Wikipedia talk:Manual of Style/Computing

From Wikipedia, the free encyclopedia
Jump to: navigation, search
WikiProject Manual of Style
WikiProject icon This page falls within the scope of WikiProject Manual of Style, a drive to identify and address contradictions and redundancies, improve language, and coordinate the pages that form the MoS guidelines.
WikiProject Computing / Networking / Software / Hardware (Rated Project-class)
WikiProject icon This page is within the scope of WikiProject Computing, a collaborative effort to improve the coverage of computers, computing, and information technology on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.
 Project  This page does not require a rating on the project's quality scale.
Taskforce icon
This page is supported by Networking task force.
Taskforce icon
This page is supported by WikiProject Software.
Taskforce icon
This page is supported by Computer hardware task force.

Needs overhauling[edit]

I've lightly copy-edited the top, but I seriously doubt whether anyone takes any notice of this guideline as it stands. One change I made needs expert review, I think: "Video game vendors that make sequels and prequels for their most notable products sometimes affix to the name of a product a different number to the version number." It was pretty clunky before. Is the meaning correct now?

I see two structural issues straight off:

  1. It starts with very specifically targeted advice, about starter packs. It should start with broad advice about style that applies across the board (with exceptions, perhaps, but they should be noted for editors' benefit). I've gnomed through a few related articles and found commonly repeated stylistic horrors, which underlines the need for a more useful MoS/Computing.
  2. There are probably opportunities for merging guidelines into this one.

I'm going to leave a note at MoS main page, at whatever WikiProject is relevant, and at the talk pages of a few people in the area who I know are interested in style.

Your thoughts? Tony (talk) 01:22, 14 September 2011 (UTC)

Thanks for the light copy-editing. Much appreciated.
As for the two other points:
  1. It starts with "common mistakes". That's the broadest topic I could find. And yes, it targets very specific audience, i.e. those who write computer articles. I hope I didn't miss anything. Did I?
  2. Unfortunately this article is an essay and not part of the "official" manual of style. (One day, it will be.) Until then, we must be careful not to merge it with anything official without gathering consensus first.
Fleet Command (talk) 06:27, 14 September 2011 (UTC)
Thanks FC. But if it's not part of the MoS, why does it have MoS/ in its title? Something's amiss here, and I'd say a computing/telecom MoS subpage is very much needed. Tony (talk) 06:35, 14 September 2011 (UTC)
Simply because I loved the title and I wanted it to become MOS one day. I asked an admin if I am allowed to use that title and he said he cannot find a clause that prohibits me and as long as I put the essay tag on it, it is fine.
And tell me about telecom. Maybe we can expand this essay. Fleet Command (talk) 06:54, 14 September 2011 (UTC)
Well, I think it needs to be a proper MoS page if MoS is in the title, and it's clearly a slip-up that whoever changed the title to our new formatting, with the slash, didn't notice the essay template. You might be interested in contributing to the thread at MoS main page, where a few other possibly similar pages have been mentioned. Telecom and computing ... is there a distinct boundary between them? What really counts is utility for editors who seek stylistic guidance. Sorry to be critical (now it's clear that it's largely the effort of one editor). Your expertise would be welcome: these topics are a bit of a mess in stylistic terms. How does this page relate to Wikipedia:WikiProject Computer science/Manual of style (computer science), in your view? Is there an overlap, and could they be conflated? Tony (talk) 07:09, 14 September 2011 (UTC)
I proposed a merger and received a non-negotiable "No"! But they do overlap. As for telecommunication, it is in direct connection with IT. For instance, take VOIP and unified communications. And yes, this article was almost entirely my work. (I felt there was a lot of unwritten consensus lying around that needed writing.) Not anymore! You copy-edited and now it is a joint work. That's how joint works start. :) Fleet Command (talk) 07:46, 14 September 2011 (UTC)

Wikipedia:Manual of Style/Command-line examples is the other obvious merger candidate. In the meanwhile, I would like to make an attempt to add some proposals for the issues we recently ran into, for example on article titles. W Nowicki (talk) 18:23, 14 September 2011 (UTC)

We definitely need guidance as to upcasing and downcasing. It's clear that protocols, trademarks, and proprietary names should be upcased, but systems and servers and equipment items such as USBs (expanded) need to be dealt with. This is one of the glaring needs. Tony (talk) 02:59, 15 September 2011 (UTC)
WP:MOS already has a section on this, so I guess that would be redundant. Fleet Command (talk) 11:47, 15 September 2011 (UTC)
Yes, you'd think so; but the devil is in the detail in computer and telecom articles, where fine judgements sometimes need to be made concerning whether an item is a proper or common noun. MoS expresses the general principle, but it hasn't been enough to stop rampant capitalisation, which in the end makes the text harder to parse, especially for non-experts. Tony (talk) 12:00, 15 September 2011 (UTC)

