I am the co-owner/developer of a software consulting and products company. I have been making a living off of Java since 2005 and during that time I've worked on large government projects, designed and build high-volume e-commerce platforms, and worked on a variety of business applications with dozens of technologies Carey is a DZone MVB and is not an employee of DZone and has posted 10 posts at DZone. You can read more from them at their website. View Full User Profile

Why Developers Keep Making Bad Technology Choices

04.23.2012
| 22109 views |
  • submit to reddit

Today, software developers are faced with a great abundance of options when choosing how to design and implement systems. We are constantly bombarded with choice and are used to dealing with buzzwords like NoSQL, the cloud, REST, Map-Reduce and so on. However, developers in charge of designing systems can be easily seduced into incorporating technologies that don’t provide a clear benefit over simpler solutions that aren’t as modern or hip. It seems like the KISS principle (Keep it simple, stupid!), while often referenced, is often neglected in favour of more “enterprisey” solutions. Why is this?

There are probably a lot of reasons, but I’ve identified a few that I think cover the majority of cases. As professional developers, I feel strongly that we have a duty to our employers to provide the best long-term solutions and therefore we need to rein in our desires when they conflict with this. Software development is not yet in the same realm as medicine or engineering, but I think we do need to make steps towards the professionalism, duty and responsibility that come with working in those fields.

Reason #1 – Boredom

Developers are often solving the same types of problems over and over. Not all of us have the privilege of working on new types of projects all the time, and even if we are, it’s often not new ground; similar problems have usually been solved thousands of times before by software developers around the globe.

It’s no surprise then that we want to try something new, even if we’ve adequately solved a problem before. We are natural puzzle solvers, and sometimes you just want to try a new puzzle. I’m sure many of you with several years of experience have seen functional systems effectively replaced with a new implementation that uses different technologies for no clear reason other than to suit the fancy of new developers.

So what do we do about this? How do we scratch that itch for something new? A relational DB is just so boring compared to trying out the latest NoSQL platform. Who cares if we don’t really have a good use for it? Well, I’d say you have a few options. For example, take the initiative and find ways to build out the platform that might actually benefit from some new technology. Other than that, why not work on a pet project in your spare time? After all, our job is to deliver high quality software – not entertain ourselves.

Disclaimer: I’m not trying to dissuade anyone from using new technologies. Just identify their benefits and see if they are the best choice for what you are doing, and if what you have doesn’t do the job, go for it!

Reason #2 – Resume Padding

This is perhaps the saddest of the reasons why developers make poor technology choices, and it mainly affects organizations with poor decision-making processes, but it’s still very common.

Contracts and positions in software development are very fluid these days – it’s not uncommon for a developer to be at a new company every year or two. Gone are the days where hopping from job to job is considered a no-no. Since this is the case, a lot of developers leap-frog from position to position to climb the ladder. It’s far easier for an average or lower-skilled developers to get ahead by doing this instead of trying to move up within a single company.

Since this is the case, developers will often try to incorporate technologies to get experience in them so as to add a bullet-point to their resume. How useful the technology will be to the platform is of secondary importance. Often it doesn’t matter how much they actually use it – nobody can pretend that people don’t exaggerate their skills when looking for new work. Therefore, platforms from small to large will often end up using untested technologies, or just technologies that nobody in-house actually understands well. Companies are then left with poor systems using too many technologies with nobody to maintain them as developers jump ship to more promising positions.

I don’t believe that most developers do this, but those of us who disagree with such actions should work to push back when presented with developers who are trying to make bad choices.

Reason #3 – Peer Pressure

Peer pressure is perhaps the most difficult cause to resist. We all like to believe that we are independent agents that make our own informed decisions, but all of us are human, and event the most prickly people are social creatures that want to have a happy social group.

When faced with new or hip technologies, a lot of us are somewhat afraid to resist implementing something that doesn’t really seem like a good idea to us. But we should suppress this feeling as much as we can. If you are in an environment where discussion and disagreement are valued (as one would hope), you should feel free to voice your concerns even if you aren’t totally familiar with the latest-and-greatest. Remember that software technologies come and go, but the basic principles pretty much stay the same. So if something doesn’t seem to add up, speak up! If you are a junior developer, you should still feel free to add your input – having experience doesn’t make one right. Plus, you could very well gain some insight into the choices that are being made.

Reason #4 – Lack of Understanding

Finally, technologies are sometimes chosen because developers don’t understand how things are actually working in a platform, or don’t want to find out.

