Don’t be a Slave to Your Tools

Abstract SlaveDevelopers attach quickly to tools because they are concrete and have well defined behavior.  It is easier to learn a tool than to learn good practices or methodology.

Tools only assist in solving problems, they can’t solve the problem by themselves. A developer who understands the problem can use tools to increase productivity and quality.

Poor developers don’t invest the time or effort to understand how to code properly and avoid defects.  They spend their time learning how to use tools without understanding the purpose of the tool or how to use it effectively.

To some degree, this is partially the fault of the tool vendors.  The tool vendors perceive an opportunity to make $$$$$ based on providing support for a common problems, such as:

  • defect trackers to help you manage defect tracking
  • version control systems to manage source code changes
  • tools to support Agile development (Version One, JIRA)
  • debuggers to help you find defects

There are many tools out there, but let’s just go through this list and point out where developers and organizations get challenged.  Note, all statistics below are derived from over 15,000 projects over 40 years.1

Defect Trackers

Believe it or not, some organizations still don’t have defect tracking software. I’ve run into a couple of these companies and you would not believe why…

Inadequate defect tracking methods: productivity -15%, quality -21%

So we are pretty much all in agreement that we need to have defect tracking; we all know that the ability to manage more than a handful of defects is impossible without some kind of system.

Automated defect tracking tools: productivity +18%, quality +26%

The problem is that developers fight over which is the best defect tracking system. The real problem is that almost every defect tracking system is poorly set-up, leading to poor results. Virtually every defect tracking system when configured properly will yield tremendous benefits. The most common pitfalls are:

  • Introducing irrelevant attributes into the defect lifecycle status, i.e. creation of statuses like deferred, won’t fix, or functions as designed
  • Not being able to figure out if something is fixed or not
  • Not understanding who is responsible for addressing a defect

The tool vendors are happy to continue to provide new versions of defect trackers. However, using a defect tracker effectively has more to do with how the tool is used rather than which tool is selected.

One of the most fundamental issues that organizations wrestle with is what is a defect?  A defect only exists if the code does not behave according to specifications. But what if there are no specifications or the specifications are bad?  See It’s not a bug, it’s… for more information.

Smart organizations understand that the way in which the defect tracker is used will make the biggest difference.  Discover how to get more out of you defect tracking system in Bug Tracker Hell and How to Get Out.

Another common problem is that organizations try to manage enhancements and requirements in the defect tracking system.  After all whether it is a requirement or a defect it will lead to a code change, so why not put all the information into the defect tracker?  Learn why managing requirements and enhancements in the defect tracking system is foolish in Don’t manage enhancements in the bug tracker.

Version Control Systems

Like defect tracking systems most developers have learned that version control is a necessary hygiene procedure.  If you don’t have one then you are likely to catch a pretty serious disease (and at the least convenient time)

Inadequate change control: productivity -11%, quality -16%

Virtually all developers dislike version control systems and are quite vocal about what they can’t do with their version control system.  If you are the unfortunate person who made the final decision on which version control system is used just understand that their are hordes of developers out their cursing you behind your back.

Version control is simply chapter 1 of the story.  Understanding how to chunk code effectively, integrate with continuous build technology, and making sure that the defects in the defect tracker refers to the correct version are just as important as the choice of version control system.

Tools to support Agile

Sorry Version One and JIRA, the simple truth is that using an Agile tool does not make you agile, see this.

These tools are most effective when you actually understand Agile development. Enough said.

Debuggers

I have written extensively about why debuggers are not the best tools to track down defects.  So I’ll try a different approach here.

One of the most enduring sets of ratios in software engineering has been 1:10:100.  That is, if the cost of tracking down a defect pre-test (i.e. before QA) is 1, then it will cost 10x if the defect is found by QA, and 100x if the defect is discovered in deployment by your customers.

Most debuggers are invoked when the cost function is in the 10x or 100x part of the process.  As stated before, it is not that I do not believe in debuggers — I simply believe in using pre-test defect removal strategies because they cost less and lead to higher code quality.

