[geeks] Seagate buckles to math ignorant consumers

[Lionel Peterson]

>From: der Mouse <mouse at Rodents.Montreal.QC.CA>
>Date: 2007/11/02 Fri AM 10:00:47 CDT
>To: The Geeks List <geeks at sunhelp.org>
>Subject: Re: [geeks] Seagate buckles to math ignorant consumers

>> [...] since many folks think 1K = 1,024 and not 1,000, despite the
>> fact that [...non-computer uses of K as 1000...]
>
>1K *is* 1024 rather than 1000, when measuring things that normally come
>in powers of two, to use the phrase from the Jargon File.  If you
>bought a "512MB" stick of memory that held only 512000000 bytes,
>wouldn't you feel cheated?  (I've occasionally toyed with fantasies of
>being a memory maker and building a small production run of 1GB sticks
>that held only 1e9 bytes, 512MB sticks that held only 512e6, etc, and
>setting them aside for orders from Seagate and Maxtor and their ilk.)
>
>Disk makers have in recent years been adding Flyspeck 2 footnotes to
>their ads saying things like "based on 1GB = 1,000,000,000 bytes"; to
>me, this can accurately be paraphrased as "we know we're being
>deceptive but we're going to continue being deceptive anyway".
>Deliberately deceiving your consumers is usually called "fraud", and
>I've long wondered why there haven't been fraud prosecutions as a
>result.
>
>You may not like it, but that *is* how SI prefixes get used in
>computers.

>From the amazing nearly-usually accurate Wikipedia:

"A gigabyte or Gbyte (derived from the SI prefix giga-) is a unit of
information or computer storage meaning either 1000CB3 bytes or 1024CB3
bytes (1000CB3 = one billion). The usage of the word "gigabyte" is ambiguous,
depending on the context. When referring to RAM sizes and file sizes, it
traditionally has a binary definition, of 1024CB3 bytes. For every other use,
it means exactly 1000CB3 bytes. In order to address this confusion, currently
all relevant standards bodies promote the use of the term "gibibyte" for the
binary definition. However, since there are no other uses for the term
'Gigabyte' apart from refering to memory or file sizes, the standards bodies'
recommendations are generally ignored amongst computer professionals, and
'Gigabyte' is used by them as if it were 1024CB3 bytes."

I like the wording (this is the opening paragraph, the article goes on to
dance around the units used to measure HD capacity).

>(There is one computer use of SI prefixes as applied to bits that uses
>their metric meanings instead of their binary meanings, that being data
>communication rates.  For example, a 10Mb Ethernet is 10e6 bits/sec,
>not 10<<20.)
>
>> I'm gonna file this under "Why Bother"...
>
>As a matter of pragmatics, it gets you a 7+ percent discount on that
>disk you recently bought.  As a matter of principle, it finally slaps
>someone down for years of deliberately deceiving their customers.

No, 5% back, not 7%, and it is the after-rebate price, not the retail sales
price (how they will arrive at a real ultimate price is beyond me, I guess it
will be a matter of trust and industry norms for pricing back when the drive
was manufactured).

This won't make Drives larger, and this won't make drive makers report
capacity in anything other than decimal values, they annoyed Seagate to the
point of writing a check to end this case.

The argument appears to be "you say this drive is 250 Gigabytes, which I
expect to be 250*(2**30) bytes of storage, but you are actually providing
250*(10**9) bytes of storage - you LIED" When asked how they came to expect
the drive to be 250*(2**30) bytes of storage, they can only point to the
numbers their OS spits out, and what others have told them (legend).

Seagate could have won, IMHO but IANAL - once a jury is involved, all bets are
off... What are the chances you'll be able to get 12 people who understand
exponents or binary math on the jury?

Lionel