Most of this article is true, but focused on too many topics surrounding DFM, whereas it gives few details about DFM itself.
Of course DFM aims at crossing the bridge between design and manufacturing, but there are 2 ways to cross the bridge: in your article, DFM (the tools for designers) and MFD (the tools for manufacturing engineers) are simply mixed whereas the goals, constraints, users and their background are totally different.
No word about the fact that DFM takes time to get mature because of the complexity of its purpose: improving yield doesn't simply means correcting new problems, it means locally deciding what has to be corrected among all kinds of problems (random particles, litho, CMP, as well as many other defects), and all that by preserving electrical performances or even even to improve it by making desing mor robust vs any process variation. The real challenge raised by the DFM is to always find "the best trade-off".
It's then quite astonished that you extend your list of fields/tools up to include regular layout verification tools... but simply ommit all products of the Mentor Calibre-based DFM platform: YieldAnalyzer, YieldEnhancer & YieldServer, which is today the only solution of the market which is able to combine the results from CFA (Critical Features Analysis), CAA (Critical Area Analysis) as well as those coming from other tools like LFD (Litho variations), then automatically propose corrections for all detected issues and finally helps designers in choosing the best ones in terms of trade-off (=highest DFM score that combines all metrics related to all problems that can be both geometrical or electrical).
Finally you provides 2005 DFM market shares whereas there were no real DFM tools in 2005 and these $700M were coming from old school technologies as you well mentioned. In 2005 there were ideas, pre-alpha codes, start-up creations, some stand-alone smart features added in existing tools, that's it... But in 2007, there are DFM tools. I believe that looking at those tools only would deeply change these market shares.