Bruce, not sure if you are interested in commercial success, but assuming you are, one suggestion is : whatever you may feel about Scala, I believe that Martin Odersky had the right strategy when he noted that Object Oriented Programming did well despite it's flaws because people could build real, working things with OO languages. This probably holds true even more for PHP.

So my suggestion is to do the same for your language : build a cool, new mp3 player that reminds folks of the days when WinAmp was so awesome. Or build a scalable server-side framework that makes WhatsApp-scale chat easy as pie. Make it easy for folks to achieve commercial success, and your language will thrive, too.