Gigster Wants To Be the Uber of Software Development (techcrunch.com)

HughPickens.com writes: Josh Constine writes at TechCrunch that a company named Gigster is trying to bring the Uber business model to software development. Simply: a user sends them an idea, Gigster passes it on to developers who sign up to build software, and when it's done they send back a functioning app. After converting product requirements into a development plan, they let their group of remote developers start hacking away at the code. It has already resulted in a dating app for Muslim millennials, a way for citizens of the developing world to buy electricity, and has over fifty more projects in the pipeline. The entire development process goes through their app, and they charge a flat fee rather than an hourly rate. Gigster developers who satisfy customers can earn karma points and qualify for higher-paying contracts. One major caveat: Gigster will still own the code to the app it designs for you, and it "leases" the software to you. They say they want to be able to reuse certain components on other projects.

Comments

Re:Requirements

[blankinthefill]

I would argue that in the agile world, it matters just as much if not more, especially with their desire to charge a flat fee. From the article: " A sales engineer discusses proposals with clients, and using the AI engine, comes back with a price quote and production schedule in about 10 minutes. Then Gigster manages the entire development process through delivery of the fully-functional app." This implies to me that they are coming up with this fee and the full schedule at the start, from someone who isn't going to be that closely invested in the actual development of the app (and who likely is going to miss a LOT of what's going to go in that development. I know some very good sales engineers, but they are almost all behind the development curve just due to the fact that they don't DO development regularly, if at all.) That's not agile. That's the antithesis of agile. The entire point of agile is that you can't know right from the start how exactly a project is going to turn out, and what kinds of roadblocks you'll encounter. And especially with the development of original apps, unexpected events are going to crop up. The agile method is built to zero in on requirements during the process of development, and to actually allow an estimate to be just that, an estimate of time and cost. And yes yes, I know that agile can't be completely open ended, but the entire point is to be flexible in your development so you can easily adjust when problems arise, which, again, is NOT what this company seems to be doing. This company is not doing that at all, they're like the epitome of waterfall style development. Which means they have to be nearly perfect at requirements and estimating, which is nearly impossible. And I think you're right that it will almost certainly come back and bite them in the ass. (And that's ignoring the bullshit sharing economy/labor issues involved.)

I have an idea for an app....

[flarb936]

Every time someone says to me, "You make apps?? I have this idea..." I'll just refer them to this site whether it's good or not. I just need an effective way to shut down these conversations immediately.

Re:what could possibly go wrong?

[Jason Levine]

Or vulnerabilities. If Gigster winds up paying their developers the least amount possible, how long until their developers realize they get paid the same whether or not they sanitize that user input. User submitted a name of "Robert'); Drop Table Students; --"? No problem. Just stick that in the query and run it.