Flash is Dead: What Technologies Might Be Next?

Last week, Adobe announced that they would stop supporting Flash by 2020. In some ways this is surprising: I still sometimes run into pages that require Flash, and you can still find a few defenders of the platform in the software development community.

But in other ways it was a long time coming. You can often see the decline of a technology in advance, by examining data on its usage in the software development community. One notable source of such data is in our Stack Overflow Trends tool, which shows questions about Flash have been declining in frequency since 2010.

This sudden shift in Flash’s fortune suggests there’s truth to the conventional wisdom that Apple “killed” Flash by not supporting it on the iPad in 2010.

If we would have been able to see Flash’s decline in advance, what other technologies might be past their peak? Here we’ll use Stack Overflow data to dive into this question.

What technology might be next to die?

Flash has had its defenders over the last decade (here’s some historical context), even though its decline in developer activity was apparent since about 2011. Might there be other technologies that appear reasonably healthy, but have also been declining in the last few years?

To answer this question, we looked at Stack Overflow questions posted over time, the same data behind the Trends tool. We considered the trend in the last five years (to include tags that were once growing, hit a peak, then started shrinking), and estimated the rate of decrease over time for each technology. This is public data (you can use this query to download it yourself), so I’d be interested in what others find.

We considered a technology to be shrinking if questions about it declined by at least 10% per year, on average. By that standard, what were the fastest-shrinking technologies?

Since 2010, the two fastest-shrinking tags on Stack Overflow were Flex (a Flash-based web application framework) and Microsoft’s Silverlight. Both shrunk over time even faster than Flash did, with Flex showing a particularly sharp decline since its peak in 2010. Each of these technologies has dropped by about two orders of magnitude in terms of their presence on the site.

One of these is already deprecated: in 2015, Microsoft announced that they would stop supporting Silverlight by 2021. Flex is technically still officially supported (since 2011, by the Apache Foundation rather than by Adobe), but considering the deprecation of Flash, along with its diminished presence in developer questions, it’s hard to say that it’s a technology with a future.

Among major technologies (those with at least 100,000 questions asked since 2010), there are a few that are shrinking to a noticeable extent. Microsoft’s ASP.NET web framework has been decreasing as a share of Stack Overflow questions since the founding of the site; this may be skewed because the site started with a disproportionate number of C# developers, but it also may reflect a diminishing relevance of Microsoft in the web developer space.

Questions about Ruby on Rails grew more frequent until about 2011, then they started a slow decline. Objective-C has been declining as well, since it’s being replaced by Swift as the language of choice for iOS development. Similarly, the “iphone” and “ipad” tags have been replaced by the general ios tags for questions about Apple’s mobile platform (see our blog post on mobile development for more exploration of those trends or see iOS listings to explore job opportunities).

There are other languages and technologies that have been shrinking rapidly in this time period. Perl has seen its relevance decline for a long time (though it’s worth noting that it still gets more questions month-to-month than truly dead tags like Flash or Flex). Questions about development on Facebook’s application platform peaked in 2012 but has been shrinking steadily since. Questions about the Eclipse IDE started declining around 2014. JQuery Mobile was a popular library for reactive web development, but judging by the number of questions asked about it the interest seems to have declined in the last few years.

Do question visits tell a different story?

Of course, the number of questions asked about a technology isn’t necessarily representative of its health. Developers often search for existing solutions to their problems before asking their own question, and it could be that once a large foundation of questions exist, users get their answers from those questions rather than asking new ones. Is it possible that for some of these tags, the questions that have already been asked are being visited more?

Instead of looking at questions asked, we could therefore consider the number of visits to questions over time. We have this kind of traffic data back to late 2011.

Can we see similar declines in traffic to these tags?

For the selected shrinking tags that we examined in this post, we generally see that the traffic tells a similar story to the number of questions asked. Traffic to questions about Flash, Flex, and Silverlight has dropped to almost nothing since 2012. ASP.NET and Ruby on Rails, while still making up a substantial portion of traffic, have undoubtedly been declining over time. Eclipse stays steady until 2014, then begins declining.

From most of the tags we’ve analyzed, we usually find question traffic is a slightly lagging indicator relative to questions. Once the share of new questions to a technology starts decreasing, question traffic starts declining shortly afterward. Note that this isn’t simply because new questions get viewed by people trying to answer them; about 98% of Stack Overflow traffic is to questions asked not on that day. Rather, this may suggest a shift where developers stop needing to ask new questions, and subsequently stop needing to visit existing solutions.

Conclusion: do questions really represent the health of a technology?

Bjarne Stroustrup, inventor of C++, said that “There are two types of programming languages: the ones people complain about, and the ones nobody uses.” I’d argue that for the most part, the same is true of asking questions.

We on the data team often examine the number of Stack Overflow questions (or visits) to get a sense of the health of a technology. I’ve heard counter-arguments that this isn’t a good measure, since what questions or traffic really represent is how many people are confused about a technology, and that more comfortable users may not need to ask questions. It’s absolutely true that we should examine the biases and shortcomings of any metric, and the comfort of users with a technology is one confounding factor to consider.

However, I think the Trends data still gives us important insights. First, I don’t think it’s possible to run out of questions that can be asked, any more than it’s possible to run out of books that can be written. There have been 1.4 million JavaScript questions asked in the site’s history, yet its pace has grown steadily. Surely if it were possible to run out of questions about JavaScript, we would have done so already.

Secondly, if everyone using your technology knows everything about it, that means your community isn’t growing. A healthy technology environment includes a stream of newcomers that are learning the basics, as well as experts testing the platform’s limits. Both of these types of users will often ask questions as part of their workflow. In 2011, a Flash developer could have argued “Yes, the number of Flash questions has been declining recently, but that’s simply because the documentation is so helpful and we already know how to use it.” Only time would tell the real story.

In a future blog post, I’ll be exploring what technologies are fastest growing as a share of Stack Overflow questions, and considering what that implies about the technology ecosystem.


David Robinson
Data Scientist

Related Articles


  1. Hilariously, I had posted a self-answered Q&A about Flash just days before Adobe’s announcement: https://stackoverflow.com/questions/45205501/warning-1090-migration-issue-despite-explicitly-registering-event-handlers

    Make of that what you will.

  2. Gerwin de Groot says:

    *What Technologies Might Be Next?* ehm … Documentation: https://meta.stackoverflow.com/q/354217/4751173

  3. does asp.net metric include the asp.net-mvc and asp.net core tags ?

    1. Alejandro Wagner says:

      I think asp is deprecated in front of restful services (web api for .net) with mvc frontend, and/or net core

    2. Camilo Blooddrunk says:

      Rather, does the asp.metric includes asp.net-mvc, webforms, asp.net-core, asp.net-core-mvc and the other hundreds of asp.net- tags?

  4. But there were reasons why Flash is being shuttered. Browsers don’t want to have users run binary blobs on their webpages (which is why the push for WebASM is weird to me) or have to install 3rd party plugins in order to see content. That’s why there have been so many new APIs that have exposed many of the things that Flash helped fill the gaps on.

    We now have access to the GPU through WebGL, we have blob access, we have sensor data, notifications, and even higher order network features thanks to WebSocket and WebRTC. There are fewer and fewer things that you can’t do with plain JavaScript. Because of that, browser vendors can push for deprecation because there are safer, more accessible alternatives. I mean, one of the biggest reasons that I never got into Flash development was because I didn’t want to have to buy a license.

    I don’t see an active push to kill the tags that you have brought up.

    1. I don’t see why binary blobs are an issue: most users never look at the code they’re downloading anyway, so whether it’s an “ASCII blob” or a “binary blob” really makes no difference.

      Whether or not the code is “source code” or “human-readable” doesn’t seem to be a huge issue, either; minified JS and JS produced by compiling other languages into JS are both widely accepted without complaint.

