I got to work with them.

[edit] for grammar [edit ii] Since it doesn't seem clear in the comments: I'd build Delicious differently today.

Here's my understanding of the history (which I hope people like joshu correct if I'm wrong):

The site was awesome and popular pre-Yahoo, but also pretty overwhelming to run. Yahoo brought it with resources, scale, and sweet, sweet promises that they understood Web 2.0. Instead, they quickly moved Joshua to a position where he couldn't really steer the product, and got sidetracked into redesigns and time sinks like integrating Yahoo ID as logins. The original, ambitious plan (integrate Delicious as a signal into Yahoo search, and make it unbeatable in real-time trending topics) was lost to management paralysis.

Yahoo disastrously announced it would 'sunset' the site before actually having a plan for it, so in winter 2010 a large number of people jumped ship. Sites like mine (Pinboard) benefitted a lot, Delicious was crippled.

The new owners at AVOS seem to have had a grandiose idea that they would repeat their YouTube-scale success, and rewrote the site accordingly, to serve billions. The handoff from Yahoo was one of data, not code (since things were written to run on Yahoo's platform) and required users to opt in. This hemmorhaged users further, while AVOS engineers wrote a bloated and ambitious Delicious clone.

But the biggest mistake was when AVOS turned off some features beloved by a core Delicious constituency, fanfic authors. In particular, they made it impossible to search on the "/" character in tags, which instantly rendered a lot of the elaborate fanfic tagging and classification scheme useless. In my mind, that's when Delicious hit the point of no return.

AVOS failed through some combination of founder grandiosity and short attention span. They managed to get the worst of all worlds—high spam, low revenue, and a $40K a month AWS bill. They sold to Science Inc., a bunch of LA tech bros who WROTE EMAIL IN ALL CAPS and owned a bizarre collection of other online apps.

When they tired of it, they sold it to Tony Aly, an SEO specialist who wanted to try his hand on a bigger project, but eventually found himself overwhelmed by the support and operational burden. At that point, he sold it to me, and I parked it on a couple of expensive servers and made it read-only.

To the question "what happened", I think the best answer is that Yahoo killed the site by neglect. AVOS took it past the point of no return by making it bloated, slow, and antagonizing the last remaining group of core users. None of its acquirers ever understood what it was for.

Except me. I understand you, Delicious! But I also compete with you. No more tears now, only dreams.

[would love to chat over coffee or email with others passionate about this space. @mceoin on twitter]

Why delicious died: - Yahoo acquisition. Death by a thousand cuts from that day forth. Some things here really pissed off core users include Yahoo sign in and shuttering some core features that power users loved (see exodus to Pinboard).

Landscape: - The attention economy has definitely shifted. Perhaps delicious could have evolved along with it (we'll never know), but now it is certainly the case that it is more difficult to gain mass adoption for products that will provide you with lists of information, since 'good enough' alternatives exist (notably, these are typically feeds, not lists) - the world has gone mobile.

Who lives, who dies, who tells their story? - Most died or were acquired (it turns out if you develop a team of search and indexing experts, you're pretty attractive to big companies). I counted 20+ here, about 50% acquired, the rest: ?? - Pinterest had the visual twist, and grew to 200m people - Pinboard generated revenue from Yr1, and has a small but profitable user base (important, because "not dying" is a core piece of the puzzle). - Kippt had a great product but team was ultimately were acquired by Coinbase, so it would be awesome to hear more about their experience. - Pocket lives on through Mozilla acquisition. In my experience they never cracked the social piece, although definitely provided the individual utility to be a daily-active-utility. Watch-this-space.

Why is this space difficult, generally? You would think that being able to explore lists of what people find interesting is appealing, and it certainly is (in theory). There's a hierarchy in information discovery for this space where: personal saves > friend saves > followee saves > random people saves. - social products require high clustering coefficients (very important). This seems to be particularly difficult to obtain in social bookmarking space. (I believe, but cannot confirm, that most products went too wide too early, and failed to nail this experience. Facebook's control of launch from college to college is a different but effective strategy that gave them saturation, so I believe a restricted entry via referral is pragmatic approach here.) - Low virality. This is a hard one to prove since non-success doesn't mean no potential for breakout success. - Low target market. Same point as above.

Monetization (three options I see): - advertising: requires mass scale, hence most attempts here die. - pay to play: Pinboard is doing this. (does this preclude wider adoption? or are other factors at play?) A balance of paying for privacy features might be an option here. - crowdfunding/non-profit: e.g. wikipedia model. Nobody has tried this yet. Would take a certain kind of team and community, but might position itself well for being a long term human-curated archive of the web.

I have some thoughts on how what this space lacks and what can be done differently that I'm happy to share privately. Suffice to say that we want this, and believe there's an opportunity to build a product that at least a few people love.

Who is taking a crack at at "social bookmarking" today? - Refind.com. Notably, they're more "show me what my friends are reading right now" and feel like a slimmer, more reading focused, twitter feed. Mobile app is best consumption vehicle. - Mix.com (from Garret Camp/Expa). Hasn't 'clicked' for me personally, but if might for you. - Us! We're not ready for a Show HN but if you're in SF and love this space, please do hit me up on twitter for a coffee: @mceoin.

You might also throw the fate of ~pg's Viaweb, once it became Yahoo Stores, into that same class of fumbled opportunities. And so you can read Graham on Yahoo & get a good idea of the issues:

http://www.paulgraham.com/yahoo.html

For Flickr: https://gizmodo.com/5910223/how-yahoo-killed-flickr-and-lost...

For Upcoming: https://medium.com/message/diary-of-a-corporate-sellout-5874...

What it failed at (under Yahoo) was navigating its product and audience around to the point where it could have become a Pinterest like Behemoth.

Facebook for all it's flaws and my general concerns with them have proven themselves the masters at incorporating and updating both acquisitions and their core product to attract a general audience that they can make money off of.

Yahoo management did a huge amount of work to kill momentum.

1) We were forced to migrate to PHP. We were not allowed to have ops support until this was complete.

2) We inherited the Yahoo Photos team that brought a backend that they had written for that platform. They estimated it would take six months to migrate delicious to that new platform. It took 2+ years.

3) We were denied having actual access to VESPA and had to reuse MyWeb's backend instead, which was designed for a different feature set.

4) Almost all the staff were placed on Yahoo Answers anyway, which ended up dying of its own accord anyway.