What Sun got wrong: Bryan Cantrill's retrospective on Sun's decline

Bryan Cantrill, a former Sun Microsystems engineer and now co-founder of Oxide Computer, published an essay on September 20 titled "What Sun got wrong" — a look back at the habits that preceded the collapse of one of computing's most influential companies.

It was prompted by a t-shirt. Ahead of Oxide's annual OxCon team meetup in Emeryville, the company printed shirts whose logos pay homage to defunct computer firms, and the Sun design by Ben Leonard drew enough attention that Cantrill decided admiration for Sun should come with a warning.

Earned fondness, and its limits

Cantrill concedes that much of the affection for Sun was earned. Oxide's own mission comes directly from Scott McNealy's epitaph for Sun, including his reflection that over 28 years at the company he "never had to hide the newspaper in shame from my children". But nostalgia, Cantrill writes, can become suffocating: people who post-date Sun reasonably want to hear what it did wrong. He himself was never ashamed to work for Sun, yet was not infrequently embarrassed by the things the company got wrong.

"Bored with the mechanics of running a business"

Fifteen years after his earlier, widely discussed take on Sun's failures, Cantrill distils the story further: Sun had become bored with the mechanics of running a business. Customers, sales and follow-through — the mundane work of a technology vendor — lost the attention of a company that preferred to think of itself in larger terms.

His main example is from 2005, when a fast-growing startup running its infrastructure on OpenSolaris wanted to buy a large amount of Sun hardware — in Cantrill's words, a vindication of open-sourcing Solaris. The sale never happened: the customer could not get Sun to pick up the phone, and when the company finally called, it tried to sell the wrong product. Dell, by contrast, called back the morning after the startup filled out a web form at night; the startup later wrote that "we all came under the impression that Steve actually worked for us". The episode is documented in a 2006 blog post, "The Sun Doesn't Shine on Me", which Cantrill read while starting the Fishworks project from a corner of abandoned Sun office space.

Studying the failure

Sun lasted a few more years; Cantrill says he and colleagues tried hard to right the ship, but it was not enough, and Sun did not make it. He left in 2010 for that same startup, and years later started Oxide with Steve from Dell. For those who never worked at Sun the homage shirts may look purely nostalgic, but Cantrill says the intent is deeper: Oxide admires what these companies got right and studies what they got wrong — "to be at once inspired and warned".