      The big difference between something like SWF (the compiled code produced by Flash tools) and WebASM is standardization and openness. While SWF could be read, taken apart, changed and rewritten (I wrote a library and software to do exactly this), there was never an open standard to work from.

      1. martinolsen says:

        Also, as far as I understand, WASM does run in a sandbox. Thus, it should be free of whoops-buffer-overrun-in-the-flash-browser-plugin-now-you-are-pwned-by-evil-website type issues.

        1. No, Web Assembly is not sandboxed machine code: it’s an interpreted language, just like JavaScript. The only difference between JS and wasm is that the latter is designed to be fast to parse and run, at the expense of being hard to read and program. (This downside isn’t a problem when you compile from another language; web developers have already been using this tradeoff for years with minify, CoffeeScript and many other tools that produce unreadable JavaScript.)

          SWF files are also interpreted in this way, so the security issues from that point of view are about the same for all three.

          1. Bálint Csala says:

            It’s sandboxed, it’s even stated on the main page: “WebAssembly describes a memory-safe, sandboxed execution environment that may even be implemented inside existing JavaScript virtual machines.”

          2. Well, ok, it’s sandboxed in the same way that “normal” JavaScript, is, and Flash, for that matter. I usually take “sandbox” to mean running native code in an environment more restricted than native code normally runs in, but certainly the general concept applies to a lot more, and could be considered to extend to almost anything (virtual machines, Docker containers, scripted languages such as Ruby which disallow operations you could do in C, every program on a Linux system other than the kernel….)

      2. It’s a security issue – in theory, someone could run a man-in-the-middle attack while you are downloading binary blobs and replace them with something nefarious like screen-grabbers or keyboard-loggers that operate at the OS/application level. With WebASM, they are restricted to what the web browser offers.

        1. You seem to be arging my point: whether the downloaded code is a “binary blob” (wasm, SWF, machine code) or not (JavaScript) has nothing to do with whether it’s allowed to run in an environment that would give it access it should not have.

          (This message has been edited; I initially misinterpreted the post to which I’m replying.)

  5. None. Just plain HTML animation.

  6. How do the graphs look with absolute numbers, not “% share of all questions”?

    1. Right, it seems odd to say that one technology is in decline just because 10 other unrelated technologies having questions asked about them.

  7. I don’t know if using the number of questions asked about a technology is a good way to track the health of a technology. For asp.net, it could mean that there is already a rich body of questions and answers available for the technology so there is a slim chance that your question is new and unique. Also stack overflow has gotten really smart, as you are typing in your question, it is searching for question with similar qualities and chances are high that your question has already been asked. Thus removing the need to ask it again as it would be marked as a duplicate and you will be given a re-directing link.

  8. Ultimately, Flash died of FUD poisoning. To this day, its most basic features–consistent cross-browser text rendering, rock-solid synchronization of audio and animation, among others–remain unmatched by native browser technologies. The web without Flash may be a more secure place, but it will also be lesser for it.

    1. Agree with everything but “rock-solid synchronization of audio and animation” — Flash is particularly notorious for going out of sync without having a silent audio track playing for the entire duration of the movie.

    2. Douwe Ouwerkerk says:

      On the whole, I’m glad

  9. You show in that first graph that questions about Flash have been steadily declining. What about other technologies? Surely you would (and should) expect the number of questions to decline naturally, due to more questions already having answers? Isn’t that the whole point of SO?

    1. But then wouldn’t the question visits stay constant?

      1. Indeed, I did a big no-no and didn’t finish reading the article before I posted.

      2. Garrett R. Morris says:

        That’s a faulty assertion. You assume the influx of new users is constant, but its not. as a job market reaches mass, the influx of new entrants declines. just as the traffic and # of questions asked. Case and point, People are still using ROR, i see an obnoxious amount of jobs related to ROR on a daily basis, yet it shows decline. Because the job market is saturated with qualified applicants, and thus less new entrants in field.

    2. Sebastian Pojman says:

      You know what? I’m not finished with you yet.

      1. “‘Why on earth?’ Top quality English. I’m joking, of course. I realise it’s typo.” Hmmm… I think there’s something that’ll top this one: literally saying, “It’s typo,” within the same paragraph of your lame ad hominem attack.

      2. You say I’m on the offense, and therefore, I’m supposed to be able to back myself up. Think again. You challenged ME, putting me on the defense, but wait a second… You’ve already admitted you have absolutely ZERO outside sources to back you up. Cambridge isn’t going to work, in case you were thinking of that, because they provide no explanation as to why they say it’s “into” for the verb “dive in” but go on to say it can’t be “into” for “turn in.” In other words, because of Cambridge’s large hole in their definitions, they’re on my side. You, being on the offense, have fallen flat on your face for not providing any reasonable support.

      3. “You haven’t directly refuted my points so I win ;)” a) No comma before “so” completely changes the meaning of your sentence. b) No period? Wow. Impressive English. I’m joking, of course. I realize it’s *a* typo. Also, what points? As I’ve brought up in number two, you have already said you have none. Again, you only *think* it can be one word.

      4. “And you have provided ZIPPO reasons why it can’t be dive into.” Well, duh, because “in” is part of the verb “dive in.”

      Finally, I WIN! Thank you for making this a walk in the park for me. It’s fascinating how fast my opinion of you changed once I realized how poorly you argue…

    3. Sebastian Pojman says:

      1. “‘Why on earth?’ Top quality English. I’m joking, of course. I realise it’s typo.” Hmmm… I think there’s something that’ll top this one: literally saying, “It’s typo,” within the same paragraph of your lame ad hominem attack.

      2. You say I’m on the offense, and therefore, I’m supposed to be able to back myself up. Think again. You challenged ME, putting me on the defense, but wait a second… You’ve already admitted you have absolutely ZERO outside sources to back you up. Cambridge isn’t going to work, in case you were thinking of that, because they provide no explanation as to why they say it’s “into” for the verb “dive in” but go on to say it can’t be “into” for “turn in.” In other words, because of Cambridge’s large hole in their definitions, they’re on my side. You, being on the offense, have fallen flat on your face for not providing any reasonable support. // EDIT: “You haven’t actually provided any external resources to back your claim up. Neither have I, of course.” Actually, American Heritage Dictionary’s defining “dive” as ONLY plunging and “dive in” as a phrasal verb and Grammar Girl’s saying that “in” should be separated when it’s part of the verb is good enough for many people.

      3. “You haven’t directly refuted my points so I win ;)” a) No comma before “so” completely changes the meaning of your sentence. b) No period? Wow. Impressive English. I’m joking, of course. I realize it’s *a* typo. Also, what points? As I’ve brought up in number two edit, you have already said you have none. Again, you only *think* it can be one word.

      4. “And you have provided ZIPPO reasons why it can’t be dive into.” Well, duh, because “in” is part of the verb “dive in.”

      Finally, I WIN! 😉 Thank you for making this a walk in the park for me. It’s fascinating how fast my opinion of you changed once I realized how poorly you argue…

      1. 1) What attack? I literally said I was joking.
        2) Yeah idk.
        3) a) In what way? I can’t think of anything else that sentence could mean. And besides that, there’s no hard-and-fast rule that says that needs a comma. b) The lack of a full stop isn’t a typo. It’s fully intentional. I generally use smilies as punctuation. Deal with it 😉
        4) I don’t care, but I’d like to take this opportunity to thank you for making this so easy for me. And also you spent way too long on that reply 😛

        1. Sebastian Pojman says:

          1. A marginally sarcastic attack of my grammar error because that’s the only kind of argument you could achieve in this conversation 😉

          2. You don’t know, huh 😉 Well, you “attacked” me, so I have to defend my claim 😉 Nature has worked this way since the beginning of time 😉

          3. If you use a comma, it’s cause-and-effect structure 😉 Ex 😉 “I had extra money [CAUSE], so I bought a TV [EFFECT] ;)” If there is no comma, you are describing the purpose of the first clause 😉 Ex 😉 “I left work early [ACTION DONE] so we could meet in the afternoon for lunch [PURPOSE OF ACTION] ;)” It’s always best to add the comma to avoid confusion 😉

          4. You’re welcome 😉 Oh well, losers will be losers 😉 (I’ll count your number four as conceding to me ;))

          Just another comment of yours that doesn’t bring you any closer to supporting your claim… 😉

  10. Sebastian Pojman says:

    Dive IN TO this question.

