Good luck to him. He sounds like a good choice. So glad they didn't find some square-jawed empty suit. The fact that he heavily participates in online communities is a very good sign.
I do hope he has a delicate touch though. It's surprisingly easy to set off communities if you come in guns blazing. It would not be difficult to replicate the The Great Digg Rebelling of 2010.
Probably the best thing to do would be to win over the community by making very small and uncontroversial improvements. Nothing that is self-serving. UI improvements, better search, nicer notifications, etc.
Increasing revenue would be easy to do but very risky if it involves any new forms of advertising. I hope they can push that one off a bit or come up with something users don't find objectionable (which will be tough).
Reddit has grown so large with only very minor improvements over the years. It would be a shame to have someone "fix" what ain't broke.
I'd love to hear your ideas for improving search. IndexTank improved reddit's search greatly when they took it over, but IndexTank is shutting down in one month. At Searchify, we would love to power reddit's search, and are willing to spend a good amount of resources to make sure the results are good.
As a reddit user here is my experience. Reddit's search tool is terrible if you want to find a thread about a specific topic. It only searches for exact keyword matches and only in the title, not in the thread body. Add to this the fact that reddit users have a habit of link baiting their submission titles such that the titles rarely have any useful information and the search function is practically useless.
If I want to check reddit's opinion on something I have to resort to doing a google search like "site:reddit.com lopping tree branches".
You're correct - the current search only searches the titles. This is one of the most common complaints, and one that we would love to help reddit fix. One challenge is there are 5 to 10 times as many comments as there are threads. So a search index for that would be much larger, require more machines and RAM, and in the end it would be quite a bit more expensive for reddit to offer.
Oh, I understand that. The "search" feature of reddit is just so underwhelming that it doesn't even really feel like a feature. It feels like it's just a temporary kludge.
External sites are so much more useful it's crazy. Google lets me search comments and I can search in subreddits using "site:reddit.com/r/subname foo". Tineye + Karmadecay lets me even search for threads about an image or gif.
I haven't thought about it that much, but I've noticed very high latency using IndexTank recently. Probably a proxying problem or something because it always reports a fast search time even when it isn't. It takes 10 seconds but says 0.02 seconds. You can't search comments at all. No date range support. No stemming.
It'd be cool if they outsourced it to someone like you who would put more time into it.
Yishan knows more than most people about (the importance of) community on discussion sites, has the experience of being a major participant on LJ, Quora, and Facebook, and the engineering/engineering management experience of PayPal, Facebook, and some startup consulting.
I really doubt there's anyone more qualified to lead Reddit. I predict Reddit will add stronger social networking features to try to leverage the existing community in new ways; there has to be a reason news sites with great engagement like Reddit, Digg, etc. sell for less than 1% of a social network with the same stats.
He also built a pretty interesting invite-only physical tech community (like a less-crappy version of a coworking space), but I'm somewhat biased.
(what I really want is a hacker news social network, but there are reasons that is unlikely)
I wonder if this technically makes Yishan CEO of a YC funded company...
One of the smartest people I've encountered, even if it is only on Quora. Great sense of humor, doesn't pretend to have answers on things he doesn't know about. Probably a great guy to work with and for. Congrats, Yishan.
I have to say, the incongruity between your accomplishments and your website [1] is mindboggling. You have listed on your resume, as one of your accomplishments, that you built the tools necessary to manage 1B pageviews/month with a single sysadmin... and your home page is 32 lines of html. Not sure I've seen such a disconnect between skills and presentation since Knuth's homepage. [2]
I don't have a lot of time to update my homepage, so it's basic. It's way down on the bottom of my todo list to update it at some point but honestly those links will tell you far more about me than anything I could write about myself.
It's a sorta-kinda complement, in a roundabout way. Were you just some random with a homepage, I would dismiss you as not being worth my attention. Knowing, however, the amount of work you'd put into keeping reddit up under some pretty unfavorable circumstances, you've been filed under "dudes with crazy skills and shitty homepages".
It's a small file with some interesting names in it.
Edit: At the moment, it's pretty much just you and Knuth.
Those charts make it look like over 50% of CEOs have engineering degrees. They're missing liberal arts and science degrees. Still, cool to see how a degree in engineering hardly limits you to the trenches.
Engineers and scientists are particularly well-suited to hold office in a communist country. The very idea of managed economies was based on the idea of informed technocrats being able to make decisions to the benefit of all. In fact, in the case of the Soviet Union, it worked brilliantly... until it didn't.
As long as they don't become more like conventional CEO's and forget they were once writing code having lot of fun building applications and solving issues on day to day basis. Now that they are CEO's asking questions which makes you think..were they engineers before? I hope he doesn't turn to conventional.
I'm by no means a reddit fan, but Yishan comes across as an amazing guy (kind, bright, and insightful) on Quora, so I don't know if reddit could have chosen a better guy for the job.
It'll be interesting to see how visible he'll be inside and outside the site.
I hope he'll be able to do something about the declining quality of posts and comments, although I wonder if there really is a solution for such a thing.
"But as I continued the conversations, I came to understand that reddit wasn't looking for a conventional CEO candidate, because reddit is not a conventional company."
Kind of scary that despite years of being a redditor, he didn't know what kind of company it was?
Perhaps he could have written it a bit more neutrally:
"But as I continued the conversations, I came to appreciate that reddit was serious about tech and wasn't looking for a conventional CEO candidate because reddit is not a conventional company."
I do hope he has a delicate touch though. It's surprisingly easy to set off communities if you come in guns blazing. It would not be difficult to replicate the The Great Digg Rebelling of 2010.
Probably the best thing to do would be to win over the community by making very small and uncontroversial improvements. Nothing that is self-serving. UI improvements, better search, nicer notifications, etc.
Increasing revenue would be easy to do but very risky if it involves any new forms of advertising. I hope they can push that one off a bit or come up with something users don't find objectionable (which will be tough).
Reddit has grown so large with only very minor improvements over the years. It would be a shame to have someone "fix" what ain't broke.