[RESOLVED] development methodologies

[binbpk]

which methodology would you consider using to develop a web based interface?

[Paul M]

Well i don't know to much about web based development but well stages of software development at a high level should go as follows:
1) Analysis
2)Design
3)Development
4)Testing
5)Documentation
6)Implementation
7)Evalutation

[binbpk]

its something similar to sdlc

[penagate]

Development methodologies are a load of nonsense. You need to just do what makes the most sense to you. Over time, you'll develop a routine for development, and refine it.

If you are working for a company then you need to follow their standard procedure, if they have one.

[Paul M]

Well that general standard procedures are as i posted. Though like you kind of said particular companies will vary in how they do things.

[mendhak]

which methodology would you consider using to develop a web based interface?

Open Photoshop.
Create interface.
Slice.
Save interface.

[bgmacaw]

Well i don't know to much about web based development but well stages of software development at a high level should go as follows:
1) Analysis
2)Design
3)Development
4)Testing
5)Documentation
6)Implementation
7)Evalutation

It usually goes like this in real world rapid development:

1) Analysis - "Marketing has an idea, get with them."
2) Design - "Why aren't you coding yet?"
3) Development - "Are you done yet?"
4) Testing - "What's the hold up?"
5) Documentation - "Send out an email to everybody."
6) Implementation - "I've been getting a lot of support calls today about your program"
7) Evalutation - "You're fired!"

The sad fact is that in most organizations development methodologies don't reflect what really happens, just what everyone hopes will happen.

With web development everything gets accelerated since most companies take a "hack and slash" approach to it since updates are so much easier.

[Al42]

Successful real-world development is something like:

Get with the end customer (your salesman, the department that's going to use the program, the end-user) and find out what the problem is and what solution the customer sees the program providing.

Estimate the time needed for the project. Double the number. Increment the units (days become weeks, weeks become months, etc.) Give marketing this estimate as "best case, if everything goes smoothly and there are no changes from here on".

Develop a few ideas (in English) and a few pictures of how you envision the screens (freehand on paper, using software - however you can do it best). Meet with the customer and see if what you came up with is what he had in mind.

Begin Loop 1

Have the customer flesh out your ideas. Add in all the details he wants, change the screens, etc. This an almost infinite loop, so be prepared to Exit Loop somewhere.

Write code.

Bring code back to customer. Enter Infinite Loop 2. Exit IL2 to Loop 1.

Wash, rinse and repeat until the customer is happy.

Write documentation, package, ship and wash hands of the whole thing. Take a vacation and forget the project ever existed.

[bgmacaw]

Estimate the time needed for the project. Double the number. Increment the units (days become weeks, weeks become months, etc.) Give marketing this estimate as "best case, if everything goes smoothly and there are no changes from here on".

Ah, the Scott-Spock estimation method. How else will you get a reputation as a miracle worker?