Jump to content

Talk:Hard disk drive/Archive 9

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia
Archive 5Archive 7Archive 8Archive 9Archive 10Archive 11Archive 15

First column of chart

Ruud, that first column of the chart, which has two cells (“source” “units”), each describing the nature of headings in the entire row to the right, is more confusing than it is helpful. If omitted, the table is self-explanatory. I tried to fix it but am not handy with tables this complex. Greg L (talk) 15:19, 14 April 2011 (UTC)

An Alternate Proposal For Table 2

I don't think the table, with it's current entries, helps our readers as much as it could. Why not use real life examples of HDD capacities. Such as: 20, 80, 120, 160, 250, 320, 500, 750 GB drives, and 1, 1.5, 2, and 3 TB drives (or some subset of them). The largest consumer drive is currently at the 3 TB point (with rumors of a 4 TB on the way). Prices of drives at and below 2 TB are at commodity price points now; $90 street for 2TB, $55 for 1TB, OEM. And there are still plenty of the smaller units around in use, especially in laptops and netbooks. If a reader sees the capacity of a drive he/she has in their system, wouldn't that provide more meaning than theoretical capacities? Thoughts? — Becksguy (talk) 02:32, 16 April 2011 (UTC)

  • I think what you are suggesting would be a good supplement to the current table (not a replacement). The current table handily drives home the concept of the divergence between the hard drive world’s decimal use of terms like megabyte and gigabyte and shows how it is a 2.4% difference and then a nearly 5, 7, and 10% difference as one works their way up the multiples.

    What is sorely missing is along the lines of what you are talking about. It could also include a column for the difference between the “advertised” capacity (again, in the decimal sense), and what you actually have left over after formatting.

    You could have “advertised” (in the same decimal measures used by the vendor), the actual formatted capacity (decimal again), and the actual formatted capacity available for files measured using the binary definition. The beauty of your approach is you can take an actual hard drive you have laying around and see what these values are. If you have both Macs and PCs (or dual-boot Mac like mine with Bootcamp), you could make one killer of a table.

    Coincidentally enough, I just had my brother drop off an off-site backup drive only 20 minutes ago and it’s getting a Time Machine backup session right now. But I’m afraid to wipe the thing. Since a colossal digital disaster a bit over a year ago, I never, ever take risks. I have my important business data in the following locations:

  1. The built-in hard drive in the iMac
  2. A Firewire hard drive in a fire-resistant safe off the end of the desk (with 16 quart-size, all-metal paint cans of paraffin wax with a melting point of 57 °C occupying the bulk of the free volume in the safe). The Time Machine allows me to retrieve versions of files I’m working on from an hour previously, or five hours previously, or a version from a month ago or more. It’s therefore especially good for human error. It can be used to make a new, fully booting hard drive volume and would be missing, on average, only 30 minutes of data.
  3. The aforementioned off-site USB backup drive (also backed up by Time Machine), with the session timing enforced by a computerized calendar reminder to make that phone call every three months so my brother drops it off after work. It is capable of completely restoring a new hard drive with everything to make a bootable volume with all OS resources and applications. But it would, on average, be missing 6 weeks of data.
  4. Carbonite offsite backup continuously running in the background. It would be current to within an hour or so on recovering all data and no fire or burglary could possibly deprive me of the data because the server is in an entirely different state on the other side of this continent. However, the recovered files don’t include important resources to make a boot disk and recover applications (just a matter of money and time).
  5. Backup of my business data onto my iPhone using Flash File on the iPhone and ChronoSync on the Mac. The timing of these sessions are also enforced with a computerized calendar reminder on both the computer and the iPhone. If I’m standing there looking at the glowing embers of my burned-down house as the firemen spray down hot spots, I can pat the pocket holding my iPhone (and cross my fingers that the firesafe and paraffin wax kept my Time Machine drive from baking).
So… no more *digitally bold* Greg L. Backups are good. Definitely. Then I watch Judge Wapner. Fate can have my data when it pries the hard drives from my cold dead fingers. Greg L (talk) 02:57, 16 April 2011 (UTC)

It was getting off-topic

The section on capacity had gotten too off-topic. I’ve trimmed it to precisely what would be in that section if there were no battles here to cloud the issue. Here is the revised wording, which cuts to the chase of what the lawsuit was about and what the industry currently does with that dual meaning. Greg L (talk) 06:59, 16 April 2011 (UTC)

Compromise: the camel's nose

BTW, I don’t have any problem with RaptorHunter’s addition of a See also: Binary prefix to the top of the Capacity section (perma-link). I think that is perfectly in keeping with Wikipedia’s best practices and best serves the interests of our readership.

I can understand, given the tendentiousness of some of the IEC-prefix advocates, that some here might fear that this edit is a harbinger of the Arab proverb that goes, “If you allow the camel’s nose under your tent during the sandstorm, eventually the whole camel will be on your bed roll.” We’ll deal with the camel if and when it occurs. In the mean time, I see no problem with its nose. Greg L (talk) 18:39, 17 April 2011 (UTC)

Greg, if you remember, making the capacity section into a summary section without IEC units, and a section tophat link to the main article, was my compromise suggestion, oh so long ago. At least it feels long ago what with all the kibibytes expended here. ;-) The combined threads on this page are now about 40K words long. It's interesting that despite all the contentiousness, the article is on less than 30 watchlists. Also, any problem with delisting the RfC from here now?
BTW, I mostly agree with your comments about Micro$oft. Even though I have used DOS and Windows since about version 3 of PC-DOS, I switched to Linux about 2.5 years ago and couldn't be happier. And despite having been a MS beta tester for a while back in the day. I keep WinXP as a VirtualBox guest OS for the few apps I need that Linux doesn't have. — Becksguy (talk) 04:18, 18 April 2011 (UTC)
No problem here with the RfC notice(s). As widely advertised as that thing was, the rest of the community either looked upon it with a collective “*Meh*”, or quite possibly looked at the length of the discussions here, crapped their pants, yelled “F*ck me!!! and ran screaming, thinking this place to be a land of Turkish butt-stabbings and jihadists with suicide belts.
As for your earlier suggestion, if that was more than… what… two weeks ago in earth-reality time, then that was the equivalent to four months in Wikipedia time, where space-time is dilated or something. I swear, some 16-year-old kid can get a year banishment at ANI and think “What?!? A year!! I could be married with three children by then!”
Speaking of virtual OSs hosted on an OS, I’ve got Mac OS X 7.5.5 that can launch as a virtual machine within OS X via SheepShaver. It is still magical to watch that thing “boot up” and the system extensions to march across. If I need to dig up stuff from the mid 80s, that’s where I go. Old stuff, there. Greg L (talk) 05:10, 18 April 2011 (UTC)

Keep a watch for vandals

I've just had to revert these two vandalism [1] [2] edits by two separate IP users. It looks like the IP user has some score to settle regarding the recent binary prefix debate so they might come to this article as well. Glider87 (talk) 08:05, 18 April 2011 (UTC)

Now user talk page vandalism. Glider87 (talk) 08:40, 18 April 2011 (UTC)
While WP:MOSNUM is a guideline which cannot mandate usage in an article, WP:VAND is a policy which does mandate that only vandalism should be described as "vandalism". Please stop making that error. Johnuniq (talk) 08:44, 18 April 2011 (UTC)
The IP has been blocked for vandalism.Glider87 (talk) 08:47, 18 April 2011 (UTC)
Nevertheless, you were wrong to use "rvv" as the edit summary in your reversions, and you were wrong to describe them as "vandalism" in the above. While the IP finished by posting nonsense on user talk pages, their edits to this article were not vandalism according to the policy I mentioned. The IP was blocked as "blocked proxy" (i.e. an IP that could be used by anyone to disguise the origin of the edit). Johnuniq (talk) 09:53, 18 April 2011 (UTC)
I am not wrong, you are wrong. From WP:VAND "Types of vandalism" see "Deliberate attempts to circumvent enforcement of Wikipedia policies, guidelines, and procedures by making bad faith edits go unnoticed." and also the section "Sneaky vandalism". Obviously the IP user was editing in bad faith given the misleading edit summary and the talk page vandalism. Glider87 (talk) 09:59, 18 April 2011 (UTC)
Your "Deliberate attempts to circumvent..." text is totally irrelevant because the two links you describe above as reverting vandalism were to undo edits where the edit summary was not marked as "minor", and was not in any way misleading (it was crystal clear: "restore Wtshymanski edit...", with the URL of the edit being restored). In the second link (the earlier revert), you reverted the three prior edits (by Preslethe and by 83.30.140.124 and by 81.105.209.229). WP:VAND does not permit use of the term "vandalism" for any of these edits. Further, the talk page vandalism came after your revert! Johnuniq (talk) 10:33, 18 April 2011 (UTC)
The edit summary doesn't have to be marked as minor for that section to apply. The vandalism edit was misleading because the edit summary started with "Undid vandal/sockpuppet revision" when it was reverting valid edits so it was also in violation of WP:VAND "reverting legitimate edits with the intent of hindering the improvement of pages". As WP:VAND also says "Some vandals even follow their vandalism with an edit that states 'rv vandalism' in the edit summary in order to give the appearance the vandalism was reverted" which is just what this IP user did, so it also falls under the "what is vandalism" section. So in summary WP:VAND does apply to those edits. Glider87 (talk) 10:52, 18 April 2011 (UTC)
Hold on a second Johnuniq I think even the first edit was clearly vandalism (sneaky but still vandalism) and correctly reverted as such so I don't know why you're giving Glider87 such a hard time. The only thing I would have done differently if I were in Glider87's shoes would have been to add a suitable vandalism warning template on the talk page of the IP that made the change. Fnagaton 11:31, 18 April 2011 (UTC)
  • Just to clarify the first edit was [3] which used a misleading edit summary and a minute after the same IP made a second edit [4] to mark someone as a sockpuppet when they obviously were not, which is vandalism. These two edits are before I reverted the first edit with "rvv". Obviously with the short misleading edit history of this IP user good faith need not be assumed and the edits are obviously vandalism. Glider87 (talk) 12:28, 18 April 2011 (UTC)

That I.P. hails from the Netherlands and may not have ever been a party to, or aware of, these discussions. If one assumes that to be the case, then the I.P. could not have deliberately attempted to circumvent a guideline. Since the edit wasn’t something like ROBERT POOED HIS SHORTS IN GYM CLAAAAS, the edit can easily be seen as being well intentioned, thinking that the finest encyclopedia ought to use the finest, cool-beans lingo just passed down from the Vulcan High Command (along with writing 99.8 % with the space before the percent symbol because that comes from Starfleet of all places).