    1. No, I think in this case “into” is correct. “Into” is a preposition. Where are we diving? Into the question. You’re thinking “dive in”, in which case the subsequent “to” is hard to make fit. You could think of “diving in” as a way to get somewhere, in which case “to” would be the ‘where’ (like “driving to such-and-such a place”), but this requires more contortion in the mind to make it work, whereas “dive into” just works.

      1. Sebastian Pojman says:

        You aren’t jumping into the question, so therefore, the author is using the phrasal verb “dive in,” meaning “to start doing something (in this case getting an answer to the question) enthusiastically.” The adverb “in” should therefore remain separate from the preposition “to.”

        We ought to dive in to this question!

        1. No. What I said still stands. If you “dive in to the question”, “to” becomes nonsensical (it _doesn’t_ describe to what you’re diving in). It’s perfectly fine to use “dive in” on its own, like so: “Why wait? Just dive in!” or “He dived in to get through the book in a day.” or “Rachel dived back in to rescue the struggling boy.” In the case of “dive into the question”, we are in fact jumping/diving into the question. It might be figuratively rather than literally but the grammar remains the same.

          While I was quite sure of my understanding of this situation, I also referenced this page: http://data.grammarbook.com/blog/definitions/into-vs-in-to-expanded/

          1. Sebastian Pojman says:

            Dude, your point doesn’t stand a chance. You say the “to” after the verb “dive in” is nonsensical, right? What kind of a bullcrap response is that? The “to” tells you what you dived in to. The verb is “dive in.” Dive in to your homework. Move in to the house. Break in to the house. Just face it. You’re wrong because unless you want to physically plunge into the question or your homework, the verb is “dive in.” Now, it’s “Rachael dived into the water,” because the verb is “dive.”

          2. “Just face it” only means something if I was intentionally ignoring the facts. I was quite certain I was right, and wasn’t ignoring facts as far as I could see. That said, I don’t know what I was thinking. It’s blatantly obvious that you’re right. It should be “in to”. Let’s just pretend I didn’t completely miss the fact that “dive in” is a phrasal verb and diving into a question doesn’t make sense.

            Perhaps we could attribute my incomprehensible arguments to me thinking quickly because I’m at work and should be working… 😛

          3. Sebastian Pojman says:

            PS: If you ever do go on english.stackexchange.com, avoid senior user Edwin Ashworth. This dumbass… Never mind. They’re just plain thick.

          4. With minimal effort I found why you would say that (https://chat.stackexchange.com/rooms/63023/discussion-on-question-by-joshua-into-vs-in-to-question). Looking at the thread that chat came from, I see several other people agree that “dive into” is valid in this context; in light of this, I’m going to partially return to my previous opinion and say that both are valid, though I will probably always prefer “dive into” (since I’m still not convinced that “dive into” wouldn’t mean the same as “dive in to”). All the sources you’ve referenced so far don’t prove you’re correct, only that you might be correct. It’s amazing how fast my opinion of someone can drop when I see how poorly they argue…

          5. Sebastian Pojman says:

            You think I argue poorly? Oh yeah, sure I do. You have provided ZIPPO reasons why it could be one word.

          6. And you’ve provided ZIPPO reasons why it can’t be. That’s not to say that proves anything, but there’s no infinitive in the sentence that the “to” could belong to so it entirely comes down to one’s own definition of “dive into”. It believe you’re saying it exclusively means to physically leap head-first into something (water, I hope), I’m saying it doesn’t. I’m also saying that “dive into”, when used figuratively, is a synonym of “dive in to”. If you can show me a reputable dictionary that shows that “dive into” exclusively means to physically dive, I will be more inclined to concede. Though you might have trouble with that because it seems unlikely that a dictionary would say what a word _doesn’t_ mean…

          7. Sebastian Pojman says:

            …And I did just that. The American Heritage Dictionary, an extremely valuable source, gives definitions of basically only plunging. Nothing even remotely close to saying the verb can have the same meaning as the phrasal verb “dive in.”

            1. “That’s not to say that proves anything, but there’s no infinitive in the sentence that the ‘to’ could belong to so it entirely comes down to one’s own definition of ‘dive into.'” Where the hell did you hear that? What kind of logic supports that?

            2. Why on earth makes you say that I argue poorly? Seriously, Level 11/10 Hypocrisy. I actually have a grammarian, a dictionary you requested, and just plain common sense. You only >think< it could be right as one word. Absolutely no reasons.

          8. Heck yeah I’m saying more! Bring it on!

            Wut, you voted up your own comments? LOL. What cheek you have to stoop to such a cheap tactic 😉

            Look I no longer care. I’m not mixing anything, it’s a verb (dive) and a preposition (into). Together they form “dive into”.

            1. I didn’t “hear” it anywhere, it’s just logic. “Dive into” is fine on its own. You’re saying that it’s incorrect in this context based on the fact that it means to physically dive. I’m disagreeing, saying that it can also mean the same as your definition of “dive in to”.

            2. “Why on earth”? Top quality English. Joking of course, I realise it’s typo. Although “have a grammarian” seems a little difficult to accidentally type. My point is that you have yet to convince me there’s any logical reason “dive into” couldn’t be used that way.

            But as I said I no longer care. Feel free to believe whatever you want. You still haven’t directly refuted my arguments so I win 😉

          9. Sebastian Pojman says:

            1. Thanks for the compliment! Plenty more “cheek” where that came from. 🙂

            2. Damn right, that’s good English (at least compared to yours). Your number one has three mistakes, number two has another three mistakes, and your last paragraph has *another* three mistakes. At least you were consistent. 😛

            3. In response to the last sentence of number two: literally 95 percent of my comments. Good try, though, good try.

            4. You win? LOLOLOLOLOLOL! I’m assuming you’re joking again because you have yet to provide me a single reason why “dive into” can be used like that. Okay, so you think “dive into” is just as correct as “dive in to.” Why? What’s the reason? I’m dying to find out. Because of these lame-ass responses of yours that you call your “reasoning,” I have won, and let me tell you something: you made it easy.

          10. 1. I certainly hope so! 😛
            2. “Why on earth makes you say[..]” is not good English, so I hope you’re referring to your “have a grammarian” phrase (which, if you insist is correct, must mean you have someone with you that is a grammarian – entirely possible, just seemed unlikely to me). Also, what mistakes did I make? I see no mistakes, unless you’re going to go super-strict and count commas (which are, by and large, personal preference) and use of conjunctions after a full stop (which is not incorrect).
            3. What about 95% of your comments? So what? I said you had yet to convince me. Clearly 95% of your comments didn’t convince me, so bringing that up is pointless. My point was that you’d have to make some better arguments if you wanted to convince me.
            4. You win? LOLOLOLOLOLOL! See the thing is, you called out the article for (allegedly) containing a specific grammatical error, which means you need to be able to defend that claim. I contested your claim, but you haven’t actually offered any external sources or references that back it up. Neither have I, of course, but I personally feel like you’re more on the offensive and should therefore back up your claim. Or something. Crap, I have to leave work and go home. I might be able to word this particular point better later, but for now, please accept this badly-worded, badly-arranged collection of random statements.
            5. (I’ll just invent another number here) You made some arguments, I made some arguments; neither of us has any better arguments to make and neither of us is going to concede that the other’s arguments are anything but hollow rubbish, so I’m not going to continue arguing the validity of “dive into” in the context of the above article.

