If this is a dilemma to you, then you don't need to be doing either. There's no point in making some software for the sake of making the software. The only point to making software is filling a need, so if you have none of those, then don't write any code. And if you have found a need, who cares who it's targeted to?

"There's no point in making some software for the sake of making the software."

This is like telling an artist there is no point of making art for the sake of making art.

"The only point to making software is filling a need" During the years I have listed to myself several "needs" that are waiting to be solved

That's actually a valid point to make. It's a perfectly valid theory that if you don't have a point to make with your art, that you're not doing anything worthwhile.

But apparently you've got an idea or two, so it wouldn't hurt for you to try them each out.

I'm going to go out on a limb here and assume you're not an artist?

My mom paints. She doesn't do it because she has a point to make with her art or because she wants to radically change society. She does it because she enjoys it: it's relaxing, mentally stimulating, and ultimately satisfying.

Which incidentally are the exact reasons why I program. Sure, there's the whole money and career and changing the world thing, but when you get to the bottom of it I do it because it's fun.

I don't seem to be the only one who thinks so (http://www.paulgraham.com/hp.html)

I'm a writer. When I write, I try to avoid things that have already been done for precisely that reason. I would hate to show somebody something that wasn't new and stimulating for THEM.

For me, a part of the fun of making art is creating things that other people truly, honestly enjoy. I think the same would occur with programming.

I definitely agree that creating something that other people enjoy is the most rewarding.

I'm not a writer, so I don't presume to tell you about your own craft, but there might be a difference between writing and programming/painting: I can definitely enjoy programs I've written (not reading my own source code, but actually using the programs), and if I were more talented, I could see myself enjoying looking at my own paintings, but I find it hard to see myself reading my own writings for enjoyment. Perhaps making something that other people enjoy is more necessary for a writer than for a programmer or painter?

No. There is no dilemma, and your question is nonsensical. You're all tied up in unrelated concepts: "Open Source", "end users vs. software developers", "money vs. chefs", "applications vs. tools"--all of these things are utterly orthogonal questions.

Open Source software can, and does, make money. Our products are 95% Open Source, and yet, our revenue is growing at a very rapid clip. MySQL was acquired for 500 million bucks. Red Hat had $300 million in the bank last time I paid attention and have made numerous acquisitions over the years. IBM makes plenty of money supporting Open Source products. Sun has Open Sourced everything they offer over the past few years--nearly everything Sun has invented is now available in an Open Source form. These companies are bigger than the majority of companies folks here at News.YC will build (not all, by any means).

Tools companies also make money. Sun and IBM and Oracle make loads of tools acquisitions...sometimes for tens or hundreds of millions. That aint chump change. There have been quite a few YC companies "betting on the come" of Ruby on Rails or Python Django or JavaScript taking a place alongside Java in the enterprise market. So far, Heroku is the most visible instance of that, but there will be others.

So, stop conflating a bunch of different questions, and things might begin to become more clear. Build something people want, and build a business on it if that's what you want to do. If you aren't passionate about end-user apps, then don't build them--you would very likely fail anyway. If you like Open Source as a development model, then emulate the success stories in that field (e.g. build something in a really big market, make the Open Source model work for you rather than against you by building a modular architecture that is easy to extend, make your users love you rather than want to fight you and fork your software, and figure out a way to monetize that doesn't piss off the community).

I guess I have a little trouble understanding how the words "more fun" can come after "big money". Big money IS more fun.

The world doesn't need YAPT (Yet Another Programmers' Tool). It needs real solutions to real problems. Find an end user with an itch and scratch it.

"Big money IS more fun."

Do you think guys like Linus Torvalds or Guido Van Rossum are having less fun than Mark Zuckerberg?

Why ask for opinions that you're just you're going to question anyway?

(I see you did the same thing to bjclark.)

Ask tzury: where do you work and are they hiring?

working for Elbit Systems (http://www.elbitsystems.com/) [I am located at Israel]

Dilemma? Why not do both? Make embeddable a simple new value-adding resource to popular text applications, targeting a well-defined and large learning community of end users.

Use a mysql-like dual license to develop and promote the software freely, then charge fees for support and/or commercial (enterprise) licensing.

I can only think of a generic answer for your generic question.