Software Development

2010 survey about classic software development mistakes

Construx is looking for participants for the 2010 update to its popular survey about classic software development mistakes.

Last year, I highlighted a 2008 Construx survey that identified 42 classic software development mistakes. These 10 mistakes were reported as most damaging:

  1. Unrealistic expectations
  2. Overly optimistic schedules
  3. Shortchanged quality assurance
  4. Wishful thinking
  5. Confusing estimates with targets
  6. Excessive multi-tasking
  7. Feature creep
  8. Noisy, crowded offices
  9. Abandoning planning under pressure
  10. Insufficient risk management

Since the post generated a great deal of interest from Programming and Development readers, I thought some of you might want to participate in Construx's 2010 survey on the same topic. Here's more information about the survey from the Construx site:

...we are solely interested in your personal experiences with these classic mistakes. If you believe a mistake to be common but you have not personally experienced it in the last three years or last five projects (whichever is shorter), please just answer "don't know" or "not applicable." We expect "don't know / N/A" to be the most common answer on this survey.

Construx says that it typically takes 20 - 40 minutes to complete the survey.

Even if you aren't interested in contributing to the survey, please share your thoughts and experiences about classic software development mistakes in this discussion.

About

Mary Weilage is a Senior Editor for CBS Interactive. She has worked for TechRepublic since 1999.

8 comments
Gabby22
Gabby22

If this was the result of the 2008 survey, there's little hope for this round. You have to wonder how the survey designers and participants defined the 'success' that the 'mistakes' screwed up. A fun place to work perhaps? How this relates to the objectives of good software development - satisfied clients and users, good profits, more contracts - is anyone's guess. Reminds me rather of a survey of school kids about their teachers. Results may be interesting but don't tell you much about how good the school is. (Hmm, the transactional analysts would enjoy this...)

john barker
john barker

it take time an no one get it right all the time mistakes are going to be made so do the best you can an fix what wrong when you find it john barker

mdunfee3016
mdunfee3016

how about failure to write an un-install, or from experience dealing with some medical software, you have to install the oldest version, then each update, then the new version, even if this is a first time purchase

Tony Hopkinson
Tony Hopkinson

A lot of them were don't knows as some of that crap never happens near me. Given the stuff that does, this can only be a blessing

Deadly Ernest
Deadly Ernest

badly written requirements (usually unstated or poorly worded) are on the list somewhere. Too often I've seen people scream when an 'absolutely needed' requirement is NOT part of the finished package, and checks show no one wrote it down as 'everyone just knew' it was an essential part of the job.

ergins
ergins

It's interesting that crowded noisy offices are cited as something that is a classic software development mistake when the agile methodology touts this as a must. I work in a company that uses agile, our team is split in to two rooms, one is very quiet (thankfully mine) whereas the other is noisy and chaotic. I prefer the quiet room because it allows me to think rather than having to deal with the chaos and constant disruption.

carlsf
carlsf

MS are very good at this, yet the changes are NOT for the betterment of the users. TWO spring to mind... 1) WIN7 the removal of the "CLASSIC" option, and a few others. 2) Office the "RIBBON". We due to these MS forced feartures that we find a hinderance to our operation have amonted to MS loosing us as a MS business user (115 seats).

QAonCall
QAonCall

If you left a provider of software that still met your business needs and required that you change (hide the ribbon) adapt, or embrace new things (at little cost considering you can run XP mode on win7 if you like) and learning the new parts of the ribbon is straight forward and intuitive for most folks (MS actually did usability studies and testing and their research showed this was a good move). As someone who has trained for the transition, the training was brief, and the users were polled after about 2 weeks, and the response were mostly (in excess of 85%) positive in that time. Add to that the option to hide the ribbon, or customize it (like previous menus). The Windows 7 thing to be seems even more shortsighted. I can understand resistance to change, but for a business to leave a vendor that otherwise (not stated in your post) seems to meet your business needs, over subtle changes, would seem extraordinarily cost prohibitive in training, lost productivity etc. My general point was that these were not changes for change sakes, MS did a LOT of research on both issues (Win7 is way more customizable and the 'ribbon' adds productivity). See associated articles: http://blogs.techrepublic.com.com/msoffice/?p=219 http://software.techrepublic.com.com/abstract.aspx?docid=768823 http://articles.techrepublic.com.com/5100-10878_11-6142596.html (note the reactions/postings under this one) http://blogs.techrepublic.com.com/msoffice/?p=120&tag=rbxccnbtr1 http://techrepublic.com.com/5208-12843-0.html?forumID=102&threadID=316043 (basically open forum to praise of hate it) (note that most of the hate is borne of frustration due to not learning new techniques or taking the time to find the answer) Again, this is all opinion (supported by fact). As a professional tester with over 25 years experience, I am simply pointing out this was not any one of the classic errors. This was well thought out, researched, and generally well accepted on both your counts. Thanks

Editor's Picks