The Story of Infocom (2000) [pdf]

As a player and fan of Infocom games back in the day they were released, I was always turned off by the very idea of them moving towards graphical titles. It seemed like they were trying to jump the shark[1], losing their uniqueness, and betraying the principle and vision of the original, purely text-based games.

Though Infocom died, their spirit lived on in a lively and incredibly innovative amateur interactive fiction community.[2][3][4]

For a tiny taste of how far interactive fiction has come since the days of Infocom and the Z-machine, see Inform 7.[5][6][7]

[1] - http://tvtropes.org/pmwiki/pmwiki.php/Main/JumpingTheShark

[2] - http://www.spagmag.org/archives/index.html

[3] - http://brasslantern.org/

[4] - http://www.ifwiki.org/index.php/Main_Page

[5] - http://brasslantern.org/writers/iftheory/tads3andi7.html

[6] - http://www.ifwiki.org/index.php/Inform_7

[7] - http://brasslantern.org/writers/howto/i7tutorial.html

Sierra and Lucasarts both turned to the same strategy of writing a VM for graphical games(AGI, SCI, SCUMM) and those engines basically remained viable for as long as those companies made adventure games - when they exited it was because of the business changing, not the technology.

With Infocom, I think the issue really was mostly with business fundamentals. The investment in Cornerstone turned out to be a product that didn't work well enough, and wasn't really within their existing business - and subsequently they couldn't keep up as the average cost of games started inflating to incorporate more audiovisual content.

But Z-Machine probably could have stuck around and adapted if Infocom had survived that period and restructured around the trends to make their games more elaborate and movie-like.

Did you read the article? Z-machine wasn't the cause of the graphical games failure. By the time Infocom ventured into graphics the games division was hamstrung by a focus on the business division (and their database product Cornerstone). They just didn't have the resources to compete. The graphical game Fooblitzky had its own issues, including that graphics quality was tailored to the lowest common denominator allowing the game to ship on multiple platforms as all of the text adventures had. This kind of thinking may have come from the write-once-ship-everywhere approach of Z-machine but it was a failure of strategy rather than technology.

The virtual machine approach did have grave consequences for Cornerstone - performance just wasn't as good as the competition. While it allowed a full-featured database product to ship on just one floppy it made performance so sluggish that one test abandoned testing their larger datasets after horrible performance with the smaller data.