views:

162

answers:

3

Our investor wants a SDLC. I've never written one before, and I don't have enough time to go and buy a book, or spend much time learning about them. From what I've been told about them, they consist of requirements (what needs to be done), and a list is done. Is this correct?

Update:

I have found this article which really helps to explain things in simple terms and very quickly. Not that I think an SDLC should be done quickly. In my case, I have no other option.

+1  A: 

Not really; that's more project management. That's what you need at the point where you've figured out how you're going to develop software.

For the 'how' of developing Software, the two 'biggies' are Agile and Waterfall; with a weird hybrid in between the two.

But that's only one part of the Software Development Life Cycle: You still have to have a maintenance and deployment plan.

My question for you: If someone's giving you money, and they want a plan, isn't it in your best interest to read a book about the SDLC and give them a plan?

George Stocker
Thanks for the answer. I don't understand why they need a plan. The software's already been made. We only need to implement two more features. And we're supposed to be giving this SDLC to whatever software dev. company we choose to let finish those two features. I've read about the Waterfall methodology on Wikipedia.org, and now I'm confused. Do you know of any sites that explain the process of writing one in simple terms? My first thought was to go anmd buy a book. But I'd have to travel into the city (4hours away). But they want the SDLC by tomorrow.
lucifer
@j-t-s: If your company is asking them for money, it's a pretty safe bet that the last line of code hasn't been written... you're probably not even close. These two features, then patches and bug fixes, then version 2 to meet new customer requirements, ... They want to know that you'll be able make good use of their money to evolve the code over time, until it's at a point where they have made their investment back with a good profit.
Eric J.
IMHO you woke up too late to do a good job. The reason is that the investor wants same work (quality) from the other company. You should basically describe how you did the development of the system.
Gabriel Ščerbák
+3  A: 

There are lots of ideas about SDLC out there. You can't swing a cat without hitting one.

What have you done to develop software that attracted your investor in the first place? Can't you describe that? Why do you have to go out and "learn one"?

There's a number of choices:

  1. Waterfall: requirements->design->build->test->deploy, all in sequence
  2. Iterative: similar to waterfall, but you break the design into smaller pieces, of 1-2 week duration, that are delivered at the end of the iteration.
  3. Extreme Programming (XP): Kent Beck's approach; no BDUF (Big Design Up Front). Everything is designed, built, and delivered in small pieces.
  4. Scrum: Agile, iterative, but not as dogmatic as XP.
  5. Rational Unified Process: Waterfall from IBM.
duffymo
...and gazillion others...
Gabriel Ščerbák
Very few of which are significant or worth knowing. And not one that can guarantee a quality program.
duffymo
A: 

If your investor wants you to describe the SDLC, he wants that you describe how looks the life of a software project which is done by you from its plan, birth, through growth to maturity and death. That is the reason why it has "life" in its name. The result of SDLC should be "software" hence the first word. The "development" part comes from the fact that you are responsible for planning, specifying, designing and implementation of the software, you should create (develop) the software. And finally, "cycle" means that, when the investor looks at you SDLC and thinks it is good (it produces quality and business value), he can ask you to use the same process once again in another project.

Gabriel Ščerbák