# How would the software industry react if Stack Overflow was no longer available for use? [closed]

Nowadays Stack Overflow is a common place for programmers to solve their problems. Most of the companies such as Microsoft, Telerik, etc. encourage discussion on Stack Overflow about coding relating issues. In this stage if Stack Overflow is no longer available to use (maybe due to some hackers group attack) means what would happen to software industry?

• Will the pace of software development be reduced? If yes, how much it will be reduced approximately?
• How will software industry manage the valuable information lost due to StackOverflow loss?
• Yes, people will try to build another Stack Overflow site. But how much time it will take to build a similar website with similar content?
• Will there be any advantage for programmers if Stack Overflow is no more?
• There are clonics of stackoverflow - even tho not as performant - creating a new site (with funding) wont take that long. Also not all the information will not be lost, there is the Internet Archive - old enough very popular answers are safe. Example of question saved. – Theraot Jan 24 '16 at 13:25
• I'm aware. Yet it is not only the sites, but their code. Some of the clones are open source - ready for grabs. Still needing money to be put on good hosting, and some more for advertising and people to seed the community. Also with SO gone, some other site will raise. Edit: I don't know to estimate the impact of SO gone, that's why I'm not answering. Yet I don't think it would be world shattering. – Theraot Jan 24 '16 at 13:36
• You state that Stack Overflow is no longer available. Is this believed to be a somewhat temporary outage (power failure maybe?), or are we talking about a situation where the site becomes permanently unavailable? I don't know Stack Exchange's (the company's) failure plans, but given that pretty much the entire reason for the company to exist at all is the content provided by others, I can't imagine they would flunk something as basic as regular backups stored off-site. At that point, even huge problems should be fixable in days (new hardware). – user Jan 24 '16 at 14:43
• It might finally give me the impetus to give technology the finger and take up sheep farming. – Whelkaholism Jan 25 '16 at 10:57
• If the whole StackExchange network went, you wouldn't be able to read answers to this question anyway. :) – biziclop Jan 25 '16 at 11:15

My job requires me to do some programming and I am a member of and frequently use StackOverflow to solve specific problems.

## Magnitude of Impact

However, I turn to StackOverflow for assistance when I generally already know how to solve a specific problem, but am having problems implementing that solution. Based upon my personal experience, I would estimate no more than a 10-20% decrease in efficiency for those tasks that I would turn to StackOverflow to solve.

Since this is a minority of the problems I face every day, I would not expect the impact to be all that great.

## Preserving the lost information

IMO, the value of StackOverflow relative to other web resources is the specificity to which it can get. I can always find general instructions on how to do various bits of code. However, if I'm trying to do something very specific and get unexpected output (usually errors), I can turn to StackOverflow to see if someone else has tried doing exactly what I'm doing and how they got past the error messages.

The other benefit is that if I can't find what I'm looking for, I can always pose the question to the group.

I'm saying the information to do this IS already there on other sites. But some of the specifics are not.

## Cost to build a StackOverflow replacement

Unknown. I don't know how much StackOverflow cost to build. I imagine a replacement would simply cost the hardware and man-power to build and staff the place.

The content comes from the subscribers.

No advantages spring to mind. I imagine it is possible that a "new" Stack Overflow might include better organization or search tools but that is by no means certain. I would expect the odds of the replacement site being better would be at most equal to the replacement being worse.

And you'd still need to accumulate the content.

• There are already clones, so we there wouldn't need to be a rebuild step. – PyRulez Jan 25 '16 at 1:05
• @PyRulez but do they have clones of Jon Skeet in there? – Tobia Tesan Jan 25 '16 at 14:46

Blimey, we'd have to read the docs for the software we use! Imagine not being able to cut and paste random bits of code off the web any more, productivity would plummet.

Mind you, quality might increase, particularly if developers spent less time on the stack exchange sites instead of working :-)

• Rather, probably the quality will probably decrease to what it was before SO. At least, now, beginners will be exposed to constructive criticism from peers. When all they had was the manuals, they interpreted them the way the looney right reads the Qur'an. – tripleee Jan 25 '16 at 11:30
• @tripleee I don't see beginners posting their code on SO for criticism. You're right that manuals are like scripture, except in the case of programming documentation they are 100% true and not open to interpretation. – gbjbaanb Jan 26 '16 at 8:40
• No, but poor code on SO will often have comments or competing answers pointing out flaws, misunderstandings, and violations of best current practices. (Probably less so in e.g. the php or regex tags, where signal/noise is worse than on SO overall.) And lots of documentation is obscure to the point where interpretations abound (witness MSDN and the elaborate guesswork it requires where marketing decided something should not be explained properly, or the documentation author was a junior intern with dyslexia). – tripleee Jan 26 '16 at 9:00
• Having worked before and after Stack Overflow and working with private APIs where you can't search for knowledge, I can confirm that quality would most certainly plummet; not knowing how to do something in a simple or efficient way does not encourage you to find that simple and efficient way. It encourages you to get anything working, no matter how hideously overcomplicated that thing might be. Yes, you'll feel guilty about it, but you have a deadline. – deworde Jan 26 '16 at 9:20
• @PhilipRowlands you kids. In my day we used to swordfight! :-) – gbjbaanb Jan 26 '16 at 10:20

While StackOverflow is one of the biggest and most convenient sites to get software issues answered, it is not the only one. There are several others out there of reasonable size and many, many smaller blogs with more limited but focused topics.