Pre-test defect removal strategies include:

  • Planning code, i.e. PSP
  • Test driven development, TDD
  • Design by Contract (DbC)
  • Code inspections
  • Pair programming for complex sections of code

You can find more information about this in:

Seldom Used Tools

Tools that can make a big difference but many developers don’t use them:

Automated static analysis: productivity +21%, quality +31%

Automated unit testing: productivity +17%, quality +24%

Automated unit testing generally involves using test driven development (TDD) or data driven development together with continual build technology.

Automated sizing in function points: productivity +17%, quality +24%

Automated quality and risk prediction: productivity +16%, quality +23%

Automated test coverage analysis: productivity +15%, quality +21%

Automated deployment support: productivity +15%, quality +20%

Automated cyclomatic complexity computation: productivity +15%, quality +20%

Important Techniques with No Tools

There are a number of techniques available in software development that tool vendors have not found a way to monetize on. These techniques tend to be overlooked by most developers, even though they can make a huge difference in productivity and quality.

The Personal Software Process and Team Software Process were developed by Watts Humphrey, one of the pioneers of building quality software.

Personal software process: productivity +21%, quality +31%2

Team software process: productivity +21%, quality +31%3

The importance of inspections is covered in:

Code inspections: productivity +21%, quality +31%4

Requirement inspections: productivity +18%, quality +27%4

Formal test plans: productivity +17%, quality +24%

Function point analysis (IFPUG): productivity +16%, quality +22%

Conclusion

There is definitely a large set of developers that assume that using a tool makes them competent.

The reality is that learning a tool without learning the principles that underly the problem you are solving is like assuming you can beat Michael Jordan at basketball just because you have great running shoes.

Learning tools is not a substitute for learning how do do something competently. Competent developers are continually learning about techniques that lead to higher productivity and quality, whether or not that technique is supported by a tool.

References

VN:F [1.9.22_1171]
Rating: 0.0/5 (0 votes cast)
VN:F [1.9.22_1171]
Rating: 0 (from 0 votes)

Size Matters

Some say that software development is challenging because of complexity. This might be true, but this definition does not help us find solutions to reduce complexity. We need a better way to explain complexity to non-technical people.

The reality is that when coding a project size matters.  Size is measured in the number of pathways through a code base, not by the number of lines of code. Size is proportional to the number of function points in a project.

There are many IT people that succeed with programs of a certain size and then fail miserably when taking on programs that are more sophisticated.  Complexity increases with size because the number of pathways increase exponentially in large programs.

Virtually anyone (even CEOs 🙂 ) can build a hello, world! application; an application that only has a single pathway through it and is as simple as you can get.  Some CEOs write the simple hello, world! program and incorrectly convince themselves that development is easy. Hello, world! only has a single pathway through it and virtually anyone can write it.

main() {
printf( “hello, world” );
}

If you have an executive that can’t even complete hello,world then you should take away his computer 🙂

CallTreeComplexity Defined

As programs get more sophisticated, the number of decisions that have to be made increase and the depth of the call tree increases.  Every non-trivial routine will have multiple pathways through it.

If your average call depth is 10 with an average of 4 pathways through each routine then this represents over 1 million pathways.  If the average call depth is 15 then it represents 107 million pathways.

Increasing sophisticated programs have greater call depth than ever and distributed applications increase the call depth even because the call depth of a system is additive. This is what we mean by complexity; it is impossible for us to test all of the different pathways in a black box fashion.

Now in reality every combination of pathways is not possible, but you only have to leave holes in a few routines and you will have hundreds, if not thousands, of pathways where calculations and decisions can go wrong.

In addition, incorrect calculations or decisions higher up in the call tree can lead to difficult to find defects that may blow up much further away from the source of the problem.

What are Defects?

Software defects occur for very simple reasons, an incorrect calculation is performed that causes an output value to be incorrect.  Sometimes there is no calculation at all because input data is not validated to be consistent and that data is either stored incorrectly or goes on to cause incorrect calculations to be performed.

Call Tree and Defects, discoveredWe only recognize that we have a defect when we see an output value and recognize that it is incorrect. More likely QA sees it and tells us that we are incorrect.