          11. Sebastian Pojman says:

            I woke up with a flu, so I’m feeling exhausted. With that being said, we’re no longer adding anything useful to prove our points. Honestly, and I’m not saying it to be snarky, I think I’ve won. My American Heritage Dictionary lists “dive in” as a phrasal verb and gives definitions of basically only plunging for the verb “dive.” Using the phrasal verb and Grammar Girl’s logic, I suggested that it be spelled as two words. But as we both said, this is going absoluely nowhere. I’m exhausted and tired of arguing over this (I’m going to take the blame, though, for this argument’s getting heated).

          12. Sebastian Pojman says:

            Btw, most of the mistakes were missing commas and periods, but at one point you said, “. . . It’s typo.”

          13. Ron Warner says:

            *Damned right.

          14. Sebastian Pojman says:

            No reasons why it can’t be, huh? How’s this: WHY ON GOD’S GREEN EARTH WOULD YOU THINK IT’S ALL RIGHT TO MIX AN ADVERB WITH A PREPOSITION?

          15. Sebastian Pojman says:

            Were you going to say something? Because I’m totally fine with tackling some more of your bullcrap.

          16. Oh my god shut up nobody cares. This is an article about Flash, not grammar minutiae.

          17. Tyler Hibbard says:

            It’s actually “Rachael dove into the water” because, you know, it’s 2017 and all.

          18. Sebastian Pojman says:

            Grammarbook.com’s sentence is irrelevant because it uses the verb “dive” with an infinitive, obviously forcing it to become two words.

        2. http://dictionary.cambridge.org/dictionary/english/dive-in-dive-into-sth
          The author is using the phrasal verb “dive into something”. It would be split in the case of “dive in to do something”

      1. Sebastian Pojman says:


      2. collaborate and listen?

  11. Christopher Bottoms says:

    How about actual number of questions instead of just percentages? You can grow in raw numbers of users and questions asked but still shrink in market share. Suppose in 2000 there are only two languages: X used by 10 developers and Y with 10 developers. Then in 2010, there are three languages, X with 20 developers, Y with 20 developers, and a newer language Z with 60 developers. If you look at percentages, then X and Y have shrunk; however, if you look at raw numbers, they have both grown.

    1. Tiago Rodrigues says:

      Those numbers are percentage changes in actual number of questions from year to year, not percentage “market share” of all questions in SO in a given year. So the scenario you postulate is already accounted for, just normalised so we don’t have to do that ourselves.

      1. I agree with Christopher: Y-Axis label says “% of Stack Overflow questions asked”.

    2. Fiddling with the query, I could find that asp.net and asp.net-mvc are pretty stable, with 33k~34k (2015-2017) and 23k~25k (2013-2017), respectively.
      The 2017 values where extrapolated (dividing by 7 and multiplying by 12).

  12. Brendon Williams says:

    I’m kind of surprised it was even at 0.8%. I don’t recall ever having seen a flash question, although that could be subject to some bias.

  13. Some of these aren’t technologies, e.g. file-io, design, image, process, menu, localization, security; and they are certainly not going away either.

  14. I never did get to work that much with Silverlight, but I loved it. It was light weight and I found the basics really easy to learn and get working. This was about 5 or 6 years ago. My experience with Flex back then was not the same at all and I quickly stopped with Flex when simple apps with a Login screen and gridview would end up with a 5mb download. Silverlight was a few kb for the same functionality and would still be under 1mb if you included some of the control libraries that were around at the time.

    I’ve only recently started working with XAML again for a wpf project and I’m loving it. I’m no expert and I’m sure others will comment with some more info, but I always felt that Silverlight had great potential. People were doing 3d stuff back then already. If Microsoft kept going with it, I think we could have seen some very interesting stuff happen with Silverlight.

  15. 20 years after Flash and I still see no script that can flip a bugger objectively 🙂 Nevertheless, one can’t do much on the browser if you have to constantly haul a compiler. There are a bunch of JS frameworks out there that make stuff easier, and even with that, one is allowed to do an amazing amount of smash up if you feel like. Couldn’t Adobe just open-source Flash or allow JavaScript to crack into its ActionScript if they both use same Language Specification /ECMAScript?

    1. Is… is this English?

      1. It’s American. It’s similar to English, but not quite.

        1. Ron Warner says:

          I have no idea what “flip a bugger” means, but it sounds like something a Brit might say! I feel similarly about “an amazing amount of smash up.” Wha?

          1. Pardon it folks! I found out that’s a displeasing word in Brit. I meant a burger.
            Illustrating the ease of animating and rendering of 2D/3D objects in Flash.

          2. Pardon it folks! I found out that’s a displeasing word in Brit. I meant a burger.
            Illustrating the ease of animating and rendering of 2D / 3D objects in Flash.

          3. I suspect it’s Nadsat.

      2. It’s neither Polish 🙂

  16. If jquery is declining in popularity, what do you guys recommend as an alternative?

    1. Browser native functions, ES6 notation or if appropriate, move to a platform like React or Vue.

      1. Mathijs Segers says:

        This exactly, jQuery originally existed because of different ways to tackle JS in separate browsers. Since JS is getting more and more standardized and ES6 being vastly more developer friendly native JS is the thing now imho.

        1. Until you have to consider supporting IE then I dunno. You can’t even use arrow functions in IE which is crazy to me and Edge (which I think supports it) isn’t really prime time yet.

          1. For now you will still have to parse your ES6 code through Babel to convert it to ES5. Still, its ease of use in development and readability make it worth it.

          2. Mathijs Segers says:

            And transpiration is something you want anyway, like ugly and minify

          3. Mathijs Segers says:

            For the record though, we have a huge js codebase where I work, most issues occur on firefox not IE. IE just has more CSS issues in their older versions.

    2. Έρικ Κωνσταντόπουλος says:

      Dunno, I don’t think jQuery is ever going to decline…especially since it’s the solution to everything. 😉

    3. Matthew Henry says:

      jQuery Mobile (http://jquerymobile.com/) isn’t the same as jQuery

    4. jQuery is declining somewhat, but only a little. It’s still good for most ‘simpler’ projects, but if you’re going to build a web application with lots of user interaction things like Angular and React might be worth considering.

    5. Tyler Hibbard says:

      I recommend you use jQuery.

    6. vanilla.js. It’s the best, smallest, most efficient framework available.

    7. Bacon Brad says:

      jQuery wasn’t listed as declining. Only jQuery Plugins and jQuery Mobile. jQuery Mobile is a UI lib which is now pretty dated especially considering their are not a lot of UI libs that focus on desktop and mobile, have more features in a smaller footprint, and visually look better.

      jQuery could eventually decline in use with the addition of more competition but considering how easy it is to use and how heavily it is used it is still a safe bet for a long time. If it works for you stick with it.

    8. Michael Stilson says:

      if you’re target browsers are all “modern”, maybe some regular ol’ ECMAScript will do the job; faster: http://youmightnotneedjquery.com/

      i believe though, one should always use the best tool for the job. lib fans are always going to hype their pick, like Flavor Flav to Chuck D and a wall of S1W’s with arms crossed.

      1. Matthew Henry says:

        Reading through http://youmightnotneedjquery.com/ just convinced me that I still need jQuery. 🙂

  17. An important wrong assumption in this article is that stack overflow questions would be tagged as “flash”, while most Flash developers would be talking about AS3 or actionscript-3. In fact there are more questions tagged actionscript-3 than flash. (The conclusion would probably still be the same though)

    1. Yeah, [actionscript-3] is declining at almost exactly the same rate as [flash]. There are some AS3 stragglers working on AIR which is still alive and well — for certain definitions of “alive and well”, anyway 🙂

  18. > Surely if it were possible to run out of questions about Javascript, we would have done so already.

    gosh, it really escaped your notice that we did in fact run out quite a while ago, so that most of our time now is spent dup-hammering and close-voting ridiculous questions? The only exception is questions about the latest feature of release of ES.

