What should Change Control Board (CCB) membes know?

ssmfone's picture
ssmfone asked on June 30, 2011 - 3:01am | Replies (10).

I’m conducting a short study to identify what awareness information does the decision maker (Configuration Control Board members) should know/aware while completing some of the tasks below.

Task 1 – Tasks to analyse the proposed change request
Task 2 – Tasks of determining what changes to make
Task 3 – Tasks of assigning of developer(s) to implement a change
Task 4 – Tasks of assigning testers to verify a change

Sample answer:
Task 3 – Task of assigning of developer(s) to implement a change
Answer - The CCB should know the expertise of the developers.

There might be a few things that CCB members should know while completing each task. Inputs from the practitioners are highly appreciated.

10 Answers

Marc Girod's picture

There should be no CCB.

This is an obsolete and counterproductive technology.

Marc

mbools's picture
mbools replied on July 6, 2011 - 8:23am.

Marc: you are wrong on both points; there should be a CCB, it is neither an obsolete nor a counterproductive technology (it's a function, not a technology, for a start).

That said, it is unnecessary to have a [i]large[/i] or bureaucratic body acting as the CCB. What is important is that the [i]function[/i] of the CCB is performed; changes are evaluated and someone takes responsibility for approving the work necessary to effect that change.

I've discussed this with many people over the years and almost always there is a misapprehension about the CCB that it must be a group of people sitting in a room every Thursday morning debating over changes. Now, it may be that some CCB are like this (certainly in large organisations the operational CCB that decides on major changes to operational systems tends to be like this), but most CCB are much smaller, often operated on a 'needs' basis, and can consist of just one person (the Project Manager for example).

The size and formality of the CCB is entirely dependent on the potential impact of the changes it considers. An operational CCB considers changes that have direct and often significant business impact. A project CCB considers changes that directly impact the projects delivery, resources (and so may approve changes that affect the project and the CCB size and formality will be commensurate to the project's size and complexity).

I suspect that what you mean, when saying the CCB is unnecessary, is that developers should be allowed to make changes on their own cognisance. I often advocate that projects operate a system that allows developers to raise and assess their own changes, and approve their own changes if that assessment shows the change is a minor defect correction, for example. In other words, they act as their own CCB; evaluating the impact of a change and approving it if appropriate. Any changes that are approved this way are notified to the main project CCB (again, for smaller projects, often the just project manager) so that they are properly monitored and tracked (ensuring that developers are not abusing the change system and are remaining focussed on maintaining the forward momentum of the project.

This form of developer CCB is not appropriate to all organisations, nor is it appropriate to all teams. As with everything, you need to decide what is appropriate to your circumstances.

Returning to the original question, I hope you can see ssmfone that 'it depends' is the appropriate answer. It depends on where your CCB is operating and what the potential impact of changes it considers are likely to be. An operational CCB is a very different animal to a small project CCB.

Richard Page's picture

I agree with most of what has been said but would highlight the point that Developers cannot just make changes. When implementing a CM system that enabled the allocation/authorisation of Tasks to Developers - one developer complained about the process stating that he just used to 'make changes' without the need for recording. To which I replied that this was probably why the Developers had an excessive amount of defects - there was no impact analysis of such changes and they were possibly affecting other areas of the system. Also some of the changes were 'cosmetic' and not 'defects' - we were missing potential revenue through 'free' enhancements that had not been asked for by the customer.
Change Control is needed on most if not all development teams.

Marc Girod's picture

The mere use of a CCB is a choice to make communications private, synchronous, volatile: discontinuous.

It is the choice of control against management.

It may have been a valid strategy in times when there was no possible tool support, and when the degree of complexity was low enough to be dealt with by people during a meeting, but neither of these two conditions hold anymore, at least in any software related industry (not exclusive).

Making this choice defeats management. It implies delays, poor quality, politics, frustration of all sorts.
Eventually, failure in the context of competition.

Marc

jptownsend's picture

Marc Girod wrote:

"It's a technology, in the same way as druids are a technology to forsee the future and slaves a technology to build pyramids (although I read recently that pyramids may not have been built by slaves after all)."

Marc you are correct a CCB is type of technology. However, CCB communications are not to be private, volatile. discontinuous or anything like that. It is also not about control its about managing your project.