Well just like in normal articles, I wanted to summarize and link to the appropriate in-depth discussion. Indeed, the problem is for topics like DSL, which started as one specific protocol (thus proper noun) but evolved into a generic term for a family of related technologies (so is being proposed now to down-case). Anyway, please review my proposal and let us discuss if it makes sense. We can then add other guidelines as we run into related issues I hope. W Nowicki (talk) 18:40, 15 September 2011 (UTC)

Sorry, Nowicki, is your proposal linked here? Tony (talk) 00:56, 16 September 2011 (UTC)

I meant my changes to the proposed MoS/Computing, for which this is the talk page. delta here W Nowicki (talk) 16:55, 16 September 2011 (UTC)

Styleguide version history / release history tables[edit]

I'd like to introduce the Template:Version template to this styleguide, with the goal to establish it as a standard. It simplifies creation of release histories, standardizes release stages and makes the content more accessible:

Version Description
Old version, no longer supported: 1.0 Old version
Older version, yet still supported: 1.8 Older version, yet still supported
Current stable version: 2.0 Current stable version
Future release: 3.0 Future release

I imported it from the german de:Vorlage:Version and I synchronize it manually. It is used widely on de:WP and begins to grow on en:WP Please comment and/or help improving the template. — Preceding unsigned comment added by Jesus Presley (talkcontribs) 20:53, 26 November 2012 (UTC)

Currently, in English Wikipedia, we have our own style. See articles like Internet Explorer 10#Release history, Firefox (actually Template:Firefox usage share) or Google Chrome#Release history.
Color Meaning
Red Old release
Green Stable release
Blue Beta release
Purple Dev release
But in my humble opinion, this is a Color of the bike shed issue. It must not be put into the manual of style without strong consensus. Please consider calling in additional input. RfC, village pump or various WikiProjects (software, computing, Microsoft, Apple, etc.) may be used to attract input.
Best regards,
Codename Lisa (talk) 21:11, 26 November 2012 (UTC)
P.S. Fragments should not end with period. Also, I forgot to mention: I don't like it. Best regards, Codename Lisa (talk) 21:37, 26 November 2012 (UTC)
Good ideas to use such a template, but why with colours? As someone with (slight) visual acuity problems, I hate that people don't think about those who have problems. There is nothing wrong with the default black on white that most of Wikipedia is published in. --Biker Biker (talk) 21:39, 26 November 2012 (UTC)
Hi. Actually, template documentation page says this template ({{version}}) populates accessibility data. (No more explanation is supplied.) In addition, as much as I appreciate your attention for visually-impaired people, I find that people blessed with the gift of sight also need attention. Wikipedia, for the most part, is black and white. (Images add color but are constantly removed with reason that amount to excuses.) Best regards, Codename Lisa (talk) 22:29, 26 November 2012 (UTC)
Thanks so far for the comments. Accessibility is provided by replacing the color codes with replacement information inside the "title" attribute. As I only translated the template, I cannot report of more accessibility features. The color scheme (of course) is replacable, it can easily be adopted to the one that CN Lisa mentioned. However, above colors are not consistant at all across en:WP. IMHO it makes sense to introduce one standard - and I didn't find any styleguide that decides on this. I am happy with differing styles for (german) de:WP and en:WP. Jesus Presley (talk) 23:25, 26 November 2012 (UTC)
Green tickY I moved this discussion to the template talk page. Please continue there. Jesus Presley (talk) 01:27, 27 November 2012 (UTC)

Cost and license should not be combined[edit]

Cost and license are entirely different things and should not be combined. Cost deals with money, licensing deals with rights. Palosirkka (talk) 19:53, 9 August 2013 (UTC)

Hi. I entirely agree with "Cost and license are entirely different things and should not be combined".
What I don't agree is what you call "cost" and "pricing information" (judging by our conversation in your talk page.) For example, you regard "freeware" as word that specifies pricing information. I don't. Most Wikipedia, by the look of it, does not. (Evidence: The volume of editing that you have done so far and is disputed. When you find an entire world in need of editing, please take into consideration that your point of view may not have consensus.)
Freeware, by definition, is proprietary software that can be used and often shared with family and friends without restriction, free of charge. "Proprietary", "used ... without restriction" and "shared ... without restriction" are use term. "Free of charge", yes, it is price. So, "freeware" is three parts licensing info and one part price info. Evidently, you have only taken "free" and "ware" and thought it is pricing. No.
So is the commercial software: "Commercial software" indicates that software may only be used by licensee, may not be shared, restrictions such as number of computers and/or number of users is in effect, a price must be paid. So, the term "commercial software" is three parts licensing terms, one part price.
One thing I find disturbing in your editing is that although you wish to eliminate the distinction between "freeware" and "commercial proprietary software", calling them all "proprietary", you take your time to separate "free software" into "GNU GPL" and "MPL".
That said, since we already disagree, I called in an RFC. Best regards, Codename Lisa (talk) 20:10, 9 August 2013 (UTC)
Looks like our little spat is a misunderstanding, a communication failure. I concentrated too much on the rousing allegations of censorship as that is something I'd never do. I guess neither of us assumed good faith. What you say about freeware and commercial software makes sense after all. You have convinced me now. Palosirkka (talk) 15:41, 10 August 2013 (UTC)
However, looks like your definition of commercial software differs from the one in the article which says that commercial software is a commentary on non-zero price alone. Palosirkka (talk) 15:47, 10 August 2013 (UTC)
Hi. About the "commercial software", I never disagreed with you. I agree completely. But it is exactly for that reason that Wikipedians always write "proprietary commercial" to indicate that a usage right is intended. (I myself write "commercial proprietary" to better deliver the message.) As the article says, "free software packages may also be commercial software". So, neither "proprietary" nor "commercial" alone sufficiently describe a licensing term because proprietary is a source model and commercial is a price model unless it is used in a context that indicates proprietary source model. (This "unless" portion is part of the key to our dispute.)
Best regards, Codename Lisa (talk) 18:11, 10 August 2013 (UTC)

