Is software industry a "bad place" to be?

Philip Greenspun's Homepage : Philip Greenspun's Homepage Discussion Forums : Ask Philip : One Thread
Notify me of new responses
I've been working in Enterprise Software for several years, and now
am finishing 1st-year at a business school. As I'm thinking about my
future career choice, I'm quite concerned about software industry:

(1) this might be a over-exaggreation, but the entire enterprise
software industry is like a big fraud to me. Overpromising, a lot
vaporware and overselling (e.g. siebel, peoplesoft, and many start-
ups), and hard to measure its effectiveness
(2) a lot of IT jobs are going to shift to india
(3) the software industry is maturing quickly and companies are
consolidating quickly

what's your view on the software industry (especially enterprise
software)? any career advice for someone who wants to create useful
products to help people and getting tired of the depressing practice
of the enterprise software industry?

thanks.

-- Allan Johnson, May 12, 2003

Answers

Definitely the software industry does not have a good reputation with customers. The small vendors go out of business and the big vendors charge monopoly prices. Generally the solutions available off the shelf are unrelated to the real business problems facing customers but custom software development is too risky to contemplate both in terms of cost and calendar.

Because there haven't been any significant improvements in tools since the mid-1970s (when the relational database management system was introduced), customers are, as you've observed, trying to move the painstaking pointless labor that results from inferior tools to countries where painstaking pointless labor is cheap.

A further problem with working for software companies that that the innovative ones tend to be fairly young organizationally. Thus they don't have a lot of good metrics in place for evaluating employee performance. Promotions may be random or based on who is sucking up to the senior executives rather than on achievement.

If you can walk out of B-school and work in any industry I'm not sure why you'd want to choose software. Health care, for example, consumes an ever-larger percentage of GDP and consumers are apparently satisfied with what they're getting because they keep dumping more money in (in fact they keep agitating for further inflation in health care prices by asking the Federales to insure everyone). The health care business changes very slowly and therefore when you're 45 you aren't at risk of your employer wanting to replace you with a 28-year-old fresh out of school. Or you could go into higher education. The chance of getting rich quick at Harvard University isn't great but on the other hand with universities accumulating $billions in assets tax-free administrator salaries approaching $500,000 per year are becoming commonplace. Every day that you go to work you'll be surrounded by intelligent, interesting people. (Look at the marketing material put out by the average software company and ask yourself if you'd want to be trapped at a dinner party next to the person who wrote it.)

Computers are indeed powerful but that doesn't mean that the computer industry generates fulfilling, highly paid, and secure long-term careers. It generates a lot of wealth that is distributed in a winner-take-all manner (per segment), which results in a lottery-style excitement. If you hear about a plastic surgeon getting rich it isn't exciting because getting an MD and completing a residency is a long and arduous process. But if you hear about an uncreative cubicle-dweller with a bachelor's degree getting rich you think "hey, that could be me!" The problem with this analysis is survivorship bias. You look at the wealth of all the middle managers and C coders at Microsoft without reflecting on the thousands of other software companies that were started in the same year and that are dead now. Virtually 100 percent of plastic surgeons are rich whereas the average fate of a software company employee is unappealing indeed.

If you just love working with computers then probably you'll be happy in the software industry because you'll be getting paid for doing what you love. The most encouraging thing I can say is that we tend to oscillate between periods of innovation and periods of monopoly. Right now, obviously, we're in a period of monopoly. Eventually customers will say "We're tired of paying big money for the same old crud; we are going to look at innovative new vendors". Perhaps this awakening will occur in 2010. By 2020 they'll start noticing that their innovative vendors have all gone bankrupt and we'll be back in a period where people only want to buy products from one or two companies, regardless of how old and tired those products may appear.

-- Philip Greenspun, May 12, 2003


I'm in the Tidewater region of VA. There is military work all over the place if you are competent. I realized about 10 years ago that I can always get a job. That is true today I believe. I've been here since 1985 and have worked for defense contractors, a computer hardware company whose customers needed custom development and I've attempted several businesses with smart (programmer) friends. One of the biggest problems I see is the 23 year olds can programm their butts off but have no concept of the context into which their work must fit. I've learned (quite painfully in most cases) the importance of requirements analysis, configuation management, the abibilty to clearly and consicely communicate with human beings of various backgrounds both verbally and in writing. I've been afforded the abiblity to be creative, intuitive and imaginative and have been subjected to much tedius, painstaking work. For the last three years, I've been working from my home for a company based in KY that I have an equity stake in. We've been growing and performing in our sector. For the same reason the work can be done in India, I can do it in VA. The industry is still too locked into the old industry mentality. I don't get into my car for days at a time. Our cities want to build more roads to accomodate the traffic. Let programmers stay home. If the work gets done, wtf. Is the software industry a bad place to be? For me, it is a safe place to be, money wise. There are lots of things you can do and be in the 'software industry.' The companies who trade 20 yrs experience for 20 year old programmers are doomed and you don't want to be there anyway. Trading 20yrs of paper pushing dead wood is another matter and I've seen plenty of that in the defense industry - technical welfare was the term I liked. Don't judge the whole industry by one firm. One thing I've learned is that nothing turns out the way you thought it would before actually trying it. If you are able to use the tool, there are lots of people who need help with it.

So much for clear and concise, thanks for the forum, I haven't spewed like that in a while. Hope it helps.

-- Greg Zumbiel, May 15, 2003


I just blogged a link, to this thread about "big bad software", which includes a link to a relevant computerworld article. at http://www.psybertron.org/2003_05_01_archive.html#200305524

-- ian glendinning, May 17, 2003

Some strange behaviour in responses to this thread.

I've had many tens (maybe hundreds) of hits on the link I added in my previous response, some reading many pages for many minutes, but not one person has made any remark or comment here or at the other end of the link. Odd.

-- ian glendinning, June 10, 2003


This is in response to the linked ComputerWorld article: web services always sound like the perfect palliative, except that businesses will always be suspicious of losing control of IP. No matter how well they are reassured, there is the fear that info inputted into a web app will leak sensitive info onto the network (internet), or will be stored at the supply point. This fear, no matter how unfounded, has been reinforced by the bad experience many departments have had of their info being held hostage by Enterprise Document Management database vendors who either raised support rates or closed down altogether.

-- Art Vandelay, June 30, 2003

Interestingly, given that confidence and trust are more important issues than technology, the top layer in the current W3C architecture is in fact trust. See http://www.w3.org/Talks/2001/12-semweb-offices/slide18-0.html

-- ian glendinning, July 4, 2003

Allan did you stick with software or switch to something else?

-- Paul Stewart, March 20, 2007