That editor probably knows that our readers will go “WTF?!?” upon encountering this New and Modern Way Of Doing Things®™©, but also likely sincerely believes our readership will actually remember this stuff and will, in turn, demand that the computer manufacturers use it on their product packaging and, perhaps too, they might write the editors of PC World demanding they adopt new-age words like “kibibits” so the computer industry can sound more “Klingon” so nerds can really make the jocks feel inferior as they’re serving up some atomic wedgies.

As Wikipedia is the encyclopedia that anyone can edit, no technical writing experience is required; only a unfailing self-confidence that he or she can teach the planet something. The high levels of self-confidence comes from our schools and all those motivational posters. Were it me reverting that I.P., I would have used something along the lines of as follows for my edit summary: Revert. Those are from a proposed "standard" that isn't industry-standard lingo. See [[WP:MOSNUM#Quantities of bytes and bits]]. Greg L (talk) 14:22, 18 April 2011 (UTC)

From the Netherlands you say? I'm sure not everyone there is rude.  GFHandel.   23:06, 18 April 2011 (UTC)
Actually, all Dutch people I remember ever meeting were pretty nice (with the possible exception of one taxi driver in Amsterdam a coupla years ago, but he had got good reasons to be rude). Anyway, one shouldn't call an edit “vandalism” unless it's obviously beyond reasonable doubt that it wasn't in good faith. ― A. di M.plédréachtaí 10:36, 19 April 2011 (UTC)
I am going to assume that Glider87 will be more careful in the future. This is a minor issue, and it would have been desirable to back out of it gracefully rather than insist that reverting three different editors (including Preslethe, see above) as "vandalism" was ok. Sure, the IP was clearly offtrack, and their edit summary was inappropriate, but experienced editors are supposed to take that in their stride (just use "not needed" as the edit summary if nothing else comes to mind). In general, it is never helpful to describe edits as vandalism, except at an appropriate noticeboard (such terminology excites real vandals and distresses non-vandals, so it is never helpful). According to the blocking admin, the IP is an open proxy so it could have been used by anyone from anywhere. Johnuniq (talk) 23:40, 18 April 2011 (UTC)
The person behind the IP used a misleading edit summary and abused sockpuppet tags, which as I explained is vandalsim according to WP:VAND. So I was justified to revert as vandalism. The person behind the IP and other related IPs was then blocked for vandalism of several user talk pages using and blocked for using multiple open proxies this confirms the person was not editing in good faith. Why should I "back out of it gracefully" when you are wrong? I hope you learn from this and be more careful in the future before leaping to the defence of vandals. The good thing is that several open proxies used for vandalism have been blocked. Glider87 (talk) 01:43, 19 April 2011 (UTC)
  • Yeah, OK… fine. Let’s not work the “I’m sooooo offended now…” routine too much here and do the pile-on with Glider87. Tempest in a teapot. If the I.P. is genuinely an I.P. and not one of the regulars around here slinking around, slashing tires with some stealth edits to blow off steam, it’s doubtful he or she is going to be all that offended by an edit summary along the lines of “rv” considering the editor’s own choice edit summaries (Special:Contributions/85.17.146.116), such as “f..k you spanish idiot - RULE BRITANNIA”. Except for that last word (Britannia), the I.P.’s comfort zone seems to be the *direct* approach with words not in excess of two syllables. Greg L (talk) 01:49, 19 April 2011 (UTC)
    The same IP address can be used by several different people (see comments about open proxies above. And FWIW, I pronounce idiot with three syllables. :-) ― A. di M.plédréachtaí 11:18, 19 April 2011 (UTC)
  • A. di M. is clearly amongst our more detail-oriented editors. From the land of Fiats. I remind myself of someone from my extended family. She was in my kitchen and said (I swear), “I hate it when people mis-pronunciate words.” Greg L (talk) 15:47, 19 April 2011 (UTC)

Consolidation of HDD market

There is a rumor, reported in the WSJ, that Samsung may sell their HDD unit to Seagate [5]. Hitachi GST and Western Digital (WD) announced in March that WD would acquire the HDD business from Hitachi, subject to regulatory approval, etc.[6]. If both spin offs go through, that will pretty much leave the consumer 3.5 inch HDD market to WD & Seagate.

  • Connor → Seagate (1996)
  • Quantum → Maxtor (2000)
  • IBM → Hitachi GST (2003)
  • Maxtor → Seagate (2005)
  • Fujitsu → Toshiba (2009) - Toshiba doesn't manufacture 3.5 drives.
  • Samsung → Seagate (rumored, maybe 2011)
  • Hitachi GST → WD (probable 2011)

Didn't research the 2.5 HDD market, the enterprise HDD market, or the SSD drop-in market. Also, Samsung announced a 4TB drive at CeBIT, and claims they will ship some time in 2011.[7]. So the 3TB drive is still king of the hill in the single unit consumer 3.5 HDD market. Stay tuned. — Becksguy (talk) 04:04, 19 April 2011 (UTC)

This rumor has been confirmed (as said on The Inquirer, for example). #!/bin/DokReggar -talk 11:22, 19 April 2011 (UTC)

An Alternate Proposal For Table

I hope we could achieve consensus on this version of the table:

Advertised capacity
by manufacturer
using decimal multiples
Expected capacity
by consumers in class action suit
using binary multiples
Reported capacity Reported capacity
using
IEC Binary Prefixes
Windows
using binary
multiples
Mac OS X 10.6
using decimal
multiples
With prefix Bytes Bytes Difference
100 MB 100,000,000 104,857,600 4.63% 95 MB 100 MB 95 MiB
100 GB 100,000,000,000 107,374,182,400 6.87% 93 GB
95,367 MB
100 GB 93 GiB
1 TB 1,000,000,000,000 1,099,511,627,776 9.05% 931 GB
953,674 MB
1 TBtic 0.931 TiB

Tom94022 (talk) 03:27, 15 April 2011 (UTC)

No it is not suitable because it goes against the consensus already discussed and agreed as being valid on this talk page and WP:MOSNUM. Stop proposing things that are clearly against consensus. Glider87 (talk) 03:39, 15 April 2011 (UTC)
While I am not surprised that you would not agree, your statement that there is a "consensus already discussed and agreed as being valid on this talk page" is not true. At best you can say is that some editors do not want to see any IEC Binary Prefixes on this page but a majority of the editors do. Tom94022 (talk) 03:46, 15 April 2011 (UTC)
What I wrote is true because there is a difference between "I want to see IEC prefixes" and "compelling arguments against IEC prefixes that use Wikipedia policy". What you are doing is saying "I want" without tackling the Wikipedia policy arguments. That is why your point of view is clearly against consensus. Glider87 (talk) 03:53, 15 April 2011 (UTC)
  • WTF?!? Do you behave this way in real life, Tom? What did your mother do when you behaved like this?? Take away your Nintendo time? Now you are just being tendentious and that’s purely disruptive. See the reasoned, wise comments from other editors, above, for why including the IEC prefixes in the table as you propose is a bad thing. Lots of those sort of comments are above the part where two admins weighed in on this page that a consensus to deep-six the prefixes exists. And those are above the part where you wrote that you’re going to go forum shopping somewhere to complain about the awful admins who refuse to see things your way (bad, bad them). Above those two things are the comments you need to A) read, B) comprehend, and C) pull a Tom94022 and dismiss it all as a government conspiracy right up there with the Roswell coverup. If you keep this up, the community can ignore you. When that happens, please don’t confuse “The community ignoring tendentious disruption” with “Well… silence from the community so the community *agrees* with me!!!” Greg L (talk) 04:31, 15 April 2011 (UTC)
    • Greg, continue like this and the only Nintendo time being taken away will be yours. —Ruud 05:29, 15 April 2011 (UTC)
    • Don't take Tom94022 too seriously Greg. After the debate and clear direction over the previous many days, Tom94022 is obviously having a bit of fun at your expense. He (and everyone else around here) knows that given the current lack of use and acceptance of the IEC units, there's no way they are getting a run on this article.  GFHandel.   06:28, 15 April 2011 (UTC)
    • Greg, please read WP:CIVIL and WP:NPA. Tom is a valued editor here, and we welcome his input.--RaptorHunter (talk) 15:40, 15 April 2011 (UTC)
    • My comments to both Tom and Greg are below. As to Tom's most recent input, while we have a going-in AGF, that is a rebuttable presumption. By violating a guideline repeatedly, an editor can lose that assumption. wp:consensus makes clear that tendentious editing against consensus is in fact not welcome; I wouldn't want Tom to get too heady with Raptor's advice, and find himself subject to sanctions. Of course, Tom's input which is not guideline-violative is very much appreciated and welcomed. Which I imagine is part of Raptor's advice.--Epeefleche (talk) 15:56, 15 April 2011 (UTC)
  • Two comments. One, I'm perhaps just a bit confused as to how Tom might have derived from the above discussion anything other than a consensus rejection of what he re-proposes, as though no discussion has been had. Perhaps I am missing something. But without further clarification, his suggestion does perhaps seem to fall afoul of wp:consensus, which if I recall correctly does have language about tendentious editing in the face of consensus that might in the eyes of some have a measure of applicability to Tom's proposal. Second comment -- @Greg -- while I understand your upset, perhaps there is a gentler way with which to communicate with Tom, that might serve to "reach" him. Though I'm sure I don't know what it is.--Epeefleche (talk) 05:11, 15 April 2011 (UTC)
  • Parkinson's Law of Triviality While I'm sure we could all be discussing the many pros and cons of (not) including a column with IEC prefixes, I somewhat doubt even a single reader is going to notice, whether it's there or not. Anyone having a strong opinion one way or the other is probably being motivated by something other than rational reasons. —Ruud 05:29, 15 April 2011 (UTC)
I enjoyed that diff. One can learn something every day here, apparently even on talk pages. You btw seem to have a strong opinion, that others should not have strong opinions here. Setting your foot in the wet cement, are you?  :) --Epeefleche (talk) 06:01, 15 April 2011 (UTC)
It’s a common behavior of third parties to disputes. They can come across as Big Picture people who poo-poo those who are passionate about silly little things that are of no consequence in the grand scheme of things that includes all manner of man’s inhumanity to man. We’re all human and that sort of effort at mediating disputes—particularly in a venue like this where it is important to not take sides on the substantive matters—is very human and understandable as mediators and leaders endeavor to garner credibility and respect.

And, no, Ruud, it is certainly not a matter of some of the mere minions who create content on Wikiedia being “motivated by something other than rational reasons”, it is a matter of rationally behaving in a particular matter with undeclared motives, which—as we all know—happens all the time on Wikipedia (such as at Race and intelligence or Anwar al-Awlaki). So there is no need to conjecture that perhaps we have editors whose behavior can be best explained as being the product of irrational thought. That sort of allegation can apparently be fun to toss out there. Does your conjecture imply that you, being the Big Picture®™© sort of way‑cool guy that you are, must be the only one capable of rational thought around here? If so, that’s quite a distasteful thing to suggest.

