Linuxoids: Butter Eff Ess is in your future. Background here.

What struck me [1] was the architectural decision to design the data structures around minimizing code complexity -- and thus maximizing reliability and maintainability. The fact that it also reduces disk activity don't hurt either.

[1] In my BSCS, if you must know.

In grand Unix tradition, "fs" is pronounced "fish"; e.g., note from the man page for tunefs: "You can tune a file syste, but you can't tuna fish." So, "Butterfish". (Mmmm, butterfish.)

My first instinct was to pronounce it "butterface".

Nathan, I like your analysis better, but Valerie disagrees.

Val is welcome to disagree with me all she likes, and it won't reduce my admiration one whit. I don't doubt that Chris Mason calls it "betterefess", but that doesn't mean he's right either. Once people hear "butterfish",the charade is all over.

I mean, all kinds of database people insist SQL is pronounced "sequel", but we all know it's pronounced "squeal" (or "squall", speaker's choice). Likewise, Sun marketing always insisted Solaris should be pronounced (per their trademark license agreements) like the movie title, but we all know that it's a thoroughgoing Unix, and Unices' names end (trademarks or no) with an "ix" sound. Thus, those of us in the know pronounce it "Solarix". Apple marketing, similarly, insists their thing is called "Mac OS X", but we all know, don't we, that it's pronounced "maccossix" like the well-behaved Unix we know it to be.

I did a lot of phone support for Microsoft SQL Server in my old job. There was nothing that bothered me more than the assumption by the Windows hosting folks that SQL Server was the *only* server out there, as implied by the name (you could throw "The" in front quite easily) -- well, nothing bothered me more than that, except for their propensity to call it Sequel Server. Something about calling it Sequel just bothered me.

So, I always made it a point to refer to it as MS-SQL. Spelling it out, in fact. Regardless of what the caller called it, I would not alter what I said. One asked why I was so careful to add "MS", to which I answered, "well, we support a number of open-source SQL servers as well, including MySQL and PostgreSQL, so calling it 'SQL Server' is not nearly descriptive enough to differentiate. And it's patently unfair of Microsoft to coopt the whole market with a clever name. It's like you invent a breakfast cereal, and call it 'Breakfast Cereal' as though it was the only one worth talking about." That customer on future calls actually made a point of calling it MS-SQL herself, so I felt somewhat vindicated for my unguarded moment and the poor QA results I got when that call was later reviewed.

My favorite techie-pronunciation story has to do with SCSI. In the very early days there was a debate about whether to pronounce it "sexy" or "scuzzy" - hard to believe now, I know. Of course, once people actually gained some experience with SCSI innards, the issue was quickly settled and "scuzzy" it remains to this day.

The SCSI bus originated as SASI, pronounced "sassy". When it was standardized they had to change it. There was never any question about what the pronunciation would be, although Shugart (or its offshoots, by then, e.g. Seagate) marketing drones must have hoped otherwise.

Jason, utterly unprompted, Ben just made a comment about "em ess ess que ell." I thought you'd appreciate it.

I'm seeing more and more evidence supporting the hypothesis that I should officially less-than-three Ben.