Basically we follow a pathway that is correct through nodes 1, 2, 3, 4, and 5.  At point 6 we make a miscalculation calculation, and then we have the incorrect values at points 7 and 8 and discover the problem at node 9.

So once we have a miscalculation, we will either continue to make incorrect calculations or make incorrect decisions and go down the wrong pathways (where we will then make incorrect calculations).

Not all Defects are Equal

It is clear that the more distance there is between a miscalculation and its discover will make defects harder to detect.  The longer the call depth the greater the chance that there can be a large distance between the origin and detection, in other words:

Size Matters

Today we build sophisticated systems of many cooperating applications and the call depth is exponential with the size of the system.  This is what we mean by complexity in software.

Reducing Complexity

Complexity is reduced for every function where:

  • You can identify when inconsistent parameters are passed to a function
  • All calculations inside of a function are done correctly
  • All decisions through the code are taken correctly

The best way to solve all 3 issues is through formal  planning and development.Two methodologies that focus directly on planning at the personal and team level are the Personal Software Process (PSP) and the Team Software Process (TSP) invented by Watts Humphrey.

Identifying inconsistent parameters is easiest when you use Design By Contract (DbC) , a technique that was pioneered by the Eiffel programming language. It is important to use DbC on all functions that are in the core pathways of an application.

Using Test Driven Development is a sure way to make sure that all calculations inside of a function are done correctly, but only if you write tests for every pathway through a function.

Making sure that all calculations are done correctly inside a function and that correct decisions are make through the code is best done through through code inspections (see Inspections are not Optional and Software Professionals do Inspections).

All techniques that can be used to reduce complexity and prove the correctness of your program are covered in Debuggers are for Losers.  N.B. Debuggers as the only formalism will only work well for systems with low call depth and low branching.

Conclusion

Therefore, complexity in software development is about making sure that all the code pathways are accounted for.  In increasingly sophisticated software systems the number of code pathways increases exponentially with the call depth. Using formal methods is the only way to account for all the pathways in a sophisticated program; otherwise the number of defects will multiply exponentially and cause your project to fail.

Only projects with low complexity (i.e. small call depth) can afford to be informal and only use debuggers to get control of the system pathways. As a system gets larger only the use of formal mechanisms can reduce complexity and develop sophisticated systems. Those formal mechanisms include:

  • Personal Software Process and Team Software Process
  • Design by Contract (via Aspect Oriented Programming)
  • Test Driven Development
  • Code and Design Inspections
VN:F [1.9.22_1171]
Rating: 0.0/5 (0 votes cast)
VN:F [1.9.22_1171]
Rating: 0 (from 0 votes)

Defects are for Losers

A developer is responsible for using any and all techniques to make sure that he produces defect free code.  The average developer does not take advantage of all of the following opportunities to prevent and eliminate defects:

  1. Before the code is written
  2. As the code is written
  3. Writing mechanisms for early detection
  4. Before the code is executed
  5. After the code is tested

The technique that is used most often is #5 above and will not be covered here.  It involves the following:

  1. Code is delivered to the test department
  2. The test department identifies defects and notifies development
  3. Developer’s fire up the debugger and try to chase down the defect

Like the ‘rinse and repeat‘ process on a shampoo bottle, this process is repeated until the code is cleaned or until you run out of time and are forced to deliver.

The almost ubiquitous use of #5 leads to CIOs and VPs of Engineering assuming that the metric of one tester to two developers is a good thing.  Before assuming that #5 is ‘the way to go‘ consider the other techniques and statistical evidence of their effectiveness.

Before the Code is Written



A developer has the most options available to him before the code is written.  The developer has an opportunity to plan his code, however, there are many developers who just ‘start coding’ on the assumption that they can fix it later.

How much of an effect can planning have?  Two methodologies that focus directly on planning at the personal and team level are the Personal Software Process (PSP) and the Team Software Process (TSP) invented by Watts Humphrey.

PSP can raise productivity by 21.2% and quality by 31.2%

TSP can raise productivity by 20.9% and quality by 30.9%

Not only does the PSP focus on code planning, it also makes developers aware of how many defects they actually create.  Here are two graphs that show the same group of developers and their defect injection rates before and after PSP training.

