Too bad all those engineers are going to be farmers by 2020 though. Because when your prototype gets popular, I doubt Bubble.is can handle all the things that actually make your app run at scale (the deployments, maintenance, database, etc management programming that is really what makes apps like Twitter run .. not just the tiny html/css/js stack that the client plays with).
Today, only about 20 million people[1] are building software for over 3 billion internet connected people[2] (and even more who can get access to software in other ways). That's only about 0.6% of users contributing to the rest. Tools like bubble.is are not replacing the very scarce resource of software engineers, they're just vastly expanding the bubble (har har) of potential developers.
The world will always need people (or at least creative robotics) capable of designing and building hardware as well as the software that interfaces with that hardware. But the amount of resources spent on those critical systems will be minuscule compared to the resources spent building very custom things on top of it.
It's sort of like how you may be able to build your own computer at home with pre-made components and plug it into the wall and have it start up, without having to know how to generate the electricity required to run it, or how to make the hardware components.
In 2020 farmers may very well be building their own application, and if that is the case then most likely the number of people using these "WYSIWYG" tools will outnumber the kind of software engineers we see today. I personally think 2020 is a bit optimistic, but it's clearly the direction in which we're going.
The best example of this happening successfully that I can think of, albeit in a much narrower scope, is the recent shift in game development. Game engines used to be super expensive to license and tricky to develop for, but just in the past year several of the biggest game engines have become open for anyone to use, and have also made their development tools work really well for people who don't write code. We're not quite at the point where you can make elaborate games just by dragging and dropping, but a lot of serious effort is being put into it. And as advances are being made, we're quickly growing the potential number of developers out there.
This change doesn't necessarily mean there will be much more higher quality content on a global scale, but it will definitely increase the amount of content and shift consumption into narrower scopes by providing many specialized solution rather than one monolith solution for the problems out there.
This may very well be a cool product - I like the idea of people being able to build things without code (how do they architect a DB?).
Of course, as a lot of people are pointing out, this approach doesn't scale very well. To be fair, neither did the original Twitter, and it was built on Rails. They're different levels of scale, but if all you really need is a proof of concept this could solve the "How do I find a CTO so we can build a quick app and see if anyone cares?" problem.
I'd be careful pitching it (yet) as a "we'll never need engineers again" product, as I don't think that's the case; it is incredibly slow, and it errored out for me three times (http://take.ms/iwa9W). I still haven't been able to get in.
If the resulting sites can't handle a traffic spike, it's not that great a tool, is it? No one wants a hammer that shatters if you swing it too hard...
Being optimistic, it might just be a hosting problem.
Ya, to be fair I've seen Wordpress blogs go down under an HN spike, as well as a lot of stuff using socket connections for each user. Not many people build products expecting 100+ people to simultaneously use the site on day one.
The level of detail in this makes me doubt the claim (0 lines of code), but assuming it is true: really well done.
For those who are wondering if its just a login page: you can register an account (login is automatic on registration, and email verification is not required). The entire UI is almost exactly same as twitter.
I really like the idea of these 0 lines of code websites, but mostly for more static sites. As you can tell with this site, it's already suffering from lag in a bunch of places, which might be hard to fix. Disclaimer: I have never used bubble.is, but similar products
> if we ever for some reason have to close down shop, we will release the Bubble source code under an open-source license so that you can set up your own Bubble server and keep your app running
That's quite a nice thing to read. I hope more people start committing to this.
Ah, you should specify that you need to make an account. For some reason I was thinking this might somehow attempt to do SSO and grab your actual twitter data
It reminds me of all those people saying wysiwyg editors would kill the need to know HTML. Really, they made it possible for developers to work on more complicated and exciting projects. Uncle Phil wants help with his flower store website? He'll be happy with WordPress.com or SquareSpace.com and you can continue working on your stuff without being a bad nephew.
Haven't been able to create an account, bugs out constantly and is very slow. Cool for prototyping but honestly it just highlights why we need engineers.
Finding a way to program computers to do what we want without requiring rare and expensive skills is probably one of the largest economic opportunities of the 21st century. Bear in mind that there are already plenty of people in other disciplines who are good at decomposing problems and designing processes. It is the syntax and structure of programming and the depth of its connection to the way the machine works that bars the door to their participation. Meanwhile the world needs more and better software every day, and the overall supply of people who are good at the way we do it now isn't growing at anything like a comparable rate.
I'm sure this is not the tool, but I'm also sure that when the tool comes along that busts this open most developers (including myself, in all likelihood) will fail to recognize what it can do and will mock its intentions, just because that's the way we humans work. So, in closing, I leave you with the lines of an old Scottish folk song:
If it wasn't for the weavers, what would ye do?
Ye wouldn't have your clothes that's made of wool!