Maybe you have worked somewhere that this wasn't the case, that doesn't mean all CCB's are bad, maybe you had one bad experience or two or ten.

CCB's do have a place in projects and complexities can be better dealt with sometimes in meetings. Email is a horrible form of communication it is at times private, volatile and discontinuous. Tools that track issues can also do the same thing

If a CCB is causing delays, poor quality, politics, and frustration, then its not a CCB its a good ole boy meeting or good ole girl meeting or whatever. If your CCB has a meeting agenda, investigates the complexity of changes and makes sure that all possible options are explored how can that be bad???

I know you won't answer me thats ok, though.

Regards,

Joe

mbools's picture
mbools replied on July 6, 2011 - 12:34pm.

"It's a technology, in the same way as druids are a technology to forsee the future and slaves a technology to build pyramids (although I read recently that pyramids may not have been built by slaves after all)."

Torturing the word 'technology' like that seems unnecessarily cruel. I'm generally more kind to my native tongue than to visit that much abuse upon it, but I guess, at a long stretch you can say a decision making entity is 'technology' (of course, that makes Parliament a 'technology' :blink: , and pretty much anything else you care to mention for that matter).

"The mere use of a CCB is a choice to make communications private, synchronous, volatile: discontinuous."

I have no idea what your notion of a CCB includes, but it seems to be the antithesis of everything I know about CCBs and their operation. CCBs, properly implemented, have always improved communication, made changes more visible, and made system configurations more stable.

CCBs make change more synchronous? Yes, when called, for but not necessarily. And 'discontinuous' (assuming you're not doing a Humpty-Dumpty again) I guess CCBs have the capacity to make the change process unnecessarily discontinuous, but, again, that's more a question of poor design and implementation than a fundamental property of CCBs.

"It is the choice of control against management."
This is a false dichotomy.

The next paragraph of your reply again reflects a misunderstanding about the nature of CCB operation. New technologies are certainly helping to make CCBs easier to run, but are far from making them unnecessary. Your insistence on referring to 'meetings' betrays a certain naivety about CCB operation.

"Making this choice defeats management. It implies delays, poor quality, politics, frustration of all sorts.
Wrong. Wrong. Wrong. Wrong. And wrong. (At least you're consistent.)

There are no reasons why a CCB should cause unnecessary delays, poor quality, politics (by which I assume you mean the CCB is used for political games?), or frustration of any sort. (Other than, once again, a badly designed or administered CCB.)

"Eventually, failure in the context of competition."
And, finally, wrong. What possible grounds do you have for this declaration?

Marc Girod's picture

[b]jptownsend wrote:[/b]
[quote]I know you won't answer me thats ok, though.[/quote]???
[quote]CCB communications are not to be private[/quote]Just a minute. Words are a device which is meant to tie to a [i]meaning[/i].
The CCB is a board of pre-defined people, right?
That is, it is [i]private[/i]. People outside this group are not asked their opinion, or their expertise, unless the members do so, and [i]they[/i] play the role of filter.
[quote]volatile[/quote]In a meeting, one [i]may[/i] keep minutes, or not. And even if one does, this is only a [i]snapshot[/i], not [i]the[/i] communications.
Furthermore, people will perform self-censorship in meetings (even if there is no censorship proper): they don't want to hold the others who want to leave. They don't want to look silly or overly eager, or whatever.
[quote]discontinuous[/quote]The meeting takes place, and then it is over. No afterthoughts, no precisions, no deeper investigations: fire and forget.
[quote]It is also not about control its about managing your project.[/quote]I use these words precisely to oppose their meanings: control means prevent from happening things which have no been planned, mandated, prescribed; it is based on top-down application of decisions based upon upfront knowledge. Management is about making sense of what happens, bottom-up. You can only manage what you first allow to happen. The more you control, the less you manage.
[quote]Maybe you have worked somewhere that this wasn't the case, that doesn't mean all CCB's are bad, maybe you had one bad experience or two or ten.[/quote]In addition to bad experiences, I have an model of why things happen the way they do.[quote]CCB's do have a place in projects and complexities can be better dealt with sometimes in meetings.[/quote]With the people present, within the allocated timeframe, etc. Quality and experience have been optimized away, they are secondary.[quote]Email is a horrible form of communication[/quote]Obviously. Email is not an SCM tool. No decent SCM tool relies upon email.[quote]Tools that track issues can also do the same thing[/quote]If you mean ticket processing tools, they are utterly unsuited to SCM. They also optimize one generic aspect of the communications at the expense of the information proper. They introduce a noise.
[quote]If a CCB is causing delays, poor quality, politics, and frustration, then its not a CCB its a good ole boy meeting or good ole girl meeting or whatever.[/quote]Quite possible in addition, but there are structural reasons for which meetings are bad, unscalable, time consuming, etc.[quote]If your CCB has a meeting agenda, investigates the complexity of changes and makes sure that all possible options are explored how can that be bad???[/quote]Making the meeting [i]better[/i] is one way to make the SCM worse: it distracts the information from being processed via the SCM tool.
Marc