Undeclared motives underlie a lot of editwarring and there is no point playing coy here about the phenomenon. Wikipedia frequently has editors throw out wikilawyered beating-around-the-bush wiki-slogans of WP:AGF and inclusiveness as a cover for WP:I DON'T LIKE IT; as was Tom94022’s above suggestion to turn right around and put back what we just got through taking out.

The rationale underlying the consensus decision was obvious and founded upon several bedrock principles underlying Wikipedia; all of which start with the galactic-grade observation that 99.8% of the industry doesn’t use such terminology. So using the terminology in this article in this manner was transparently a simple matter of trying to promote a nice idea that didn’t catch on by using them in a table in an “Oh… Didn’cha know??”-fashion. Wikipedia is never allowed to be exploited in that manner, which was a clear violation of WP:SOAP and its #1 prohibition against advocacy. The proponents’ own arguments on this talk page betrayed that motivation; that the IEC prefixes are superior and the industry ought to adopt them. Notwithstanding that deep desire, the simple fact is that the computing industry near-universally does not use the IEC prefixes and merely defines the common prefixes two different ways; ergo, almost none of our readership has a clue what “kibibit” and “KiB” means when they encounter it and it is a near certainty they would never see it again after they left this Wikipedia page.

So it’s over, now. We write using the same terminology in the same way the vast majority of the computer industry uses it. Tendentiousness may not be used as a tactic to undo that. Greg L (talk) 22:20, 15 April 2011 (UTC)

  • RaptureHunter appreciates Tom’s “valued” input. I don’t because the consensus is clear and Tom was just being tendentious by turning right around on the heels of deep-sixing the IEC prefixes to suggest they be put right back in. Greg L (talk) 17:25, 15 April 2011 (UTC)

Been busy since yesterday and missed most of the above, none of which seems to add much. Let me explain why I think the additional row both is fully compliant with MOSNUM and adds to the article:

  • The so-called consensus paragraph now explicitly discusses IEC Binary Prefixes and therefore the article is eligible for the MOSNUM exception. Note that I did not propose the paragraph, but I do accept it. I believe it has also been accepted by all the IEC Binary Prefix Thought Police.
  • The pedagogical question is then whether the article is better by discussing the existence of IEC Binary Prefixes AND using them or as it is currently proposed, discussing their existence but not using them. IMO, the former is preferred since it both describes their existence and shows a context. The current construction leaves the reader ignorant and possibly confused. I think any teacher amongst our editors would state that the preferred teaching method is to both discuss and use.

All should recall that this fire storm started with a discussion of whether the article was or was not compliant with MOSNUM; the consensus seems to have brought the article into literal compliance and now the question is do we use the prefixes or not. I would really like to hear from someone other than Greg L, Fnag, Glider and the unidentified IP why my reasoning is flawed. I'm going to be off line for another 24 or so hours. See u then Tom94022 (talk) 05:17, 16 April 2011 (UTC)

  • Okay, since I'm not defined (rudely by you) to be part of the "IEC Binary Prefix Thought Police", I'll have a go: until IEC prefixes are accepted (by an significant quantity of hard disk and OS manufacturers) as a way to quantify the size of hard disk drives, this article will not be mentioning them. "see u then".  GFHandel.  
  • Thanks for responding. Turned out I was more than 24 hours off this dialog. I fail to understand any requirement that "a significant quantity of ... OS manufacturers" use IEC Binary Prefixes in order to discuss them in an encyclopedic article (FWIW, I doubt if the HDD manufacturers will ever use IEC Binary Prefixes - no reason for them to do so). It seems to me that a use by reliable sources such as the Ubantu and GNU utilities to describe HDD capacity should be sufficient to discuss them in the capacity section of the HDD article. I can't think of another MOSUM item being censored in this manner, particularly one endorsed by an extraordinarily reliable source, the IEC (this is not furlongs/fortnight). I also note you didn't respond to my teaching argument about usage in the context of the so-called consensus paragraph - if we were to mention them why not use them? Unfortunately, paragraph was subsequently completely rewritten to obviate my argument (by guess who). I am sorry you find my characterization of "IEC Binary Thought Police" as rude; it is an accurate shorthand for the gang of three (maybe four) and far less rude than the several things said by them about me personally and my arguments in favor of using IEC Binary Prefixes in this article. Tom94022 (talk) 23:10, 20 April 2011 (UTC)
  • There must be thousands of things more important things to do on WP than to extend the debate about units which have been abandoned by 99.x% of the industry. There is a link in the article for the interested reader to discover more about the discrepancy in HD capacity (which will lead them to the IEC units), and that's sufficient.  GFHandel.   23:26, 20 April 2011 (UTC)


  • The article discusses the difference between GB/TB in the decimal sense and binary sense. It does not explicitly discuss IEC prefixes, it only mentions an attempt by the IEC to create a standard. The article does not need to use IEC prefixes to discuss this difference because it already talks about the relative sizes of the bytes involved. The job of discussion of the IEC prefixes is specifically the domain of the other linked article, not this article. The article itself mentions that the IEC prefixes have seen little adoption. Therefore to disambiguate a number of bytes with prefixes that themselves are not often used and are not familiar to the readers and therefore need further explanation in this article does not improve the article. Since there exist other more familiar disambiguation methods (number and power notation are more familiar) that do not use IEC prefixes and do not need so much explanation, then that method of disambiguation is better. Not to mention using IEC prefixes causes problems with WP:NPOV and WP:UNDUE because the majority of reliable sources for the article don't use IEC prefixes. Since the reliable sources do not generally use the IEC prefixes then you do not have a case to use them here. The other article shows conext, it is not the job of this article to show "context" just because in your opinion you think they are better. Wikipedia doesn't accept WP:ILIKEIT as a valid reason to add something to an article. Wikipedia reports how the real world is by using reliable sources and then reporting significant points of view. Fnagaton 06:04, 16 April 2011 (UTC)
  • Indeed. Tom94022’s logic is self-serving rhetoric that would vanish into thin air if he turned off the reality distortion field surrounding it. Sure, the article mentions that the IEC proposed their prefixes and the computing world virtually totally ignores the proposal. I just got through rewriting that section and likely expanded the discussion of it as I made that passage more useful to our readership. And for my trouble, I get a face full of silly-reasoning pie. The article mentions their being essentially unused in the real world and from that he makes the logical leap that we now magically need to use them in the table. It’s a leap of logic I can’t fathom. Setting aside the wikilawyering for a moment and exposing the 800-pound underlying-motive gorilla in the room, Tom94022 must craft logical nonsense like that because he is afraid to openly admit the real reason: He advocates their adoption by the computing world as he yells “Pay no attention to that undo weight behind the curtain!

    If Tom94022 is going to continue to agitate based upon the fact that the article now discusses them, I am perfectly content just leaving out those two paragraphs. Those two paragraphs amount to a big pile of discussing something that *could have been* and aren’t in the least bit necessary when discussing the class-action lawsuit, which was entirely over how the industry had two different ways of defining “megabyte” and came up with the lower-capacity definition for self-serving reasons because it was smaller. Nothing about the lawsuit was centered about the industry’s use or non-use of the IEC prefixes. Greg L (talk) 06:46, 16 April 2011 (UTC)


    P.S. Taking a real objective look, touching upon the IEC prefixes like that was entirely unnecessary to that section and clearly was an awkward shoehorning borne out of an attempt at a compromise solution. But the reader (which is what Wikipedia is supposed to be about) was the loser with that attempt. Now fixed. As much as I liked my well-crafted text I had there, that section is clearly better off without it. See below. Greg L (talk) 07:14, 16 April 2011 (UTC)

Still, I do not see why a purely factual mention of a real world reported quantity would be harmful. Even if only 1% of the readers would recognise it.
Advertised capacity
by manufacturer
using decimal multiples
Expected capacity
by consumers in class action suit
using binary multiples
Reported capacity
Windows
using binary
multiples
Mac OS X 10.6
using decimal
multiples
some strains of Linux
using IEC
binary prefixes
With prefix Bytes Bytes Difference
100 MB 100,000,000 104,857,600 4.63% 95 MB 100 MB 95 MiB
Woodstone (talk) 07:31, 16 April 2011 (UTC)
Because in Wikipedia articles do not report minority points of view to the same extent as those in the majority and this article doesn't need to use IEC to make a point. Like in the article about Earth it barely mentions the flat Earth theory, it certainly does not go on to deeply discuss the issue. Why? Because that article is about the Earth and as such it does not report or discuss minority points of view. Using IEC prefixes faces the same problems of WP:UNDUE and trying to expand the section of a flat Earth in the Earth article. To include IEC prefixes in the table when they are hardly used presents a misleading report on how the world is. Wikipedia reports how the world is,, certainly not how a minority barely used standard says it should be. To be honest the IEC prefixes are so uncommon that the IEC is lucky to get a mention at all in this article. Fnagaton 08:10, 16 April 2011 (UTC)
Chemical composition of the crust
Compound Formula Composition
Continental Oceanic
silica SiO2 60.2% 48.6%
... ... ... ...
water H2O 1.4% 1.1%
carbon dioxide CO2 1.2% 1.4%
titanium dioxide TiO2 0.7% 1.4%
WP:UNDUE is about (alternative) viewpoints, not (additional) facts. Compare for example the table from your mentioned article earth, of which an extract is shown here. According to your reasoning the lines about water and further should be scrapped, because they form less than a few percent. Am I right in supposing you would not avocate that? How is mentioning one more operating system and its other way of reporting different?. −Woodstone (talk) 09:31, 16 April 2011 (UTC)
Your assumption is incorrect for the following reasons: For the Earth article there are lots of reliable sources that go into such detail regarding the chemical composition of the crust, which is why that table contains such detail. Obviously the subject of chemical composition of the crust belongs to a different field to that of storage capacity so you cannot actually make a like for like comparison about table entries. For the subject of this article the majority of reliable sources do not go into such detail to use IEC prefixes. Which is why this article does not include them. Therefore it would be WP:UNDUE to use them and WP:NPOV to claim you think you need to include IEC prefixes here. Fnagaton 10:33, 16 April 2011 (UTC)
A big difference is that the earth is not flat and that some systems really report mebibytes. That is not a point of view, it is a provable fact. Neither UNDUE nor NPOV apply. Reliable sources are available. To name just one, see Hard Disk Size Barriers. −Woodstone (talk) 11:45, 16 April 2011 (UTC)
WP:NPOV and WP:UNDUE always apply to articles and you are wrong to claim "Neither UNDUE nor NPOV apply". The real world is not flat (even though a minority of sources say they think it is) however the real world doesn't use IEC prefixes either (even if a minority of sources say they think it does). The bar for adding something to an article is not that it is a fact or that someone can find a source or two. If that were the bar then article would be a mess of minority points of view. The bar for adding content to an article according to WP:NPOV and WP:UNDUE is that it must fairly, proportionately and without bias represent significant views of reliable sources. You claim "Reliable sources are available" but it is biased and a violation of WP:NPOV and WP:UNDUE to ignore the fact that many more sources do not use IEC prefixes. Once it is considered that many sources do not use the system you think should be used then the onus is still on you to make a compelling argument for their use despite the sources that stand against your point of view. You have still not made a compelling argument. It is no good trying to make WP:ILIKEIT arguments because WP:NPOV and WP:UNDUE are policies. Fnagaton 12:13, 16 April 2011 (UTC)
1.0000 rem = 1000.0 mrem = 1 rem = 0.010000 Sv = 10.000 mSv = 10000 µSv
0.0010 rem = 1.0 mrem = 1 mrem = 0.000010 Sv = 0.010 mSv = 10 µSv
100.0000 rem = 100000.0 mrem = 1 Sv = 1.000000 Sv = 1000.000 mSv = 1000000 µSv
0.1000 rem = 100.0 mrem = 1 mSv = 0.001000 Sv = 1.000 mSv = 1000 µSv
0.0001 rem = 0.1 mrem = 1 µSv = 0.000001 Sv = 0.001 mSv = 1 µSv
  • I completely agree with Fnagaton here. Woodstone is comparing apples and oranges. The point of a table of the composition of the earth’s crust is to find out what-all is in it. Constituents like iridium, which is rare on earth but was enriched by an asteroid in a layer at the end of the Cretaceous period, is of great interest to readers. The point of such charts are to show great detail; the best ones show constituents at finer levels. What people are not interested in is learning new terminology to use as part of their vernacular if it is used by only by the computer-equivalent of Mongolian yak herders and isn’t to be found on the store shelves at the computer store nor in the magazines they buy.