  19. It’s spelled JavaScript, with an uppercase S.

    1. In browsers it would still work with lowercase.
      Not negating that spelling though. 🙂

    2. Félix Gagnon-Grenier says:


    3. It’s pretty common to spell it both ways, and I would say Javascript it acceptable as well.

  20. André Terra says:

    Where can we look up info on tags that are growing rather than shrinking? That’s an equally good indicator of the health of these technologies. I’m really curious to know how django is fairing, for example.

  21. Tyler Hibbard says:

    So, for all these graphs that show a downward trend as a % of total SO questions asked, but have those graphs been adjusted for the steady increase in popularity of SO as a whole? I’d like to see the graphs next to a graph showing the % change in total volume of questions, period, for each tag/language, rather than as a % of the total. Because there could be more questions asked in jquery-mobile this year than last year, but if there were more questions asked about non-jquery-mobile this year than last year, too, then a “% of total” graph would make it seem like jquery-mobile is decreasing in popularity, rather than the actuality, which is that it is *increasing* in popularity.

    1. forresthopkinsa says:

      I dunno, I think relative popularity on a site as huge as SO is a lot more useful than absolute popularity. I think it’s likely that most tags have had increasing absolute popularity for a long time; it just doesn’t seem like a relevant metric.

  22. Why is PHP never included in these comparisons?

    1. forresthopkinsa says:

      Well, I don’t see it in the scatter plot. Maybe it’s not decreasing? That would be interesting.

      1. You can see for yourself: https://insights.stackoverflow.com/trends?tags=php. It seems it might be declining, but only since 2015 or 2016 and it didn’t lose enough yet to be included here.

        1. cool, PHP is going strong! 🙂

    2. Bacon Brad says:

      Love it or hate it PHP is used very heavily due to it’s ease of use and being to language of choice by the world’s most used CMSs, forums, and other projects. PHP also doesn’t face the same risks as Flash did. Nor is the PHP Foundation operated in the same manner as Adobe. So it may decline in use as people switch to Node.js and other technologies but it will remain relevant.

    3. PHP is too busy determining why the hell (‘1 monkey’ + ‘2 oranges’ == 3) returns true.

    4. Well, because it’s growing…

      PHP was never dying, it just matured, and the seniors didn’t move to new jobs as often, while the companies working with php tried to retain as much of their devs as possible, so as to not lose the minds they trained. Asking questions to other people from the company reduced the number of questions, as well as not having many big frameworks for the last few years before Javascript’s (recent) boom.

      And even if VR and AR will kill most of the other languages, PHP will survive, eventually replacing html&css or using newer versions of those.

      What I think php needs now is easier default_on access to multitasking, a better compiler/interpreter, a few frameworks for AI work, and implementations thereof.

  23. Ron Warner says:

    I always toyed around with the idea of learning Flash, not to do any web work, but merely to produce cartoons/animations. My understanding is that Flash could be used in this way without any intention to publish to the web, as the resulting animations could be exported to a digital video format and displayed in any way that digital video can be. I’m wondering if anyone can tell me what the new defacto technology will be for simplifying the making of animations? Is there a void left for those who want to learn to draw, tween and animate without one of the powerhouse Hollywood desktop platforms? What would, say, the Homestar Runner guys have used, had they come of age in the post-Flash world?

    1. Alain Bellemare says:

      I’m not an animator, though my animator friends tell me that 2D animation has Toon Boom as a standardized tool. Also, I believe “Adobe Animate CC” is sticking around (I haven’t looked too deeply into this myself)

      1. yeah, I second this. I used Adobe Animate CC lately and this is basically Flash but it outputs html5 content. Also the concept there is the same like keying and scripting. I can’t tell if Animate uses JS or some kind of action script like in flash.

    2. Oliver Bisztyga (UnholySheep) says:

      Flash (the Actionscript framework + Runtime) and the animation tool (which has been renamed to Animate) are two (mostly) separate things (the naming did cause a lot of confusion though, which is why it was renamed). You can/could easily create animations without ever touching ActionScript or the Flash runtime – and you still can, because Adobe is only stopping update for the runtime/framework, not the (animation) software. And of course the other big name product for 2D animation is Toon Boom Harmony

    3. Herbert123 says:

      AnimateCC (formerly known as “Flash”) is sticking around, and Adobe has been improving the drawing tools. The frame-by-frame animation are quite good, although it falls flat on its face for the unpredictable and unusable cut-out animation bone (IK) tools, and the graph editor is a nightmare to use.

      Toonboom Premium is more or less the standard in Western animation studios for 2d work, although it is very pricey. Forget about the ‘Essentials’ edition: you are far better off with other alternatives (free/open source ones included).

      If you are interested in acquainting yourself with 2d animation, consider OpenToonz – it is production-proven open source 2d animation software, which is on a level somewhere between Toonboom Advanced and Premium. The frame-by-frame and inbetweening tools are arguably a bit better. Get the latest builds from the OpenToonz Github. The first Japanese feature-length animation made with this version was released a couple of weeks ago (Mary and the Witch’s Flower).
      A commercial supported version Toonz is also avalailable for a couple of hundred $.

      Krita is also quite nice, and includes frame-by-frame animation tools now. Open source.

      For skeleton based cut-out animation Moho (formerly known as Anime Studio) works extremely well, and tops Toonboom Premium for its IK bone tools.

      TVPaint is another alternative, although fully frame-by-frame painting.

      All these tools are being used in actual production for TV and film productions, often in combination. For example, Futurama is produced with Toonz and parts with Moho and 3d applications.

      3d applications are also used for 2d animation – flat cut-out character animation works well in those too. Try Blender and the cutout animation tools addon – open source and free again.

      As for Flash/AnimateCC: it is still in use in studios – just not as popular anymore as it used to be years and years ago. Adobe dropped the ball on that, and it is no longer the top dog.

      1. This is a very informative and thoughtful reply. Thanks!

  24. I thought you meant Flash “storage” and was a little shocked.

    1. Ha, I didn’t see it that way. That would indeed be a shock 😛

    2. Bob Jarvis says:

      Yes – this cross-contamination of tech names, abbreviations, and acronyms can be very confusing.

      Mr. President! We must move at once to close the ACRONYM gap!!!!

      1. Joshua Koudys says:

        I thought they meant people weren’t using their camera flash anymore, instead opting for longer shutter speeds.

        1. Bob Jarvis says:

          Or, y’know, it could be some guy in a raincoat…

  25. Is there is an animation version of the top graph?

    Did anything happen with the 2D graphics extensions to the C++ standard library?


    1. Olzhas Zhumabek says:

      There was a talk about it on ACCU 2017, so probably it is alive. Though waiting for it to come in C++20 would be optimistic in my opinion.

  26. Félix Gagnon-Grenier says:

    From the title, it was impossible for me to surmise wether the “next” was about the next technology to die, or the one that would replace it 😉

    1. Joshua Koudys says:

      That’s why we have other languages to program in, instead of simply using English. Too ambiguous!

    1. At the top spot for language usage.

      1. Tycho Grouwstra says:

        That only tells its x coordinate in the graph

  27. Michael Stilson says:

    what’s Flash? 😛 seriously, though … just recently died? 😉 i’m sure “scholars” will argue it was in 2015 when Youtube dropped it as it’s default player; however, imho, Flash has been dead since 2010. heh, that’s like 49 or more “internet years” ago, correct?

  28. David King says:

    Python seems to quite popular and steadily climbing. You go Monty!