So while it would be a pain to have more places to look it wouldn't be crippling. Like Jim2B said, often I use it to remember how a specific implementation is done that I use infrequently and haven't memorized yet. There would be a period of mourning, and then we'd just keep on doing our jobs. I've been a software engineer for twice as long as stack overflow has existed, and I was still able to do my job back then.

It's happened before, or near enough: Ten (OK 15-20) years ago you could have asked the same about usenet, or at least the relevant parts of it. To give an example with a near-direct equivalent: tex.stackexchange has pretty much replaced comp.text.tex. I'll admit to a nostalgia for the old distributed ways, but the model here is much more useful, with inline images, rep., etc. I used to be quite active in equivalents to diy.se, superuser (many newsgroups), etc.

What you suggest in the question is pretty much what happened to usenet -- ISP stopped providing servers, Google groups was more of an attempt to take over than a decent interface (though its search facilities got it lots of traffic). People switched to forums of varying quality, some free, some paid (expertsexchange). Eventually SE gained a dominant position. With hindsight it's easy to say that an open model was always going to win out.

Communities are more resilient than web services; something would grow up. If aliens took out SE there would be a dozen hacked-together replacements within a week (all those programmers stuck on their paying projects would have to do aomething to prevaricate).

If Stack Overflow Inc. went out of business tomorrow, there'd be a momentary disruption in access to the existing data and a larger disruption in getting new questions answered.

Questions and answers on Stack Exchange sites are licensed by their authors under the Creative Commons Attribution-ShareAlike (CC BY-SA) license. Stack Overflow Inc. provides data dumps of Stack Overflow and the rest of the Stack Exchange network to Internet Archive, which in turn makes them available to the public through BitTorrent. Someone with a halfway recent data dump would probably toss up a read-only interface. In fact, such mirrors already exist. So many exist in fact that Meta Stack Exchange has a procedure to report Stack Content Republishers Attributing Poorly and/or Excelling at Ranking (SCRAPERs), or mirrors that are incompletely fulfilling their obligations to Stack Exchange users under CC BY-SA.

But because of the levels of traffic that Stack Exchange gets, it might take a while for a site to offer read-write access, or the ability to ask and answer questions. It'd have to replicate all the caching and moderation "magic" that happens behind the scenes. There exist workalikes of Stack Exchange software, but as far as I can tell, none are intended to scale as large as, say, Stack Overflow. Survivability, ultimately, is one of the pitfalls of using secret, non-free software. Compare to reddit, which makes everything but the spam filter available through its GitHub repository.

Is this Worldbuilding or Meta Stack Exchange?

• Underrated response. This is not only amazing but also provides roadmaps today for real work arounds in case SO would cease to exist. – nilon Sep 3 at 12:20

I don't think that SO helps senior developers - at least the help is not critical. If SO stops working, many junior programmers will slow down their work but globally this will not be noticed. After few months, the same junior programmers will be fired or will become more experienced and better thinking. They will not expect external help for every little bug.

• Do you never change disciplines/platform/libraries/languages? I started with Java 1.0.2 in 1996, Hadoop proficient in 2012, contributor to Spark in 2014, etc. But I'm a noobie in swift as of the beginning of this month. While not the only resource, SOF is a great boon. But back to your original premise: there used to be true experts on this site and still are here and there. Tough questions can and do get answered here. – javadba May 25 at 21:29
• @javadba When I change platform/language, I read books, PDFs, etc. sources. I do not start asking very basic questions in forums like "help with my homework". You are not losing time while trying to find answer of question in books or other non-interactive sources - you learn other things. – i486 May 26 at 7:56
• There was a short lived attempt at Documentation Stack Exchange for the very reason that documentation is generally sparse. Yes you go read what is out there but what is out there is limited. – javadba May 26 at 13:06

## Pace of development

In the initial hours I would believe the pace of software development will be decreased. Mostly because

1. Searching with Google for any programming related problem will most likely continue to serve StackOverflow. You have to filter those results yourself, which takes time.
2. People posting on social media that StackOverflow is down and that the efficiency of programmers is decreased.

For some programmers this may result in a larger decrease than others. This may even vary per project.

## Data Loss

The loss will be at most three months of data. Since 2009 StackExchange dumps its data every three months and makes it available at the internet archive.

## Clones

Since the data is available. Whenever the news occurs that SO will not return, (illegal) clones will pop up rather soon. I would think that within a week or-so a clone will exist. However it may take longer, since SO itself is probably required for people to create the clone.