Before PSP training After PSP training

The other planning techniques are:

  • Decision tables
  • Proper use of exceptions

Both are covered in the article Debuggers are for Losers and will not be covered here.

As the Code is Written

Many developers today use advanced IDEs to avoid common syntax errors from occurring   If you can not use such an IDE or the IDE does not provide that service then some of the techniques in the PSP can be used to track your injection of syntax errors and reduce those errors.

Pair Programming

One technique that can be used while code is being written is Pair Programming.  Pair programming is heavily used in eXtreme Programing (XP).  Pair programming not only allows code to be reviewed by a peer right away but also makes sure that there are two people who understand the code pathways through any section of code.

Pair programming is not cost effective overall (see Capers Jones).  For example, it makes little sense to pair program code that is mainly boiler plate, i.e. getter and setter classes. What does make sense is that during code planning it will become clear which routines are more involved and which ones are not.  If the cyclomatic complexity of a routine is high (>15) then it makes sense for pair programming to be used.

If used for all development, Pair Programming can raise productivity by 2.7% and quality by 4.5%

Test Driven Development




Test driven development (TDD) is advocated by Kent Beck and stated in 2003 that TDD encourages simple designs and inspires confidence.  TDD fits into the category of automated unit testing.

Automated unit testing  can raise productivity by 16.5% and quality by 23.7%

Writing Mechanisms for Early Detection

Defects are caused by programs either computing wrong values, going down the wrong pathway, or both.  The nature of defects is that they tend to cascade and get bigger the further in time and space between the source of the defect and the noticeable effects of the defect.

Design By Contract

One way to build checkpoints into code is to use Design By Contract (DbC), a technique that was pioneered by the Eiffel programming language   It would be tedious and overkill to use DbC in every routine in a program, however, there are key points in every software program that get used very frequently.

Just like the roads that we use have highways, secondary roads, and tertiary roads — DbC can be used on those highways and secondary roads to catch incorrect conditions and stop defects from being detected far away from the source of the problem.

Clearly very few of us program in Eiffel.  If you have access to Aspect Oriented Programming (AOP) then you can implement DbC via AOP. Today there are AOP implementations as a language extension or as a library for many current languages (Java, .NET, C++, PHP, Perl, Python, Ruby, etc).

Before the Code is Executed

Static Analysis

Most programming languages out there lend themselves to static analysis.  There are cost effective static analysis for virtually every language.

Automated static analysis can raise productivity by 20.9% and quality by 30.9%

Inspections


Of all the techniques mentioned above, the most potent pre-debugger technique is inspections. inspections are not sexy and they are very low tech, but the result of organizations that do software inspections borders on miraculous.The power of software inspections can be seen in these two articles:

Code inspections can raise productivity by 20.8% and quality by 30.8%

Design inspections can raise productivity by 16.9% and quality by 24.7%

From the Software Inspections book on p.22.

In one large IBM project, one half million lines of networked operating system, there were 11 development stages (document types: logic, test, user documentation) being Inspected.  The normal expectation at IBM, at that time, was that they would be happy only to experience about 800 defects in trial site operation.  They did in fact experience only 8 field trial defects.

Evidence suggests that every 1 hour of code inspection will reduce testing time by 4 hours

Conclusion

Overworked developers rarely have time to do research, even though it is clear that there is a wealth of information available on how to prevent and eliminate defects. The bottom line is that if your are only using technique #5 from the initial list, then you are not using every technique available to you to go after defects.My opinion only, but:

A professional software developer uses every technique at his disposal to prevent and eliminate defects

Other articles in the “Loser” series

Want to see more sacred cows get tipped? Check out:

Make no mistake, I am the biggest “Loser” of them all.  I believe that I have made every mistake in the book at least once 🙂

References

Gilb, Tom and Graham, Dorothy. Software Inspections

Jones, Capers. SCORING AND EVALUATING SOFTWARE METHODS, PRACTICES, AND RESULTS. 2008.

Radice, Ronald A. High Quality Low Cost Software Inspections.

