Talk:Hard disk drive/Archive 22

Splitting
The "Hard disk drive" article is currently 106,339 bytes. According to the WP:SIZERULE guideline, it almost certainly should be divided into smaller articles.

I suggest we WP:SPINOUT each of the the main sections of this article into sub-articles, so the "Hard disk drive" article contains only summaries of those articles.

In particular, it seems to me that the following main sections have enough " Notability" to spin out into stand-alone articles:
 * hard disk drive --> hard disk drive technology
 * hard disk drive --> hard disk drive capacity
 * hard disk drive --> hard disk drive form factor
 * hard disk drive --> hard disk drive market
 * hard disk drive --> external hard disk drive

Is there a better way to divide this article into smaller articles? --DavidCary (talk) 16:06, 18 May 2015 (UTC)


 * Hello! Quite frankly, I'd suggest we keep the article in one piece.  Splitting it into multiple articles as outlined above would result in a truly uncomprehensive mess. &mdash; Dsimic (talk | contribs) 06:55, 19 May 2015 (UTC)


 * I can see splitting out maybe External HDDs and Form Factor, but otherwise I think we need to keep the article mostly together. Tom94022 (talk) 16:19, 19 May 2015 (UTC)

Most computers also have memory chips
I find this statement somewhat fatuous:

Computers internally do not represent either hard drive or memory capacity in powers of 1,024; reporting it in this manner is a convention.

I've personally never seen a BCD-encoded memory chip addressing scheme. The convention of using binary MBs and GBs better aligns with how memory capacity is expressed.

A system with 4 GB of memory and a 4 GB disk drive (perhaps a firewall with a bootable thumb drive) would definitely have 4 × 10243 bytes of memory, so the question then becomes whether the other mention of 4 GB should mean something different.

It might be a convention, but there was a certain amount of sanity involved that some wish to forget.

That said, it's just better to adopt GiB across the board and be done with it.

It's not just the computer that has memory, but the disk drive itself, which usually expresses its cache memory as 16 MB or something like that. So really it comes down to which consistency one wishes to prefer. &mdash; MaxEnt 08:56, 19 June 2015 (UTC)


 * Hello! Regarding the quoted sentence, it's all about how the memory capacity is reported.  With your example, the system would actually have 4 × 10243 bytes of DRAM and 4 × 10003 bytes of hard disk drive storage, plus some DRAM for the HDD buffer, possibly some more DRAM for the video card memory, some more SRAM for the CPU and HDD's microcontroller caches, and some flash/EEPROM storage for the system and HDD firmware.  Speaking of those 4 × 10003 bytes of HDD storage, the drive manufacturer would advertise it as a 4 GB drive, some operating systems would report it as 4 GB and others as 3.72 GB, while the computer would only know and care about the total number of addressable sectors on the drive.  Furthermore, it isn't that we can or should adopt GiB, storage manufacturers use and have been using powers of ten instead to express the capacity of their products. &mdash; Dsimic (talk &#124; contribs) 11:32, 19 June 2015 (UTC)

These sort of issues crop up all the time on Wikipedia. There are a variety of practices in the computer industry when representing the magnitude of different computer capacities. Lately, it's often powers of ten when expressing storage, and powers of two when expressing memory. However—as many others have pointed out here and in innumerable archived discussions throughout the Mandelbrot-like landscape that is Wikipedia discussions—practices vary depending on when in computer technology, what exact technology it is, who made the equipment, and which publication one reads on a given subject.

This article is a general-interest one directed to a general interest readership. Wikipedians best serve our readership when we are careful to use the prevailing computer-speak of the day, abandon any notion that our adoption of neat ideas like “kibibytes” will help speed adoption of such terminology (it didn't, doesn’t, never will, confuses most readers, and elicits bewilderment if the reader repeats it elsewhere), and be careful to avoid stating that “something is this way” if in fact, alternative practices and measures are also widely used in the discipline. If there was a historical practice in measuring capacity that would mislead our readership unless one clarifies the matter, then proper technical writing practice is to clarify.

Throughout Wikipedia, verbiage and units of measure should—where practical—be those commonly used today in the respective industry or discipline so readers are properly prepared for further reading elsewhere. Prose that least calls attention to itself is superior prose. Links should adhere to the principle of “least astonishment”; which is to say, readers should be able to reasonably anticipate the type of article to which they are being taken. Greg L (talk) 18:44, 28 June 2015 (UTC)