    If we wanted to really run with Woodstone’s logic, we should burden our tables with obscure lingo used by one ten-thousandth of the computing world (Koko the gorilla?) because radiation tables go down to one-millionth of a REM.

    The place for our readers to be exposed to topics about the units of measure that the computer industry uses (and for the most part, doesn’t use) to measure the digital capacity of computers is in articles on those subjects, such as Megabyte; not by adding unneeded and flat-out unnecessary obscure terminology that hardly anyone uses right into body text and tables showing capacities. The underlying intent of pulling such a stunt is to use Wikipedia to advocate the use of a practice in an industry that is soundly ignoring that practice my making the table come across as “Oh… Didn’t-cha know(!?)…this is the sort of language *we* use in the computer world.” Well… no; such terminology is not used by over 99.8% of the computing world. That would be a clear violation of both WP:SOAP and WP:UNDO WEIGHT. And those two inviolate principles are upheld and reduced to detailed, sound technical writing practice on WP:MOSNUM at Quantities of bytes and bits, which provides detailed guidance on how we can best communicate to readers using words and terminology with which they are familiar. Greg L (talk) 15:03, 16 April 2011 (UTC)

A comparison made above with flat earth, stating "the real world doesn't use IEC prefixes either (even if a minority of sources say they think it does)", shows the the flaw in reasoning. The real world does use IEC prefixes, only not in great numbers, and the (minority) sources do not "think", they factually "use" those units themselves and point to systems that factually use them. Just like the Irridium content of the Earth's crust may only interest a few readers, so do the MiB's. The NPOV and UDUE only talk about viewpoints, not about established facts. Furthermore, what could be the harm of adding the proposed information. Readers not interested in Linux, will just skip the relevant column. I have no understanding of your exclusionist views. −Woodstone (talk) 15:20, 16 April 2011 (UTC)
There’s no point jumping up onto the dinner table and pointing a finger at Fnagaton while yelling “Ah HAAA… you said ‘doesn't use’ !!!” As amply documented above—I won’t repeat it again—well over 99.8% of computer users aren’t exposed to such terminology by their computers. As for “exclusionist views”, that is a principle of Wikipedia that is outlined at WP:UNDO WEIGHT. Along with “follow the RSs”, the intention is to produce good articles that are easy to understand and don’t confuse with trivia or unusual terminology. Try reading WP:WEIGHT some time and you will understand the exclusion. Greg L (talk) 15:36, 16 April 2011 (UTC)
Woodstone the comparison with flat Earth is valid, it actually appears as an example in the policy about how sources must be considered with WP:NPOV and WP:UNDUE which is exactly the same issue as here. You claim to have a source, or some sources, but the fact is that is a minority of sources whereas the majority of sources are against your point of view. It doesn't matter if it is a "fact" or not, the point being that WP:UNDUE and WP:NPOV always apply when using content of sources in articles. You atually used the phrase "established fact" now then that misrepresents the situation because as you know the IEC prefixes are not established in the computing industry, they are very much in the minority and are not familiar to our readers. You mention "your exclusionist views" as if that is my thinking, it isn't, it is the way Wikipedia works as amply described in Wikipedia policy where it goes on at great length to explain why anyone cannot add any old thing they like to articles and how even what some people consider to be facts still have to be sourced and considered with a neutral point of view. To continue to want to use IEC prefixes in this article means you have to argue against WP:NPOV andWP:UNDUE. Instead of getting personal with comments like "your exclusionist views" that misrepresent the issue I ask you to try to make arguments based on Wikipedia policy instead. You have already incorrectly stated you think UNDUE and NPOV policies do not apply which I think it the root cause of why you are making incorrect assumptions here. Fnagaton 20:57, 16 April 2011 (UTC)
I have double checked again and NPOV and UNDUE only talk about viewpoints, not about facts. The proposed addition does not claim that IEC prefixes are common; that would be a viewpoint. Instead it says that a particular operating system uses IEC prefixes; that is not a viewpoint. It could only be construed to be a viewpoint if there are reliable sources that state that particular operating system never uses IEC prefixes. −Woodstone (talk) 09:36, 17 April 2011 (UTC)
You think because the IEC prefix is a "fact" then it can be included, but you are wrong with that assumption. You are wrong because WP:NPOV (and therefore WP:UNUDE that redirects to NPOV) do talk about facts, I cite from the policy page and bold the specific section: "Even where a topic is presented in terms of facts rather than opinions, inappropriate tone can be introduced through the way in which facts are selected, presented, or organized.". Since you are wrong about that then the rest of your argument is also wrong. This is because WP:UNDUE and WP:NPOV apply to the selection of facts as well as viewpoints from reliable sources. As I was saying to select and use a "fact" (IEC prefixes) in an article when that "fact" is not familiar and hardly ever used presents a misleading and inappropriate tone because of the way you want to select that "fact". As such it violates WP:UNDUE to want to include in the article a minority "fact" when the majority of reliable sources do not use that "fact". In any case an IEC prefix is not just a "fact" as you claim, it is also a viewpoint of the IEC that they think IEC prefixes should be used. As such even if WP:NPOV did not specifically talk about facts my argument would stand because obviously the viewpoint of the IEC is not being followed by the majority of the computing industry. As such that viewpoint would still fall under WP:NPOV and WP:UNDUE. Fnagaton 10:22, 17 April 2011 (UTC)
  • Quoting Woodstone: The proposed addition does not claim that IEC prefixes are common; that would be a viewpoint. Instead it says that a particular operating system uses IEC prefixes. Indeed, using them in the table wouldn’t be *claiming* that the use of the IEC prefixes are common in the industry; the practice would be using them in a fashion as if they are far more common than they really are. Since 99.8% of the industry doesn’t use them, nor does one single computer magazine directed to a general-interest readership, nor is such terminology found in the Microsoft® Computer Dictionary, the vast majority of the RSs on this topic don’t use the IEC prefixes. Wikipedia must follow the RSs. MOSNUM’s advise sets the bar appropriately high and states that if a majority of the RSs on the subject of hard drives use the IEC prefixes, then we may do so in this article. That obviously isn’t the case. You know that, Woodstone.

    Your desire to flout bedrock-like principles of Wikipedia pertaining to “always follow the RSs” amounts to SOAP-like advocacy of a cool-beans practice you think the world should know more about. We never allow Wikipedia to be hijacked so. The community was asleep at the switch when a very small cabal of editors—one or two of whom very likely literally still lived in their mothers’ basements as they had a lot of time on their hands—and a rogue admin pulled this very same stunt of advocacy five years ago and literally changed scores—if not hundreds—of articles in one single evening to create thousands of instances of The Dell Inspiron 1501 originally came with 256 MiB or RAM. That idiotic practice that violated several tenets of Technical Writing 101 persisted for three long years. Please recall that Wikipedia’s Wise and Unfailing Leadership By Example®™© for three years did not inspire the rest of the world to follow our practice. So, give it up please; unless it is on an article such as Megabyte or Binary prefix, the IEC prefixes are dead and buried for general use on Wikipedia’s computer-related articles to describe binary capacity in an “Oh… Didn’cha know?!?”-fashion. It’s time to play taps, wipe a tear, and move on.

    Now, WP:MOSNUM#Quantities of bytes and bits is perfectly clear and perfectly applicable in this case. Your reasoning, Woodstone, and that of Tom94022 and RaptorHunter crumbles under even cursory inspection because the underlying motive is advocacy. You are therefore wasting dozens of man-hours on both sides of the fence here as you beat around the bush with wiki-slogans in a vain attempt explain away inconvenient truths. There is nothing in you guys’ arguments that is in the least bit compelling to flout the applicable guidelines on MOSNUM, which are themselves founded upon fundamental, bedrock principals of WP:RS and WP:WEIGHT.