VN:F [1.9.22_1171]
Rating: 0.0/5 (0 votes cast)
VN:F [1.9.22_1171]
Rating: 0 (from 0 votes)

Testing departments are for Losers

Loser, smallI understand that this is a very strong statement and I’m definitely not trying to insult anyone; so apologies in advance.  I’m trying to challenge the belief that testing is mandatory and that there should be one testing resource for every two developers.  Quality development is about quality assurance and zero defects not about having testing departments.

One testing resource for every two developers is not a good solution

Quality Assurance (QA) is about making a positive statement about product quality.  QA is about positive assurance, which is stating,  “We are certain that there are few, if any, defects in this product.”  Airplanes are a product with quality assured, the manufacturers will stand by their quality and the statistics back them up.  Contrast this with this article or this article which wonders what would happen if Microsoft made airplanes — would you fly in them?

The reality is that most testing departments simply discover defects and forward them back to the engineering department to fix.  By the time the software product gets released we are basically saying, “We got rid of as many defects as we could find before management forced us to release this product, however, we really have no idea how many other defects are in the code”.  This is not assuring quality; at best you get negative assurance out of this.

Everyone understand that buggy software kills sales (and start-ups :-)), however, testing is often an after thought in many organizations.  When software products take longer than expected they are  forwarded to the testing department.  The testing department is often expected to test and bless code in less time than allocated.

To compound problems, many testing departments don’t even receive proper requirements against which to test the code and/or sufficient tools to work with. Large testing departments and/or large amounts of manual testing are not healthy or efficient.

Humphrey Watts was emphatic that calling defects “bugs” trivializes the issue and downplays the negative impact that defects cause on a development organization.

Calling defects “bugs” trivializes an important issue

Goblins and Elves

Defects are not introduced into software by goblins and elves.  Defects are injected into the code by developers that:

  • don’t understand the requirements or architecture
  • misunderstand how to use their peer’s components
  • misunderstand 3rd party libraries
  • having a bad day because of home troubles or work environment
  • are careless because someone else will test their code

Defects are injected by the team

No one is more aware of how code can break down than the developer who writes it.   Any line of code that is written without concentration and planning becomes a potential defect.  It is impossible for testers to understand every pathway through the code and make sure that every possible combination of variables is properly taken care of.

There are many techniques that can increase code quality and dramatically reduce the amount of testing that is necessary:

Test Driven Development

Properly written tests require a developer not only to think about what a code section is supposed to do but also plan how the code will be structured.  If you know that there are  five pathways through the code then you will write five tests ahead of time.  A common problem is that you have coded n paths through the code when there are n+1 conditions.

TDD is white box testing and can reach every pathway that the developer codes.  TDD is proactive and can test pathways from end to end, it does not just have to be used for unit testing.  When TDD is hooked up to a continuous integration engine then defects are located and fixed before they make it to testing.

Database Driven Testing

Using actual test data to test existing routines during development is an excellent way to make sure that there are fewer production problems.  The test data needs to be a copy (or subset) of production data.

Database driven testing can also be hooked up to a continuous integration engine and prevent defects from getting to testing.

Design By Contract

The Eiffel programming language introduced design by contract (DbC).  DbC isDbC orthogonal to TDD because its goal is to ensure that the contract defined by the preconditions and postconditions for each function call is not violated.  DbC can be used in virtually any language for with their is an Aspect Oriented Programming (AOP) solution.

During development, the minute a developer violates the expected contract of any function (his or a peers) then the developer will get feedback to fix the problem before it gets to testing.

Inspections

Since the 1970s we have statistical evidence that one of the best ways to eliminate defects from code is through inspections.  Inspections can be applied to the requirements, design, and code artifacts and projects that use inspections can eliminate 99% of the defects injected into the code.  Se Inspections are not Optional and Software Professionals do Inspections.

Each hour of inspections will save you 4 hours of testing

Pair Programming

Pair programming can be selectively used to prevent and eliminate defects from code.   When developers work in pairs they not only review code as quickly as possible but also learn productivity techniques from each other.  Pair programming should only be done on complex sections of code.