Tenses in articles about discontinued software[edit]

This has come up in the FAC for The project stopped in 2011. I see that, for example, RSTS/E is entirely in past tense, Microsoft Works has a not-entirely-clear mix of tenses. I put entirely into past tense - but is there a perceptible general practice about how to address dead software? - David Gerard (talk) 19:52, 17 October 2013 (UTC)

There are the school teachings, you know. Facts about software that are true regardless of time should be written in simple present tense while timed events should be in other tenses. For example, Windows 95 is an operating system; it is a discontinued operating system but an operating system nonetheless. Discontinuation does not turn it into a tomato. So, "Windows 95 was an operating system" is wrong.
That said, tense consistency is an issue of story books. Wikipedia articles are not story books to have a consistent tense. There is nothing wrong with mixed tense.
Best regards,
Codename Lisa (talk) 20:39, 17 October 2013 (UTC)
I see what you mean. OTOH, for RSTS/E, was an operating system would not be unreasonable. (The software, company and hardware are long dead; old geeks might fire it up in an emulator once or twice just because they can and go "ooh, lookit that!") But OOo is not quite that dead. I might try recasting it along the lines you suggest - David Gerard (talk) 21:13, 17 October 2013 (UTC)
Or not. Microsoft Works is basically along the lines you suggest and comes across as a mess in terms of tenses, often switching in the same sentence. It's not striking me as an example to emulate - David Gerard (talk) 21:35, 17 October 2013 (UTC)
Hi. You are treating software like humans! It is an interesting take, one that shows your sensitivity to accuracy; but metaphor and Wikipedia are not really in good terms. Take this example: "Julius Caesar was a ruler". To our knowledge, he was not born a ruler but became one and ceased to be one when he died because rulership is invalidated with death. Now, which one of these two sentences are correct about a dead wife? "Lisa Hayes was my wife" or "Lisa Hayes is my wife"? That depends on the culture; some take death as an end to existence, while others take it as another great adventure. And it is generally not a good idea to say "Lisa Hayes is my wife" peremptorily to a woman one is dating.
But consider cold hard facts: "Halberd is a two-handed pole weapon." "Sword was once a widely-used weapon." "Gun is a modern weapon."
If we go by a rule-based approach, "To be" verb in all these sentences is a state verb and changing it to past tense requires the state to be invalidated. "Dead" is also an absolute adjective; one thing is either dead or not. That's why in legal context, which is strictly rule-based, Lisa Hayes is treated as a deceased wife, not an ex-wife, and Julius Caesar is a deceased ex-ruler. Treatments like "was my wife" are not rule-based; they are emotional. Saying Windows 95 was an operating system because we don't see it around is an emotional treatment of Windows 95. A rule-based treatment says "Windows 95 is an operating system which we don't see around much." (You and I both could argue that "don't see it around much" is also an emotional treatment too.)
IMHO, the fact that Microsoft Works is badly written and is using rule-based treatment of the tenses does not sanction emotional treatment of the subject. The solution is to resolve the badly written state through a rewrite that still supports rule-based treatment of the tenses.
Best regards,
Codename Lisa (talk) 22:52, 17 October 2013 (UTC)
tl;dr "write well". Well, yes ;-) - David Gerard (talk) 07:26, 18 October 2013 (UTC)

──────────────────────────────────────────────────────────────────────────────────────────────────── I'd like to fire up this discussion again. I just got corrected at Intel Binary Compatibility Standard, which I had written in the past tense, because this is a dead standard. MOS:COMPNOW currently states that:

TrueCrypt was a disk encryption program, released by TrueCrypt Foundation.

This sentence suggests that TrueCrypt is not a disk encryption program, although it once was. (Assume TrueCrypt has never changed its nature during its development lifecycle.)