Ye wouldn't have your coat of the black or the blue
If it wasn't for the work of the weavers.
Regarding software engineering won't be done by software engineers in 5 years time:
There are engineers who seem to be focused on the mere plumbing behind applications: DBs, ORMs, web frameworks, REST, SOAP, you name it. When asked to implement a particular solution to a real world problem these developers tend to first argue about whether to use PostgreSQL, MySQL or WhateverDB without even trying to understand the actual problem first. I know this sounds very clichéd but quite a few developers seem to try very hard to live up to the cliché.
Believe it or not, in some areas this attitude is almost pervasive. I'd say it's the main reason why recruiting freelance developers mostly works by recruiters matching TLAs instead of developers pitching their services to clients directly (like design agencies do).
Anyway, this kind of developers might indeed have a hard time finding a job in 5 years but there will - save for a possible technological singularity - always be the need for people who can fathom a real problem and design a solution.
EDIT: To clarify that, I don't mean that in a nasty, snarky way. VB drove a lot of business software and got huge numbers of people writing software, many of whom weren't employed as developers.
The hard bit of software engineering is not knowing the syntax and typing the lines of code, but the design and problem solving, that abstractions can't solve.
You can make it easier to produce software with better tools, but the fundamental intellectual requirements of software design and implementation will never go away (until you have AIs genuinely powerful enough to write code themselves)
The idea of lowering the learning curve for website development has fascinated me for the past 10-ish years during my employment in the content management system industry. Every once in a while the idea that a tool can replace a developer surfaces, energizes a few people, and then falls flat for one reason or another. Yet it always resurfaces, likely due to the revolutionary nature of a successful solution.
The most common reason for failure I've seen seems to stem from the idea that a website requires a developer due to a website requiring massive amounts of meticulously written text in an esoteric language; by replacing this text with a graphical interface, a website no longer requires a developer who understands how to write this text. Experience leads me to believe a developer serves a much different purpose in building websites.
Consider applying this logic to another problem, such as writing. If one builds a tool allowing non-writers to drag correctly spelled words and punctuation onto a page, then a non-writer now has the ability to write gripping novels, impactful research papers, and influential editorials -- and they can do this without knowing how to spell. This doesn't make sense, of course, since writing a book requires a certain understanding of how to communicate effectively, an achievement demanding a much more clever tool.
Much like writing a book, writing software also demands a much more clever tool than a graphical, drag-and-drop interface. Developers understand more than how to write text in odd languages; they understand how to solve problems. Creating a graphical interface for building websites doesn't remove the necessity of a developer. On the contrary, it creates more jobs for more developers who specialize in building applications using a drag-and-drop interface.
I agree with you mostly, but let's also consider the value of lowering the bar of entry. Once upon a time "software engineer" meant someone punching cards and running them through one out of a handful machines on Earth to execute a piece of logic. In 10 years from now "software engineer" might mean someone who practically draws something on a touchable, wireless surface that weighs less than a couple of those punch cards that used to hold a single logical instruction, and their software is instantly available to billions of people.
Or to translate to your analogy: there might have been a lot of brilliant writers in the world that simply never learned how to write, or they never had access to distribution so the only person who read their masterpiece was themselves. Today, with the Internet and computers and better access to schools, we have a lot more writers. There might also be more lower quality content because the bar of entry has been lowered, but I'm absolutely sure there will be a lot more high quality content as well.
It's seriously slow and it lost my user account picture and status after I clicked away.
Maybe this could be a good thing in the future for simple prototyping (to show off to investors, not a general public kind of thing) but the bubble.is pricing advertises "Grow" "Scale" and "Flourish" plans for $460, $1700, and $4200 respectively.[1] I would definitely wait until the architecture here becomes more advanced before trusting them with your entire app.
I noticed their architecture was built on JS. JS is a pain in the butt when it comes to sandboxing. Either you run multiple contexts in a V8 engine that can crash each other if one OOMs, or you start up a new process for every instance. Single-threadedness is also not fun to deal with, in terms of efficiency.
(Yes, I've explored building a service like this on Node.js. I summarily discovered that it's probably not the best plan.)
I honestly don't think I will ever use something like this because you have to stick to the resources provided. Also they might have released it too soon. I tried signing up and this is the message I get "Sorry, we ran into a temporary bug and can't complete your request. We'll fix it as soon as we can; please try again in a bit!"
The side effect of such products (that I have to love day to day with) is that it really devaluates the work done by software developers when it needs to be done by someone of the trade. When a client comes with a feature request and I say "that'll be 10 days, sir", many scoff at the value and don't understand why, pointing at how easy and quick it is to do A or B in Z assisted development system, glossing over even the skin-deep details.