Pair programming not only eliminates defects but allows developers to get enough feedback that they can prevent defects in the future.

Minimizing Cyclomatic Complexity

There is evidence that routines with high cyclomatic complexity will have more latent defects than other routines.   This makes sense because the number of code pathways goes up dramatically as cyclomatic complexity increases and increases the chance that the developer does not handle all of them.   In most cases, testing departments can not reproduce all of the pathways in routines of high cyclomatic complexity.

Use Dynamic and Static Code Checking

There are many code problems caused by a careless use of pointers and other powerful language constructs.  Many of these problems can be detected by having the development team use dynamic and static code checking problems.

Proper Code Planning Techniques

There are developers that try to write code at the keyboard without planning, which is neither efficient nor effective.  This is like having to do errands in 5 locations and driving to the locations randomly — you might get your errands done, but odds are it won’t be efficient.

Watts Humphrey talks directly to the idea of planning in the Personal Software Process.  In addition techniques like diagramming with UML or using decision tables can go a long way to thinking through code structure before it is implemented.

Conclusion

Developers are the ones who inject defects into the code and therefore they are the best line of defense to remove them.  The developer has the best information on what needs to be tested in his code at the time that he writes it.  The longer it takes for testing or a customer to discover a code defect the longer the developer will spend in a debugger chasing down the problem.

PreventEliminateDefectsDevelopers need to be trained in preventing and eliminating defects.  Developers who learn to get the code correct the first time will reduce and eliminate effort in testing.

The goal of development should be to catch defects early; this is the only way to assure quality.  Hence quality assurance starts and finishes in the development department, not the testing department.

VN:F [1.9.22_1171]
Rating: 4.7/5 (3 votes cast)
VN:F [1.9.22_1171]
Rating: -2 (from 2 votes)

Inspections are not Optional

Every developer is aware that code inspections are possible, some might have experienced the usefulness of code inspections, however, the fact is that inspections are not optional.

Without inspections your defect removal rate will stall out at 85% of defects removed; with inspections defect removal rates of 97% have been achieved.

Code inspections are only the most talked about type of inspection; the reality is that all artifacts from all phases of development should be inspected prior to being used.  Inspections are necessary because software is intangible and it is not once everything is coded that you want to notice problems.

In the physical world it is easier to spot problems because they can be tangible.  For example, if you have specified marble tiles for your bathroom and you see the contractor bring in a pile of ceramic tiles then you know something is wrong.  You don’t need the contractor to install the ceramic tiles to realize that there is a problem.

In software, we tend to code up an entire set of functionality, demonstrate it, and then find out that we have built the wrong thing!  If you are working in a domain with many requirements then this is inevitable, however, many times we can find problems through inspection before we create the wrong solutions

Let’s look at some physical examples and then discuss their software equivalents.

Requirement Defects

The requirements in software design are equivalent to the blueprints that are given to a contractor.  The requirements specify the system to be built, however, if those requirements are not inspected then you can end up with the following:

Balcony no Door Missing Landing Chimney Covering Window
Stairs Displaced Stairs to Ceiling Door no Balcony/th>

All of the above pictures represent physical engineering failures.  Every one of these disasters could have been identified in the blueprints if a simple inspection had been done. Clearly it must have become clear to the developers that the building features specified by the requirements were incompatible, but they completed the solution anyways.

Balcony no Door

This design flaws can be caused by changing requirements; here there is a balcony feature that has no access to it.  In Balcony no Door it is possible that someone noticed that there was sufficient room for a balcony on the lower floor and put it into one set of plans. The problem was that the developers that install the sliding doors did not have their plans updated.

Here the changed requirement did not lead to an inspection to see if there was an inconsistency introduced by the change.

Door no Balcony

In Door no Balcony something similar probably happened, however, notice that the two issues are not symmetric.  In Balcony no Door represents a feature that is inaccessible because no access was created, i.e. the missing sliding door.  In Door no Balcony we have a feature that is accessible but dangerous if used.

In this case a requirements inspection should have turned up the half implemented feature and either: 1) the door should have been removed from the requirements, or 2) a balcony should have been added.

