Subject Matter Experts vs true expertise

[cdhames]

I read the following quote from Vern Edwards today, under a different discussion and I wondered:  Should we always make the assumption that our Government SME's are capable in their field?  I know this might be a profound (or even unintelligent) statement but... 

Quote

Any government acquisition office that must hire a contractor to write an S.O.W. is staffed by incompetents. I can understand hiring a contractor to write a system, subsystem, or component specification. But an S.O.W.? Please. If an office needs that kind of help it shouldn't be allowed to conduct the acquisition.

Bear with me a moment.  I realize the question might immediately beg for a quick and decisive answer, but my thought is this:

Our law and regulation is crafted in such a way that it automatically assumes a sure and competent hand of the Government executes something in a sure and competent manner.  There's really no room for error.  Our FAR regulation is stated in a specific manner, to be interpreted (for the most part) a specific way and implemented in such a way.  Sure, there is sometimes, difficult to discern language but we normally reach consensus on how that something should be interpreted.  GAO could loosely be described as such a consensus builder.  The Supreme Court as another.  My point being, in the general context of day-to-day activities, regulation is somewhat unforgiving.  It assumes expertise.  Vern's statement above does the same.  Now generally I would agree with the sentiment; a Program Manager "should" be able to draft a recently reliable SOW.  But in operational reality, most of us know this just isn't the case.  Error and fallacy is built into human nature.

Awhile back, I read a Quora post by Kostadis Roussos, former Chief Software Engineer at Zynga (article here:  https://www.quora.com/Why-would-anyone-want-to-work-for-Zynga ).  A short time after Zynga went public their common stock fell off the proverbial cliff, from somewhere around $15 to $2 (where it's been sitting nearly ever since).  During the fall, an exodus of talent took place as top executives left the company.  Zynga attempted to counter the exodus.  Roussos' article on Quora was one such attempt. 

In the article he discusses market disruption in the field of software engineering in the mid to late 1990's.  How one group at MIT who thought provocatively and in a different way about writing software, changed the industry.  It's about a 50 year model of determinism that was disrupted by new, randomized thinking.  As the context goes, what if software engineers planned for failure?  Instead of ignoring failure, and attempting to reduce it through a deterministic model, what if they planned for failure instead, and respond to failure through intelligent modeling.

So as a former, aspiring software developer, this article had a profound effect on me and I've since thought back often about it.  I look at the FAR now, and the way our Government programs and policies are set up and I can't help but ask, why do we have such a hard time with failure, and wouldn't it be better to plan for it in our acquisitions?  To write contracts that attempt to error check and deal with it, instead of forcing failure into the standard model of deterministic outcomes? 

When I read the quote above, I both agree and disagree at the same time.  I'd like Government experts to be experts, but I know the vast majority (myself included) are not; and we make mistakes.  Some of us don't know what we don't know yet.  Some of us can't write a SOW yet (or we can write one, just not a capable version of one).  There's an inflexibility with regard to the incapable, or untrained.  A built in intolerance that I think we might be able to redefine.

 

So that might sound a bit lofty, but I'd be interested in starting a discussion on it, in more than a candid fashion.  What if we allowed a contractor to write a SOW, or take part in it's development?  What are the takeaways?  Would it make it better, and would it put the Government in an unfavorable position?  I'm not suggesting the contractor write the SOW, and then the Government accept it as is.  I'm referring to a collaborative effort.  What's best for all parties, the public included. 

[ji20874]

Almost any engineer will benefit from the services of a technical writer.  

[Boof]

Our lack of expertise (plus HR refusing to provide more FTEs) is why Product Service Code Category R - Support (Professional/Administrative/Management) is very high in spend at most Federal Agencies.  It is number one at mine. We use lots of consultants.  That is not counting how much gets sent to non-profit FFRDCs to plan things for us.   

[here_2_help]

Recently a fellow consultant posted something inane on LinkedIn. I mean something almost unintelligible and quite obviously wrong. But because that fellow consultant had more than 25 years with DCAA there was instant credibility. It sparked me to think about the difference between expertise and the application of expertise. If you don't apply your knowledge to further the objectives of your entity, then your knowledge is kind of worthless, in my view. Or, as my friend once said, "I am full of useless superior knowledge."

