Why would anyone Scrum?

by JoeStagner 10/16/2007 4:45:06 AM

I just read “The Enterprise and Scrum” by Ken Schwaber and I’m left with a single burning question.

WHY WOULD ANYONE DO IT ?

Don’t get me wrong, I’m not talking about starting a new company or product unit and basing that new entity on Scrum practices. To the contrary, if I were starting a product development effort from scratch I WOULD.

I’m talking about re-factoring an existing software development organization who has historically used conventional Software Development Lifecycle practices (which means some variation of Waterfall methodology) to use Scrum.

Schwaber states in the first couple of chapters that you will loose a fifth of you staff, including a fifth of your management and your organization will be in turmoil for a year.

In this day and age, “Internet Time” where a year is three months long, what existing organization of any size could tolerate such an interruption to their business.

I’ve used what I’ll call “Scrum Like” process before on small teams over which I had complete control, which, I think from a Scrum purist perspective is an oxymoron. 

I’ve been taking some time to re-inform myself after the software development team Microsoft’s Developer Community Web Properties (www.asp.net, www.iis.net, www.windowsclient.net  , www.silverlight.net) suggested that they adopt a Scrum process. (That team is an independent third party.)

After reading the book I can’t figure out how we could do it and all keep our jobs.

One of the other fundamental tenants professed in Schawber’s book is “Never CHANGE Scrum”.  Don’t adopt it to your company’s process, don’t ease into it, don’t even change the vocabulary (role names, etc.) 

This all leaves me very conflicted ad Ken Schwaber is a very smart guy, a definitive voice on Agile Software Development Processes and one of the Developers of Scrum.

Joeological belief : Agile Software Development == GOOD

In this day and age for most software development purposes Agile is not only good, it’s necessary. We simply don’t have time to spec and test things in advance to the Nth degree the way we did in the 70s and 80s and I even believe that Agile architectures at both the Macro and Micro levels are an imperative in today’s successful software shops.

But Scrum ?

Not that it’s not a good idea, I’m just wondering if there is any evidence on the survivability of refactoring existing development organizations, especially those of any real size.

So subscribing gurus, what is YOUR experience on converting a development team to Scrum ?

Be the first to rate this post

  • Currently 0/5 Stars.
  • 1
  • 2
  • 3
  • 4
  • 5

Tags:

Related posts

Comments

10/16/2007 5:38:25 AM

Joe

Don't knock it until you've tried it!

In under a year, we did have to replace some people, and—in the beginning—there were as many downs as ups, but in the end, our company quadrupled (no exageration) our output—and we now maintain a zero defect average for the first time in our company's 20 year history. Additionally, one year ago, developers worked an average of 60 hours per week. No one works overtime today.

What Scrum has done for us is astronomical. We did not change any of our engineering practices in the last year. Using Scrum, we simply changed the way our teams were managed.

I started this comment with a cliche', so I will end with one as well: The proof is in the pudding, Joe!

Regards,
HL Arledge
Software Development Manager
Decade Software Company

HL Arledge

10/16/2007 6:28:27 AM

I would only agree that it's questionable to inflict scrum on an established team's process if that process is successful. And that opens a can of worms - if management thinks the team could be producing more, than it's not a successful process in their eyes.

In my opinion Scrum works, but the software process has so many variables that Scrum alone cannot save a team. A much bigger factor is "good developers write good software" and if your team doesn't work well together, or you've got some ego problems or just single developer productivity issues - then no process in the world is gonna save you.

Scrum does a good job of letting you know when to fish vs. cut bait. It provides visibility into a project that is doomed, long before the final deliverable deadline. That alone is worth billions of dollars.

M@

M@

10/16/2007 7:54:11 AM

pleae update me the procedures to adapt with we dev project with changing requirements. with average team

Aanandarajan P

10/16/2007 9:00:50 AM

might be a good idea to spell and grammar check before you post Smile

"loose" should actually just be "lose" in this context.
"you" should be "your"

Partha

10/16/2007 9:32:23 AM

You are thinking short sighted vs far sighted. Scrum can give a company a competitive advantage when it comes to building software. It's not a silver bullet and some people just can't make the adjustment to a self managing work environment. Not necessarily their fault - they have just been brought up the "waterfall" way. Scrum also highlights issues that impede progress - it does not solve them. This is hard for people. Many of these issues were simply swept under the rug and now they are being exposed! That is scary for lots of folks.

The beauty of Scrum is that it is not presecriptive. It can be adapted to almost any organization provided you understand the philosophy it offers. Let's face it, you can't implement Scrum without changing the way things have been done. Change is hard. On the flip side, if you feel your "current" process is fine - then you have no reason to switch to Scrum!

Long term, companies that write software in a fast and efficient manner will win out versus those that don't. My bets are on companies that can make the switch to better align their work with business strategy and shorten cycle times between releases.

Paul

10/16/2007 9:54:18 AM

I can see if people "jump" into scrum with separate projects but not as an entire company rooted in waterfall. In the current climate in my area, it seems very difficult to change the practices of a company without losing valuable time. The unemployment rate is ridiculously low around DC, which makes it hard to find good candidates, even if we had the time to set aside to do interviews to replace the 20% that leaves according to Schwaber.

Han

10/16/2007 12:02:12 PM

Joe, I agree with you. I was at a place that tried to convert a large existing project on a fairly large team to Scrum and it failed miserably. And there wasn’t one point of failure; it was all over the place. First the technical challenge to get the application under agile/scrum technical guidelines would have taken months. And the managers quickly reverted back to their old ways and ignored the iteration schedule and piled on scope creep work. It led to some long days and nights and many developers jumping ship. Scrum lasted about a month; we never even made it past the first iteration.

Justin

10/16/2007 12:27:26 PM

HL Arledge

Could you elaborate a little more on how Scrum quadrupled your output and eliminated overtime?

Partha

Powered by BlogEngine.NET 1.3.0.0
Theme by Mads Kristensen

About your host.

Name of author Joe Stagner
?????

E-mail me Send mail

Calendar

<<  November 2008  >>
MoTuWeThFrSaSu
272829303112
3456789
10111213141516
17181920212223
24252627282930
1234567

View posts in large calendar

Pages

Recent comments

Disclaimer

The opinions expressed herein are my own personal opinions and do not represent my employer's view in anyway.

© Copyright 2008

Sign in