Hi Paul,

 

+1 on agile as pretty common language for describing projects these days. From what I've seen, middle/upper management I have worked with still tended to prefer more rigid waterfall inspired styles, where as leaf teams usually were open/eager to adopting agile practices. Not sure if that's generational, based on hard won exp, etc...

I haven't seen any team which was able to sustain a "pure" process methodology for very long. Not that I disagree with the principle, just that teams are complicated machines, composed of a bunch of humans with very differing needs. A non-trivial sized org that doesn't want to be a monoculture has to compensate for those competing needs, and tends toward compromises on those things, which tend toward the "average".

Personally, I wouldn't suggest you get too hung up on a specific process management methodology beyond understanding the core techniques, unless you hope to fast track on a mgmt focused path. My experience is that the specific methodology employed isn't nearly as important as that the methodology used matches the requirements of your team and stays relevant. 

For instance, in any software team, Agile really counts on respected, experienced practitioners and meaningful trust between mgmt and peers. In the steady state, a team's process generally conforms to the style of management. A team may commit to agile, incremental high-quality principles, but mgmt defined goals/shipping schedules can put a crunch on that in practice - at that point people have to make painful tradeoffs against their ideals to decide what is important. Methodology can be a casualty of crunch time, as some folks regress towards more ad-hoc approaches when the process "gets in the way".

Advice - be aware of the different schools of thought and be open to balancing what works in your situation:

• Study the discipline of "Organizational behavior" - this is an important early pre-requisite for folks in business school that outsiders sometimes miss out on. It covers concerns like leadership styles, how to successfully introduce change into an org and other such things which provide a foundation for project management. Ignore the wisdom of this discipline at your peril - doing so can lead to good ideas failing for dumb reasons.
  
• Learn and understand the principles and practices of the waterfall style methods - like it or not, they still have a big influence on idealized, centrally managed processes. Folks with "PMP" credentials or backgrounds in places like NASA, critical systems (ie- embedded, medical, defense industries) tend to be a reputable source of expertise here... Notice - these disciplines have an explicitly high cost of failure, which means these folks can't really employ with Agile practices at the same scale. Pretty much any book on Project management covers the core bases here.
• Read up on agile - I'd argue it's not rocket science. If a team of bright, agile-aware people are empowered to make decisions themselves in a healthy environment, things can get self organizing. If a lead doesn't trust people to do the right thing or wants to micro-manage all the decisions, it's hard to make agile work.
• Learn by doing - Apply what you know- demonstrate better ways of doing things in a way that is positive and non-subversive of mgmt.

Peter

On Tue, Mar 10, 2009 at 3:43 AM, Fabrice Lété <drealmer <at> skynet.be> wrote:

Hi,


Agile software development methods are quite popular nowadays in game development, especially Scrum.

http://en.wikipedia.org/wiki/Agile_software_development
http://en.wikipedia.org/wiki/Scrum_(development)

Or just search for "agile game development" on Google, and you will have enough reading material till the sun turns into a red giant.


++
Fabrice

Paul Hollingworth wrote:

Hi all,

I'm a software engineering student hoping for a future career in gamedev, and was wondering;  what's the level of uptake on good standards and practices in the industry?  Are you guys on the forefront for trying out new methodologies, or stuck on traditional stuff?  Do the company use any process methods, such as PRINCE2?  The reason I ask, is because I cannot seem to find any good reading materials on software enginering in game development.  I know Rucker wrote a book, but the previews seem to indicate it is more of a programming tutorial book, than a disucssion.

Does anyone have any recommendations for reading that looks at Software Engineering (or even Project Management) in Game Development?


Thanks,
Paul Hollingworth

_______________________________________________
Sweng-Gamedev mailing list
Sweng-Gamedev <at> lists.midnightryder.com
http://lists.midnightryder.com/listinfo.cgi/sweng-gamedev-midnightryder.com

_______________________________________________
Sweng-Gamedev mailing list
Sweng-Gamedev <at> lists.midnightryder.com
http://lists.midnightryder.com/listinfo.cgi/sweng-gamedev-midnightryder.com

Thanks for the replys everybody.

I'm aware of, and have studied agile development (especially Scrum).  My main concern is the lack of information about how it fits in to the game development industry.  Maybe my thinking is wrong here, or I am idolising the industry, but it seems to the outsider that game development deals with many of the issues covered in other domains (criticality, cutting edge, etc) whilst also having it's own distinct set of issues. 

Apart from some articles I've found in Gamasutra/GD.NET/etc, there seems to be very little information or analysis on how Scrum fits in to the types of projects performed.  I suppose I was looking for a "Pressman on GameDev" or something

I'm not buying that all methods are portable, I agree with Peter that each should be taken upon it's possible merits to the project in hand.  Does this happen in the 'real world' though or do companies tend to stick with what they know?  How big of a deal is the whole Quality Standards documentation to the devs and to management?

Oh, and it is my firm aim to stay in front of a computer, programming, for as long as I possibly can.  Unless they let management fire some code out in the near future, I'll try to aim for a Software Engineer's spot. )

Paul


Windows Live just got better. Find out more!

_______________________________________________
Sweng-Gamedev mailing list
Sweng-Gamedev <at> lists.midnightryder.com
http://lists.midnightryder.com/listinfo.cgi/sweng-gamedev-midnightryder.com