    Please don’t confuse a lack of further response by me to your posts as acquiescence and agreement with your logic and motives. A lack of response would be best explained as my seeing nothing new and persuasive out of you. If that proves to be the case, I might be feeling your persistence to be redundant and potentially tendentious. One tactic I can guarantee won’t work for you is getting your way by refusing to let up in the face of a finding of a general consensus here by a couple of admins. I suggest you move on to other, more productive ways to build the project. Happy editing. Greg L (talk) 18:22, 17 April 2011 (UTC)

Sorry Woodstone but as per Jeh's and Fnagaton's explanations your arguments are specious and contrary to established consensus. Glider87 (talk) 04:55, 18 April 2011 (UTC)
Jeez. Does that post have the dual purpose of making me look loquacious? Greg L (talk) 05:13, 18 April 2011 (UTC)

The few editors right above here claiming that the use of IEC prefixes as proposed is against MOS should realise they are in a minority. See the count of comments below. −Woodstone (talk) 17:43, 18 April 2011 (UTC)

Count of comments
is the use of the IEC binary measures in the table
compliant or in violation with MOS
- In violation Greg L 15:18, 9 April 2011 (UTC)
+ Compliant RaptorHunter 18:46, 10 April 2011 (UTC)
+ Compliant Tom94022 16:19, 9 April 2011 (UTC)
+ Compliant Woodstone 16:37, 9 April 2011 (UTC)
+ Compliant Martinvl 16:47, 9 April 2011 (UTC)
? Compliant but not appropriate Arthur Rubin 17:00, 9 April 2011 (UTC)
- Not Compliant Jeh 20:18, 9 April 2011 (UTC)
- In Violation SteveB67 21:36, 9 April 2011 (UTC)
- In violation Fnagaton 22:52, 9 April 2011 (UTC)
+ Compliant Becksguy 12:32, 10 April 2011 (UTC)
- In Violation NERVUN 18:38, 10 April 2011 (UTC)
+ Compliant Amit6 19:25, 10 April 2011 (UTC)
+ Compliant Ruud 22:35, 10 April 2011 (UTC)
- Not Compliant Glider87 23:47, 10 April 2011 (UTC)
+ Compliant Charles E. Keisler 00:58, 11 April 2011 (UTC)
? Misguided RFC Johnuniq 09:51, 11 April 2011 (UTC)
- Not Compliant Guy Macon 10:45, 11 April 2011 (UTC)
- In violation SWTPC6800 13:48, 11 April 2011 (UTC)
+ Compliant HaPpY EdItInG! 12:40, 12 April 2011 (UTC)
compliant: 9, in violation: 8, undecided: 2

Mr. Woodstone, would you care to join the rest of the class and learn? (That was my best imitation of a 2nd-grade teacher). Some admins have opined that we have a general consensus here that isn’t to your liking. That comes under the heading of “so sad – too bad.”

Now, if you care to read up on Wikipedia:Consensus, an RfC isn’t just head-count. Any notion of relying upon head count was thrown out of a 4th-floor window after an editor who shall remain nameless directly contacted over 100 individuals in a humongous, marathon canvassing session. And, for whatever it’s worth, if we omitted the responses from those editors who had been directly contacted, you’d lose on a pure head-count basis (as if that mattered anyway).

Beyond head-count, a big factor in considering a consensus is the weight and validity of the arguments and their adherence to Wikipedia’s principals. Those are even more important than head-count; WP:Consensus states Consensus is ultimately determined by the quality of the arguments given for and against an issue, as viewed through the lens of Wikipedia policy, not by a simple counted majority. Did you notice that? “Not by a simple counted majority.”

Furthermore, the guideline on WP:MOSNUM governing this issue is clear and unambiguous and was the product of a very lengthy process that was extensively and meticulously conducted and which enjoyed substantial participation on a wide variety of pages and subpages. As WP:CONSENSUS states: Consensus among a limited group of editors, at one place and time, cannot override community consensus on a wider scale.

It simply comes down to the fact that those who came down on the side of not using the virtually unknown IEC prefixes to denote binary capacity consistently cited WP:UNDO WEIGHT and WP:RS as bedrock principals that were being violated. That consensus view had the added virtue of being solidly “through the lens of Wikipedia policy.” That stacks the deck against the other side, whose arguments ranged from “The IEC prefixes are unambiguous and should not be ignored by the world” to “I like them… smell my finger!” Greg L (talk) 19:45, 18 April 2011 (UTC)

Woodstone even if 100 people came along and voted WP:ILIKEIT against one person who makes a strong argument using Wikipedia policy it would still not be consensus for those 100 people. This is because consensus is determined by strong argument and not a simple majority. You know this, or at least you should know this by now given how long you've been editing and since others have directly told you this. So I don't see how you can continue to misrepresent what Wikipedia policy says. Fnagaton 00:06, 19 April 2011 (UTC)

Arbitrary break: Why not?

I didn't read the whole discussion, but I can't see what the big deal is with mentioning and exemplifying IEC prefixes, provided that we also clearly state that their adoption by the media and industry has been negligible. I can see the point about undue weight, but we're talking about one sentence and one table column in an 84-kilobyte-long article. (By comparison, 99.9% of the times the unit of volume equal to 0.568 litres is called just “pint” rather than “imperial pint”, but if we were talking about when you promised me ten pints of beer, I expected to get 5.68 litres but you only gave me 4.73 litres, I would see no reason in banning the i-word altogether.) On the other hand, the section as it is right now looks clear enough to me even only using common prefixes, so I can see no big deal in leaving the IEC prefixes out, either. ― A. di M.plédréachtaí 11:12, 19 April 2011 (UTC)

Your analogy to “quarts” is interesting but doesn’t hit the spot. A list detailing all the world’s “quarts” is similar to a chart detailing the crustal composition of the earth: one detailing the rarer elements and minerals is better; the list detailing the most very rare constituents is best. In stark contrast to that, the issue at hand here is one of the very vernacular used in the industry. When the vernacular and the units of measure customarily used within a particular discipline is the topic at hand, Wikipedia uses a specific standard: one of following the practcies of the majority of the most reliable RSs. That principle is delineated in great detail at WP:MOSNUM.
The whole point of with all of MOSNUM’s advise is to uphold the principal of what any good encyclopedia does: to educate our readership and properly prepare them for their continuing studies elsewhere. One does not accomplish that by using weird terminology in hopes that readers will waltz into a computer store and ask for a SIMM card with “two gibibytes of RAM”. That advocacy and hoped-for grass-roots adoption was a stated goal of the advocates here on this page. We don’t do that.

The lawsuit and confusion was over the industry’s dual definitions of “megabyte” because the IEC prefixes are so very soundly ignored by the computing world. The IEC prefixes are used not by “Linux users”, nor by “Ubuntu uses”; they are used by some partitioning tools (a particular GUI-based one as opposed to the built-in command-line one) available for some users of Ubuntu, which is one of a dozen or so flavors of Linux, which in total has less representation than Apple’s mobile iOS (~2 %). They are a curio and relic of what might have been.

The best way to educate our readership and allow them to learn and do research is to do as we are now doing in the article: Have a table that doesn’t attach UNDO WEIGHT to a most-ignored bit of the very vernacular used by the industry, and to provide three links (MB, GB, and TB) and to provide a See also tophat link.

Think of it this way, A. di M.: imagine that a small (<0.2 %) group of SI purists in the astronomy world wanted to use metres as the only measurement of length in astronomy. Are we to now add a table to Hubble constant so that the very vernacular supposedly used by the astronomy world includes a measure they don’t really use (kilometers per second) per attometre?

Wikipedia will always have science cadets bent on violating WP:SOAP and its admonition against using Wikipedia for the purposes of advocacy. Notwithstanding that the vast, vast majority of RSs in astronomy do not use the SI unit s–1 (reciprocal second) to describe the magnitude of the Hubble constant, which is a measure of how velocity changes with distance, and notwithstanding that doing so is contrary to Wikipedia:Mosnum#Which units to use, which is perfectly clear on the matter (and for good reason), our Hubble constant article—as of this version—provides the way-cool, highly-SI-compliant value along with the measure actually used in the industry: H0 = 70.4 ± 2.4 (km/s)/Mpc (or 2.39×10−18 s−1 ± 3.3%). Note that the citation to which they link (The Astrophysical Journal) didn’t provide the conversion; ‘twas an invention of a wikipedian somewhere (likely a college student giddy with how wise and knowledgable he or she had become in only four short months) trying to promote the adoption of all-things-too-cool and change the world. Indeed, to a certain extent, they did. Because now there are numerous Web sites that also mention this ridiculous unit of measure the BBC, here, which merely copies the text from our article.

We need to be ever-vigilant to prevent Wikipedia from being hijacked for such advocacy; in the case of the IEC prefixes being used on Wikipedia, that practice went on for three whole years for one and only one reason: because the proponents were so animated and so tendentious and because the majority of the rest of the community just doesn’t have the time and patience to fight those who do the Wikipedia-equivalent of showing up at security check points wearing suicide vests.

Please note: that last metaphor about suicide vests isn’t too far off; two editors responsible for spreading the IEC prefixes essentially overnight across Wikipedia’s computer articles fell on their Wikipedia swords after they lost on MOSNUM. One of those two editors who quit was an admin (who had used his powers to revert revertings of what they were doing).
By the way, when I wrote “<0.2 %”, above, I placed a space before the percent symbol. That too is in conformance with the clear requirements of the SI. However, I see that whoever chose to write 2.39×10−18 s−1 ± 3.3% (as if “two quintillionths of a reciprocal second” clarifies anything for a single human being on this pale blue dot) was choosy with what parts of the SI (and MOSNUM) they were going to religiously conform to because he or she used the so-very scientific and SI-compliant (but utterly unheard of) reciprocal second as a conversion and then eschewed the BIPM’s mandate that a space always goes between the numeric value and a unit symbol. Greg L (talk) 16:29, 19 April 2011 (UTC)


  • P.S. Regarding that asinine reduction of (kilometers per second) per attometre into s−1 (on the false premiss that “two quintillionths of a reciprocal second” clarifies anything in the slightest for anyone), the same thing is done with viscosity. I happen to have a Brookfield viscometer (like this) about ten feet behind me as I type this so I quite familiar with the subject. The whole underlying principle of viscosity is resistance to shear forces and the formula for calculating viscosity—when it is actually understandable—is as follows:
µ = K·f / (m/s) / m2·m
…where µ is viscosity
K is a factor for a particular fluid
f is force in newtons
(m/s) is velocity
m2 is the area of the surface being used in the shear-force measurement, and
m is the distance separating the the measuring surface and the stationary one.
Thus, this formula is illustrative of the actual physics underlying how viscosity works; a measuring area in close proximity to a fixed surface and the force required to slide the area at a given velocity.
So guess what Ph.D instructors at colleges and engineers trying to impress do in order to demonstrate why they have claim to special knowledge regarding viscosity. Why, the reduce all those “meter” terms relating to area, distance, and velocity so the formula becomes abstracted beyond all comprehension. They end up with this:
µ = K·p ·s
…where p is pressure in pascals. They had taken every imaginable term and reduced and canceled until they ended up with force divided by area (pressure) even though “pressure” doesn’t have jack to do with the measurement of viscosity.
As we all know, there is far too much of that going on here on Wikipedia as budding technical writers do drive-by shootings trying to show how they have mastery of mathematics and can reduce crap to unholy abstraction until what ought to be clear to our readers is as clear as mud. Too many wikipedians write to impress, not to educate. Greg L (talk) 17:44, 20 April 2011 (UTC)

My personal favorite for many decades has been furlongs per fortnight when talking about obscure or nerdy UofM. As an example of inside humor, VMS references microfortnights (about 1.2 seconds), although apparently actually uses seconds. Also see List of unusual units of measurement.</humor>. — Becksguy (talk) 19:11, 20 April 2011 (UTC)

