Fog Creek Software
Discussion Board

Where is the industry going?

I am trying to get an idea of where this industry is going so that I can plan what I am going to do next, what language to look at next etc.
If you look at all the news sites and magazines you'll read all the marketing departments gumph, like .net is the best no Java is best, web services are the next best thing and going to change everything we do. So I thought to get a clearer (marketing free) view I'd ask here.

WHere do you think the software industry and the web industry going, what's going to be big, is .net all that or is Java going to rule now it's going to be bundled with XP, are we going to be unemployed for years to come, so hopefully you guys can give me a clearer picture as long as your not all marketing guys using this forum as a place to spread your message.


stephen adams
Wednesday, January 29, 2003

It is a bit like the stock market: If we all knew where the industry was moving in terms of lucrative software opportunities, we would all go there and ruin the opportunity by over supply.

With this in mind I advice you to specialize in Java ;-).

Just me (Sir to you)
Wednesday, January 29, 2003

Well, it so happens that I DO know where the industry is going, but I would prefer that the rest of you not follow me there.

So I'm afraid I can't tell you.

Wednesday, January 29, 2003

Since Im not a coder by profession anymore I shall dare to speculate a bit.

Database driven apps will continue to be a big deal.

.Net will become a big thing and a lucrative market. To much is riding on it. It wont kill Java in the forseable future though. Considering the work that is being done on Mono and the advantages it could bring to nix developers, the .Net API would probably survive even if MS gives up on the entire thing. Not that they will.

Java will still be a reasonably safe bet for a long time simply because there is such a huge amount of systems out there already.

The demand for simpler webapp development will continue to drop. PHP and ASP are simple enough to make it a passtime for huge quatities of geeky kids and a great many pro developers have web dev as a sideline or complemetary skill.
The result is often not 100% quality but nobody cares if the price is right.

Eric DeBois
Wednesday, January 29, 2003

I don't think there will be one technology that is going to overtake the rest.  Java will be around for a long time.  .NET use will increase.  I expect ASP.NET to become more and more popular.  I think a lot of new windows apps will be created using .NET.
  Other than .NET I don't think there will be any major new concepts to learn.  As for employment, I have no worries.  There will always be a need for good developers to think up and implement new stuff, and fix stuff that bad developers made.  Just make sure you are one of the good ones.

Wednesday, January 29, 2003

Over the past years, I expected a bigger push to better software management than there has been. I read Steve McConnell's "After the Gold Rush" and said "Yes! Yes! This is the same way I feel." Our industry has to mature, software is produced just to haphazardly. There is such gross inefficency. I thought surely there will be more and more demand for those people who can produce better quality software and do it quicker and more cheaper.

Unfortunately, I haven't seen the industry mature too
much in the past few years. "Cowboy Coders" still run amok and many folks in management still don't understand how solid software development practices can pay dividends in the long run. Reading Fred Brook's "The Mythical Man Month" didn't fill me with confidence either. The situation Brooks described in his book hasn't changed much in 30 years. I suppose we're talking about change occuring on an evolutionary pace.

But, I'm an undying optimist. I believe eventually our industry will enter a stage where a software developer will be a true professional, not just some guy that was promoted to technical-lead because he was able to build an Excel macro.

So my advice is invest some time in learning more solid software development practices. Learn more about project management. Build better leadership and communication skills. These will always be valuable.

The technical "flavor of the month" will change. Be flexible. Keep a finger on the technical pulse, but don't ignore the bigger picture.

(I know that you were more interested in specific technologies, but I had to get on my soap-box regarding software management....)

Mark Hoffman
Wednesday, January 29, 2003

This may be seperate thread material, but I think a more intersting question than, what is next years language?, is What kinds of things the industry will be doing that it hasn't done in the past (whether we do it in java or .NET), yes database backed web apps will be popular, but for instance:
Will software for robots take off  or expert systems
or LDAP based web-sites or Cell Phone Video Games(this actually looks promising already)...

Daniel Shchyokin
Wednesday, January 29, 2003

Actually to add one other thing, I think that companies that follow the "language du'jour" philosophy as a substitute for actual innovation will (and should) die out.

Daniel Shchyokin
Wednesday, January 29, 2003


Your question strikes me as a bit odd; like asking, 'What's going to be the next big thing at the Boston marathon? Nike or Adidas? Or is it Saucony?' I can think of only one way to answer that question. The next big thing at the Boston marathon with be the same thing it's been at every Boston marathon: running. Likewise, the next big thing in the software industry will be the same thing it's always been: thinking.

Study. Practice. Learn.
If it doesn't bring you employment,
I hope it brings you enjoyment.

