Outsourcing Killed By Django And Ruby On Rails | Django Aware.
This is an interesting perspective. I think it will be really hard for people to get used to the 20% being the whole project. Experienced project managers have an expectation that a project of a certain size will take a certain amount of time to complete. Even if you tell them that it’ll only take 20% of the time, they’ll still plan for the full 100%. I think this mindset is so strong in some circles that those managers will find ways to draw the project out just so that they feel like they still have some grasp over the project. To have that 80% totally evaporate in an instant just by switching technologies would blow their minds, in a scary way.
Things always seem so easy from a developer perspective, where you just have to focus on building things as quickly and as well as you can. Those applications don’t exist in a vacuum though. It’s when you build an organization around an app, where not everyone is as capable, optimistic and knowledgeable about what’s possible with the tools currently available that things start to bog down.
I’d love to hear success stories though. By all means, prove me wrong.
I've got a masters degree in computer science and over 10 years of experience building web-based systems using Java/J2EE, Ruby, Rails and PHP. I'm a strong believer in the effectiveness of Agile Methods.
One Comment
Fascinating article, so glad you blogged on it, as I’m not sure I would have run across it otherwise.
I completely agree with your take on things Laran, but I would qualify it to say that its never really that simple.
Production of a non-trivial commercial web application has many parts. The code is a big part, but its not even close to all of it.
You’ve got to meet the client, sell the client, develop enough specs/stories to get started, and then do iterations with the client. These are often in-person with a product manager, or someone wearing that hat.
Then you’ve got to do deployment to production, hosting, and often, do onsite training.
And then there’s the management and organization of all of this.
I’d say the coding is no more than 50-60% of the cost of a typical application. And the more complicated the application is, the more intensive the ‘product manager’ role is, and the bigger share of overall costs which that requires.
Dont even get me started on how one or two rare (but critical) edge conditions can end up costing you as much to discover/define/develop as the rest of the 20%.
In addition, the staff you put in front of clients for the initial closing, and ongoing product manager roles are often quite high-end folks. They’re the face your clients associate with your company, are often owners, and their time isnt cheap.
And the extraction/conversion of ideas in your client’s mind to concrete features is not a trivial process.
The best consultants have a real skill in taking what people think they want and developing what will really be compelling features or approaches. It’s anything but a passive process where the client knows exactly what they want and can describe it clearly and unambiguously. Quite often the opposite, actualy.
So overall, while I agree that frameworks like RoR (which we use extensively now) are amazing and powerful and industry changing, the aspects that separate a good consulting company from a great one lie mostly in the non-coding work.
Good discussion, and despite my tangent above, there are truly some amazing things happening in frameworks that are changing the cost structure. no question.
Andrew Hollamon