I wholeheartedly disagree. This text does not, of itself, suggest that TrueCrypt changed into a tomato, as Codename Lisa puts it. It just says that at some point in the past, TrueCrypt was an encryption program, just like "Uruk was an ancient city of Sumer". That's not saying it turned into anything else. Its present status is simply left open; it might be lost, ruined, put to other use, etc. (I also find it hard to believe that "to be" is a dynamic verb here. "TrueCrypt was being an encryption program" strikes me as ungrammatical, but I'm not a native speaker.) QVVERTYVS (hm?) 20:19, 13 November 2014 (UTC)

Hello! FWIW, using present tense per MOS:COMPNOW makes sense to me, as software remains what it used to be even after its discontinuation. For example, you can still download Slackware 3.6 ISO and use its packaged iBCS kernel module as if iBCS hasn't been discontinued long time ago. On the other hand, Uruk, as an ancient city of Sumer, can't be brought back to life that way. In other words, iBCS still exists in its original form, which was pretty much virtual all the time. In comparison, some ancient city also still exists in a virtual form, as a set of records, while its original, physical form, has been lost many years ago. Just my $.02. :) — Dsimic (talk | contribs) 21:37, 13 November 2014 (UTC)
Hello, Qwertyus
  • I also find it hard to believe that "to be" is a dynamic verb here. Well, because it isn't. It is a stative verb.
  • it might be lost, ruined, put to other use, etc. The first two almost never happen to computer software. The last is a valid state in which you can use "was", e.g. Final Cut Pro was a professional video editing app. It is now a consumers video editing app because the last version Final Cut Pro X, changed things a lot.
It seems you and I agree more than you think.
Best regards,
Codename Lisa (talk) 22:04, 13 November 2014 (UTC)
Hm. Aside from the whole discussion, can we maybe remove the stative/dynamic verb stuff from MOS:COMPNOW? It seems really beside the point. QVVERTYVS (hm?) 10:11, 14 November 2014 (UTC)
I wish it was that easy. They are the source of the problem, not a comment. Also, as I told David above, they are the rules, the basis of our rule-based treatment of the subject. Best regards, Codename Lisa (talk) 21:18, 14 November 2014 (UTC)


Can I add to License section clarification and standardization on license tags. For example:

  1. GPL instead GNU GPL or GNU General Public License
  2. BSD instead BSD License or Berkeley Software Distribution or Berkeley Unix
  3. MIT instead MIT License or X11

--Rezonansowy (talk • contribs) 20:34, 2 November 2013 (UTC)

No, you can't; the guideline reflects community consensus and currently there is no wide consensus on restricting any of these correct forms. The community is free to choose whichever it likes. In fact, per WP:STABILITY, ArbCom has previously blocked people who made such edits.
Best regards, Codename Lisa (talk) 20:51, 2 November 2013 (UTC)
In fact, I would only to make a provocation to discussion about it. Anyway thank you for quick info, let us begin the discussion to reach the consensus. --Rezonansowy (talk • contribs) 21:51, 2 November 2013 (UTC)
First, let me tell you my feelings about such a mandate: I see no harm in using either "GPL" or "GNU GPL" and bluntly spoken, I think I'd be a very mean and sadistic person if I forced other people to surrender their freedom and stop using "GNU GPL" in favor of "GPL".
Second, A simple talk page discussion with a few dozen participants is not enough to override ArbCom sanction.
Best regards,
Codename Lisa (talk) 10:01, 3 November 2013 (UTC)
I forgot to mention an important thing that this will be useful only in infobox. I agree with it, editors may write as they wish. Determined writing of the license in infobox will be also useful when infobox entries will move to Wikidata. Besides, I don't want to override any sanctions. If this is a wrong place to propose it, then just please tell me where i have to do it. --Rezonansowy (talk • contribs) 12:05, 3 November 2013 (UTC)
Hi. Policy proposal ordinarily go to Village Pump. But in this case, you must first contact an arbitrator to check with him. Basically, you should state what you want to do and provide a very strong reason as to why this case is so urgent that should be exempt from past ArbCom sanctions, e.g. trying to prevent a huge damage to Wikipedia which cannot be prevented by any less restrictive measures. If he or she confirmed that you are indeed right and something should be done about it, then you proceed with the proposal in the Village Pump. There, you must collect a huge consensus.
Best regards,
Codename Lisa (talk) 13:20, 5 November 2013 (UTC)
OK, thank you. I'll try to take this challenge :). --Rezonansowy (talk • contribs) 12:27, 6 November 2013 (UTC)

Linux vs GNU/Linux[edit]

I basically agree. I've rephrased it a bit as a guideline rather than a rule (because the exceptions do exist, and legislating against editorial judgement is silly). I've also toned down the second paragraph, which came across as condescending. Do we have handy links to the discussions, since we're referring to them? - David Gerard (talk) 13:57, 4 January 2014 (UTC)