  29. Is questions asked really a good metric? StackOverflow actively discourages asking duplicate questions, and all the easy newbie questions haven’t changed in a while, it seems inevitable that certain technologies appear to decline. I feel like traffic to questions, searches or other metrics would be a better indicator.

    1. David Robinson says:

      The second section, “Do question visits tell a different story?”, addresses this directly. All of the technologies listed are shrinking in visits as well, not just new questions.

  30. I’m be interested to see the rise of swift against the fall of objective C.

  31. Javascript ! Okay, just kidding…

  32. quantomworks [TeEm] says:

    One thing you’re not taking account for is how AIR is linked to Flex. So to say Flex is declining based on that isn’t accurate. We tend to use the actionscript-3 tag as well which is tied to both Flash, Flex, and AIR questions at times. Adobe has said they are continuing the use of AIR for mobile and desktop applications.

  33. Isn’t the decline of asp-net caused by the prevalence of asp-net.mvc to web development in the .NET environment?

    1. David Robinson says:

      It’s possible, but I’m not sure it fits the data. ASP.NET-MVC has been pretty steady at 1% of questions for a while; usually when one technology replaces another we see it growing alongside the other shrinking (like Swift vs Objective-C). https://insights.stackoverflow.com/trends?tags=asp.net%2Casp.net-mvc

      1. Could the decline be caused by the fact that eventually users run out of ‘new’ questions, i.e. many questions have already been asked so less would need to be asked.

        1. David Robinson says:

          This is discussed in the article; both in the conclusion and the section where we examine question views. The visits to existing ASP.NET questions have also been declining as a share of Stack Overflow traffic.

  34. quantomworks [TeEm] says:

    This article does not cover the fact that developers on the flash platform often use the actionscript 3 tag. This article also doesn’t go over the fact that AIR/Flex/ Flash share a relationship between the language that they use. Small communities such as StackOverflow are also being outperformed by independent framework communities such as StarlingForums when it comes daily engagement. Adobe has said the plugin is dead. Not the Flash Player (SWC) itself. AIR is also a powerful platform when it comes to compiling native code for Desktop and Mobile applications using its JIT compiler. On that note, I don’t understand why the community wants slow executable code running in your browser that tends to freeze, fail loading with other glitches and malformations in the first place. Though I am tired of misrepresentation when it comes to this platform.

      1. quantomworks [TeEm] says:

        Great. So then take the ones that only use one tag and add them together to get the total to be less baised. You could also make the argument that stackoverflow’s community population grew over time thus affecting all language’s popularity in the site if you go by percentage. That and again, this is only a representation of Stackoverflow.

        1. It is to show a pattern, both trends on that graph show the same pattern (decreasing interest in flash and actionscript). Also what about posts which use both Flash and Actionscript tags?

        2. Joshua Koudys says:

          This article’s about the percent decrease, and they show an identical trend. If actionscript-3 remained steady/rose while flash dropped, that’s one thing, but this only supports the article’s conclusion more. Why not just say thanks for gathering useful data and teaching us something?

          1. quantomworks [TeEm] says:

            Let me explain this to you slowly so you understand it. The stackoverflow community has grown since 2009 and surely has developed more active users. By that sense, other popular languages are sure to dominate therefore decreasing the % of questions. You could argue that, if the popularity of the other languages scaled enough and flash & actionscript stayed the same in number of questions asked a day, it would show a drop when compared to overall questions on stackoverflow. The authors argument is flash’s popularity as a whole in the developer community. I could contrast this chart with a flash creator community such as Newgrounds since 2009 and show you an increase in popularity. You’re not remembering what they taught you in school when it comes to graphs are you? Having browsers bend to end-users idiot will and block flash does NOT equal a decrease development from us.

          2. quantomworks [TeEm] says:

            Re-read the title. The article is about flash being dead and the author uses a % of users on stack overflow to present it. It’s not taking account for population increase since 2009 to 2017 as well as the fluctuation in popularity of other languages. Other communities as well. I could put a chart up that shows an increase in the number of content creators from other sites such as Newgrounds and it would counter this easily.

          3. Never thought I’d be a Newgrounds statistic in my life but that sounds awesome and I’m all for it.

  35. For many years I worked as a Flash Developer (not normally something I admit on forums because of all the hate and abuse I’d get 😉 ) but I now happily work with Javascript amongst other technologies.

    Having experienced the decline and fall of Flash from the inside I’d have to say that I’d be surprised if any other major technology is facing a similar crisis situation because not many events have been anywhere near as disruptive as the arrival of the iPhone and the truly mobile web that followed.

    1. quantomworks [TeEm] says:

      Show some pride. It’s not a bad technology…and never heard to raspberry pi forums asking anything flash related. Even the mods there will mess with you.

    2. Joshua Koudys says:

      I wouldn’t say it’s even as big a transition as many assume it to be. ActionScript’s a superset of ecma, and a lot of what it offered either came later (e.g. class syntax), or people transpile today to get (e.g. strict-typing, like TypeScript). Flash devs were really ahead of the curve in a lot of ways. The only big thing to change is now your ecma is manipulating the DOM/a canvas instead.

      1. Totally agree about ECMA script and Typescript with ES6 and 7 really aping the progression between AS1, 2 and 3.

  36. There are three kinds of lies: lies, damned lies, and statistics

  37. Etor Madiv says:

    It may also mean that the stackoverflow community itself is shrinking.


      The y axis is % of questions for the x axis, which would normalize against the community size.

  38. Jonathan White says:

    Does this take into account the fact that the number of new questions on a particular topic may be declining because they have already been ask? E.g. Java is not showing as popular in new questions but this may be because pretty much every Java question that can be asked has already been answered on here.

    1. Michael De Keyser says:

      Read the article.

  39. Reason javascript has increasing questions is because there are new frameworks created in JS everyday. So there will be always something new to ask. But other languages do not have so many variety of frameworks.

  40. What’s next ? Java, hopefully. Yes it may be heavily implemented, ingrained in curriculum, vigorously defended, and exploited statistically, but denial’s of reality only last so long before an eventual catch up. The timing really depends on the foresight of developers choosing to adopt better, more future-proof technologies, and leading enterprises who’s practices are copied by young’n’fearfuls, making the move. Hypotheticals .. what if Apple supported Flash on the iPhone all those years ago ?? What a disaster. What if Google abandoned Java today ?, further, what if everyone that was once forced to learn Java was forced to learn an alternate emerging technology today .. I wonder how steep the trend graph would start to look

    1. Karol Depka says:

      How about Kotlin to replace Java?

    2. Michael Occhipinti says:

      I don’t see Java going anywhere anytime soon. As much as some might like to think, its not all about Apple. For backend engineering, I don’t see how Java is anything but THE choice.

      1. At the risk of starting some kind of fanboy flame-fest (that’s not what I want, at all), are there reasons people choose Java over C# other than “I already know Java” and “This is not going to be implemented on a Windows system?” The second reason is already eroding away, and I’ve never seen any evidence that C# is less powerful than Java, yet it seems easier to use, more feature-rich and has better tooling available. To me, C# has always seemed like Java++, and with the advent of .Net Core and Microsoft’s efforts to go opensource and cross-platform with .Net, I am unaware of any reason, besides familiarity, why one would choose Java for a greenfield system over C#. I’d love to hear any such reasons, just for educational purposes. I agree, though, that it doesn’t seem to be going anywhere, anytime soon. And for what it’s worth, I’d probably change industries before accepting a job where I have to work exclusively with weakly typed languages. TypeScript has been a wonder, in that sense, but still seems to have a long way to go to be anywhere near as easy to use as, say, Java or C#.

        1. HonoredMule says:

          There’s certainly still a very healthy faction of people whose choices are guided by memories of being burned by Microsoft at every turn, and protectionist/anticompetitive practices in general.