Tag-bombing the Capacity section

I submit that tag-bombing the section right after all this debate, as Tom94022 and RaptorHunter did (∆ edit Tom) and ∆ edit R.H.) after so much protracted debate (and CANVASSING during an RfC) amounts to IDON'TLIKEIT. The section has been worked on by A. di M., Woodstone, Greg L, and RaptorHunter. Tom94022’s allegation that the current text “advocates” the customary prefixes strikes me as just absurd. Greg L (talk) 01:14, 21 April 2011 (UTC)

  • I removed the tag since it does not have consensus. Tom94022 and RaptorHunter please follow consensus and stop being disruptive. Tag bombing is the addition of multiple tags to an article or adding one tag to multiple articles. (talk) 01:20, 21 April 2011 (UTC)
  • This is a classic example of why the IEC prefixes stayed all over Wikipedia for three whole years. The proponents were willing to resort to such extreme measures to keep the dream alive. They simply wore down the rest of the community until it finally came to a head. Now those two are treating this article like their own Pacific island in 1953 so they can continue to fight WWII. Greg L (talk) 01:45, 21 April 2011 (UTC)

"Tag bombing is the addition of multiple tags to an article or adding one tag to multiple articles." Adding a single tag to an article is by definition not tag bombing. Furthermore, there is no consensus that the tag should be removed since right now the only rebuttal to my contention the article now contains POV are the rather weak opinions of "just absurd" or "beyond me" coupled with personal attacks ("not editing the article to improve it," etc) rather than the discussion as called for by Wiki policy. These personal attacks and violation of Wiki policy probably deserve an WP:ANI but since neither Greg L nor Glider87 seem interested in discussion I will just edit the paragraph into an improved form, one more like the original so-called consensus paragraph, and see what then happens. Anyone want to speculate how long it takes one of the gang of three to revert? Tom94022 (talk) 21:20, 21 April 2011 (UTC)

  • I would suggest that there not be a tag. And that discussion either be had (though I thought it already was) or, better yet, concluded.--Epeefleche (talk) 22:38, 21 April 2011 (UTC)
    • I took your suggestion, even though when you read my now posted concerns about the section and my proposed revision you would agree that the tag was legitimate and the changes made were beyond the consensus paragraph and not discussed. Tom94022 (talk) 00:36, 22 April 2011 (UTC)
  • Please desist, Tom, with your “pay no attention to that obvious truth behind the curtain” stuff; I find your protestation of your tag-bomb protest to be unconvincing. Any experienced editor around here knows it’s a common tactic used on Wikipedia by our most tendentious editors for them to slap IDON’TLIKEIT tags on articles after they don’t get their way by discussing things on talk pages, in RfCs, and in AfDs.

    Your threat earnest pledge to start editing the section in question is fine by me. Just make sure it is abides by the recommendations of MOSNUM and is well cited for those bits that will—or are likely to be—challenged. If you plan on more of your POV-pushing to use Wikipedia as a means of promoting an industry practice that is currently not an industry practice, I’d suggest you think twice.

    Minor detail: please ensure your edits are truly in keeping with the community consensus. You know exactly what I’m talking about here so please don’t play coy on that bit. Poo-pooing “Greg L and his friends” as not being sufficient to overcome your exceedingly wise judgement won’t work.

    And, to pre-empt your poo-pooing of MOSNUM and how it is very bad and smells like a Frenchman’s armpit, per Wikipedia:Consensus#Level of consensus you and RaptorHunter’s little antics here cannot override community consensus on a wider scale. The relevant guideline on MOSNUM was the product of a huge, thorough, protracted, well-advertised discussion. The consensus is that WWII over this “IEC prefix” issue is settled. The current text to which you object is pretty much “just the facts ma’am” and is unremarkable in describing the simple realities.

    The guidelines on MOSNUm are extremely explicit and provide ample guidance on the issue. You and RaptorHunter can may not use this article as your private little Pacific island to continue your battle for applying the Chicago Rule For Voting: “Mention the IEC prefixes early and often.”

    Oh… like Epeefleche wrote, I would suggest that there not be a tag. GFHandle and Glider87 also clearly feel that way too; ergo a clear consensus not to have tags. Greg L (talk) 00:58, 22 April 2011 (UTC)

Capacity measurements section POV

In about 30 edits commencing 23:56, 15 April 2011 Greg L rewrote the Capacity measurements section from the so-called consensus paragraph into a version which is written from the POV of a advocate of customary binary prefixes in that it includes much extraneous material relating to customary binary prefixes and is written in a tone that appears to favor customary binary prefixes over decimal (SI) prefixes. I propose reverting to the so called consensus paragraph. Tom94022 (talk) 23:39, 20 April 2011 (UTC)

  • How in the world you figure that section advocates anything is beyond me.

    You also seem to focus too much of life’s grief and all that gives you discontent as being all-things-bad wrought by User:Greg L. Note that the history of the page is quite clear. That section wasn’t just edited by me; it has been worked on by me, A. di M., Woodstone, and your good friend RaptorHunter, who decided to abide by the general consensus and edit collaboratively on the project.

    Your response by contrast was to tag-bomb that section with a POV-tag (∆ edit).That stunt was purely disruptive and arguably vandalism. If you continue to disrupt Wikipedia like this, you can count on being the subject of a well-deserved ANI in a kibisecond. Greg L (talk) 00:27, 21 April 2011 (UTC)

Greg I advise you to remember WP:CIVIL. Threatening editors you disagree with is not acceptable.--RaptorHunter (talk) 00:53, 21 April 2011 (UTC)
I’m not threatening an editor over a *disagreement*; I’m promising an ANI if he continues to disrupt. He’s not *new* here and knows better. Oh, GFHandel, 1 Kis ≈ 17 minutes so 1 second = 0.0009765625 kibiseconds Greg L (talk) 01:01, 21 April 2011 (UTC)
  • Those two are working in consort to put that tag in. I think this is actionable because the allegation of “POV-pushing” is so amazingly absurd. Greg L (talk) 01:21, 21 April 2011 (UTC)

Why am I not surprised by the response. Would one of the gang of three mind explaining why adding, for example, "However, in computer memory (random-access memory, or RAM), and some other computing contexts, binary math is used ..." is not POV justification and/or TMI of customary binary prefixes in an article on HDD capacity. There was consensus on a paragraph and then GregL radically changed it, why isn't that editing against consensus? I am content to go back to the so-called consensus paragraph, why aren't you? Tom94022 (talk) 01:24, 21 April 2011 (UTC)

  • Greg is not editing against consensus because his edits are intended to improve the article. You (the Tom94022 and Raptorhunter accounts) are not editing the article to improve it, you are placing tags based on a non-neutral point of view that is against consensus. Stop it. Glider87 (talk) 01:28, 21 April 2011 (UTC)
Quoting Tom94022: Would one of the gang of three mind explaining… Very good; thank you for explicitly stating that you can clearly see that at 3 to 2, with the “2” being Tom and RH, there is no consensus for tag-bombing the article. Wikipedia works by consensus. Without that, chaos would reign supreme. Happy editing. Greg L (talk) 01:40, 21 April 2011 (UTC)

My objections to the current revised Capacity measurements section are:

  1. The added sentence "However, in computer memory (random-access memory, or RAM), and some other computing contexts, binary math is used, which only approximates decimal math but where the decimal prefixes such as kilo and mega were given nonetheless in the early days of computing science." is irrelevant to the subject of this section, is TMI and only represents one POV regarding the evolution of computer capacity measurements.
  2. The added list of binary values is redundant, adds irrelevant or non-existent values (kilo, tera, and "so forth") and is POV in that it overemphasizes binary values.
  3. The material on file system reporting is irrelevant and therefore TMI in this section.
  4. The unqualified "had financial consequences" is POV since it is not clear that the settlement costs were of any substantive financial cost to either HDD company.

Rather than just change the section as Greg L did, I have created a proposed new section which I believe corrects the deficiencies of GregL's edits and improves the article over the so-called consensus paragraphs. I would point out the following about this proposal which I believe to be superior to both the prior paragraphs:

  • The first two paragraphs have reliable sources and are essentially a paraphrase of the WD article with the addition of the Fox news comment on familiarity. They are symmetrical in treatment of decimal and binary systems.
  • I chose to use the WD terms "decimal (base 10)" and "binary (base 2)" through out rather than introduce other and potentially unfamiliar terms such as 230=10243.
  • The third paragraph follows from the first two, explains why there is confusion, and comes from a reliable source.
  • The three bulleted examples come from the consensus paragraph but are shorter. There is no extraneous file system material.
  • I moved the actual differences to referenced link to shorten the article and remove what to me is TMI. I suppose I could make this an inline link.
  • The lawsuit paragraph has the POV removed.

FWIW, I think the article would be even better if the third bullet mentioned IEC Binary Prefixes (as does the consensus paragraph) and the associated table used their symbols (as I have always held) and I wish we could have a serious discussion instead of the shouting, frivolous and ad hominem arguments to date. To the extent this proposal will bring on such nonsense (unfortunately I expect such) I invite you to comment on the talk page associated with my proposal where I have the right to excise such nonsense. Tom94022 (talk) 00:06, 22 April 2011 (UTC)

Having skimmed through the discussion so far, I agree that the proposed version at the sandbox is an improvement over the current version for the arguments you have just explained, because it provides a better exposition of the concepts at hand, and because it has more sources which then follows more closely. I have some caveats though: the usage of "base 2 measurements" in the Western Digital source is technically incorrect; the reported quantity 1,073,741,824 is not using base 2 (which would use only the digits 0 and 1) but powers of 2. All references to (base 10) and (base 2) should IMO be replaced by (powers of 10) and (powers of 2) respectively.
  • Good point; I have no problems with this proposed change. 02:18, 22 April 2011 (UTC)
I agree that the IEC Binary Prefixes paragraph, which had consensus to be included, would be a welcome addition as it's directly relevant to the topic of hard drive capacity, explaining that it was a (failed) attempt to resolve the confusion by a high-profile institution. I'd also like to see a wp:summary version of the Binary_prefix#Disk_drives section; the current treatment of Capacity measurements is painfully lacking any context as of the reasons why this confusion appeared in the first place, and that information is actually more relevant in Hard disk drive than in Binary prefix.
Finally, your proposed version should include the current table as it concisely explains the concepts at hand and has already been refined through consensus.
Disclaimer: I arrived to this talk page through the RFC_on_the_use_of_the_IEC_prefixes. I largely agree that there's no need to use the prefixes to describe quantities and that MOSNUM is to be respected (I didn't participate to its creation, but I feel it's a good guideline), but also think that the IEC binary prefixes should be mentioned as a topic in this section. The See also wikilink to Binary prefix feels otherwise unexplained IMHO, which creates undesired wp:astonishment. Diego Moya (talk) 01:39, 22 April 2011 (UTC)
Sorry I didn't make it clear that I intend to use the current table and tag, I was just addressing the text of the section. Tom94022 (talk) 01:36, 22 April 2011 (UTC)
Ok, that's been clarified. Since I seem to come here with a clean slate as to my motivations WRT editing the section, I'll boldly edit the section to include what I deem worthy of your proposal, likely including some additions of my own. Other editors are welcome to modify my edits with rational arguments for their modifications. Diego Moya (talk) 01:48, 22 April 2011 (UTC)
I'm mostly done. I've not included your description of the different OSs; although more detailed than the current one, it doesn't include sources so I'm not confident to use it given the current disputes over the section. Two other details irk me a little:
  • The decimal capacity section uses the prefixes MB, GB and TB, which are not symmetrical to the binary capacity ones (KB, MB and GB); I think KB should be included because it better explains the sequence of powers of two. I would add KB also to decimal capacity, but that's not included in the source.
  • The last paragraph links in wp:summary style to the relevant sections in Binary prefix; this could be seen as a form of wp:overlink (although that guideline doesn't explicitly mention this case). Should I include a more detailed summary of those sections including their references, instead of linking to them? Diego Moya (talk) 02:50, 22 April 2011 (UTC)

