User:Tom94022/5MB RS

A Google search of "The HDD weighed over a ton and stored 5 MB of data" should get about 31,100 hits that, at first glance, might contain a reliable source that the IBM 350 disc storage, also known as RAMAC had a capacity of up to "5 MB of data" There are other sources that state 5 MB capacity. None of the one's I've checked explain how they derive their 5 MB and although I am an active Wikipedia participant, I don't have the time or energy to investigate all possible sources.

The history section of the Hard Disk Drive article and the IBM 350 section of History Of IBM Magnetic Disk Drives article both report 3.75 MB. There are sources that state 3.75 MB.

There is a tedious discussion of this at RAMAC Price and Ratio. RAMAC in this context means the IBM 350 disk storage device.

The question is, is this a disagreement between sources such that all sides should be presented or is the value asserted in one or more source obviously factually incorrect by some objective test and should be excluded.

The IBM disk storage device was the first HDD and it last shipped in the early 1960s but because it was the first HDD it is frequently and repeatedly discussed in many sources to this date, some reliable and some not so.

There is no dispute that the IBM 350 had a capacity of 5 million characters. The question is which if any are reliable sources for converting 5 million IBM 350 characters into bytes as HDD capacity is commonly expressed today and if there is more than one factually correct number is it necessary for the source to provide the basis of the conversion?

There is no dispute that the IBM 350 recorded each character in 8 bits, which IBM further described as 6 data bits, one parity bit and one space bit, with neither the parity bit or the space bit having any numeric or alphabetic value (see RAMAC 305 Customer Engineering Theory Of Operations, IBM Corp, © 1959, p.7-8

Since the clearly reliable contemporaneous source states the IBM 350 character had 6 data bits in seems
 * (5,000,000 char/IBM350) * (6 data bits/char) / (8 data bits/byte) / (1,000,000 byte/MB) = 3.75 MB/IBM350

is a routine calculation that confirms any source using 3.75 as a reliable source for such a value.

The question is does the use by a source of another and substantively different value for the capacity of the IBM 350, long after the last one retired, and without explanation qualify as a obviously factually incorrect value and therefore sources quoting such values without explanation are not reliable for Wikipedia? For example, a source might use the "recorded" 8 bits and arrive at 5 MSomething/IBM350 but IBM says 6 data bits per character so unless the source explains the calculation of any number other than 3.75 it is on its face obviously factually incorrect.

I suspect this situation happens with any urban legend, but in this case there is no reliable source that says so, but a routine calculation demonstrate the obvious factual inaccuracy of any data capacity other than 3.75 MB. So can such sources be excluded on this basis without examining each and every one, an impossible task?