4 ms·

I know it's fashionable in certain circles to rag on it lately, but I'm not too big to admit it has saved me a lot of time here and there over the last 4-5 year

I know it's fashionable in certain circles to rag on it lately, but I'm not too big to admit it has saved me a lot of time here and there over the last 4-5 years. I would bet a lot of devs use it more than they would admit to check down on things throughout the course of any given project. Billions seems like a lot, but I wouldn't call BS on it out of hand.

InclinedPlane 13y ago
Personally I think StackOverflow is pretty good. It's always nice to see it popup in search results when googling for a technical problem, for example. My main complaint is that they've been resting on their laurels for quite a while, there's a lot of potential that's been wasted with SO/SE.
- jc4p 13y ago
  Just curious, what do you think we could be doing that we aren't? Also thanks for teaching me a new idiom :)
  - InclinedPlane 13y ago
    In general, experimenting more. You've tackled the narrow use case of a single question with answers, but is that all SE can be, or should be? The thing that I think SE is most lacking is better ways to tackle the sheer amount of data on the site. There's a lot of information there but in general it's not very well organized. Go back to the original announcements about the launch of SO and see what it was supposed to be and how it has either lived up or not to those ideas. Try this thought experiment: if a stack exchange competitor showed up tomorrow, what sorts of features would it have to have to make you start waking up in a cold sweat every day?
    - jc4p 13y ago
      Luckily a lot of us are on the same page as you :) One of the things I consider a main detriment of our current product is that for the vast majority of users it's built to be transactional. Search, land on SO, read, go back to work. On our currently in development mobile apps we're trying out something different, for the first time at scale we're pushing content to users rather than having them search for it as the number one utility of the product. The apps have a "Feed" which is machine learned interesting for you questions. I love it since I search SO at work but when I get home or am walking around town I can open my app and instantly lose 30 mins (at least) to reading and interacting with very interesting content. I can't wait to see if others like it.
      - startuup 13y ago
        Thats a great start but there are other ideas too. Why isn't SE as useful for people in other professions, like product managers or data scientists or marketers who are also looking for answers to their work problems? The key is that just replicating the site is not the answer, you need to really focus on community dev and getting the thought leaders from those professionals involved. Lastly, along with the 'feed' implementation, you can also start curating other content from the web that could be useful to the users of your site. example: If I am particularly interested in Javascript, I get a feed of curated resources from around the web that will help me learn about the latest and greatest in Javacript. Thats whats missing today and thats why SO is a transactional site
        jc4p 13y ago
        I actually originally had something related in my post but removed it: I won't personally consider the site I work on a success unless my mom uses it. She's a physics PhD and she once in a while stumbles onto physics.SE when searching online but she doesn't actually use the site. I think SO and Server Fault are the size they are because they have an intrinsic advantage over our other sites. People in the target audience work and live in the same tech ecosystem as a website, with something like physics that's an initial burden. SE itself also has a high barrier to entry. Every one of our sites has it's own rules and definition of what type of content is acceptable. This is for a very good reason, seeking out absolute answers to questions. Most other websites on the internet live to foster discussion, but we're not a forum. Making stack exchange a common name (in more than our sizable niche) is one of our company goals, but I don't really know what we can do to get past the initial issue I mentioned and reducing the second would hurt our product. It's quite an interesting idea to be more proactive about bringing thought leaders to out lesser known sites, I don't know if we could do it just directly since most thought leaders are very busy people who already have a dozen places people ask them questions but it's something to think about. We already consider expertise of contributors in Area 51, our incubator for new website ideas. Perhaps while we may not be able to get the top 1% of thought leaders onto our websites without giving them a really hard question to answer (my mom loves difficult questions, god knows how many times she used to run into my room and say she solved a question I asked her in passing days prior) but it might be possible to bring in up-and-coming thought leaders. A lot of my friends from the health and fitness industry (were I used to live) might want to use our Physical Fitness Stack Exchange site to foster their own personal brand.
  - CJefferson 13y ago
