Alan J. Perlis
                        American Computer Scientist 1922-1990

The business of software: The cliché defense

Phillip G. Armour
Communications of the ACM - Barbara Liskov: ACM's A.M. Turing Award Winner, 2009

A management Cliché is a fine device.  Using it, an initiator (the “cliché-er”) can simply and effectively attain superiority over the unfortunate target (the “cliché-ee”).  I know of whole organizations that are run primarily on clichés though not, of course, run effectively.

But from a systems and data structure point of view what are the key attributes of a cliché?

• Brevity: must be sound bite size.
• Scope: covers a wide range of human activity with few words.
• Specific = Non-Specific: it should sound specific and focused, but really not actually say much.
• Style over Substance: see “Affect over Effect” below.
• Affect over Effect: see “Style over Substance” above.
• Truth: it should have a veneer of truth, just to make it hard to argue against.
• Effort: It should require significant effort on the part of the recipient (the cliché-ee).
• No Effort: it should not require any effort on the part of the deliverer (the cliché-er), except for spouting the cliché.
• Success: any success in resolving the cliché should accrue to the cliché-er.
• Failure: that would be an implementation problem and is clearly the fault of the cliché-ee.
• Status: it must strongly reinforce the status of the cliché-er over the cliché-ee.

How to Defend Against a Cliché

The Cliché: Do it right first time
If it really is the first time, we probably cannot do it "right" because:

• We don't know how to do it―because it is the first time.
• We may not know a priori what "right" is―because is is the first time
• Doing it "wrong" may be the best (or only) way to figure out how to do it "right"

The Cliché: Work smarter not harder
This is a pretty sneaky cliché.  The defense is:

• If we were smart enough to know we're not working smart enough, and...
• If we were smart enough to know there was another, smarter way of working, and...
• If we were smart enough to know how to transition from the not-very-smart way of working to the smarter way of working...
• We'd already be doing it.

So clearly, we are not smart enough to work smarter.

The Cliché: Quality is the most important thing

• Just what is quality?
• Who can provide us with guidance on this in advance (as opposed to second-guessing after the event).
• Are we prepared to do what we need to do to obtain the quality we say we must have?
• Are we prepared to balance the other "most important" attributes of the product against quality (such as cost, delivery date, etc.)?

The Cliché: Our customer are the most important thing

• Are we identifying all the customers?
• Have we built the appropriate customer data-collection and customer facing systems and relationships necessary to treat the customer as the most important thing?
• Are we appropriately balancing supporting the future customer needs against the current customer needs, say by building capability in our people and extensibility in our systems?

The Cliché: Our people are the most important thing
Few clichés have the power to generate a cynical response quite like this one.  It often telegraphs a blatant attempt at control that insults our intelligence.  Defend by:

• Do we really provide our people with all they need to do the job?
• This includes: material, information, resources (including time and budget), motivation, support, guidance, clear goals, appropriate rewards, consistent support processes, etc., etc., etc.

If not then it's just, well, a cliché