Missing Landing

The Missing Landing occurs because the requirements show a need for stairs, but does not occur to the architect that there is a missing landing.  Looking at a set of blueprints gives you a two dimensional view of the plan and clearly they drew in the stairs.  To make the stairs usable requires a landing so that changing direction is simple.  This represents a missing requirement that causes another feature to be only partially usable.

This problem should have been caught by the architect when the blueprint was drawn up. However, barring the architect locating the issue a simple checklist and inspection of the plans would have turned up the problem.

Stairs to Ceiling

The staircase goes up to a ceiling and therefore is a useless feature.  Not only is the feature incomplete because it does not give access to the next level but also they developers wasted time and effort in building the staircase.

If this problem had been caught in the requirements stage as an inconsistency then either the staircase could have been removed or an access created to the next floor.  The net effect is that construction starts and the developers find the inconsistency when it is too late.

At a minimum the developers should have noticed that the stairway did not serve any purpose and not build the staircase which was a waste of time and materials.

Stairs Displaced

Here we have a clear case of changed requirements.  The stairs were supposed to be centered under the door, in all likelihood plans changed the location of the door and did not move the dependent feature.

When the blueprint was updated to move the door the designer should have looked to see if there was any other dependent feature that would be impacted by the change.

Architectural Defects

Architectural defects come from not understanding the requirements or the environment that you are working with.  In software, you need to understand the non-functional requirements behind the functional requirements — the ilities of the project (availability, scalability, reliability, customizability, stability, etc).

Architectural features are structural and connective.  In a building the internal structure must be strong enough to support the building, i.e. foundation and load bearing walls.

insufficient foundation Insufficient structure Connectivity problem

Insufficient Foundation

Here the building was built correctly, however, the architect did not check the environment to see if the foundation would be sufficient for the building.  The building is identical to the building behind it, so odds are they just duplicated the plan without checking the ground structure.

The equivalent in software is to design for an environment that can not support the functionality that you have designed.  Even if that functionality is perfect, if the environment doesn’t support it you will not succeed.

Insufficient Structure

Here the environment is sufficient to hold up the building, however, the architect did not design enough structural strength in the building.

The equivalent in software design is to choose architectural components that will not handle the load demanded by the system.  For example, distributed object technologies such as CORBA, EJB, and DCOM provided a way to make objects available remotely, however, the resulting architectures did not scale well under load.

Connectivity Problem

Here a calculation error was made when the two sides of the bridge were started.  When development got to the center they discovered that one side was off and you have an ugly problem joining the two sides.

The equivalent for this problem is when technologies don’t quite line up and require awkward and ugly techniques to join different philosophical approaches.

In software, a classic problem is mapping object-oriented structures into relational databases.  The philosophical mismatch accounts for large amounts of code to translate from one scheme into the other.

Coding Defects

Coding defects are better understood (or at least yelled about :-), so I won’t spend too much time on them.  Classic coding defects include:

  • Uninitialized data
  • Uncaught exceptions
  • Memory leaks
  • Buffer overruns
  • Not freeing up resources
  • Concurrency violations
  • Insufficient pathways, i.e. 5 conditions but only 4 coded pathway

Many of these problems can be caught with code inspections.

Testing Defects

Testing defects occur when the test plan flags a defect that is a phantom problem or a false positive. This often occurs when requirements are poorly documented and/or poorly understood and QA perceives a defect when there is none.

The reverse also happens where requirements are not documented and QA does not perceive a defect, i.e. false negative.

Both false positives and negatives can be caught by inspecting the requirements and comparing them with the test cases.

False positives slow down development. False negatives can slip through to your customers…

Root Cause of Firefighting

When inspections are not done in all phases of software development there will be fire-fighting in the project in the zone of chaos.  Most software organizations only record and test for defects during the Testing phase.  Unfortunately, at this point you will detect defects in all previous phases at this point.

QA has a tendency to assume that all defects are coding defects — however, the analysis of 18,000+ projects does not confirm this.  In The Economics of Software Quality, Capers Jones and Olivier Bonsignour show that defects fall into different categories. Below we give the category, the frequency of the defect, and the business role that will address the defect.

