David S. Read

I routinely receive emails, tweets and snail mail from IT vendors that focus on how their solution accelerates the creation of business applications. They will quote executives and technology leaders, citing case studies that compare the time to build an application on their platform versus others. They will make the claim that this speed to release proves that their platform, tool or solution is “better” than the competition. Further, they claim that it will provide similar value for my business’ application needs. The focus of these advertisements is consistently, “how long did it take to initially create some application.”

This speed-to-create metric is pointless for a couple of reasons. First, an experienced developer will be fast when throwing together a solution using his or her preferred tools. Second, an application spends years in maintenance versus the time spent to build its first version.

Build it fast!

Years ago I built applications for GE in C. I was fast. Once I had a good set of libraries, I could build applications for turbine parts catalogs in days. This was before windowing operating systems. There were frameworks from companies like Borland that made it trivial to create an interactive interface. I moved on to Visual Basic and SQLWindows development and was equally fast at creating client-server applications for GE’s field engineering team. I progressed to C++ and created CGI-based web applications. Again, building and deploying applications in days. Java followed, and I created and deployed applications using the leading edge Netscape browser and Java Applets in days and eventually hours for trivial interfaces.

Since 2000 I’ve used BPM and BRM platforms such as PegaRULES, Corticon, Appian and ILOG. I’ve developed applications using frameworks like Struts, JSF, Spring, Hibernate and the list goes on. Through all of this, I’ve lived the euphoria of the initial release and the pain of refactoring for release 2. In my experience not one of these platforms has simplified the refactoring of a weak design without a significant investment of time.

Speed to initial release is not a meaningful measure of a platform’s ability to support business agility. There is little pain in version 1 regardless of the design thought that goes into it. Agility is about versions 2 and beyond. Specifically, we need to understand what planning and practices during prior versions is necessary to promote agility in future versions.

(more…)

Michael Delaney, a senior consulting software engineer at Blue Slate, commented on my previous posting.  As I created a reply I realized that I was expanding on my reasoning and it was becoming a bit long.  So, here is my reply as a follow-up posting.  Also, thank you to Michael for helping me think more about this topic.

I understand the desire to rely on unit testing and its ability to find issues and prevent regressions.  For TDD, I’ll need to write separately.  Fundamentally I’m a believer in white box testing.   Black box approaches, like TDD, seem to be of relatively little value to the overall quality and reliability of the code.  Meaning, I’d want to invest more effort in white box testing than in black box testing.

I’m somewhat jaded, being concerned with the code’s security, which to me is strongly correlated with its reliability.  That said, I believe that unit testing is much more constrained as compared to formal reviews.  I’m not suggesting that unit tests be skipped, rather that we understand that unit tests can catch certain types of flaws and that those types are narrow as compared to what formal reviews can identify.

(more…)

Last week I was participating in a formal code review (a.k.a. code inspection) with one of our clients.  We have been working with this client, helping them strengthen their development practices.  Holding formal code reviews is a key component for us.  Part of the formal process we introduced includes reviewing the unit testing results, both the (successful) output report and the code coverage metrics.

At one point we were reviewing some code that had several error handling blocks that were not being covered in the unit tests.  These blocks were, arguably, unlikely or impossible to reach (such as a Java StringReader throwing an IOException).  There was some discussion by the team about the necessity of mocking enough functionality to cover these blocks.

Although we agreed that some of the more esoteric error conditions weren’t worth the programmer’s time to mock-up, it occurred to me later that we were missing an important point.  What mattered was that we were holding a formal code review and looking at those blocks of code.

Let me take a step back.  In 1986, Capers Jones published a book entitled Programming Productivity.  Although dated, the book contains many excellent points that cause you think about how to create software in an efficient way.  Here efficiency is not about lines of code per unit of time, but more importantly, lines of correct code per unit of time.  This means taking into account rework due to errors and omissions.

One if the studies presented in the book relates to identifying defects in code.  It is a study whose results seem obvious when we think about them.  However, we don’t always align our software development practices to leverage the study’s lessons and maximize our development efficiency.  Perhaps we believe that the statistics have changed due to language construct, experience, tooling and so forth.  We’d need similar studies to the ones presented by Capers Jones in order to prove that, though.

Below are a few of the actions from the book’s study of defect detection approaches.  I’ve skipped the low end and high-end numbers that Caper’s includes, simply giving the modes (averages) which are a good basis for comparison:

(more…)

One of the aspects of agile software development that may lead to significant angst is the database.  Unlike refactoring code, the refactoring of the database schema involves a key constraint – state!  A developer may rearrange code to his or her heart’s content with little worry since the program will start with a blank slate when execution begins.  However, the database “remembers.”  If one accepts that each iteration of an agile process produces a production release then the stored data can’t be deleted as part of the next iteration.

