How about if we extend the market to include programming frameworks, libraries and database systems?
From a qualitative standpoint, think of this way ... all companies are now tech companies and developer jobs are always in any recent "hottest job" list you can find. So, it stands to reason that tools for those companies and jobs will be a big market.
If you want real numbers, here's 2 examples: (1) Atlassian is now a public company and pretty much only makes developer tools. They started with one product (Jira) in 2002. They now have a market cap of $11B and annual revenues of $620M [0]. (2) If you read Indie Hackers, you'll find people like Mike Perham who makes Sidekiq. Sidekiq is an open-source tool for background processing in Ruby on Rails apps. Mike makes $80K/month on premium licenses/support. Think about that, it does just one thing for just one developer framework and he makes close to $1M per year with no employees.[1]
So, yes, I think there's a big market (in the billions) and plenty of room for small companies. Off the top of my head I can probably think of a dozen developer tool ideas that I'd consider paying for.
Would you care to share and discuss those? There are quite a few categories of developer tools, a non-comprehensive list of which could include:
- generic, one-size-fits-all tools that reach a very wide audience: IDEs, issue trackers, CI and code quality tools
- frameworks and libraries, which also are fairly generic
- one-off / niche tools and scripts for very specific purpose: Those are usually developed in-house to solve a specific problem and out-of-the-box they often don't have a lot of reproducible value outside of a particular organization.
Most of those don't real lend themselves to being realised as commercial products, in part because of existing tools, particularly open source ones and also because the target audience is more likely to implement a tool they need than any other audience (http://threevirtues.com ...).
Regarding open source, I agree that's tricky. But its been proven it can be done one way or another. I think ansible and docker have shown that in the ops space.
Regarding one-off/niche tools, that's where glue tools like zapier become commercially viable.
There still is a small market (in terms of volumes) for dev-tools sold to enterprise clients. But AFAIK enterprise clients prefer to buy from "enterprise strength" vendors. Which makes it very difficult for a smaller developer to compete.
The biggest stumbling block to being a tool vendor is the demand for support. Even if you can make money on selling the tools in the first place, the on-going support is likely to cost you more than you can charge in on-going fees.
That's a pretty unethical attitude. A developer, who by definition makes her living by creating software, more than anyone else should be able to appreciate that developers have to make a living.
I'm not exactly sure what this widespread attitude is caused by. Perhaps a mixture of a lack of empathy, a feeling of powerlessness and not having significant say in an organisation's direction.