To bad they're on the hyper-growth train now as well, but I guess their investors want to get some money back. I'd preferred seeing them grow slow and stable, because I feel that focusing solely on growth often leads to decisions not in the interest of the users.
- Find thing that gets traction and growth.
- Demand more growth than the thing you do well is providing.
- Focus not on the thing you do well....
Seems to happen time and again. I always think of the story on HN about the guy who bought back his startup from the VCs for $1 after it "failed" and he went and ran it without the same expectations, and it is doing just fine.
There are maybe just some good ideas out there that aren't meant to be unicorns, but are productive none the less but they get buried in other expectations.
I can’t imagine someone messing up Stackoverflow, but Digg managed to do it so there’s always a way (although SO doesn’t have a high quality direct competitor like Reddit...yet.). The best time to take on a business is when they’ve peaked and get distracted, they forget about being the best in the market. I should note I don’t think SO is near that point yet.
Prashanth is a grave robber trying to find a few valuables to snatch for investors.
But SO is still pretty damn useful, like Wikipedia.
I'm genuinely curious, I'm not familiar with SO besides having answered a few top questions and getting some points long ago.
Except SO for businesses could probably work quite well. I work for BIGTECHCO and the natural language query interface into institutional knowledge is very ad-hoc, being distributed amongst Slack histories, internal mailing list archives, etc. It would definitely be improved if consolidated with a decent UI.
Agree with the general point, but the guy burned 8 million dollars in investor money to get to the point where the business is "just fine". It's not really a good example IMO.
Also here is where the project started on Show HN:
Really to cool to see a Show HN and know the story years later.
SO, like Trello, took a long time to find a business model (honestly Trello never really did); it seems clear that SO Jobs didn't work, but enterprise sales is the key, just based on what they emphasized in the announcement. Using the SaaS website as a loss leader advertisement for the shrinkwrap software is an interesting model: I see it as somewhat similar to the old 'give away Unix and C free to universities, sell System V to the companies those students work at' model, modernized for the internet age.
Genius is in the same boat. A decade old, failed premise (turns out people don't actually want to annotate everything), failed expansion plans (so... back to lyrics then), no business model. At the mercy of the VCs, they'll eventually be passed around for $10m or $20m like MySpace or Tumblr.
These companies are this start-up generation's About.com (or the latest eHow).
Simple rule for anyone doing a knowledge service: if you put the profit motive over the knowledge motive (which is the only choice if you take a lot of VC), your service will end up in the dustbin of history, no exceptions.
It could disappear tomorrow and leave barely a memory.
Worse yet, the site is full of dumb answers as well.
The problem IMO is the scale of the opportunity relative to how much money is consumed.
There are great businesses at every increment of enterprise value from $5 million up to $5 billion. We just need to realize this, and stop judging greatness by the pure size of the exit.
I think Silicon Valley could learn a lot from Hollywood in so many ways. One is how we think about "greatness". Is there the equivalent of a "great indie flick" for software companies? Or will we continue to use "exit value" as the only yardstick?
: "especially Teams, where we are starting to close many huge deals and becoming a hyper-growth enterprise software company very quickly" "And we want to make it possible for knowledge workers everywhere to use Stack Overflow to get answers to the proprietary questions that are specific to their organizations and teams."
I imagine that many companies experience a sort of death spiral once they start focusing too much on profit/growth/monetization at the expense of actually catering to their users' needs.
On the other extreme, focusing too much on users and not enough on profit probably isn't a good way to remain in business for long...
Plenty of other stuff in past year or two...
(below): I shared it but no comments: https://news.ycombinator.com/item?id=21053475
One: is Rackspace really such a great company? Hard to say. Maybe this CEO was able to achieve great results there. From where I'm sitting, Rackspace is a third-tier cloud infrastructure provider at best, falling farther and farther behind competent "supermarkets" like Azure/GCE/AWS, as well as niche providers like Digital Ocean, Hetzner, or Softlayer.
As for the announcement itself, this feels _incredibly_ "photoshopped". I give Joel a lot of credit for stepping back from something he wasn't excited about, but this press release reads as just this -- a press release. This is the guy who wrote "Hitting the High Notes", "The Guerilla Guide to Interviewing", and many other fantastic essays. The entire tone of this release just screamed "PR".
(See the image alt text)
Edit - there it is: http://gkoberger.github.io/stacksort/
I always hated EE. They were an early adopter of dark UI patterns, and it obviously made them vulnerable to less user-hostile competition like SO.
Looking at Microsoft's strategy I doubt they would care whether it runs on .Net or something else. They would look into integrating with Azure (lots of Linux) and GitHub (R-o-R) as well as SO Jobs with LinkedIn (Java(?))