    My main with stack overflow, as it gets older, is answer rot. A couple of times I have come across answers which have become out of date, but it is non obvious to me the best way to reopen the question (I, along with most readers don't have the karma to edit questions / answers )
    - pjbrunet 13y ago
      Maybe write a comment, "This is out of date. Can someone update this?" You might even want to flag the question. You can also add a new answer and downvote the obsolete answer. Getting enough points to edit questions is pretty easy.
    - sklivvz1971 13y ago
      Everyone can edit questions and answers (if you don't have the full-on privilege yet, your edit will go in a queue and be vetted by more experienced users).
- sklivvz1971 13y ago
  (*) where resting on laurels is defined as launching 111 sites running on 9 web servers with traffic doubling every 12 months while writing two native mobile apps and localizing the site :-)
  - InclinedPlane 13y ago
    Hey, that's great, but to be brutally honest most of those sites are borderline inconsequential. The fact that SE has yet to launch a site that's gotten bigger than StackOverflow is pretty damning. Everything else you describe is just work, it's not innovation.
    - spolsky 13y ago
      stackexchange.com is bigger than stackoverflow.com. Source: https://www.quantcast.com/stackexchange.com?country=US https://www.quantcast.com/stackexchange.com?country=US https://www.quantcast.com/stackoverflow.com?country=US https://www.quantcast.com/stackoverflow.com?country=US
      - jc4p 13y ago
        True and Stack Overflow is #1 in North Korea.
      - InclinedPlane 13y ago
        That's good but SE is a federation of sites, so it's an unfair comparison. From the start I've always thought that the biggest danger to SO/SE wasn't failure, you guys are smart enough to see and recover from outright failure. Instead it's reaching a level of success far below what you're capable of. It seems to me that SO/SE spent a lot of work innovating in the early days but once they found their comfort zone they've stopped innovating. They are good sites for their intended purpose but how much of the original vision has been abandoned? I remember back when you and Jeff were talking about taking over the way all online discussion happened. Nothing like that is even remotely possible with the way StackExchange is today. To be bluntly honest I think you guys are just lucky that most of the ambitious and competent folks in the industry don't view discussion as "sexy" so you don't have as much ompetition as you would otherwise (quora is practically a joke). But realistically it's just a matter of time until SE will be out innovated.
fein 13y ago
I can learn and know one language incredibly well, excluding the html/css/js world, as those are effectively required at this point. For me, that's PHP. However, when I need to go to work, and my boss decides "So that position we hired you for, now its going to be ruby as well, but later on ObjC and Java.", SO comes to the rescue. The core concepts of programming languages were cemented in my years deciphering terrible notes and learning to code in college, however the little nuances of each language inherently evade me. This is where SO has shined the most. Most of the time, solutions with multiple approaches suggested show more about the language itself than the problem asked.
duaneb 13y ago
What on earth is bad about using StackOverflow? You're expected to google around to find solutions anyway, if anything I would think much less of the programmer who tried to avoid it. It's stupid pride when it's such a valuable resource.
- rrich 13y ago
  I agree with you, but several people I have encountered don't feel this way. For instance I had someone try to explain how using HAML made their developers more valuable because of all the keystrokes they were saving. When you're counting keystrokes as a developer productivity metric I can only wonder how toxic the work environment becomes when you need to stop to look at documentation yet alone use Google or SO.
  - subsection1h 13y ago
    Maybe that person's team is more productive with Haml not simply because of the small amount of time that's saved by less typing, but because of the huge amount of time that's saved by not needing to replace developers who are institutionalized or commit suicide as a result of being forced to type HTML tags like it's still 2005[1]. [1] The year before Haml was introduced.
    - rrich 13y ago
      I'm not saying HAML isn't of value. Please don't misunderstand me, I like HAML. I use it for my template engine with Sinatra. But in the grand scheme of things if using HAML to increase productivity by saving keystrokes is what the company thinks is its competitive advantage, which is what they told me, then how do they deal with someone spending time just researching (i.e. stackoverflow, google) and thinking a problem and solution through thoroughly? The two seem at odds to me.
    - mattmanser 13y ago
      Uh, not to rag on HAML which I'm sure you find great but all the text editors I use these days auto-complete HTML, including closing tags. I spend a fraction of the time typing html compared to the time I spend thinking about why the hell the damn image/span/font/button isn't in the right place.

