Sun Microsystems' loss-making woes are the result of poor strategy for the last six years, according to a former Sun executive, and he implicitly blames the current leadership, headed by company chairman and CEO Scott McNealy.

That's the view of John Shoemaker, who until 2002 was Executive Vice President & General Manager of Computer Systems at Sun, writing in Business Horizons [pdf], a publication of the Kelley School of Business at Indiana University.

His specific criticisms of the leadership start while Shoemaker was still at the company. For instance, Sun's reaction to the dotcom boom was too little too late. He describes how Sun was growing at a phenomenal rate at the peak of the boom, hiring some 3,000 employees per quarter. But while new hirings were halted immediately the downturn was spotted, few were laid off. Sun should have fired a lot more people a lot more quickly - in line with other major IT companies such as Cisco HP and IBM.

Shoemaker writes that layoff decisions, no matter how hard they are for the individuals concerned, are required both for the company's survival and the long-term security of most employees. "This single failure to make a tough decision to reduce headcount at Sun was, I believe, the critical event that precipitated Sun’s now infamous decline", he says. "I think Sun Microsystems has failed to rebound in recent years because it couldn’t make the S-curve transition and would not re-size immediately and dramatically. You have to hand it to leaders, such as Steve Jobs at Apple and John Chambers at Cisco, who acted fast and made the tough decisions to turn the negative technology outlook into a positive competitive position for their companies. The lack of like decisions at Sun led to a major impact on morale, a loss of competitiveness, and a brain drain."

The problem was that, as a result of an inability to make that hard decision, says Shoemaker, employees became focused on their own fate rather than that of the company. "Without the energy and spirit of the people, nothing else works."

He cites the decision to allow Sun's COO, Ed Zander to leave as poor decision-making. Zander is now CEO of Motorola. "I look at Motorola’s rising stock price and can’t help but think of what might have been for Sun. When Ed left Sun, an already low morale hit bottom. Sun lost an opportunity to go outside the company and bring in a proven, senior, high-profile replacement for Ed. Instead, they brought in a junior, unproven, internal person to the COO position."

Shoemaker also criticises last year's decision to buy storage vendor StorageTek, which took place after his retirement. He reckons that the $4 billion purchase price would have been better spent buying back stock and reorganising the company.

He also provides a view of Sun's technology strategy, saying that the company's chip was never competitive on its own, only in conjunction with the systems and Sun's Solaris OS.

"SPARC was never competitive. Its RISC architecture was an advantage in technical applications, and it made certain tradeoffs in input/output and cache sizes to achieve this targeted application success; however, it was not a competitive general purpose uni-processor, ever!"

Shoemaker says that the move to large numbers of smaller computers, managed by open source system software, Google-style, caught Sun on the hop. As Linux gained in popularity, Sun's uni-processor approach was in trouble:

"Suddenly, Sun was whacked on the side of the head by software that enabled the inherent cost and use advantage of many small boxes. While we actually saw this shift coming and had effectively moved major R&D spending to the low end business back in 2000, our SPARC developers couldn’t find a SPARC chip answer that wasn’t a completely new additional architecture. "

This would have cost too much, burdened as the company was by the cost of developing and maintaining Java, which had become an industry standard:

"The cost burden was staggering: hundreds of millions of R&D dollars per year, plus the huge opportunity cost of all the highly skilled technical people who could have been working on direct revenue producing products. Had some of these resources been devoted to Solaris, for example, it would have potentially made a big difference."

Shoemaker doesn't say what his role in the company was in those years -- and what he suggested the company should have done at the time.