SoylentNews

Cowboy Engineers in the Software Industry

posted by janrinok on Sunday March 02 2014, @11:59PM   
from the was-it-much-harder-in-my-day? dept.

shabadoo writes:

"Will software engineering always be a cowboy's game? Or is it just a case of when you're a passionate expert the pimples stand out more clearly. This guy has clearly had enough. His vents are amusing, but also raise some good points about the state of the industry."

• What do you expect? What do you expect? (Score: 5, Insightful) by prospectacle on Monday March 03 2014, @01:04AM

  by prospectacle (3422) on Monday March 03 2014, @01:04AM (#9848) Journal

  Some software can kill people if it goes wrong. Most of it can't, it will just waste time and money (which it may have been doing anyway).

  Therefore it's not regulated in the way medicine, engineering, etc, are regulated. Therefore anyone can start a software company and they can hire anyone they want to write the software.

  So what do you expect?

  On top of this, software is a million different jobs: Security, database design, performance optimisation, interface design, test automation, commenting and documenting. Ok six, but there are probably others as well. Someone who may be competent at one of these is often hired to do all of them, especially in smaller companies.

  The third problem is that most software written is, in some key aspects, completely new and different. So you can't just do it the standard way.

  Your only chance is to have a project manager that understands how software development works, and is allowed to allocate sufficient resources to testing, refactoring, and code reviews. These are the things that appear most wasteful if you're unfamiliar with software development: no new features are added, they take a lot of time, and need to be done over and over again.

  They are also, in my opinion, the most vital steps to creating effective, reliable and maintainable code.

  --
  If a plan isn't flexible it isn't realistic

  Starting Score:	1		point
  Moderation		+4
  Insightful=4, Total=4
  Extra 'Insightful' Modifier		0
  Total Score:		5

• Re:What do you expect? (Score: 4, Insightful) by mth on Monday March 03 2014, @01:39AM

  by mth (2848) on Monday March 03 2014, @01:39AM (#9859) Homepage

  The third problem is that most software written is, in some key aspects, completely new and different. So you can't just do it the standard way.

  I agree. If you need the same functionality again and again, you put it in a library, framework or code generator. So the parts that you actually end up writing are the parts that are different from other projects. I think this is one of the reasons why software projects are difficult to plan: you're going to be spending most of the time on things you haven't done before.

  Parent

• Re:What do you expect? (Score: 2) by Thexalon on Monday March 03 2014, @01:08PM

  by Thexalon (636) on Monday March 03 2014, @01:08PM (#10014)

  The problem in a nutshell:
  1. Bad techies are somewhere around $45K cheaper per year than good techies.
  2. When there are problems, they can easily cost more than $45K.
  3. However, just saying "hire good techies then" doesn't solve your problem, because to someone who doesn't understand technology a bad techie looks identical to a good techie.

  And, as it turns out, in many organizations the people who are responsible for hiring and supervising techies aren't techies themselves and don't understand technology. These people making hiring decisions about techies do so in roughly the same way that you probably make decisions about which auto repair shop to use: Do the results seem good enough, and are the prices in line with what I can afford? You might check the mechanic's certifications (which may or may not indicate anything), but you don't really know if you have a great car guy or a lousy wrench monkey until they screw something up.

  --
  The only thing that stops a bad guy with a compiler is a good guy with a compiler.

  Parent