I completely disagree with that actually.
What I think is that Sony has realised that as games have been driving technology harder than everything else, the technology required for games can also be used for a number of other things that are in its portfiolo and even things that aren't. You could see this coming from a mile away back when the first PS1 came out. The CD-ROM player used for storing game data could also be used to play back music, and it was even used as such in games at the time. So, why not add in software to actually just play CDs?
And later on, when Video CDs become popular in Japan, the white version of the PS1 was released which supported that also (though this was never released in other regions, as Video CDs never were that hot there).
Especially with Sony having traditionally being in the hi-fi business, it must have been even more obvious than usual ... Games need more storage capacity, and we have this thing called DVD. So, when not use it for both? Hey, we happen to own a lot of music licences, why not make our own karaoke game that can use it? Etc.
Really the only thing that you can blame them for is not doing it more and more efficiently than they have. This has partly been because software wasn't its strong point from day one. Always the best software came from third parties. They made up for it partly by at least recognising that talent, giving it a warm welcome, and make them into either very close third parties, or buying them and making them into first parties.
But as they are learning, they get better at the software part, and this is where it gets interesting.
To contrast, enter Nintendo. Forced by the unexpected success of Sony to review their strengths and opportunities, they looked at what their stand-out quality, their talent: playability. Nintendo can make awesome games. They can look rubbish, or not stand out at all, and the hardware may not always be the best, but their first-party material is awesome. So, as they focus on gameplay, they look at the market and think: "what do we need most to improve our games? what is currently the biggest limitation on the quality of our gameplay?" And so they focus on the controller. They do so for the DS, with the Styles, and they do so for the Wii, with the wii-mote.
This minimalist approach, asking themselves 'what is the single most important thing that we, given the qualities we know we have, can do to distinguish ourselves' allows them to innovate very specifically, and therefore make a smaller step into the future in terms of investments and getting the product on the shelves at an acceptable price. They effectively managed to get out of step of the traditional console generation cycle even more than Microsoft tried to do, and they succeed.
I think the shift we will start seeing, above anything else, is that the lines between the generations will blur. This has already happened during previous generation shifts, but this time I think it will stick. The reasons for this are varied: Xbox Live Arcade, PSN, Virtual Console, PSP, Wii, DS, PS2, PC 'Casual Gaming', MMOs and so on blur the lines between different hardware.
Certain games and game assets can cross-over very naturally between different formats, such as we are seeing already for PSP, PS2, Wii. We already had a lot of PC and 360 crossover, but of course there will be the more traditional 360 and PS3 crossover. The DS and Wii have a gameplay crossover in that the Styles and Wiimote share certain gameplay principles that can carry over fairly well. The PS3 shares in that genre also - a game like Mercury started out for PSP hoping to get tilt sensing, then is released for Wii, and there's no reason for it to be released on PSN.
I could go on and on. It's a big happy mess, and just goes to show that the games industry is still growing and maturing, and gaining a level of mainstream that can sustain various companies and business ideas at once easily.