Hello, David. Thanks a lot for the review, though I'd rather we supply the reasons based on which the discussions are ended, i.e. WP:COMMONNAME and WP:NPOV.
The first paragraph, however, is where I tilt my head a little. "Generally," is redundant. (All MOSes have generally written on them.) "Exceptions include proper [...]" does not apply here at all because when you say "proper names", you are automatically walking outside the domain of "family of operating systems". So, this not an exception.
I am going ahead and fixing it, but if you didn't like it, you know where the revert button is, right? Bless you.
Best regards,
Codename Lisa (talk) 23:16, 4 January 2014 (UTC)
Looks good, cheers! - David Gerard (talk) 09:28, 5 January 2014 (UTC)

Debian GNU/Linux is specifically cited here, although in practice Debian seems to be used throughout WP. A clearer example is perhaps Trisquel, which specifically describes itself as Trisquel GNU/Linux on its front page, unlike Debian. The Trisquel article was recently moved to Trisquel GNU/Linux, and then moved back again, citing the MOS, although the MOS seems to indicate Trisquel GNU/Linux is preferable. See talk Greenman (talk) 10:00, 7 November 2015 (UTC)

@Greenman: Is the fact that "Trisquel GNU/Linux" is the common name established beyond all doubt among the community, or is it challenged too? And I hope this isn't one of those cases when the editor practicing WP:BOLD has forgot to add an edit summary and riled up everyone! (That's pretty much always the case. *sigh*) Fleet Command (talk) 05:46, 8 November 2015 (UTC)
Fleet Command. no, Trisquel would be the common name. MOS:LINUX appears to be ambiguous then, as it begins by stating that GNU/Linux should be used for the proper name of certain individual OS's, such as Debian, and only later mentions WP:COMMONNAME, but doesn't go into when each suggestion applies. Article names? Use in text? Use in contexts where a full name is necessary? I suggest rewording the section to emphasise the use of common name in article titles and in general text, and then give an example (if any exists) of when Debian or Trisquel GNU/Linux would actually be used (I can't see any right now, so not sure where the suggestion applies). Greenman (talk) 08:24, 8 November 2015 (UTC)
@Greenman: Hi. About this: "MOS:LINUX appears to be ambiguous then, as it begins by stating that GNU/Linux should be used for the proper name of certain individual OS's". No, it doesn't. It says "call the family 'Linux', not 'GNU/Linux'". The part about the proper name does not have "should" in it.
Still, I will try to make the sentence clearer because it is using a negative whose positive form might be a blurry shape. AVRS, for instance, was at a loss as to what I mean when I said "violation of MOS:LINUX", because MOS:LINUX says it does not concern itself with the use of GNU/Linux in proper names of individual operating systems. But to think this way, he must have taken it for granted that I think "Trisquel GNU/Linux" is the proper name, while I do not. I took the "GNU/Linux" in it as a descriptor. In this case, the descriptor should be "Linux", not "GNU/Linux". Now, I changed the MOS:LINUX section to mention WP:COMMONNAME directly. This will resolve the "does not concern itself" issue.
Best regards,
Codename Lisa (talk) 19:21, 8 November 2015 (UTC)
So, after much back and forth, we're back to the original version, which states that "This practice does not apply to proper names of individual operating systems, such as "Debian GNU/Linux". Expect a further stream of edits changing article text and titles to GNU/Linux based upon MOS:LINUX :) Greenman (talk) 20:14, 14 November 2015 (UTC)
@Greenman: It is amazing how you hold no credit for the human nature responsible for the edits. When a person makes up his or her mind to do such edits, he or she would do it even if the MOS, through clairvoyance of the writer, call him (or worse, her!) by name say "Don't do it". "GNU/Linux" fans do not come to Wikipedia to read MOS. They come to change names!
IMHO, all this forth and back alterations were pointless, the section still says the same thing.
Best regards,
Codename Lisa (talk) 23:06, 14 November 2015 (UTC)

@Codename Lisa: this is all about confusion between article titles, which should follow WP:COMMONNAME and not generally use "GNU/Linux", or for that matter even "Linux", even in OS/distro names, and mentions in article bodies, which would depend on context: namely, if talking about the specific Debian variety called "Debian GNU/Linux" (there are other Debians!), it makes perfect sense to say "Debian GNU/Linux", and similarly, when explicitly mentioning the official name of an OS/distro, for example "Trisquel GNU/Linux", the actual official name should of course be used. The fact that we don't use official names, but common names, in article titles and elsewhere to refer to a topic does of course not mean that we're prevented from giving information about official names. LjL (talk) 15:37, 25 November 2015 (UTC)

