• Add a Comment
  • Edit
  • More Actions v
  • Quarantine this Entry

Comments (2)

1 localhost commented Permalink

Good points, Grady.I think the software industry right now is in correction mode, just like the stock market goes through from time to time. A correction is not necessarily a bad thing. What was happening in the 90's was unsustainable -- it was just crazy. Hopefully, as a result of the correction, many so called "IT professionals" -- professionals without the required skills and education -- were forced to get out. As a result of the correction, I think the IT industry will be better off than it was before. It is still early in the correction, and companies are still afraid to start hiring but eventually they will have to. And they will hire people with real software skills and education. As a result, the quality of the software should improve.What I’m trying to say is that it is too early to call software industry dead. C’mon, software engineering field is still new! As companies start producing better products -- by using software engineering processes -- the demand will increase. As demand will increase, so will the need more people. Software industry is not dead. Software industry is just starting its life. Software industry will live a long time and what goes with it, innovation.Just expressing my points on it…Stas Kubasekhttp://it.kubasek.com

2 localhost commented Permalink

Hi Stas,Thanks for your interesting reply to Grady's interesting post. Grady spoke about "software professionals" (architect, developer, tester, etc.) and you spoke about "IT professionals". I am also going to talk about software professionals, which is a subset of all IT professionals.I agree with you that the quality of an average software professional probably went down during the dotcom boom (though I'm not aware of any studies on this). I think it's for the reason you mentioned - many people who lacked the requisite knowledge and/or experience in the industry gained positions that required this knowledge/experience. During the bankruptcies and downsizings that followed, many software professionals lost jobs, and it's likely (in the case of downsizings) that more frequently the non-core software professionals were the ones to go first, though not always. So it's probably safe to say that over the last two years, the quality of the average software professional has gone up.In the future, I think the level of quality of software professionals will go up, but not primarily for the reason that you mentioned (maturation of the field). I think the level of the average software professional will go up for a less happy reason: there will be fewer positions available and the best will compete for them. The reason I say this is not because I doubt the importance of software; rather because as the level of abstraction at which software is created and modified goes up, there will be less and less "software professionals" affecting software but rather more domain experts affecting software.When we talk about the sub-field of software development, this may actually have a beneficial effect on the overall state of software. It's generally accepted based on many studies that the top developers are anywhere between 2 - 28 times more productive than the worst developers (2 - 5 times seems more typical, which is still quite a difference). This suggests that as there are less pure software development positions, the level of quality in developers will go up at a greater than linear rate. I have no way of proving this, but I think it's got decent logic behind it.Unfortunately, this means that all of the kids taking computer science today (or other IT related courses) will face stiffer competition as time goes on and they may have to leave the field to look for greener pastures, but such is life.PS - If anyone reading this doubts the comment about the best developers being 2 - 28 times better than the worst developers, see "Fact 2" of Robert Glass' excellent "Facts and Fallacies of Software Engineering" for a description and historical references and also section 22.5 of Steve McConnell's original "Code Complete" for another set of references (again this is the 1st edition - I don't have the new edition ... yet).

Add a Comment Add a Comment