InclinedPlane 13y ago

Personally I think StackOverflow is pretty good. It's always nice to see it popup in search results when googling for a technical problem, for example. My main complaint is that they've been resting on their laurels for quite a while, there's a lot of potential that's been wasted with SO/SE.

jc4p 13y ago
Just curious, what do you think we could be doing that we aren't? Also thanks for teaching me a new idiom :)
- InclinedPlane 13y ago
  In general, experimenting more. You've tackled the narrow use case of a single question with answers, but is that all SE can be, or should be? The thing that I think SE is most lacking is better ways to tackle the sheer amount of data on the site. There's a lot of information there but in general it's not very well organized. Go back to the original announcements about the launch of SO and see what it was supposed to be and how it has either lived up or not to those ideas. Try this thought experiment: if a stack exchange competitor showed up tomorrow, what sorts of features would it have to have to make you start waking up in a cold sweat every day?
  - jc4p 13y ago
    Luckily a lot of us are on the same page as you :) One of the things I consider a main detriment of our current product is that for the vast majority of users it's built to be transactional. Search, land on SO, read, go back to work. On our currently in development mobile apps we're trying out something different, for the first time at scale we're pushing content to users rather than having them search for it as the number one utility of the product. The apps have a "Feed" which is machine learned interesting for you questions. I love it since I search SO at work but when I get home or am walking around town I can open my app and instantly lose 30 mins (at least) to reading and interacting with very interesting content. I can't wait to see if others like it.
    - startuup 13y ago
      Thats a great start but there are other ideas too. Why isn't SE as useful for people in other professions, like product managers or data scientists or marketers who are also looking for answers to their work problems? The key is that just replicating the site is not the answer, you need to really focus on community dev and getting the thought leaders from those professionals involved. Lastly, along with the 'feed' implementation, you can also start curating other content from the web that could be useful to the users of your site. example: If I am particularly interested in Javascript, I get a feed of curated resources from around the web that will help me learn about the latest and greatest in Javacript. Thats whats missing today and thats why SO is a transactional site
      - jc4p 13y ago
        I actually originally had something related in my post but removed it: I won't personally consider the site I work on a success unless my mom uses it. She's a physics PhD and she once in a while stumbles onto physics.SE when searching online but she doesn't actually use the site. I think SO and Server Fault are the size they are because they have an intrinsic advantage over our other sites. People in the target audience work and live in the same tech ecosystem as a website, with something like physics that's an initial burden. SE itself also has a high barrier to entry. Every one of our sites has it's own rules and definition of what type of content is acceptable. This is for a very good reason, seeking out absolute answers to questions. Most other websites on the internet live to foster discussion, but we're not a forum. Making stack exchange a common name (in more than our sizable niche) is one of our company goals, but I don't really know what we can do to get past the initial issue I mentioned and reducing the second would hurt our product. It's quite an interesting idea to be more proactive about bringing thought leaders to out lesser known sites, I don't know if we could do it just directly since most thought leaders are very busy people who already have a dozen places people ask them questions but it's something to think about. We already consider expertise of contributors in Area 51, our incubator for new website ideas. Perhaps while we may not be able to get the top 1% of thought leaders onto our websites without giving them a really hard question to answer (my mom loves difficult questions, god knows how many times she used to run into my room and say she solved a question I asked her in passing days prior) but it might be possible to bring in up-and-coming thought leaders. A lot of my friends from the health and fitness industry (were I used to live) might want to use our Physical Fitness Stack Exchange site to foster their own personal brand.
- CJefferson 13y ago
  My main with stack overflow, as it gets older, is answer rot. A couple of times I have come across answers which have become out of date, but it is non obvious to me the best way to reopen the question (I, along with most readers don't have the karma to edit questions / answers )
  - pjbrunet 13y ago
    Maybe write a comment, "This is out of date. Can someone update this?" You might even want to flag the question. You can also add a new answer and downvote the obsolete answer. Getting enough points to edit questions is pretty easy.
  - sklivvz1971 13y ago
    Everyone can edit questions and answers (if you don't have the full-on privilege yet, your edit will go in a queue and be vetted by more experienced users).
sklivvz1971 13y ago
(*) where resting on laurels is defined as launching 111 sites running on 9 web servers with traffic doubling every 12 months while writing two native mobile apps and localizing the site :-)
- InclinedPlane 13y ago
  Hey, that's great, but to be brutally honest most of those sites are borderline inconsequential. The fact that SE has yet to launch a site that's gotten bigger than StackOverflow is pretty damning. Everything else you describe is just work, it's not innovation.
  - spolsky 13y ago
    stackexchange.com is bigger than stackoverflow.com. Source: https://www.quantcast.com/stackexchange.com?country=US https://www.quantcast.com/stackexchange.com?country=US https://www.quantcast.com/stackoverflow.com?country=US https://www.quantcast.com/stackoverflow.com?country=US
    - jc4p 13y ago
      True and Stack Overflow is #1 in North Korea.
    - InclinedPlane 13y ago
      That's good but SE is a federation of sites, so it's an unfair comparison. From the start I've always thought that the biggest danger to SO/SE wasn't failure, you guys are smart enough to see and recover from outright failure. Instead it's reaching a level of success far below what you're capable of. It seems to me that SO/SE spent a lot of work innovating in the early days but once they found their comfort zone they've stopped innovating. They are good sites for their intended purpose but how much of the original vision has been abandoned? I remember back when you and Jeff were talking about taking over the way all online discussion happened. Nothing like that is even remotely possible with the way StackExchange is today. To be bluntly honest I think you guys are just lucky that most of the ambitious and competent folks in the industry don't view discussion as "sexy" so you don't have as much ompetition as you would otherwise (quora is practically a joke). But realistically it's just a matter of time until SE will be out innovated.