@LjL: Fully agree! But the title of the article at this time Debian, not Debian GNU/Linux. The example was just counter-productive and sub-optimal. I am open to replacing it with another actual and working example. Also, is a stalker whose contribution to Wikipedia has been exclusively WP:HOUND. Best regards, Codename Lisa (talk)
So why can't we simply spell out that usage in article titles vs other uses don't necessarily have to match? Which is what we had more or less come to an agreement to do, with a text I found decent, before most of it was stripped out... LjL (talk) 15:58, 25 November 2015 (UTC)
Because, LjL, I think you are overreacting. You remind me of Mission: Impossible – Ghost Protocol: William Brandt was confused as to why people shooting towards them suddenly starting shooting a floating body with a flare strapped to it. "What were they thinking?" he asked. Ethan Hunt responded that they weren't thinking; they were shooting! Only they didn't have a tangible target before.
Do you know why PerfectBike was opposed so hardly? The essence of what he did was probably right. He sensed the full name had "GNU/Linux". But he overdid it. He renamed every nook and cranny of Wikipedia, even a sandbox, making people he has ulterior motivations. (Maybe he has.) If instead of renaming, he had thought a bit as to what he is doing, none of this would have happened. Likewise, you should stop pushing so hard for ultra-clarifying a text that the longer it gets, the less likely people would read it and the less likely they divine its purpose. So, while what you think is correct, you are in the wrongest place in the universe. So, stop worrying about MOS and get down to business instead; you know, establish precedence.
Also, there is one more thing: We have an operating system called "Microsoft Windows", right? But people call it "Windows" for short. Have you thought that even though "GNU/Linux" might be part of the title, people might still want to shorten it? One more reason not to emphasize.
Best regards,
Codename Lisa (talk) 16:36, 25 November 2015 (UTC)
Wow, so now having a civilized discussion about a guideline (after the previous consensus about its contents was actually overridden and reverted by one editor) is "overreacting" and being "in the wrongest place in the universe"? I did also try to "establish precedent" by pointing out to you and the other (pretty rude) user that the reason you used for reverting PerfectBike's edits was badly-grounded in the policy being invoked.
As to Windows, you can easily see that while the article exists, it is a redirect to Microsoft Windows: the actual article uses the full, official name (in the title and in several other places, while - sure - mentioning that it can be shortened to "Windows"), so this seems like a pretty bad example on your part. LjL (talk) 16:49, 25 November 2015 (UTC)
@LjL: There is little doubt in your civility. But the Victorian era in which longwinded speech was celebrated is over. Civil and charming as it maybe, it is of little consequence. WP:STICK and field work. (Also, don't think so badly of the other editor; plain text is occasionally open to being misinterpreted due to lack of emotional cues that disambiguate the person's intent.)
As for "Microsoft Windows", no the example is deliberate one: The reason "Windows" is not the main title is WP:DAB's mandate; it should not be confused by windows. As for the redirect, that's pure common sense.
Best regards, Codename Lisa (talk) 17:03, 25 November 2015 (UTC)
Uh, what logic is that? "windows" is not a WP:DAB page which says "Windows can be the thing that stop you from getting cold, or Microsoft Windows". It is an immediate redirect to Microsoft Windows. WP:DAB does not (and cannot) mandate anything about non-disambiguation pages.
Anyway, if you don't like to have discussion (or "longwinded speech") about policy, well then perhaps Wikipedia is not the most appropriate thing for you, as there is plenty of that here, and justifiably so. You seem to be (very civilly) trying to shut me up when, in fact, the topic is actively being debated, and it wasn't me challenging WP:Consensus about this guideline in the first place, but the person or people modifying it from what it originally stated. I have every right to object, and there is no stick to drop, as there is no carcass yet. LjL (talk) 17:08, 25 November 2015 (UTC)
Actually, I was trying make you less uncomfortable about people who don't appreciate your concerns and turn your attention to those who do. I seem to have gone the wrong way. Well. Ironic how you also told me not to overreact to PerfectBike's edit and I didn't think you were trying to shut me up. In a way, I stayed my hand. Best regards, Codename Lisa (talk) 20:46, 25 November 2015 (UTC)
@Codename Lisa: Here, I've created your precedent. How long do you think before someone reverts it, possibly based on the very fact that this guideline presently doesn't provide any example like it did before the non-consensual change that removed the longstanding "Debian GNU/Linux" example? LjL (talk) 17:17, 25 November 2015 (UTC)

How is "Freemium" a software license?[edit]

"Freemium" answers the question of "do I have to pay for it" not "what are my rights related to its use". The latter is the concern of licensing. 02:07, 15 January 2014 (UTC)