SO will still be there but maybe it will start the journey of it's eventual slow death.
LinkedIn is making bank with its subscription tiers. So they’d do a good job, financially speaking, of doing the same thing with SO. Large companies and recruitment services will be able to buy premium subscriptions to see who is doing what on the platform. Professionals will pay $50/month to have better access and limit access to non-paying members.
EDIT: Oops. Now I know that Microsoft owns LinkedIn.
I've never contributed anything to SO, but this is one of the most valuable answers they could give.
Worst case, the people helping you can better understand the context around what you're trying to accomplish. Best case they recognize that there's a much easier way to do what you're trying to do, when you're too busy bashing your head against the wall on the particular implementation that you chose.
So often when people get stuck, it's because they were shaving a Great Brown yak.
But the Great Brown Yak can only be shaved with a razor from Bhutan. So they go to get the Bhutaneese razor but first they need to prove their worth by climbing the nearby mountain.
In order to climb that mountain they need the best climbing gear in the world, which is only available from a small shop in Honduras.
To get to Honduras from Bhutan they need their passport, but they lost the page with their photo on it.
Since they can't get back to their home country they need to get a passport photo some other way.
So they go to Stack Overflow and beg for someone to help them forge a passport photo.
SO asks what they're trying to accomplish, and they mention they need some Yak fur, and it needs to be brown.
SO suggests that they shave the Minor Brown Yak which can be shaved with an off the shelf Gillette but produces indistinguishable brown Yak fur, and could be done in five minutes.
A better way to respond would be: “Here is how to do precisely what you’re asking, but there may be a better way if you more clearly stated your end goal.”
Maybe the real issue is finding yak fur, or maybe its a razor, or maybe its climbing gear, or maybe its the passport. Or maybe the yak fur turns out to also just be a way to solve an even realer problem?
Rackspace - 7 years
Senior Vice President & GM, Cloud & Infrastructure Services 2019 – Present
Senior Vice President & GM, Global Managed Public Clouds Business 2017 – 2019
Vice President & GM, Global AWS Business Unit 2015 – 2017
Senior Director & GM, SMB Marquee Business Segment 2013 – 2015
Director & Chief of Staff 2012 – 2013
Compared to chat or wikis I can see an argument that it is good for a distributed team - although in my company we tried it for a bit and didn’t find it to fit with our team.
/rant of an on-prem diehard
And then the Question2Answer app is never updated by infra, it falls 402 releases behind mainline, and gradually becomes unusable due to constant crashes.
The company then spends ten times the install cost for an external consultant to migrate the content to cloud. (It's a living)
I think it probably has a lot to do with team size, industry, and also with how distributed/remote the team is whether it's worthwhile.
That's a pretty big deal and probably saved them from looming irrelevance.
Didn't intend to spend an hour on it, but we all know how these things often go.
Also Jeff Atwood, the other cofounder of SO has a blog that is very informative called Coding Horror.
Awesome awesome talk
It always failed (failed = does not continue on the same trajectory; at best is stable and in the long term eaten up by more innovative companies).
I wonder if he regrets not taking VC money earlier - maybe then FogBugz would be the new jira, and FogCreek would be as big as Atlassian.
I love reading the docs and using issues.
More generally, I use async Python features and libraries, which are developing and changing very quickly. Most stack overflow answers on things like this are at least partially outdated, many more-so. I tend to go to prominent async libraries source code & Github issues to see how the leading users of async features are doing it.
I work in web so basically anything NPM related (most of the web in my field) is all on GitHub. Much faster than StackOverflow for sure over the past few years. I now read SO mostly to shoot the crap. GitHub has a hold of the latest issues by far. JS being the wild west with guns blazing in the air, has done really well for GitHub.
I also think people raise good points about Microsoft acquiring SO. They are buying up all the developer circles.
Personally, I hope the license thing bites Mr. Spolsky. I think he deserves it after speaking out publicly about getting companies to squash our creative freedoms at home. I'm sorry Joel, not everybody wishes to work 14 hour days for somebody else's vision.
StackOverflow's surplus of dickhead programmers doesn't help either. We don't hide behind usernames on GitHub.
For help with libraries, it makes sense kind of. But I personally wouldn't want to file an issue unless I was sure there was an actual issue. "I don't know how to use it" isn't an issue.
I got a clear, concise, and correct answer. I've heard a lot of noise about the toxic community on SO, but I never see it whenever I'm there. Or maybe it's me. Anyway, SO is a very useful resource for me, even at this late date.
I don't really see what is so horrible about these answers?
What did he say / where could I read it?
That being said, your hypothetical is materially different in that it's literally not increasing the type of diversity they were referring to. This is assuming a reasonable metric of diversity, probably similar to something I've encountered, Shannon entropy, etc.