For example, if you don’t have experience with highly performant relational databases, you may be inclined to go the NoSQL route, out of fear that you may implement something that won’t scale. Often though, this fear can be unfounded. If you are using a tool improperly, of course it won’t work well. But don’t let lack of understanding or knowledge force you into an unwise course of action. If, in reality, a solution could be implemented well in a relational database, and your platform already uses one, it would be foolish to introduce a new dependency simply because you aren’t familiar with what you have.

To avoid this, read and learn! If you are making choices, examine your assumptions and see if they hold up. Consult with senior developers who have worked with the tools in question and ask specifics about what they can and can’t do well. It’s never a waste to learn more about the tools that are available to us, and it will very likely pay dividends well into the future if you take the initiative.

Reason #5 – Misunderstanding or Solving Non-Existent Problems

This point ties into my previous point a little bit, but it really deserves its own discussion since it is such a big problem.

A common theme when developers pitch a new technology is that it does X and Y and protects against Z. But a lot of the time, X, Y and Z were never issues in the first place. For example, if we have a read-only data set that needs to be cached on multiple nodes in a cluster, someone may pitch a caching technology that offers distributed data sets where elements are not duplicated on each node. But what if the data set is small and we don’t anticipate any change that would necessitate distributed caching? We’d be introducing new technology that is inherently slower, more brittle and more complicated for a problem that doesn’t exist!

To guard against this, developers need to make sure that they understand the problem domain all the way through, and they also need to cross-check their assumptions to make sure they are correct. Sometimes we assume things that actually aren’t the case, so the latter step is important. Avoid the temptation to cover “what-if” situations. Chances are, you ain’t gonna need it, and if you do, we usually overestimate the cost of making changes at a later date, not realizing we are basically committing to the same effort now to avoid a slim chance of having to do the same amount of work later.

So What Should We Do?

So what are the rights things to do when choosing technologies? To start with, you might want to review the following points, and try to make it a team decision. The more input you have, the less likely you are to miss a piece of information that might alter your decision.

  • Review the requirements – consistency, failover, performance, etc.
  • Evaluate if what you have can meet the need well. If so, this is almost always the right choice.
  • Investigate how other technologies would meet the need, and factor in the costs of extra dependencies and potential failure points (nothing is free, and every new technology can have significant maintenance costs).
  • Find out your team’s expertise – favour things that you know well.
  • Factor in any other concerns like pricing, timelines, etc.
  • Discuss with the team, and make a pros and cons list.

These are just guidelines, and you can approach it any way you like, because the main thing is that you do make the decision carefully and rationally.

I hope that nobody takes this article to mean that new technologies are scary or that they should just be avoided! For instance, I’ve used NoSQL as an example already. I believe it definitely fills a need that exists and I’ve used it before to solve specific problems, but sometimes I think we get caught up in the fun stuff, and forget our ultimate goals. Just keep your objectives in mind and try to make the best long-term choice.

Published at DZone with permission of Carey Flichel, author and DZone MVB. (source)

(Note: Opinions expressed in this article and its replies are the opinions of their respective authors and not those of DZone, Inc.)

Comments

Reza Rahman replied on Mon, 2012/04/23 - 3:41pm

Excellent insightful post...

Andy Till replied on Tue, 2012/04/24 - 6:33am

I think your post unfairly portrays developers as selfish and unprofessional.  In my experience there are simpler and reasons:

  • Had to pick technology by a deadline which meant that there was not enough time to properly measure performance, weigh up alternatives etc.
  • The team went with the tech that they know best for short-term productivity even though it is not best suited for the task.
  • The developers did not foresee how the directions that the software would need to go in the future.

...And many more.  We can only make decisions with the information that is available to us at the time.

Bruce Christie replied on Tue, 2012/05/01 - 11:49am in response to: Andy Till

Although I get what you are saying, I think your comment sort of confirms the position of the article.

"Had to pick technology by a deadline which meant that there was not enough time to properly measure performance, weigh up alternatives etc."
All the more reason to go with proven solutions.

"The developers did not foresee how the directions that the software would need to go in the future."
This seems to be what Carey is saying in #5.

Cheers, Bruce

Chris Clark replied on Wed, 2012/05/02 - 3:10pm

I tend to agree with Andy - although I am sure Carey never intended to portray developers in a bad way (he is just highlighting a malaise in IT after all).

Maybe the spotlight should not shine on developers - after all, there are so many variations of developer ability. When considering technology solutions, the focus should be on Technical Architects. It is the role of the TA to have knowledge of technical solutions, and appraise them based on the need of the company.

My 2 eurocents.

Paul Russel replied on Sun, 2012/06/10 - 9:26am

I think that its great to know why Developers keep making bad technology choices. I think that this choices should be avoided in the future. I know that everything happens for a reason and purpose.

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.