The view from stupid

When tech writers lie to you - when they tell you with a straight face that a couple of IBM Power processors can carry the same load as 500 x86 servers - they're making the assumption that you'll be too stupid and lazy to call them on it.

It's my belief that all of us have a general responsibility to correct public falsehoods and act against both real and perceived injustices. In its application to IT what I think this means is that we - you, me, everyone - have a duty to correct mis-information when we see or hear it, to periodically review our own beliefs, and to clearly differentiate both in our own minds and in what we say, opinion from fact.

I try to do this - obviously not always successfully - but in general I hope that regular readers will recognize that I try to differentiate fact from opinion and reasoned conclusions from demonstrated reality.

Unfortunately this assumes that both sides of a conversation have equivalent information - recently, for example, I stated as fact that "a Windows server application that will just barely fit in 8GB will run measurably more slowly in 16GB". This is true, but to know that it's true you have to know how the Windows server page file management process works and understand in context that if you hardwire both the OS setup and throughput on a small application so the combination just fits below the 8GB boundary, then it will continue to do so no matter how much additional memory you put in the box.

Absent the shared referents, however, it looks like a perfectly bogus, and completely unsupported, statement that really needs to be questioned - as indeed Shadetree promptly, and to his credit, did.

As a practical matter I don't believe I can take the time needed to demonstrate both parts of the statement to his satisfaction, but I perceive the discussion to be valuable because I know from his comments that he's more than smart enough to think through the logic himself and then use internet information resources to understand the impact unused memory has on Windows Server 2003/XP.

Unfortunately a lot of other writers and speakers take the opposite approach: they view the reader as rather stupid, seriously uninformed, and too intellectually lazy to question conclusions sold on an appeal to group membership or other emotional decision drivers. That's really, for example, what's going on in that phony Unix vs Linux advisory I reviewed on Tuesday: the title leverages a misleading sales pitch, the Sun bashing is an appeal to a data processing "us", and the real message is the absurd contention that a few Power processors dressed up in IBM livery can generally outperform 500 or more x86 processors.

The worst thing about this kind of writing is that it works: tug at the right heart strings, speak loudest to the already convinced, and few will notice that you're simultaneously insulting your readers and lying to them - and don't kid yourself, this kind of thing isn't limited to a few IT agenda writers: in our own field think of the groklaw hordes endlessly conning themselves; think of the all the hacks pretending that Xeon is somehow performance competitive with SPARC, Power, or even AMD; think of all the Machacks effortlessly burning intellectual U-turns from their previous informed and sincere support for PPC over x86 to their currently profitable, but equally informed and sincere, support for Apple's technically disastrous (and commercially successful) switch to x86.

Bottom line: I regard the perspective shared by writers willing to do this kind of thing as the view from stupid because that reflects the assumptions made about readers eager to give credence where none is due - and because this kind of thing is both endemic and disgusting I think that you and I, and everybody we know or can talk to, should be out there yelling "enough already" at the top of our voices.