That said I am totally in favor of the "in 2020 most programmers won't be engineers, they'll be [whatever trade uses programming as a tool]". Empowering the users should be the number one goal of every system builder, whether it is via simpler systems (as seen on mobile), more accessible interfaces (from colorblind to sensory to kinetic disabilities), or powerful coding (R, NumPy, iPython, Matlab, Mathematica...) or non-coding (Hypercard...) environments.
I'll wait until the site is responding faster/not erroring to be sure but if it's really as it appears, this is amazing. Attractive customizable design is one thing, but real logins, real graph relationships, real feeds, real filtering for mentions, user photo uploads, potentially real messaging, all without any coding would be (is?) really amazing and covers 99% of all the site creations any laymen could want.
It doesn't need to scale especially well for the vast majority of customers. If your first reaction is "this doesn't scale" or "this doesn't eliminate programmers" you have to be a professional point-misser. I'd want to see how intuitive the interface for making this is but it's cool stuff, and as it matures I'm sure will only get more interesting.
Except for the fact that it's barking up the wrong tree. This likely won't change anything. It'll just make a new programming market. Now, instead of people going "I don't want to learn to code, I'll just hire a developer." they'll go "I don't want to learn to write a website, I'll just hire a developer".
If someone doesn't want to dig in to building something for themselves they won't no matter how easy you make the tools. WYSIWYG editors have been around for decades but programmers still have a job. They've gone from writing crazy cobol to doing .net but they are still doing it.
I've been using Zoho Creator for years (http://creator.zoho.com) as a tech-savvy non-developer. Anyone had any experience with both Creator and Bubble and can compare the two?
Why does Bubble.is make me confirm every single time I try to navigate to another one of its pages? Extremely annoying, and makes me think the product doesn't actually work.
Don't post on HN of all places and claim we don't need programmers...
On a serious note, cool product but we've seen many many many of these sorts of platforms before. They never work outside a niche market simply because programming isn't about typing, but assembling logic.
I think lots of people on HN are programmers, so probably a good place to post it, no? Good to get the people that will be 'replaced' in 20 years or whatever to start thinking about other ways to use their knowledge or at least thinking about how the industry could evolve in the future!
Yay for frontpage 2.0. All this seems to do is make it easier for people to not better themselves. "What's this? I can make a website without having to learn anything?! Great! Let me put a million animated gifs on here!"
If in 2020, all those people are programmers, we are going to have some shitty software! If the people I worked with wrote software, they might fix their own itch and not think about the bigger picture nor the others in the organization.
Nice! I often face the problem of wanting to demoing prototypes of ideas, and coding takes too long. Eventually bubble could add auto-scaling when the app is used by many users, and have a pay model similar to aws.
I got a good laugh out of this- definitely very creative and a good idea. I'd just be careful about the whole "please don't sue us" line. Companies like Twitter take these things very seriously.
The poor guy! If I look at the bubble pricing sheet then this month is going to be a bit costly for him (My assumption is ~100000 hits = 920 USD). Not that he would have any advertising on it ;)
Very soon you might realize, that Twitter is good not bcoz of frontend but the major tech expertise is in the backend systems to handle super high traffic and dispatching messages quickly.
>Because in 2020 most programmers won't be engineers.
They will be doctors, teachers, farmers, executives...
I do not like this narrative at all, and it's becoming pretty common. Do you imagine a doctor saying "Don't worry, in 2020 most physicians won't be doctors, they will be engineers, teachers, farmers, executives"?
* The majority of doctors are self-employed
* Farmers are, by definition, owners of the business
* Executives tend to be large stakeholders in the companies they manage
Institutional school teachers tend to be employees, although I'm not sure about all teachers. However, I did have at least one high school teacher (from years ago) tell us that in the future people would learn on demand over the internet, which is coming true, and largely being provided by people who do not have teaching backgrounds.
I expect programmers also tend towards being employees, and thus there is a strong interest to automate/eliminate that work.
I would love it more if it's doctors teachers farmers executives and engineers. Programming as a tool is apparent now. All these frameworks help people to easily realize their ideas. People can be more multi-disciplined than ever before.
Do you call a carpenter every time you want to put a nail in the wall to hang a picture? Do you call a doctor to put Neosporin and a band aid on your boo-boo? Do you call an accountant to calculate tips?
I liked their demo, but honestly we've seen this before. Even with tools like Unreal you still need to write code if you want to make genuinely good games. Same thing applies to stuff like GameMaker.
If any of the site's devs are browsing the comments, you might want to address this problem I found while navigating the site: http://imgur.com/h0iUIVw
Too bad all those engineers are going to be farmers by 2020 though. Because when your prototype gets popular, I doubt Bubble.is can handle all the things that actually make your app run at scale (the deployments, maintenance, database, etc management programming that is really what makes apps like Twitter run .. not just the tiny html/css/js stack that the client plays with).