Réf. : [scrumdevelopment] Comments on Silver Bullets
My understanding of Ken's article : there is a risk that agile methods be used as a silver bullet and follow the fate of the previous ones. He warns : adopting agile is no silver bullet, it is hard but the reward is high visibility and the cost is the necessity to face bad news early and often.
I found Ken too cautious in this article :
"Agile processes come in with a different tenet: software development is difficult work and here is a framework and set of practices within which to engage and manage teams in this difficult work." : there is no promises to be in a better shape, so what's the trouble worth ? Agile do make a promise : to produce software more successfully, though possibly under certain conditions. Visibility is only one part of the reward.
"The good news is that you can build great software predictably with agile processes. The bad news is that it is really hard." Well, non agile processes have proven successful (under certain conditions) and hard to apply as well. If we stay with this conclusion, few people will be willing to embark the change for agile.
The fact is that, as other readers said, I don't know the intended audience.
My take about silver bullet :
One rendition of the the story of Pandora in ancient greek mythology runs as follows. Zeus was enraged that Prometheus gave fire to man. He decided to inflict a terrible punishment on both man and Prometheus. He had Hepheistos shackle Prometheus to the side of a crag, high in the Caucasus mountains and each day, Prometheus would be tormented by Zeus' eagle as it tore at his immortal flesh and tried to devour his liver.
Zeus' anger did not stop there. He intended to give the mortals one more gift and undo all the good Prometheus had done. He fashioned a hateful thing in the shape of a young girl and called her Pandora. Her name means, 'giver of all' or 'all endowed'. Her body was made by Hepheistos. Athene gave her dexterity and inventiveness. Aphrodite put a spell of enchantment around her head.
Zeus gave Pandora to Ephemetheus (brother of Prometheus). Ephemetheus had been warned by Prometheus never to accept gifts from the Olympians, especially Zeus. One look at Pandora and Ephemetheus was rendered helpless. He could not resist her, he accepted her willingly. Pandora had a small box supposed to be containing her marriage presents. But when the box was opened, evil and despair entered into this world. Mistrust and disease spread over the wide earth. After the whole contents of the box had escaped, only one thing was left inside. Hope. Unreasonable, groundless Hope that makes the curse of life into a blessing.
In a modern version of the story, we could add high technology among the gifts that Athene and Zeus gave to Pandora.
But, wait, how could hope can be considered an evil thing ? Well, let us consider the harmful silver bullet syndrome. This is one of the evil sides of hope. Hope grounded on wrong models, that like superstition, makes us suffer.
Ken refers to Fred Brooks famous article " No Silver Bullet ", where Brooks contended that there were *essential* reasons that no silver bullet would be found in the future of the software development. To address the essential parts of the software development he even promoted some of the agile fundamental practices : " growing software organically, adding more and more functions to the system as they are run, used and tested ". But in my reading of the Brooks article, he failed to identify one essential part : the people factor. This is what agile brings (see the agile manifesto). But what was left out of the manifesto (voluntarily I think) are the essential reasons *why* the agile movement have these preferences.
We will lower the risk to suffer the fate of silver bullets if we clearly expose a understandable theory of the essence of software development that supports the different agile practices, explains why they are arranged that way in the different agile flavors and what are the reasonably expected rewards.