Two guys walking into an empty country club, sitting down in the wing-back chairs, and going “Campy idea… indubitably” back & forth to each other isn’t what Wikipedia calls a “consensus.” Greg L (talk) 15:52, 22 April 2011 (UTC)

  • I disagree with everything you wrote. The section has been the focus of intense scrutiny by the community because of your hissy fit. The section has been worked on by a handful of editors now. The better explanation for your sweeping objections is that you vociferously dissented the entire time the current MOSNUM guideline was debated two years ago, still don’t like it, tried to explain how MOSNUM didn’t apply here, you still BELIEVE®™© in the IEC prefixes and earnestly think they should be frequently used and discussed on Wikipedia in hopes that our doing so will somehow rescue them from near-absolute obscurity. The consensus view is not in alignment with your wishes, which are seen as being out of the mainstream and at odds with several of Wikipedia’s core principals. It’s also safe to assume that a number of other editors (you can guess who those are) feel the same way. Let’s see what they think… Greg L (talk) 00:59, 22 April 2011 (UTC)

Misleading assertions

User:Diego Moya had this bit added at the bottom of the Capacity section:


POV-pushing

What he had looked like this. This was editorializing and POV-pushing and revisionism and just flat-out inaccurate. The confusion has not “been addressed” in part by the IEC prefixes. Except for an odd, GUI-based disk partitioning utility used by a particular flavor of Linux, no dominant player in the computing world (not O.S.s, not RAM, not computers, not hard drives) uses the IEC prefixes. So the IEC prefixes are no part of any solution that addresses anything for 99.8+ percent of users. And even for that 0.2% of users, everything they buy at the computer store and every computer magazine they buy that is directed to a general-interest readership doesn’t use such terminology. At all. So the confusion has certainly not “been addressed” in part by the IEC prefixes. So to lump them in with the reality of the situation (the computing world uses the conventional prefixes two different ways) was totally misleading.

Overly aliased links, both advertising the same article on the binary prefixes

Moreover, the links in Moya’s above-quoted paragraph had been aliased to the point they provided the reader no flying clue as to the nature of the article to which they were being taken. Aliasing the living daylights out of our links is an old practice that is being deprecated. Editors are to take extra effort to use no aliasing or to use aliasing that does not totally obscure the nature of the article; readers shouldn’t be confronted with mysterious Easter Eggs and have to hover their cursor over links to find out where they’d be taken or (for most users) click on them to find out. The first one was this:

[[Binary_prefix#Standardization_of_dual_definitions|standardization of the dual definitions]]

And the second was this:

[[Binary_prefix#Unique_binary_prefixes|unique binary prefixes]].

Thus, there were two, highly aliased links in one single sentence advertising the wholesome goodliness of the IEC prefixes and their history. That too amounted to transparent, industrial-strength POV-pushing.

Resolution

I’ve now revised that paragraph so it describes the current state of affairs just like a modern, most-reliable RS address the present state of affairs, which is without POV-pushing that was clearly in violation of WP:UNDO WEIGHT and WP:SOAP, NO ADVOCACY.

The closing paragraph now reads (as of this revision) as follows:

Short, sweet, pure, and simple. And now that there is an explicit mentioning—linked, no less—of the binary prefixes, there is no need for the redundant linked See also tophat directing readers to the Binary prefixes article.

Appeal

To the IEC prefix-advocates: Although reality sucks, reality is what it is. You’ve got your mentioning of the utterly failed and ignored binary prefixes. If we’re gonna mention them, then it will be clear that the dominant players in the computing world have totally ignored them so their relevancy is properly and honestly and accurately conveyed (they were something that could have been but never were). I suggest you not try to push this point any further. Greg L (talk) 14:45, 22 April 2011 (UTC)

P.S. I added the short bit explaining the progression of the binary math. We, as editors, understand this stuff fine. But binary math is not used in daily life by most people and readers shouldn’t have to go read another article to glean the important basics. It would be unclear to many readers how one obtains these powers of 1024 with each prefix. The short section describes how each ten powers of 2 equals powers of 1024 that themselves can be in powers. That section also makes it exceedingly clear how the 2.4% compounded departure with each prefix comes about. User:Tom94022’s objection that explaining this math equates to POV-pushing is utterly absurd and simply without foundation. Greg L (talk) 15:35, 22 April 2011 (UTC)

Binary Math 101

Yes, Diego. Regarding your edit, and accompanying edit summary (8 bits is not the same concept as 2^8, 10 bits is not the same as 2^10)

…please dig out your calculator. 28 = 256 and 210 = 1024. Greg L (talk) 15:48, 22 April 2011 (UTC)

"Eight bits" means "eigth binary digits". "eight bits" doesn't mean "the eight power of two" even though it can carry that much amount of information when used together to represent a single number, which is what my edit summary means. See my expanded rationale below please. Diego Moya (talk) 16:03, 22 April 2011 (UTC)
  • (*sigh*) Try reading before writing. Here’s a link: Bit. The distinction you seem to be drawing is that 10 bits does not *necessarily* mean that they are logically grouped together into a word and therefore not logically even capable of having 210 possible combinations to them. Quibbling. Any plurality of 10 bits is mathematically capable of forming 1024 combinations; all you have to do is point to them or speak of the plurality. So I find your distinction to be razor-thin equivocating far beyond necessity. Nevertheless, I revised the text to belabor about how if one logically groups 8 or 10 bits into a word. Doing so wasn’t in the least bit necessary to drive home the simple concept of 210 = 1024 and introduces insider-speak (“word”) that normally means something entirely else to a general-interest readership. Greg L (talk) 16:55, 22 April 2011 (UTC)

Addressing the latest concerns in Capacity measurements

I'm glad to see that this section is reaching a shape that solves the major contentious arguments in the above discussion. I have some remaining details that still bother me though. I'd like other editors to discuss these points in order to improve the quality of the section.

  • I'm worried about use of the term binary math to describe the values of the prefixes. At Wikipedia that term refers to a binary numeral system, i.e. a base 2 positional number system using only two digits, but then all capacity measurements are expressed in base 10 through powers of two; "binary math" is not a synonym. I know computer scientists have no problem converting between the two, but the article should be accessible to other audiences so there's no excuse to introduce lousy technical language. I think 'binary capacity' as explained in terms of combinations of bits, without any mention of binary math, would reduce confusion and eliminate the current inaccuracies, but I'm open to other proposals.
  • The last paragraph about unique prefixes in its current form contains wp:weasel words: "a trade organization" (which one?) and "the dominant players in the computer industry" (who are they?). Either the IEC should me mention by name, along with the various standardization bodies that did adopt the proposal; or the paragraph should be included to explain that this proposal was not unique but there existed other similar earlier or later proposals by other agents.
  • I'd like to source the assertion that no version of unique binary prefixes have been adopted by the industry. Would User:Greg L or any other editor who participated in the discussions of WP:MOSNUM be kind to provide a reliable source for this? I hope this should be relatively easy for you as you surely must be familiar with the material brought to that discussion. I'm genuinely curious about the history behind the IEC prefixes and their lack of success (I agree with Donald Knuth calling them "funny-sounding" and being taken by surprise by their "official" adoption and later de-facto relinquishment; I think it's a colourful anecdote in the history of computers). Diego Moya (talk) 16:01, 22 April 2011 (UTC)
“Trade organization” isn’t a “weasel word” and your linking to it in that “I made it blue so it must be true”-fashion doesn’t make the allegation correct. Wikipedia is directed to a general-interest readership. Hardly anyone has heard of the International Electrotechnical Commission or Institute of Electrical and Electronic Engineers so “trade organization” is more accessible. If readers want to find out such details, they can click on the link; that’s what those are for.

I don’t agree with the rest of what you wrote and that doesn’t surprise me given the poor technical-writing you did while simultaneously POV-pushing. Your confusion over basic issues like bits and powers of 2 means I’m in for too much time-consuming, time-wasting business than I am prepared for given that I have real-life duties that call. Wikipedia is the encyclopedia that anyone can edit, but not everything that is written here sticks; being factual, true, and balanced not only matters, but is a necessity. Greg L (talk) 16:14, 22 April 2011 (UTC)

I will not answer to ad-hominem arguments since motivations of editors are irrelevant at Wikipedia, only discussion about content and sources matters; I don't want your agreement anyway, only that my arguments be addressed by some willing editor in a rational way.
As for the paragraph you added, you'll excuse me for disagreeing, but there's absolutely no indication in it that the provided link (by the words "binary prefix") will contain a subsection about the discussed proposal; the link can only be reasonably interpreted as linking to a definition of binary prefixes. Do you expect readers to be prescient and know that the information they want will be waiting for them at the other end of a link they haven't followed yet? Unless you do, the words "a trade organization" remains an unsupported attribution (a.k.a. weasel word). Is there a problem in changing the sentence to "the IEEE, IEC and ISO trade organizations attempted to address these dual definitions..."?
There's the alternate possibility of linking directly to Binary_prefix#Unique_binary_prefixes expanding the link to the words "by proposing unique binary prefixes and prefix symbols". This would provide enough information scent for the reader to find out more about the existence of this proposal that was mentioned in the included paragraph; but you don't seem fond of this kind of wikilinks, so I'd opt for mentioning the name of the organization making the proposal and copying here some of the references available at the other article. The request for a reliable source in particular for "the proposal has seen no adoption by the dominant players in the computer industry" you should not ignore, as this assertion is likely to be removed. Being verifiable is even more important, and as you said, not everything that is written here sticks. Diego Moya (talk) 16:58, 22 April 2011 (UTC)
  • Let me know when you find a dominant player in the market using the IEC prefixes. Not even one single manufacturer of RAM, such as Crucial (the most logical segment of the computer industry that would adopt such terminology) uses the IEC prefixes. The proposal failed. Don’t you understand that yet? I’ll find a citation later. Or you can add the citation (it’s called “collaborative writing”). Don’t delete an indisputable truth just because you don’t like the truth and I don’t have time to dig up a citation at the moment. It’s going to be tough to cite because about the world’s RSs don’t even talk about the whole issue of the IEC prefixes; POV-pushers on Wikipedia are the only ones discussing them as they try to Keep the Dream Alive®™© by getting them mentioned and linked at every turn and aliased beyond all comprehension so other editors can’t easily discern that there are multiple links in the same sentence to the same article. Pure silliness.

