Just because spam is an "old" problem doesn't mean it's not solvable. For example, if someone were to build a working prototype or something that demonstrates

Just because spam is an "old" problem doesn't mean it's not solvable. For example, if someone were to build a working prototype or something that demonstrates to advertisers that a pull method actually works better for generating sales/ interest/ revenue than spam, companies might be more willing to invest in infrastructure conducive to pull. Spam is "profitable" because of the low cost of spamming a bunch of random people at once. I think the article is making the point that there are alternative low-cost solutions which can be targeted more precisely to an audience that is actually interested in the source of the "spam," thus making it much less annoying.