Category: Software Engineering

Great article on distinction between Software Engineering and Software Development


Even though this article is five years old, it is just as salient and relevant today as then.  The author’s main argument is that software development is applied software engineering.  What a great way to think about the two concepts!!!  I think he’s exactly spot on, but it’s hard to get to this idea because of our experience of engineering is applied physics and physics is applied math.  I’ve never thought about applied engineering!  I always thought it should be applied computer science, but I was never comfortable with that idea.  I rarely use concepts from CS.  Maybe algorithms, but that’s about it.  This really fits!!  I encourage you to read this article if you’re interested in how our field is shaping up in this century!

http://www.ibm.com/developerworks/rational/library/dec05/pollice/index.html

 

CMMI version 1.3 and Agile


Well, the Software Engineering Institute (SEI) just published the newest version of the venerated CMMI and there are many side notes in the CMMI-DEV version that touch on using an Agile methodology with CMMI.  I’ve always thought this was possible and I’m glad to see the SEI has taken the lead.

For example, here is the “Agile note” in the configuration management section:

In Agile environments, configuration management (CM) is important because of the need to support frequent change, frequent builds (typically daily), multiple baselines, and multiple CM supported workspaces (e.g., for individuals, teams, and even for pair-programming). Agile teams may get bogged down if the organization doesn’t: 1) automate CM (e.g., build scripts, status accounting, integrity checking) and 2) implement CM as a single set of standard services. At its start, an Agile team should identify the individual who will be responsible to ensure CM is implemented correctly. At the start of each iteration, CM support needs are re-confirmed. CM is carefully integrated into the rhythms of each team with a focus on minimizing team distraction to get the job done.

I especially like this one from the Requirements Development section:

In Agile environments, customer needs and ideas are iteratively elicited, elaborated, analyzed, and validated. Requirements are documented in forms such as user stories, scenarios, use cases, product backlogs, and the results of iterations (working code in the case of software). Which requirements will be addressed in a given iteration is driven by an assessment of risk and by the priorities associated with what is left on the product backlog. What details of requirements (and other artifacts) to document is driven by the need for coordination (among team members, teams, and later iterations) and the risk of losing what was learned. When the customer is on the team, there can still be a need for separate customer and product documentation to allow multiple solutions to be explored. As the solution emerges, responsibilities for derived requirements are allocated to the appropriate teams. (See ―Interpreting CMMI When Using Agile Approaches‖ in Part I.)

Download the latest CMMI-DEV and enjoy!

Software Engineering isn’t dead yet, BUT….


Tom DeMarco wrote an article a couple of years ago proclaiming that Software Engineering was dead that caused a real stir in the software field.  I opposed his view then; but sadly I’m beginning to think he had seen a glimpse of the future.

Today I did my last search for SWEBOK (the Software Engineering Book of Knowledge) in hopes of actually finding something coming out of this incredibly slow and opaque group.  My expectations were met, as they once again neglected to tell anyone where exactly they are in their unpublished process.  If any democratic process was run like this, we’d see Wisconsin style protests.  But alas, this is a dying beast that is finding its way to the bottom of irrelevance.

I think it started with the Agile Manifesto.  Almost none of the signatories came from Academia and it has shown.  They don’t publish in IEEE Software.  They don’t need to.  They get things done.  Which, of course, is what many academics in Software Engineering can’t do (see Exihibit A – SWEBOK).  I don’t want to paint too wide of a stroke.  There are of course exceptions like Mary Shaw and Tao Xie.  But many academics published articles are much like science fiction, somewhat entertaining, but not really useful.

This stems from the fact that most are tenured and have never worked a day in their life as a REAL software developer.  Its like writing about sex while still a virgin.  Exciting for the writer, but woefully invaluable for anyone else.  Let us look at the present month’s edition of IEEE Software, their leading journal. Oh wait, that will be $20 an article (because we all know they are just that valuable).  A requirements article, that looks potentially useful.  Oh it’s about GORE and SORE.  You don’t know about GORE and SORE !?! Oh you better get in the know; because that’s the new rage in software development.  Yeah, Dr. Seuss invented the terms.  Wait, an article on Architecture, now that should be promising.  “Unusable Software is…”  wait for it….. “Useless”.  Woah!!!  What a find!  20 years from now we will definitely be talking about this one.  And wait, you get more,  it’s only Part I!!!!

We don’t even use the term Software Engineering any more in the field where we actually create software.  We call it software development.  We can’t even agree on the same term!  IEEE Software went downhill after Steve McConnell (someone who HAS written real software) left and it may have been an omen that Software Engineering is going down too.

 

Code Reviews


Code reviews are a best practice developed back in the early 1970’s by Michael Fagan.  He was a computer hardware engineer by training and knew from experience that reviews in his previous field found quite a few defects before a hardware spec was sent to the factory to be burned onto silicon.  It was expensive to undo defects once the spec had been sent to the factory and reviews were a great tool in catching defects early.  His original paper describes the “Fagan Inspection”.

As time has gone on, more papers and books have been published on the subject; most notably Handbook of Walkthroughs, Inspections, and Technical Reviews by Daniel Freedman and Gerald Weinberg in 1990.  The best current source for how to do reviews is the IEEE Standard 1028 for Software Reviews and Audits.  Regrettably, the IEEE does not make their standards free to access, which I think diminishes their use and ultimately their importance.  In it they describe four types of reviews: a Management Review, Technical Review, Inspection, and Walkthrough.  I have taken these four descriptions and amalgamated them into a review procedure that, I think, brings the four procedures together nicely.  I post it here for your edification and use.

Code Review Policy & Procedure

From Art Form to Engineering Discipline?


This is the title of a new paper that is to be published in the IEEE Annals of the History of Computing by Christopher McDonald.  This paper starts to answer how software processes became so formal and heavy-handed in the 70’s and 80’s.  It is interesting to see Barry Boehm advocate waterfall in the 70’s and then back away from it in the 80’s with his spiral process.  It is also interesting to see that iterative software was alive and well in the late 80’s and the Agile movement did not invent it (as many Agile zealots will want you to believe).

Following this paper’s narrative; you can see that Software Engineering was born out of the so-called “Software Crisis” of the late 60’s and early 70’s.  The answer was scientific management espoused by Frederick Winslow Taylor.  This did not work as they were comparing apples and oranges.  Taylor was dealing with the mindless work of blue-collar factories workers; while software is engaged in by highly educated, creative white collar workers.

It seems that a lot the first ideas of “Software Engineering” came from this factory analogy.  Interchangeable parts and software components is another example.  There is a reference to the quality movement in the early 1900’s that I’d like to explore more fully.

Overall, this is a good read and elucidating history of Software Engineering.  As the military’s influence on software development wanes; it becomes less relevant.  But the history shows you how we got to where we are today in software development.

CAST Study shows software industry still has a long way to go


CAST’s recently released “Worldwide Application Software Quality Study” is the first study to give real numbers, based on a good amount of data, to code quality.  This is a notoriously difficult problem to address given the millions of lines of code most applications have.  They do a very admirable job of not going too far in their assumptions; but also delivering valuable results.

A key finding for me was that code quality does not go down with increasing lines of code (p. 10); unless you are using COBOL.  Thankfully I don’t :).  What did cause code quality to go down in modern languages (Java, .Net, etc.) was module fan-out.  It seems that David Parnas’ theory of modularity has won out in modern languages; as it is encouraged in the language design.  But this has produced the un-intended consequence of developers creating too many complex dependencies between modules.

The study also puts a price tag on maintaining code that all bean-counters will love.  But is also gives real research for developers to point to for getting more time and money to re-factor code.  I highly recommend this study.