Ken Dyck
Wednesday, January 29, 2003

Oops. My apologies, stephen, for misspelling your name.

Ken Dyck
Wednesday, January 29, 2003

My opinion: it doesn't matter what will be the next big thing.

What matters is: what will be the big thing after the next big thing. Next year's big hype is already on its way out the door. If you haven't already anticipated the need for it and learned the necessary skills (or learned them for some other reason, without anticipating their demand), then you've missed getting in at the beginning. And that's where the money is.

If you learned HTML in 1990, before anyone else, you could have earned 6 figures easily. Now, the same skillset won't even pay the rent. Or if you learned Java when it was first introduced, you could have earned a higher than average salary for at least a few years.

Never mind that web development and Java application development have become vastly more complex. Complexity doesn't make you money. Scarcity of skillset makes you money.

Arguing whether Java or .Net will be the next big thing is silly: the were the next big thing two years ago. You missed it.

So, here's my advice: try to anticipate what will be important ten years from now, and start learning it right away. Of course, the language du jour ten years from now probably hasn't even been invented yet. So don't bother scampering around trying to learn fashionable languages just because you think you'll make more money with that particular language. Instead, try to think of what problem domains will be on the cutting edge in ten years.

Unfortunately, that's where I leave you to your own devices. I have some ideas for where I want to be in 10 years (problem-domain-wise), and I'm setting my course right now. Since I don't want other people to already be there when I arrive, I'm not advertising my opinions.

Benji Smith
Thursday, January 30, 2003

" Since I don't want other people to already be there when I arrive, I'm not advertising my opinions."

That is such a laughable statement, it hardly warrants further comment.

I think you overestimate two things:

1. The weight of your opinion.
2. The effect of this web site on the industry.

Wherever it is that you hope "to arrive", if you are the only one there then chances are you aren't going to make much of a living doing it.

Likewise, if someone has to use a web site as a compass for where they invest their talents, rather than figuring it out for themselves, then you really should have no fear of them stealing your job away. They are the sheep. They aren't threats.

Bottom line: Everyone should be planning their future and their careers. But to think that you've got some inside track and letting the cat out of the bag will spoil your future indicates the relative weakness of your plan.

Go Linux Go!
Thursday, January 30, 2003

Go Linux Go! siad:
>> That is such a laughable statement, it
>> hardly warrants further comment.

Ok. He's right. I was being silly. Here's my person plan for the future of my computing career:

I think that simple business software development will become even more componentized than it already is and that custom business software will be mostly developed by the end-users of that software in a super-duper-drag-and-drop development system. If you're an accountant and you need to put together a specialized bookkeeping system for a client, you'll do it yourself rather than pay a programmer to do it.

I think this drag-and-drop development will apply to business software and web development software (to name just a few industries). Since anyone will be able to do the work that used to require a programmer, the developers in those industries who don't move on to something else beforehand will suffer for it.

Another priority for me is to pick a field of computing that will remain active and intellectually stimulating for the next five decades (you never know how long you'll keep working). In addition, I want to concentrate in a field of study where the stuff that I learn this year will still be important in five or ten years.

So, I'm going into simulation research. Specifically, I want to simulate complex biochemical drug reactions for pharmaceutical research. Eventually (read: 30-50 years), I think that large-scale simulation of cellular life will be possible, and that these chemical reactions can be simulated in the context of the actual cells that need treatment.

So, to make a long story short, I want to create computational models that can solve cellular chemistry problems by moving around tens of billions of simulated molecules within simulated cells.

Now, none of you guys are going to steal my idea, right....?


Benji Smith
Thursday, January 30, 2003

"Now, none of you guys are going to steal my idea, right....?"

Actually, I am. I only said all that to get you to spill the beans. <g>

Go Linux Go!
Thursday, January 30, 2003

Here is where the industry is going:

A lot more development is going to move to India and other countries as it has started to already. Makes sense. Cheaper labor.

A few product companies are going to develop next generation ways of creating software. That is a level above what we see now. Eventually software should be so easy to program that analysts/consultants throw together diagrams in some GUI and out pops software. Makes sense. Humans make too many errors right now. Computers should program computers as much as possible.

The need for software engineers as we know them now (coders) will start to decline as software development tools and methodologies finally mature.

Over time software engineering as a practice will become more standardized. Just as building brick buildings was at one time a difficult thing to understand, finally people figured it out and now there is not much argument as to how to do it. Same goes for software. Remember its only about 30-40 years old as an industry. Eventually people will "get it".

No Brainer
Wednesday, January 14, 2004

*  Recent Topics

*  Fog Creek Home