Interesting -- however, just because this headline claims that it failed to become mainstream, doesn't mean it's not a useful methodology.

Let's look at the list on this page.

• TDD is Expensive / There may be an additional time cost, but I believe there's also an increase in the quality of the final product.
• TDD Will Retard Your Project Launch / This depends; if you have a hard date for your launch, then TDD may cause you to launch with a slimmer product.
• You Will Change Your Projects and Old Tests Become Waste / Perhaps some top-level tests will be invalidated as the project changes, but I would imagine that the lower level stuff code and tests would continue to be useful.
• Testing the Means is a lot more Work Than Testing the Outcomes / I would rather have decent tests all the way up the food chain, than having something at the end that says Go/No Go with no idea where the issue might lie.
• Extensive Testing is Boring / Sure, sometimes developing software is boring. However, if you bounce between writing tests and code, I don't imagine that would be boring for long. Writing documentation for two weeks straight is something I did recently that was a little boring -- but necessary -- for the contract I'm on right now. I wouldn't be happy if I were doing it full-time, but for me it's come after four months of development. So I'm OK with that balance.
• Many TDD Preachers Do Not Use it Most of the Time But Do Not Admit it / This sounds anecdotal to me -- to each their own.
• Many Reputed Developers Do Not Use TDD at all / Another anecdotal claim. The reputed developers may have good enough levels of craftmanship that they don't need tests. I'm happy to use tests myself when I develop software, as it gives me confidence in my code.

And it's good to also read in this article that TTD is not dead. I think it's still a valuable development methodology.

Everything Has to be an Object, however .. ugh.