:O

There are tools out that do this user interface building you talk of. The custom logic that pulls it together though is not simple a drag and drop affair. Software will never be so trivial to produce. I hope anyway or I will be out of a job.

Have you seen multimedia builder?

The company I worked for actually has something similar: http://www.simx.com. What they did was create a visual integration platform, but its sophistication was still fairly high - that is, for custom logic you would need to script.

I think this is the direction the entire industry is moving in.

When you make the programming software easy, there can only be small/easy programs made with (I think...)...

Its not so easy to create software. As somebody already said you can use programs to creates the visual UI. The same kind of program might make it possible to create simple interaction that takes data entered into the UI and stores the data somewhere (in a DB lets say) and retrieves it from there for display in the UI.

Anything beyond this would require you to write custom code.

It's a common misconception that the only hard thing about software development is the syntax of the programming language, and if you just got the language out of the way anyone could do a good job. This is like saying that if you just learned to turn a wrench, you could build your own car.

Fact is, even a lot of people who know how to code aren't qualified to design software. UI design is one field where the harm done by amateurs is very apparent, but you only have to skim sites like dailywtf.com to realize that even knowing how to code doesn't make you good at designing solutions and applications.

I don't mean to sound elitist. There's certainly a market for allowing non-programmers to throw together quick solutions. Products like Frontpage and the Microsoft Access forms creator are examples where a user with little or no programming knowledge can put together simple solutions. Still, this is a niche market suitable for creating limited-use customizations. You don't want to create real apps this way without considering the impact on productivity and stability.

I'm not saying there's not a good business here. Just saying you need to take a realistic look at the potential market.

I think it would be hard to get programs to program something that could put them out of work....

GL: on the contrary, I think a lot of experienced coders like to work on this kind of problem. For one thing, most good coders already automate a lot of the grunt work they do, so they see expanding on those tools to do simple things as useful work.

They also see this as a good scope-definition tool. I can't tell you how many jobs I worked on where the company originally thought they could do everything without a programmer. Whether it's building a website in Frontpage or building an enterprise-wide accounting system with Access forms and Excel spreadsheets, they try, they fail, and then they hire someone qualified. It's not that they lack adequate tools. It's that they lack the problem-solving expertise to design a proper solution. And when they hire someone who is capable, the company has at least defined the problem space in their attempts to solve it. Otherwise, they have to resort to a lot of hand-waving and whiteboarding to try to explain the problem to the hired help.

I'm actually thinking along the lines of NeoBook, only instead of implementing e-books, and instuction guides, using this same method to create the program. If this is developed it would allow people to customize their pages and so forth. The programming would already be done, and they could use the software for various resons relating to one specific catagory. For example to keep up with inventory, or finacial matters. Not that they could create multiple software programs, just one major one, that could be customized for thier needs. The risk for developers going out of business is extremely low. In general most people are lazy, and want everything done for them. Not to mention, if they had another software idea they would be required to purchase the Software Generator again, since it would only focus on one specific subject. If you look at NeoBook, it is very similar to the design I have in mind. After looking at NeoBook, what do you think?

What is Neobooks?

If you're now proposing software that is customizable, I think that is often the case. Or if you are looking for templates, they are common also.

This has been tried time and again. There must be hundreds of make your own software software packages out there and they are all largely useless.

Software, in whatever form, is never simple. Everyone has different requirements. In order to produce software that creates software, the product needs to be able to avoid:

a) over generalisation (making the product useless because it actually ends up doing nothing)
b) over specialisation (making it do one thing well but nothing else very well at all)
c) over complication (even scripting is beyong people who would be most likely to use these products).
d) over simplification (not being able to do complicated tasks easily - even the most simple applications have to do complicated things, they just make them look easy)

The trouble with fulfilling these criteria is that in order to avoid all of the above you actually end up over generalising (a) no matter how hard you try.

To add to all those negative things, I as a computer programmer myself am always aware of software that says it create me or my friends "software applications". Most of the time the product is snake oil and they might as well learn a programming language themselves.

I see where you're going with the Neobooks analogy and the specific topics. The example of inventory or financial systems is especially appropriate I think. A lot of companies start with an off-the-shelf accounting system, but then they either need to customize the software to fit their business practice or modify their business practice to match the software. I've seen cases where companies seem to have built their businesses around some quirky accounting software and when they're forced to upgrade, they can't align their software with their process. Having an accounting system where you didn't have to pay expensive consultants to add a field here or a constraint there would certainly have benefit.

However, to make that work, you first have to have the back end in place. If you're going to have a software generator for an accounting system, you first have to have a world-class accounting system to put it on top of. From a business standpoint, this would be a tough development because you'd be playing catchup with existing competitors in every category and you'd have to match them on basic features and stability before your customization options became a selling point. Neobooks works because the user-designed front end is the content. There's not a lot of back end business logic. But to put a Neobooks-type interface on an accounting system, you need an accounting system first.

You might make this work by designing plugins for existing software, but as _Matt_ points out above, these types of products tend to be either so specialized that they have very limited appeal or so generalized that they're useless.

Sounds nice i'd love to see this for mobile programming.smthing like controlled animation for making games.it will be cool.

The company that create PRO-C here in Waterloo tied to sell it as a wonder-tool that would allow totally untrained people to produce quality code. They even advertised for "software factory workers" that would crank out fine programs for customers. Alas, as mentioned by several here, there is more to programming than syntax. There is design, implementation, algorithms, real math etc. This doesn't come easily without much experience or education. Maybe in 30 years you will be able to tell an A.I. what you want done by describing it, and it will write you code to do it. But with a sentient A.I., you wouldn't need the program!

I think this is a great idea, but not original. I originally tried to implement something like this on a home computer before PCs became mainstream, and have had a few failed attempts since. I come back to it every now and then, but each time technology has moved on and the job is even bigger than it was the last time!

I don't doubt the industry will get there. With plugable components and drag-and-drop designers, the technology is there, it's just a case of being able to pitch it right and find the resources to implement it. Then find the market that can justify the investment...

Check out http://www.popfly.ms just been playiing with the alpha version, you'll like it, its pretty close to your idea.