I am not a government employee but I interact with government employees nearly every day. To me, they all seem to be working really hard but very few seem to know why they are doing anything. They have taskers and metrics and priorities that have to be reported all the time, but nobody seems to think there is any strategy in which they may be playing a role. They seem to be a collection of individuals doing their best, but they have no overall direction. That's my perception, anyway.

To the initial question, regarding whether the acquisition system should be changed to prevent or detect mistakes, what I see is that the system is already doing that by adding process on top of process on top of process. There are peer reviews, management reviews, headquarter reviews. There are review boards. How long does it take a CO to write a PNM? Not long -- but it can take weeks to get it approved. Then there is another approval cycle for post-negotiation memoranda, right? It literally takes two separate processes to get approval to, for example, negotiate an FPRA. And if DCAA disagrees with the DCMA ACO, there is yet another process to mediate and resolve that disagreement, because nobody wants bad feelings (or a phone call to the IG).

Not sure where I'm going with this, except to say that process doesn't seem to be the way to fix the problems associated with a lack of expertise. It would seem that you fix those problems by giving people more expertise and letting them use it -- and then holding them accountable for mistakes.

 

[Guest Vern Edwards]

14 hours ago, cdhames said:

What if we allowed a contractor to write a SOW, or take part in it's development?  What are the takeaways?  Would it make it better, and would it put the Government in an unfavorable position?  I'm not suggesting the contractor write the SOW, and then the Government accept it as is.  I'm referring to a collaborative effort.  What's best for all parties, the public included. 

Nothing wrong with that approach in a sole source situation. (I think it's impractical in a competitive situation. Too much work doing it with multiple firms.) But that's not the same as hiring a company to do your job.

A statement of work is a contractual document. It should describe the tasks that the contractor is offering to perform. Agency personnel must know what a statement of work is supposed to be and how it should be written in order to engage in the process during a negotiation. So I still hold that an agency that has to hire someone to do a task that if fundamental to its reason for being is staffed by incompetents.

I'll add this: Statements of work are not technical documents.They are not process instructions or technical manuals. They are contractual documents with some technical content. They are not addressed to the people who will perform the service, but to agency and company managers and their lawyers. Most worker bees will never see it. The people who should be most qualified to write statements of work are contracting personnel, because they are supposed to know about contracts. They should learn the technical lingo of the people they support and how to interview and work with them in order to discern their needs and produce a document that describes what they want. Of course, in order to write a good statement of work you must know how to write, that is: think, develop ideas, and express them clearly in appropriate language. It appears that only a small number of the workforce are able to do that well. I say again--an agency that has to hire someone to write an S.O.W. is staffed by incompetents. They are incompetent because they are unable to perform a crucial function.

Becoming professionally competent is really, really hard work. Much harder than college. And it is never over. You never graduate. It's like rock climbing. You have to have a good strength to weight ratio and good stamina, you have to be able to accept the reality of the abyss, and you have to be able and willing to expose yourself to falls. And after you finish one hard climb, you'll want to find a harder one in order to stay in the game. You'll want that or you'll want to quit.

Edited February 24, 2016 by Vern Edwards

[apsofacto]

Question:

Quote

Should we always make the assumption that our Government SME's are capable in their field? 

Answer: No.To be fair, they should not assume that of us either. 

Quote

The people who should be most qualified to write statements of work are contracting personnel, because they are supposed to know about contracts. They should learn the technical lingo of the people they support and how to interview and work with them in order to discern their needs and produce a document that describes what they want. Of course, in order to write a good statement of work you must know how to write, that is: think, develop ideas, and express them clearly in appropriate language.

I really enjoy helping our project managers with their statements of work.  My only rule is that they generate the first draft- I don't even care how bad it is.  We sit together and try to eliminate passive voice, convert it to language that compels the contractor to do things, re-arrange sentences so the background material isn't cluttering up the tasks, and usually come up with some questions.

It's not perfect but it helps.  If it was all that I did all day my employer would be getting their money out of me.  However, we must invent exciting new paperwork and go to Very Important Meetings^TM (VIMs). 

[prodigalko]

I think Boof's funding issue is quite prescient.  When you have senior leaders mandating a 25% cut in FTE's or otherwise reducing the manpower below realistic levels it is very difficult to get all the work done (and done competently) without hiring professional services to assist; especially when contract dollars are provided to soften the blow of the FTE decrease.  This is compounded by poor career field management (mainly DoD) resulting in large decreases in average years of experience in any given office over the last ten years.