I'd appreciate if you kept this discussion in one forum instead of multiple. (Just add notices to the forums to publicize the topic).
"Freemium" is a license type (similar to shareware, freeware, free and open-source and commercial software), not the name of a pre-written licensing text such as GPL or GFDL. As for "what are my rights related its use", yes, the word does imply that: Freemium type EULAs grant rights similar to freeware, but stipulate that with a payment, the user can receive additional features or services. For the exact nature of your rights, you must read the license agreement itself. (But again, this is true when you know a program is free and open-source but do not know whether it is under Apache license or GPL.)
Perhaps it helps alerting the participant what brought you here: The issue of what to put into the infobox to describe concisely and as much as possibly accurately, the licensing terms of a piece of software that allows one to use its basic features freely but requires payment for its additional features and services. Surely, we cannot copy and paste the license agreement in there. "Freemium" is an excellent compromise.
Best regards,
Codename Lisa (talk) 02:22, 15 January 2014 (UTC)
I'll keep the discussion here since it has the correct audience and greatest visibility.
Shareware, freeware, FOSS and commercial software aren't a licenses either, although 'Proprietary commercial software" is. GNU General Public License, GNU Lesser General Public License and BSD License are clearly licenses and are some of the choices that FOSS may use. The former are distribution models and not licenses.
If a product does not have a standard license type it's a proprietary license. So "Proprietary freemium" would be a better description. Walter Görlitz (talk) 02:52, 15 January 2014 (UTC)
Hi. I do not disagree with your proposal in the last paragraph (because it is a good compromise) but I myself won't go with it because when an article goes to WP:FA, they ask "why proprietary freemium? Do we have non-proprietary freemium too?" The answer is "no" and the author is forced to remove "proprietary".
As for your first paragraph, correct, they are not licenses (I said the same myself); they are license types (mind the word "type" at the end). But also, their use in |license= field is the best compromise to accurately and yet concisely tell the reader what are the licensing terms without risking a revert on the grounds of publishing extremely detailed info that interests a minority.
Best regards,
Codename Lisa (talk) 03:05, 15 January 2014 (UTC)


Apparently only Microsoft (and those rehashing their press releases) use term "x64". Term "x86-64" appears to be used much more widely, and the article about the platform is called x86-64 either. I propose replacing all "x64" suggestions with "x86-64". Opinions? — Dmitrij D. Czarkoff (talktrack) 12:36, 30 July 2014 (UTC)

I am afraid your proposal is already overruled by WP:ArbCom. (See footnotes in MOS:STABILITY.) Both forms are currently in widespread use, hence both forms are acceptable.
Best regards,
Codename Lisa (talk) 00:31, 31 July 2014 (UTC)

Command-line elements formatting[edit]

I've changed the Wikipedia:Manual of Style/Computing § General guidelines section so it suggests use of the {{Mono}} template for the purpose of displaying command-line elements, instead of using <code>...</code> tag pairs. IMHO and based on my experience, that reflects the way command-line elements are actually formatted in many computing articles, and allows a clear distinction from source code samples. Of course, I'm more than happy to discuss this change. — Dsimic (talk | contribs) 12:13, 21 January 2015 (UTC)

Hello, Dsimic
I'm not sure I agree. For a single filename or inline name that one wants to retain the case and make sure no one mistakes 1 with l, {{mono}} is acceptable. Most of times, plain sans serif text is good. For one ore more lines, <code>...</code> is the way to go, if <syntaxhighlight>...</syntaxhighlight> is not available. The gray background is definitely a plus, IMHO.
Best regards,
Codename Lisa (talk) 15:10, 21 January 2015 (UTC)
Hello! Sure thing, {{Mono}} template would be used only for displaying short command-line elements, such as command names or their switches, as described in the Wikipedia:Manual of Style/Computing § General guidelines section. Using <pre>...</pre> or <syntaxhighlight>...</syntaxhighlight> would be the way to go for longer command-line elements, such as examples of command outputs.
What I had in mind while referring to "a clear distinction from source code samples" is something like this:
When running command foo, function bar is called to perform data manipulation.
In this example, foo is the name of a command, while bar is the name of a subroutine. Unfortunately, I can't recall the exact article in which I've seen something like that, but it is present. Thoughts? — Dsimic (talk | contribs) 15:34, 21 January 2015 (UTC)
@Dsimic: Very close to what I have in mind. (In real world, bar is something like CreateFileEx1. So, yes.)
But (prompt) command parameter-name should be in <code>...</code>. What's your take?
Also, please make sure you notify the original author, FleetCommand. I think he'd like it if you called him, than I did with a ping.
Best regards,
Codename Lisa (talk) 15:40, 21 January 2015 (UTC)
To me, something involving a command prompt and a set of switches/paremeters, what should apply to longer command-line examples in general, would fit the best in a separate line using <pre>...</pre> or <syntaxhighlight>...</syntaxhighlight>. Sure thing, let's hear FleetCommand's opinion as well; thank you, Codename Lisa, for pointing that out. — Dsimic (talk | contribs) 15:55, 21 January 2015 (UTC)
@Dsimic and Codename Lisa: I just logged in and it seems I must already go. So, I'll be brief; please ask for any needed clarification. (Codename Lisa knows how to find me outside Wiki.) <Code> and <SyntaxHighlight> tags are more than just some formatting tags. They are HTML5 semantic tags, something that {{Mono}} (essentially a <Span> tag) and <Pre> tag aren't. You might not immediately feel their importance but many technologies that provide for your good life today use them, including your favorite search engine. Its companion is <Samp> tag, used to format sample output. <Kbd> and <Var> tags are for key combinations and variables.
Code snippet: Out ("Hello, World!")
Output: C:\> Hello, World!
Keyboard: CTRL + Alt + Delete
Sample text: Use MaxRows variable
So, any piece of code whose sense for a computer is similar to the sense of a sentence or fragment for human must be put inside <Code> tag, inline or otherwise. If you don't like the gray background, you must appeal to the MediaWiki developers (or something to that effect) or use alternative CSS styling. It is my understanding that if your personal opinion is different from that of the public, you can write your own personal Common.css in your user space which overrides Wikipedia's. Fleet Command (talk) 16:30, 22 January 2015 (UTC)
Thank you for a detailed explanation, FleetCommand! I see what you mean, but could having another template, which would apply additional CSS to the <code> tag to make its appearance different, be a possibility? I'm having readability of something like "When running command foo, function bar is called to perform data manipulation" in mind when referring to that additional template; it might be found to be similar to what we have available with the {{Key press}} template (which produces things like Ctrl+Alt+Del), for example. — Dsimic (talk | contribs) 17:16, 22 January 2015 (UTC)
@Dsimic: Your example is okay AND is not constrained by the MOS in any way because a single name of a function plus a name of a command are hardly exclusive command-line items. And yes, you can apply CSS to <code>. So, even if you don't change the MOS, you can still do your formatting things in the articles. Fleet Command (talk) 17:46, 23 January 2015 (UTC)
Well, Ok, guess that editors are given their freedom of choice when it comes to different fixed-width styles. :) — Dsimic (talk | contribs) 00:35, 24 January 2015 (UTC)
@Dsimic: Indeed they do. There is only a consistency requirement that shouldn't be a problem. Fleet Command (talk) 17:54, 25 January 2015 (UTC)

