Re: "parallelize"

[Chris Kowalchuk <chris -at- bdk -dot- net>]

> By all means post terminology questions. But, please don't demo a superior
> attitude, a willingness to assume, disrespect for SMEs, disrespect for
> technical language, or disrespect for technical cultures that you don't
> belong to.
>
My goodness, a lot was read into my boast. I won't even attempt to
answer half the charges I've been accused of; I have projects to finish
in the next few days...

I have never in my life changed an SME's original meaning either. Nor
was I recommending that anyone do so. For the record, I usually have a
very cordial and respectful working relationship with my SMEs (and that
goes both ways), often to the point where I *am* accepted into those
technical cultures, often on account of my insistence that the members
thereof explain exactly what they mean by their peculiar terminology. I
am technically proficient enough to keep digging and asking the right
questions until we have a clear definition that anyone can understand.
Turns out to be a useful exercise for all concerned. Turns out that half
the time the technical experts do not agree with one another on precise
meanings, and the business of putting it into "lay" terms adds clarity.
I am a strong believer in plain language writing, regardless of the
audience. I dislike words such as "parallelize" because they are
necessarily vague. Of course the SME meant something by what he wrote,
and may well have coined the word for want of a better word to
efficiently describe his process. As a technical writer, it is my job
and my prerogative to find a better word, or explain the process in more
words if it adds clarity, or if try as we might we cannot find a better
word, then to define the unusual word in a glossary etc. Of course this
would be done in consultation with the SME. I am not an idiot, although
apparently I think that everyone else is.

 This is not disrespectful to SMEs. I fail to see where I accused anyone
of speaking nonsense, being illiterate or stupid. Nor was my assertion
based on the fact that I don't understand the term. It was based on the
fact that I have no reasonable way of understanding the term (short of
asking the person who wrote it, an option not open to most readers of a
published document). Naturally if you invent something, it will be full
of previously-unnamed parts that will require a specialized vocabulary
if you wish to refer to them efficiently and specifically. Creating new
verbs for your new invention (be it thing or process) is a more
problematic endeavour, however. And often it isn't necessary.

To get back to our specific example, I would prefer:

"In some situations, <program> can simultaneously process multiple [or
two, or however many it can process at the same time] operations on
partitions that it can't on tables..."

or, more succinctly, assuming your audience knows what parallel
processing is:

"In some situations, <program> can parallel-process operations on
partitions that it can't on tables..."

or "process operations in parallel..."

Now finally, if you know that the audience of the white paper will know
that "parallelizing" means parallel processing, then I suppose it is
acceptable to use the term, as Mr. Locke argues so vehemently. But
frankly, given that the word could mean anything to do with
"parallelness", or not much at all, in a great variety of contexts, I
would prefer a verb that more closely corresponds to the action being
taken by the program, and leave "parallel" as the good old descriptive
adjective (and sometimes noun) that it has always been.

Respectfully,

Chris Kowalchuk