    Mind you, all this fuss is about trying to mention the IEC prefixes in an article on hard drives of all things, which is an industry that uses decimal math to define the magnitude of the conventional prefixes and wouldn’t even use the IEC prefixes if the terminology were actually used elsewhere. Words like “mebibyte” aren’t even in the Microsoft® Computer Dictionary. It’s been 15 years and all the players in RAM are still using the terminology we all know. Be happy that the IEC prefixes are mentioned in this article at all. My stripping out any mention was because it wasn’t necessary for that section or this article; adding it back was a concession to make peace with the IEC proponents (I know: overly optimistic). As the topic of the IEC prefixes is currently addressed, it is at least informative and accurate—if not a bit off-topic. Greg L (talk) 17:10, 22 April 2011 (UTC)

As I said, what I understand or not is irrelevant to the discussion; I'm only trying to ensure that Wikipedia content policies are met in the article, given that those for collaboration are not being followed in the talk page. I don't care whether the IEC prefixes are mentioned or not, I'm not here to advocate their usage by the industry; what I care is that the Capacity measurements section at the Hard disk drive article at Wikipedia explains that measuring capacity of hard drives has lead to significant confusion through the years, and that prominent agents in the industry tried to solve it in various forms (one of which is the IEC proposal, but I don't care whether that particular one is used here); those are significant facts about the topic, and trying to suppressing them amounts to some bit of POV-pushing of your own. We don't need your condescension for their inclusion, only that regular consensus-forming processes are respected.
Nor I have the least intention to delete the sentence myself, I'll just add a [citation needed] tag until someone finds a good source for it. As I already said I'm not familiar with the MOSNUM, so I assumed good faith and presumed that you would be eager to support that reference given that you've been at this talk page for weeks, waving that "undeniable fact" (irrelevant to Wikipedia pillars, btw) as justification for your inordinate defense of your posture; you should have found that reference much earlier, and not dared to introduce that assertion into a contended section without it. Now that the section is taking good shape I hope that you stay cool and in the future you will strive to try to use some WP:WikiLove. Now if you go and re-read my three original concerns under this light you'll find that what you've added to this section so far is completely orthogonal to them, since I'm not trying to push for the IEC prefixes. ;-) Diego Moya (talk) 17:50, 22 April 2011 (UTC)
  • Very well. Peace. Hopefully someone can find a statement by an RS that proves a negative about an issue not even being discussed in the computing world. It seems to me it would be easier to show through elimination, the links to all the major players proving they don’t use the IEC prefixes than find an RS discussing this obscure subject stating they don’t. SWTPC6800 is amazingly good at this sort of thing. It’s like my writing “Klingon is a constructed fad language that hasn’t seen any adoption in mainstream populations.”[citation needed] What the heck?!? What RS to whom we might cite would want to write on such a topic? I still suspect the proper way to prove obviousness is to challenge you to show me one single major player in the computing world that uses the IEC prefixes. But the cite tag is fair enough for the moment. Ample white space is provided below for you to show me a major provider of consumer operating systems or RAM who is using the IEC prefixes when communicating to a general-interest consumer. Just a suggestion… Greg L (talk) 20:00, 22 April 2011 (UTC)
The verifiability policy is crystal clear that the burden of evidence relies on the editor who adds material to the article; if you want to add content about the major HD players or the popularity of Klingon in mainstream populations, you better have a reliable reference that can be cited as the source. That said, I'm perfectly fine with using an enumeration of some major HD builders which can be verified to not use IEC, instead of an unsourced assertion about all the (unknown to the reader) major players. This way we would be able to verify exactly how relevant is that assertion to the article instead of inducing ourselves in wp:OR. Diego Moya (talk) 20:25, 22 April 2011 (UTC)
I agree with Greg L. It shouldn't be a driving concern to provide a source for something that is observationally-verifiable—as is that assertion. We don't need a source verifying the statement "the human hand has five fingers", and making the observation that the IEC units have not been adopted by the industry is only slightly less observationally-verifiable. If that sentence needs a "citation needed" tag, then I can point out dozens of sentences in about 3,000 kibiarticles that need a similar tag. Perhaps if this article ever gets to FAC it would need addressing, but I'm comfortable with the uncited assertion (as would be 0.097646484375 kibipercent of knowledgeable readers who drive through these parts).  GFHandel.   20:54, 22 April 2011 (UTC)

I found a sufficiently authoritative citation for the obvious point that the computer industry has for the most part thoroughly ignored the IEC prefixes.

While staring at that paragraph, I realized we were doing our readers a disservice by not providing an example of what these mysterious IEC prefixes looked like. So I added an example by name and all written out ‘n’ stuff (*sound of audience gasp*) Why?, because not providing an example after mentioning their existence stood out like a sore thumb; no proper encyclopedia would tantalize like that. Our failure to address this omission made the damned things as mysterious as those Playboy magazines atop the toilet tank at your friend’s parent’s master bedroom bathroom: OK, pubic hair; now, onto something else for excitement. Greg L (talk) 04:33, 23 April 2011 (UTC)

Is this book the one you're using as a reference? It certainly talks about the IEC prefixes and mentions that "many readers are unfamiliar" with them, but I can't find where it addresses the "little adoption by the computer industry". Can you point to the sentence that supports that assertion?
I'm glad that you clarified the discussion of the mentioned prefixes with the MiB example; it does better explain the concept at hand, and at its current size it isn't given undue weight. Diego Moya (talk) 10:07, 23 April 2011 (UTC)
there has been little adoption etc. Glider87 (talk) 12:54, 23 April 2011 (UTC)
Yes, “sadly” so, Glider; little adoption—as everyone knows, even the IEC proponents; no one would be so foolish as to suggest otherwise. As for Diego’s question, the book’s title, author, page number, and ISBN are sufficient. I would normally suggest he learn how to better exploit the power of Google Books or go to the library on his own without demanding that others educate him on how to improve his craft at being a wikipedian. Nevertheless, the sentence is as follows: “Even so, these new prefixes have yet to be widely adopted and confusion still reigns.” Given that no one could possibly point to evidence that the IEC prefixes have enjoyed anything other than little adoption (because they haven’t), Diego’s wikilawyering doesn’t impress and is akin to demanding citations that Raquel Welch is female. One wouldn’t think that point of fact would be “likely to be challenged” but some editors will surprise you. Even when Diego lands on this talk page, common sense talks and bureaucracy walks. Greg L (talk) 14:05, 23 April 2011 (UTC)
Given that I suggested how to introduce the facts in a way not open to interpretation and which you directly ignored, that you've been mocking my concerns for this article based on some paranoid feeling of yours, and finally you introduced an overly general assertion for which you didn't care to provide references, I rest assured that my attention to detail has served to improve the article and bring it to compliance with a core policy. Diego Moya (talk) 15:26, 23 April 2011 (UTC)
Yes, you are entirely correct and I was perhaps a bit too harsh, Diego, and was unreasonably put off by your demand to cite a point of fact that no one around here was disputing (near-zilch adoption in the industry). But then, perhaps you saw the handwriting on the wall; particularly in light of how tendentious some of the arguments had been on this page. Your collaborative contributions here resulted in a clearly improved article. You got your way—if one can call it that—on certain points and graciously yielded on others. Rather like how both civil and criminal law are practiced in the West, out of the struggle comes truth and the reader is the benificiary. Thanks. Greg L (talk) 15:36, 23 April 2011 (UTC)
Now that's a welcome change of tone! :-) I ask forgiveness that I was becoming a bit harsh myself. Diego Moya (talk) 16:10, 23 April 2011 (UTC)
Two wikipedians are usually better than one. Greg L (talk) 16:28, 23 April 2011 (UTC)



P.S. To a particular editor who shall remain nameless: Please don’t use my good-faith effort at endeavoring to exercise best technical writing practices (as described in the above paragraph) as an excuse to now start agitating on getting that column back into the table that was dedicated to conveying capacity using the IEC prefixes. Nothing has changed about that issue. Doing as you have previously (strongly) pushed for would be wildly in violation of WP:UNDO WEIGHT as well as flouting the wide community consensus that resulted in the perfectly clear guidance on MOSNUM. That last paragraphs in the article discusses the IEC prefixes and shows the reader what they look like and tells of how the computer industry has for the most part ignored them. They are history. That’s the end of it now. It will not be a good thing if you try to exploit good-faith, best-writing practices by trying to turn it into a “camel’s nose under the tent, so why not the whole camel in the tent”-sort of thing. I’d be perfectly content to sacrifice what I think best serves the interest of our readership and remove the example of what the prefixes look like (an “i” in the middle) if that’s what it takes to restore peace and stability around here. Greg L (talk) 05:30, 23 April 2011 (UTC)

  • "It shouldn't be a driving concern to provide a source for something that is observationally-verifiable" - Agree. Everyone with a neutral point of view knows the IEC prefixes are hardly ever used in the real world. Glider87 (talk) 07:08, 23 April 2011 (UTC)

I've reworded the sentence that introduced the correspondences between decimal prefixes and their assigned power-of-two values. That one was again using the term 'binary' to describe a decimal value that happens to be a power of two. I've also changed there the link to binary prefix, since that article is not only about the IEC prefixes but about this precise usage of decimal prefixes when used to represent powers of two.

I really think the last paragraph would be improved if the sentence "...by proposing unique binary prefixes and prefix symbols to use for denoting powers of 1024" would directly link to the unique binary prefixes section; even if that article is already linked to, this second link is really about a different concept and a direct link would help the interested reader. But this time I don't want to boldly push it without consensus given the heated debate about this paragraph. I've prepared a redirect to avoid a piped link to a section. What do you think? Diego Moya (talk) 17:35, 24 April 2011 (UTC)

Sure, Diego, the concept of linking to the IEC stuff is fine and best serves the reader. I simplified it so the link comes about more naturally (linking to article sub-sections is a tad contrived). I just linked “mebibyte”. Once there, the reader can explore the world of that stuff. Greg L (talk) 18:39, 24 April 2011 (UTC)