          Using technologies owned or even just primarily guided by a single commercial entity never feels very safe. It usually takes a very compelling reason to overcome that risk, which is typically lack of an alternative that doesn’t exclude a massive segment of the potential market.

          1. Interesting. I don’t have those memories. Regardless, is it not the case that Sun/Oracle were/are to Java what Microsoft is to C#? Also, to perhaps make somewhat of a point to the contrary, I’ve sometimes felt having a big player behind a technology to be helpful. For instance, as a fledgling developer, when I switched from using PHP/MySQL to using the various flavors of ASP with SQL Server, it felt like I went from playing with a squirt gun to driving a tank.

  41. For the question visits, why not look at raw number of visits rather than % of SO questions? The overall makeup (both number of tags and types of users) of SO has certainly changed significantly over that time also, so the interpretation of the percentage over time must also change. Looking only at percentage implies that developers can only handle so many questions, as though there is some sort of threshold… and if users as a whole ask more question about other topics, then they don’t have time or care to ask about existing topics. That’s doesn’t seem right. If raw visits were inspected for each, I suspect we’d see a different picture, maybe even an increase in visits for some tags shown as decreasing in the last few graphs.

    1. Sankalp Sharma says:

      That would not account for reducing Stackoverflow traffic in this period, if any (although most probably it increased instead of decreasing). % of hits would IMO normalize for overall traffic fluctuations.

      1. Michael Gorman says:

        not if the scope of stack overflow questions broadens, every new tag added is less percentage to go around even if absolute traffic to a specific tag goes up. if the overall traffic is going up faster than specific traffic, the % would trend downward on a percentage scale leading to a false claim of the technology losing traffic.

        1. HonoredMule says:

          Such effects would at least universally affect all tags/technologies. The metrics are only comparatively meaningful anyway.

  42. asp.net is indeed obsolete. A more interesting tag would be asp.net-mvc which indeed does not decline.

    1. Mark Carpenter Jr says:

      It’s like code-igniter, just won’t quit.

      1. BearerOfHarshWords says:

        Codeigniter is being maintained by a new owner and is regularly updated. It’s not as popular as it once was when EllisLab owned it, but still loved and used by many.

        1. I though it was a bug in Disqus, as I was listening to Thrice’s Red Sky on Deezer and then I saw your avatar. =D

  43. Doni Trums says:

    Do you still alive by 2020? Nuclear will kill you and the earth will become Like in the movie MadMax Fury Road!

  44. Mark Carpenter Jr says:

    “but it also may reflect a diminishing relevance of Microsoft in the web developer space” –
    I think this is true and I think Microsoft has come to realize this, if you look at Microsoft’s presence in the open source community along with it’s growing presence on Github.

  45. Zack Macomber says:

    These trends are interesting but not good indicators of the future. So many factors contribute to the rise and fall of things and programming trends can rapidly change. Who would have thought JavaScript would be what it is today when eich created it? Anybody in their right mind back then would not have predicted it to become what it is today. How about node as well for that matter?

    1. Jeff Mcneill says:

      Trends are indicators of the near future. That is by definition. Sure, technologies can be reborn or revive for some reason or another, but a trend is a trend.

  46. Dhe Constantine says:

    You don’t need flash anymore to run your web, because nearly future technology is hologram and virtual world where your website will run in each human brain

  47. An obvious thing not covered by this, for example is that you have used one tag to represent an entire technology and ignored all offshoots. For example you say the “asp.net” tag has declined but what about the “asp.net mvc” tag and other “asp.net” related tags. I’d say you have to collate all tags that relate to a single tech like in my example to gauge any true decline.

    1. Andre Platov says:

      In addition to Peter’s argument, I would add two more. One, Microsoft’s official documentation is improving over years, and thus there is less need to google questions around. Two, when you google for specific issues, there are now many more quality articles written by individual developers, which also drops the traffic for stack overflow.

      1. Jeff Mcneill says:

        Two things: stack overflow traffic has increased, and we are dealing with big number tags. Sure not all tags are the same or perfect but they provide a useful approximation. Sure official documentation might be getting better, but stack overflow is used more as a resource therefore relative changes in use are a useful approximation to popularity. Its the data we have. And it does say something useful.

  48. SVG, HTML5? I’ve written about the impending demise of Flash back in 2006 so no surprise here.

  49. Michael Kay says:

    Interesting data, but I think there’s another factor that ought to be considered, which is that tag usage isn’t the same as technology usage. For example, we generally pounce on people who use the “xml” tag for questions that are not really about XML per se, but in the case of Android developers the battle appears lost. Perhaps one day, though, Android developers will stop mis-tagging their questions as “xml” questions, and then suddenly XML will appear to be in decline.

  50. Is it possible that technologies that don’t evolve into vendor independent standards all have an expiration date? PDF has gone one way, it will stay. Flash the other way, it is long gone.

  51. It is interesting to see that there was a decline in searching for design patterns. Either developers are getting better to the point that they won’t ask for help regarding that subject, or they just stopped caring (or never cared).

    1. HonoredMule says:

      I’ve found that the adoption of functional programming approaches has replaced a lot of design pattern usage in my own work, particularly the ones focused on enabling testability, loose coupling, and program composition (i.e. initializing/assembling components for specific uses/workflows). It seems now that many of them were OOP workarounds to the shortcomings/complexities of OOP itself (or more specifically the tighter coupling OOP approaches tend to implicitly encourage, and the application of its own mitigating constructs such as interfaces).

      I don’t mean to suggest that I no longer use design patterns. In fact the factory pattern is practically standard in many situations (especially application configuration/composition) and enjoys more widespread application than ever before. Adapter layers, decorators, facades, etc. are also seeing increased usage – but they’re barely distinguishable as special cases of `function` and not at all distinguishable from each other. I may use fewer distinct patterns, but the main change is that it no longer matters how I use them or what they are.

      What they are no longer is a source of integration impedance. Instead, they are more frequently able to be encapsulated (i.e. implemented by 3rd-party libraries without dictating 1st-party architecture). They’re often reduced to coding style choices where whatever I like generally just works and whatever someone else likes happily coexists without the combination producing a schizophrenic mess. And they’re mostly present/recognizable in thin OOP layers built around the increasing volumes of underlying FP code.

    2. Mikey Awbrey says:

      From what I can tell looking at the data, it appears that the design-patterns tag declined, but more specific tags regarding design patterns rose, such as tags including MVVM or MVC. If I were to draw a hypothesis to test, it would be that the searches and questions have become more specific regarding design patterns over the years, resulting in a decline of the more generic “design-patterns” tag.

  52. Jeffry Houser says:

    Disclaimer: I’m a member of the Apache Flex PMC and have answered the most Flex questions on Stack Overflow.

    The Apache Flex team has done a lot to port the framework to HTML5 with their FlexJS initiative. It’s good work. My personal impression is that the market appears to have filled that gap with other technologies, such as React, Angular, and Vue.

    Since Adobe is not killing Adobe AIR, Flex could potentially live on as a desktop or mobile application framework. My personal opinion is that Flex will become more niche than it already has.

    So long, and thanks for all the fish! I had a fun ride.

  53. Those graphs prove absolutely nothing, as they are not about the _amount_ of questions asked, but show the _percentage of questions_ asked about a certain topic on SO. Since the site has gotten more users and more topics, since new technologies have been invented. Most notably the explosion of smart phone use. Technologies that had not yet been invented were for obvious reasons not very popular in the early days of SO. Just because new users ask questions about those new technologies, you can’t run off and say that for example x86 assembler language has become less popular, because it is not related to smart phone technologies what-so-ever. But since the graph shows a percentage of the total, pretty much every single technology seems to be in decline if you just look at the graph without understanding what it means. I already pointed out this flaw in reasoning when SO Trends were introduced. In order to get meaningful statistics, you have to look at the _amount_ of questions asked and nothing else.