mbools's picture
mbools replied on July 7, 2011 - 12:59pm.

Very quick one. Joe and Billy, I think we're all in agreement. Hurrah!

One thing I do to avoid the CCB name over function issue is.... not use it (it's not a very good name and does imply in many people's minds a 'large bureaucratic body'). I define a Change Authority. This can be read as either a single person (as in 'he's the Authority on this matter') or a body (as in Port Authority). If that doesn't satisfy then describe a Change Approver and then the Change Approval Forum is a support group helping the change approver understand exactly what they are approving.

Yes, it's somewhat irritating to have to use 'safe' words, but better that than get embroiled in debating the merits of running a 'Board' because they're more focussed on the name than the function (I suspect that's Marc's problem).

bglangston's picture

Oh, yes. We're in violent agreement. I too try to avoid calling it a CCB. Unfortunately, I'm a contractor. Most of my clients over the years have insisted on a CCB by that name. They don't have the knowledge to know otherwise (the book says they must have a CCB ... and by gadfry, "We will have a CCB!" ("Oh, and my buddies will be the members because it will look good on their resumes/evaluation reports.")

I disagree about the Marc comment. I don't think it is confusion on his part, but lest I be putting words into his mouth...He has written several posts and articles (here in CMCrossroads) where he has expressed that CM is an archaic and restrictive practice. It logically follows that he would consider the CCB, an integral element of an effective CM program, also to be antequated and disruptive.

bglangston's picture

Marc,
At the risk of being nauseatingly redundant, "To that I can only repeat what I once heard - 'For those that choose to believe, no proof is necessary; for those that choose to not believe, no proof is enough.'"

You said:
[quote]This has never really led to a constructive exchange, which would require to accept a disagreement, respect a different opinion ...[/quote]
First, no one has to accept or respect a disagreement. One only must recognize that there is one. I think we can all accept that you disagree with our point of view. More to the point, I think it is you who cannot accept that we disagree with your point of view.

In order for a "constructive exchange" to take place, there must first be agreement on the basis of the argument. Here, the argument is about the existence of the basis. In this instance, the disagreement between you and me is not about the fulcrum, or about where to put it, or about any of its characteristics. Rather, the disagreement is about whether or not it (CM) should exist.

I once witnessed an argument about the existence of God. The believer used all sorts of biblical quotes to prove his point of view. After a while I pointed out to him that he was using quotes from a Person or Entity that to the nonbeliever did not exist. It is much like using quotes from Prester John to create a map to his palace. It logically follows that if one does not accept the existence of Prester John, then one cannot accept the validity of his quotes or the resulting map.

Next, it is not difficult to discount your "disagreement." So far, your arguments have been about the benefits of your so called "SCM" and the detriments of "CM."

As others before me, I ask, "What is this SCM of which you speak?" You have never described it or defined it. The best you have done is sort of dance around its benefits and told us what it is NOT. In fact, if memory serves, you have stated that it cannot be defined because to define it would be to destroy the very lack of structure necessary for its existence. Did you ever notice in the story of "The Emperor's New Clothes" that everyone described the clothes only in qualitative terms -- beautiful, elegant, bright, light, airy, high quality.

As for me, unless and until you can provide something meaningful to argue and provide it in meaningful terms, your disagreement will be noticed, but not seriously considered (or not respected, if you prefer).

CMCrossroads is a TechWell community.

Through conferences, training, consulting, and online resources, TechWell helps you develop and deliver great software every day.