fein 13y ago

I can learn and know one language incredibly well, excluding the html/css/js world, as those are effectively required at this point. For me, that's PHP. However, when I need to go to work, and my boss decides "So that position we hired you for, now its going to be ruby as well, but later on ObjC and Java.", SO comes to the rescue. The core concepts of programming languages were cemented in my years deciphering terrible notes and learning to code in college, however the little nuances of each language inherently evade me. This is where SO has shined the most. Most of the time, solutions with multiple approaches suggested show more about the language itself than the problem asked.

duaneb 13y ago

What on earth is bad about using StackOverflow? You're expected to google around to find solutions anyway, if anything I would think much less of the programmer who tried to avoid it. It's stupid pride when it's such a valuable resource.

rrich 13y ago
I agree with you, but several people I have encountered don't feel this way. For instance I had someone try to explain how using HAML made their developers more valuable because of all the keystrokes they were saving. When you're counting keystrokes as a developer productivity metric I can only wonder how toxic the work environment becomes when you need to stop to look at documentation yet alone use Google or SO.
- subsection1h 13y ago
  Maybe that person's team is more productive with Haml not simply because of the small amount of time that's saved by less typing, but because of the huge amount of time that's saved by not needing to replace developers who are institutionalized or commit suicide as a result of being forced to type HTML tags like it's still 2005[1]. [1] The year before Haml was introduced.
  - rrich 13y ago
    I'm not saying HAML isn't of value. Please don't misunderstand me, I like HAML. I use it for my template engine with Sinatra. But in the grand scheme of things if using HAML to increase productivity by saving keystrokes is what the company thinks is its competitive advantage, which is what they told me, then how do they deal with someone spending time just researching (i.e. stackoverflow, google) and thinking a problem and solution through thoroughly? The two seem at odds to me.
  - mattmanser 13y ago
    Uh, not to rag on HAML which I'm sure you find great but all the text editors I use these days auto-complete HTML, including closing tags. I spend a fraction of the time typing html compared to the time I spend thinking about why the hell the damn image/span/font/button isn't in the right place.