Again, this differs per person. Those who search every issue they have, even multiple times and fail to remember solutions will have to remember those solutions. Initially this will give them a time penalty, however I would like to think that the cache in the brain has a lower latency than the cache on the web. With sufficient training you could increase the brain-cache in size, reducing the need for high-latency lookups.

There will be someone trying to fix this issue.

But for a while, users feel less motivated on sites that clone the data on StackOverflow. And users cannot easily find sites that doesn't clone those data, and it would be somewhat paradoxical to deal with low quality contents there: a site must firstly live with that, but StackOverflow has shown they will be unwelcome sooner or later, and users always debate about this. It's very likely they all will be forgotten slowly, as the informations on them expire and there are not enough users motivated to maintain them. The final clones that are not down are likely read-only or converted into other formats such as a wiki. The same person or company who runs it may also run another separate QA site, but they must have a different path to be successful and to get enough users.

Existing QA sites that are not exactly the StackOverflow model might have some advantages here. Some new sites will also try to advertise their sites in different ways, such as being specialized, having hired answerers, or being incorporated into another model. Many of them may not work well as a model, but they only work because they distinguish themselves from exact StackOverflow clones. And users can have a higher expectation based on how they did their works.

This will be discussed in other types of sites such as forums, SNSes, chatrooms or even mailing lists, maybe even in the comment area of a news sites, to improve their social cohesion. Specially, someone will claim the things you can get from StackOverflow are not the (as they thought) most useful ability a programmer should have. Though it is irrelevant when StackOverflow exists because it supposedly doesn't hurt to also have StackOverflow no matter what you value the most.

Others may criticize the users on StackOverflow that may try to push their common sense to the industry, and say that we should agree to disagree. Those who don't agree to disagree may think misconceptions exist everywhere, and it doesn't worth to clarify anymore. This might not seemed to be that harmful. But new sites that intended to replace StackOverflow struggle to find why they cannot replicate what was on StackOverflow exactly, and don't realize they must do extra works inventing a different way to encourage some of the users to do the dirty, hard and repetitive jobs.

New sites sometimes might also be considered jokes there. Otherwise, users aggressively want to help a new site, bringing in different ideas about how the site should work, only making it less manageable. As there might be many similar sites, users will also abandon a site that they think not good enough rapidly.

And the possible final results are:

• There might be an already big, prepared site advertising too much about their new things and take the role of StackOverflow. They already had their users, so they wouldn't have the difficulty. It's very likely the will refuse to talk to their users too much as that would be likely the way they would have succeeded.
• It's possible that someone finally worked a way mixing contents from different locations and successfully preserving their quality before someone becoming the new StackOverflow. And StackOverflow would be irrelevant.
• With those problems, the path of being another StackOverflow is effectively blocked for a while. But seeing it as a chance, sites owned by a company with an irrelevant model might all want to also have some QA features. And they may develop faster without StackOverflow. One site may finally grow to a complete StackOverflow, with a lot of legacy fascinating new features that didn't exist in QA sites.

The pace of software development would not be reduced significantly immediately, as the data won't be lost, and they would remain being useful for a long time. And everything new asked there is still answered by a human. Because the problems described on StackOverflow should be isolated enough, I'd say it won't be too harmful to the overall efficiency if a company just hire more people to deal with those problems as a last resort, if that really becomes a concern. And 8k questions/day still doesn't compare to the total number of developers around the world.

But without a default way solving problems for those who cannot solve problems themselves, ugly workarounds might be more common and potentially might increase the number of bugs a little. Or something on StackOverflow is already ugly workarounds, and people find a way to improve without StackOverflow. Who knows?

• and maybe, if some software packages are too difficult or undocumented to get working without SO answers, people will simply strop using them. A lot of software libraries or languages or products would no longer be used, and perhaps... this would be a good thing. – gbjbaanb Dec 16 '17 at 0:40

If Stack Overflow was no longer available today lots of “clone” sites will start up populated with the old questions/answers from the data dumps. However the expert will be divided between these sites and it will take a very long time until one site becomes the clear winner.

BUT

It will take a change in the world for StackOverflow to no longer be available (the investors will not allow their money printing machine to close, if it is still getting lots of users) if this was to happen, it would not be today, but in some time. When it happens the world would have change in a way that stops StackOverflow from being useful. We can no more predict what the outcome will be like, then newnet users could predict the details of StackOverflow.

So what sort of changes could it be.

• Programmers no longer wish to help each other
• All programmers work for a small number of companies and are not allowed to talk to anyone outside of their own company.
• Computers learn how to program themselves.
• The world is no damaged, there we no longer have power to run computers.
• Legal issues make StackOverflow impractical.
• Something a lot “smaller” then the above, that none of us have thought off.

Personally I will put my money on the last item in the above list.

• Will pace of the software industry decrease?

Yes.we could expect some short-term decrease in the pace of software industry.

Maybe it would not affect well-experienced developers but it will surely affect freshers and less experienced developers who always looks straight forward answers for their problems and surely this issues will transform to small scale industries who heavily depend on these developers since they cannot always hire market leaders.As a result small scale industries, freelancers will be affected more and they need to give more training to freshers to understand the language documents.