    1. Yeah, this is a good point that seems to invalidate the usefulness of the graphs above, and wasn’t apparent during my read-through.

  54. Thomas Williams says:

    Surely the lack of questions does not prove that it isn’t being used, but proves that it is so easy to use people don’t have to ask questions. For instance EVERYBODY uses a steering wheel in a car, but it is so simple to use that we don’t need to ask a question on how it operates. If we use your analogy nobody ever uses a steering wheel in a car, because nobody has asked a question about it.

    1. Milan Brezovsky says:

      Are you suggesting that Flash development has been growing steadily more easy since 2010?

      1. I think he simply meant that the absence of evidence is not the same as evidence of absence.

        1. Milan Brezovsky says:

          I don’t think he did, otherwise I’d agree with him. He literally said “Lack of questions… proves that it is so easy to use” which would actually be arguing the opposite of your theorem.

          1. Thomas Williams says:

            I didn’t say lack of questions proves it is easy to use. I am just saying it doesn’t prove anything either way. A lack of questions might mean it is so easy to use that people don’t have to ask questions, or it might mean that nobody is using it. Either way it proves nothing.

          2. Milan Brezovsky says:

            I agree – correlation isn’t causation. Just in this case, ActionScript didn’t get any easier to use over time. I think they had a good point when they mentioned JavaScript – it’s not that people would run out of questions, even with arguably a lot more people knowing JS, and a lot more questions having been answered over time, there’s still an influx of questions. It doesn’t make a statement about how easy/hard/trick JS is to use, but it does make a certain statement about it _being used_.

    2. Jeff Mcneill says:

      Flash is definitely not as easy to use as a steering wheel. This is a red herring argument, waste of time.

  55. anaya tikekar says:

    you can’t run off and say that for example x86 assembler language has
    become less popular, masticoopon.com because it is not related to smart phone

  56. sharleenjoynt says:

    At whatever point we talk about print industry one name flies up in our brain and that is HP printers. HP was built up around 1980 with the best technology in print industry. Some of the products are outstanding with HP mark like LaserJet, DeskJet, Inkjet and multipurpose printer. Clients can take assistance from HP printer customer support number +1-800-787-2406

    1. Teemu Leisti says:

      I, too, am quite concerned about the disturbing lack of attention and respect shown for HP printers around these parts. Why, back in the day, many of us got our first taste of programming on HP’s fabled dot-matrix printers. Oh, I still remember the sweet crackle-crackle-screech sound of a successfully running program!

  57. Mikey Awbrey says:

    I took a preliminary stab at analyzing some of this data. For those interested, here are the top 5 technologies I found on the rise.
    1. Android
    2. Javascript
    3. IOS
    4. Python
    5. PHP

    Here is the linq I used to get that so you can check it too https://pastebin.com/eFp2CKEp

  58. Artis Zelmenis says:

    “…this may be skewed because the site started with a disproportionate number of C# developers…”

    Umm.. how about using absolute numbers, rather then relative-to-other-data percent?

    1. David Robinson says:

      If we used absolute numbers of questions, almost every technology would look like it was growing until about 2014, when many would change direction (since the number of questions per week leveled off at about that point; see image below). Furthermore, there’d be a drop each December, when many countries have a winter vacation.

      I don’t think there’s any sense in which this would give a more accurate depiction of the popularity and usage of various programming languages. If the absolute # of questions were a true representation of its popularity, and if a technology stayed at a constant popularity over time, we’d expect that one to show a pattern much like the one below. But almost no technologies do (and many that we’ve seen steady over time, like e.g. Django, don’t show any kind of sudden shift in 2014).

      Thus, it seems a more honest reflection of the data to show each as a percentage of all questions asked that month.


  59. The article didn’t really touch on it, but it seems to me that the number of new questions will also be affected by new features being released to a language/platform. For example, if JavaScript is increasing in questions asked, it likely involves es6, es7, nodejs, etc. So it’s also a partial measure of health.

    If the Flash platform has not been actively developed for a few years (I have no idea if this is the case) you might expect questions to slow down whether or not users decline. Of course, the numbers in this case make it obvious that developers have stopped using it.

  60. William Davies says:

    Did Steve Jobs kill Flash with his open letter in 2010?

  61. “We considered a technology to be shrinking if questions about it declined by at least 10% per year, on average.”

    Maybe that just means that all the low- to mid-level questions that can be effectively asked on SO have already been asked and answered. That certainly seems true of Java; these days it’s only worth asking a new question if the problem is relatively obscure. It seems to me that this would be true of any mature platform.

  62. sharleenjoynt says:

    HP printer is known for its finest technology based products and it is used worldwide. We are in the era of technology and HP printers are completely unique in features. Users may sometime face issues related to HP printer but we have a solution to resolve the issue they face through HP printer Helpline Phone Number +1-800-787-2406. Our tech support team will resolve the issues within no time.


  63. sharleenjoynt says:

    HP is worldwide used printer by millions of customers in the entire world. Its features have provided it an edge over other printers. Sometimes users face the issues but we have the solution for them. Users can avail it through HP printer Customer Service Number +1-800-787-2406 24×7 online

  64. sharleenjoynt says:

    We are in the era of technology where everyone is using modern techno gadgets. HP printer is one of the finest printers across the world and it is user-friendly printers yet some time faces the technical issues and they can be resolved through HP printer customer support number +1-800-787-2406.


  65. sharleenjoynt says:

    The name of HP is quite a big in the print industry and we are in the era where technology matters a lot. Users of HP have the opportunity to avail the features of HP printer and at the same time, they face some technical issues. We have solution to resolve their issues through HP printer customer care number +1-800-787-2406 which will be available 24×7 online.


  66. sharleenjoynt says:

    HP printer can’t be compared with any of the brands in the print industry as it is highly advanced technology-based products. Users may face some issues related to that in spite of some wonderful features it has but we have a solution for every technical issue they face. Users can contact directly to the HP Printer Customer Care number +1-800-787-2406.


  67. sharleenjoynt says:

    We are in the era where everyone needs quality products with durability and HP products are the proven products in print industry. HP printer has lots of features with some issues for the users. Users can resolve those issues directly from the HP Printer Support Services +1-800-787-2406.


  68. HP printer doesn’t require any introduction as it is an established brand worldwide. HP printer was launched with a mission to provide the best printing technology in the world and it is fulfilling the same. HP printer constantly works on advanced technology and provides the best in the industry. Customer can call 24×7 on HP Printer Installation Support +1-800-787-2406


  69. Whenever we discuss print industry one name pops up in our mind and that is HP printers. HP was established around 1980 with the best technology in print industry. Some of the products are well known with HP brand like LaserJet, DeskJet, Inkjet and multipurpose printer. Customers can take support from HP Printer Helpline Number +1-800-787-2406


  70. HP printer is a well known and an established brand in the world. HP printer is based on an advanced technology and its all products are comprised of many features and functions. Whenever any customer needs support for the technical issues while using HP printer they can have support from HP Printer Tech Support +1-800-787-2406


  71. HP printer is one of the best among all in print industry. Whenever we required a good quality printer HP is one of them. HP printer is based on the advanced technology with some unique features and functions. Customers have the facility to take the support from HP Printer Support Services +1-800-787-2406


  72. The current scenario belongs to the hi-tech world where the product which is based on the advanced technology will stand out. HP printer is one of them which is based on modern technology with some advanced features and functions. Customers have the facility to call on HP Printer Support Number +1-800-787-2406


  73. What did you use to make those graphs? They look nice!

  74. Really liked your blog. It was very informative. I din’t really liked flash. Lets see what happens next. I’ve been using yahoo and saw somewhere that Yahoo! Maps Flash APIs.is being used.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.