The refactoring of a database becomes less and less trivial as project development continues.  While developers have IDE’s to refactor code, change packages, and alter build targets, there are few tools for refactoring databases.

My definition of a database refactoring tool is one that assists the database developer by remembering the database transformation steps and storing them as part of the project – e.g. part of the build process.  This includes both the schema changes and data transformations.  Remember that the entire team will need to reproduce these steps on local copies of the database.  It must be as easy to incorporate a peer’s database schema changes, without losing data, as it is to incorporate the code changes.

These same data-centric complexities exist in waterfall approaches when going from one version to the next.  Whenever the database structure needs to change, a path to migrate the data has to be defined.  That transformation definition must become part of the project’s artifacts so that the data migration for the new version is supported as the program moves between environments (test, QA, load test, integrated test, and production).  Also, the database transformation steps must be automated and reversible!

That last point, the ability to rollback, is a key part of any rollout plan.  We must be able to back out changes.  It may be that the approach to a rollback is to create a full database backup before implementing the update, but that assumption must be documented and vetted (e.g. the approach of a full backup to support the rollback strategy may not be reasonable in all cases).

This database refactoring issue becomes very tricky when dealing with multiple versions of an application.  The transformation of the database schema and data must be done in a defined order.  As more and more data is stored, the process consumes more storage and processing resources.  This is the ETL side-effect of any system upgrade.  Its impact is simply felt more often (e.g. potentially during each iteration) in an agile project.

As part of exploring semantic technology, I am interested in contrasting this to a database that consists of RDF triples.  The semantic relationships of data do not change as often (if at all) as the relational constructs.  Many times we refactor a relational database as we discover concepts that require one-to-many or many-to-many relationships.

Is an RDF triple-based database easier to refactor than a relational database?  Is there something about the use of RDF triples that reduces the likelihood of a multiplicity change leading to a structural change in the data?  If so, using RDF as the data format could be a technique that simplifies the development of applications.  For now, let’s take a high-level look at a refactoring use case.

(more…)

Last week I had the pleasure of attending Semantic Arts’ training class entitled, “Designing and Building Business Ontologies.”  The course, led by Dave McComb and Simon Robe, provided an excellent introduction to semantic technologies and tools as well as coverage of ontological best practices.  I thoroughly enjoyed the 4-day class and achieved my principle goals in attending; namely to understand the semantic web landscape, including technologies such as RDF, RDFS, OWL, SPARQL, as well as the current state of tools and products in this space.

Both Dave and Simon have a deep understanding of this subject area.  They also work with clients using this technology so they bring real-world examples of where the technology shines and where it has limitations.  I recommend this class to anyone who is seeking to reach a baseline understanding of semantic technologies and ontology strategies.

Why am I so interested in semantic web technology?  I am convinced that structuring information such that it can be consumed by systems, in ways more automated than current data storage and association techniques allow, is required in order to achieve any meaningful advancement in the field of information technology (IT). Whether wiring together web services or setting up ETL jobs to create data marts, too much IT energy is wasted on repeatedly integrating data sources; essentially manually wiring together related information in the absence of the computer being able to wire it together autonomously!

(more…)

I recently overheard a development leader talking with a team of programmers about the trade-off between the speed of developing working code and the effort required to improve the run-time performance of the code.  His opinion was that it was not worth any extra effort to gain a few hundred milliseconds here or there.  I found myself wanting to debate the position but it was not the right venue.

In my opinion a developer should not write inefficient code just because it is easier.  However, a developer must not tune code without evidence that the tuning effort will make a meaningful improvement to the overall efficiency of the application.  Guessing at where the hotspots are in an application usually leads to a lot of wasted effort.

When I talk about designing and writing efficient code I am really stressing the process of thinking about the macro-level algorithm that is being used.  Considering efficiency (e.g. big-O) and spending some time looking for options that would represent a big-O step change is where design and development performance effort belongs.

For instance, during initial design or coding, it is worth finding an O(log n) alternative to an O(n) solution.  However, spending time searching for a slight improvement in an O(n) algorithm that is still O(n) is likely a waste of time.

Preemptive tuning is a guessing game; we are guessing how a compiler will optimize our code, when a processor will fetch and cache our executable and where the actual hotspots will be.  Unfortunately our guesses are usually wrong. Perhaps the development team lead was really talking about this situation.

The tuning circumstances change once we have an application that can be tested.  The question becomes how far do we go to address performance hotspots?  In other words, how fast is fast enough?  For me the balance being sought is application delivery time versus user productivity and the benefits of tuning can be valuable. (more…)