>> On Wed, 28 Mar 2007 18:01:46 -0700, Lyn Stewart-Hunter <lynsh AT US.IBM DOT 
>> COM> said:

> I work on some HSM documentation at IBM.  So we can improve our next
> doc release, could you please be more specific about which documents
> you read that either led you to believe something that you found
> later may not be true or the documents you read that you felt should
> have described the 'caveat' you mention below?


HSM has what seems to me a very clear set of meanings which run
consistently through different vendors' product lines.  I haven't yet
seen one of them include corner cases where live data eventually drops
off the end of the world, or where the use of a migrated file
generates permanent new storage use.  In this product, those "corner
cases" are your entire universe: you must pick one.

In order to effectively prepare your customers for the very
substantial differences between this product and everything else
labeled "HSM" out on the market, you need some red-and-black trimmed
READ ME FIRST pages at the head of the manual, and bullet points like

* We handle [topic X] this way, instead of that way.

Making the differences blindingly clear is going to be critical to
keep customers from feeling misled.


Changing the name from HSM to something without so clear a set of
expectations might be the best way to address this.


- Allen S. Rout