Note, only the bolded rows below are assigned to developers.
Defect Role Category Frequency Role
Requirements defect 9.58% BA/Product Management
Architecture or design defect 14.58% Architect
Code defect 16.67% Developer
Testing defect 15.42% Quality Assurance
Documentation defect 6.25% Technical Writer
Database defect 22.92% DBA
Website defect 14.58% Operations/Webmaster

Notice that fully 25% of the defects (requirements, architecture) occur before coding even starts.  These defects are just like the physical defects shown above and only manifest themselves once the code needs to be written.

It is much less expensive to fix requirements and architecture problems before coding.

Also, only about 54% of defects are actually resolvable by developers, so by assigning all defects to the developers you will waste time 46% of the time when you discover that the developer can not resolve the issue.

Fire-fighting is basically when there need to be dozens of meetings that pull together large numbers of people on the team to sort out inconsistencies.  These inconsistencies will lie dormant because there are no inspections.  Of course, when all the issues come out, there are so many issues from all the phases of development that it is difficult to sort out the problem!

Learn how to augment your Bug Tracker to help you to understand where your defects are coming from in Bug Tracker Hell and How to Get Out!

Solutions

There are two basic solutions to reducing defects:

  1. Inspect all artifacts
  2. Shorten your development cycle

The second solution is the one adopted by organizations that are pursuing Agile software development.  However, shorter development cycles will reduce the amount of fire-fighting but they will only improve code quality to a point.

In The Economics of Software Quality the statistics show that defect removal is not effective in most organizations.  In fact, on large projects the test coverage will drop below 80% and the defect removal efficiency is rarely above 85%.   So even if you are using Agile development you will still not achieving a high level of defect removal and will be limited in the software quality that you can deliver.

Agile development can reduce fire-fighting but does not address defect removal

Inspect All Artifacts

Organizations that have formal inspections of all artifacts have achieved defect removal efficiencies of 97%!  If you are intent on delivering high quality software then inspections are not optional.

Of course, inspections are only possible for phases in which you have actual artifacts. Here are the artifacts that may be associated with each phase of development:

Phase Artifact
Requirements use case, user story, UML Diagrams (Activity, Use Case)
Architecture or design UML diagrams (Class, Interaction, Deployment)
Coding UML diagrams (Class, Interaction, State, Source Code)
Testing Test plans and cases
Documentation Documentation
Database Entity-Relationship diagrams, Stored Procedures

Effective Inspections

Inspections are only effective when the review process involves people who know what they are looking for and are accountable for the result of the inspection.  People must be trained to understand what they are looking for and effective check lists need to be developed for each artifact type that you review, e.g. use case inspections will be different than source code reviews.

Inspections must have teeth otherwise they are a waste of time.  For example, one way to put accountability into the process is to have someone other than the author be accountable for any problems found.  There are numerous resources available if you decide that you wish to do inspections.

Conclusion

The statistics overwhelming suggest that inspections will not only remove defects from a software system but also prevent defects from getting in.  With inspections software defect removal rates can achieve 97% and without inspections you are lucky to get to 85%.

Since IBM investigated the issue in 1973, it is interesting to note that teams trained in performing inspections eventually learn how to prevent injecting defects into the software system.  Reduced injection of defects into a system reduces the amount of time spent in fire-fighting and in QA.

You can only inspect artifacts that you take the time to create.  Many smaller organizations don’t have any artifacts other than their requirements documents and source code. Discover which artifacts need to be created by augmenting your Bug Tracker (see Bug Tracker Hell and How To Get Out!).   Any phase of development where significant defects are coming from should be documented with an appropriate artifact and be subject to inspections.


Good books on how to perform inspections:


All statistics quoted from The Economics of Software Quality by Capers Jones and Olivier Bonsignour:

Capers Jones can be reached by sending me an email: Dalip Mahal


VN:F [1.9.22_1171]
Rating: 0.0/5 (0 votes cast)
VN:F [1.9.22_1171]
Rating: 0 (from 0 votes)