Linux vs GNU/Linux (2)[edit]

The MOS:LINUX guideline is a shame, specially as it stands now.

According to it, the Android (operating system) page (among others) should not even exist or should be merged into Linux. It is also worded to imply that "GNU/Linux" is only used by the FSF and thus must be refused on those grounds, but this is an outright lie as can be witnessed in a particular section/paragraph at Linux and a more extensive list I have prepared during the last years.

If you are going to put forward a guideline that favors the common usage of a term over its correctness and accuracy at least formulate it properly. --isacdaavid 05:35, 4 March 2015 (UTC)

Red information icon with gradient background.svg Not done: We've discussed this issue ad nauseam; discussing this with one who opens the discussion with "you are a liar" simply can't happen. The part about Android is pure nonsense. Best regards, Codename Lisa (talk) 20:55, 4 March 2015 (UTC)
Isacdaavid didn't call anyone liar. He told a lie (i.e. his first two sentences of the second paragraph) and then said it is a total lie. It is called a straw man, which is an extension of lying.
Fleet Command (talk) 01:40, 5 March 2015 (UTC)

I've restored the Debian GNU/Linux example because the removal appears to be motivated by a recent dispute where the example was cited in discussion and now participants on one side of the discussion are trying to change the rules against consensus. (talk) 13:11, 25 November 2015 (UTC)

Well, hello! If it isn't my favorite stalker!
I already said why I removed the example: Because it is not practice. But since you are a stalker whose contributions so far have been exclusively hounding, I say your restoration is probably motivated by harassment and nothing else.
Best regards,
Codename Lisa (talk) 15:53, 25 November 2015 (UTC)

Capitalization of Internet/internet[edit]

Apologies if this is re-opening a can of worms. I've done some searching but can't find much.

Wikipedia, strangely, has an article about the history of the capitalization of "Internet", but the Manual of Style doesn't seem to have any guidance on it. Am I missing something?

... My own two cents: capitalizing internet is archaic and looks stuffy - and as far as I can tell, there was never any good reason to treat it as a proper noun in the first place. The article on the history of its capitalization helpfully provides a list of publications that don't capitalize:

"A significant number of publications do not capitalize internet. Among them are The Economist, the Financial Times, The Times, the Guardian, the Observer, the BBC, and the Sydney Morning Herald. As of 2011, most publications using "internet" appear to be located outside North America, but the gap is closing. Wired News, an American news source, adopted the lower-case spelling in 2004. Around April 2010, CNN shifted its house style to adopt the lowercase spelling."

Thoughts? Popcornduff (talk) 15:42, 18 May 2015 (UTC)

Hello! To me, it should be capitalized when referring to the Internet as a global network, and left uncapitalized when referring to individual or generic interconnected networks. Just my two cents. — Dsimic (talk | contribs) 00:41, 6 June 2015 (UTC)
Thanks the reply. Care to explain why?
Anyone else want to chip in? Popcornduff (talk) 16:36, 6 June 2015 (UTC)
Well, when referring to the Internet as a global network, IMHO it's about a unique entity, something you can pretty much point a finger to and say "that's it". With individual interconnected networks, internets, it's not about unique entities but rather about the concept of internetworking. — Dsimic (talk | contribs) 21:50, 6 June 2015 (UTC)
I appreciate that it's a unique, individual entity. But that isn't grounds for capitalising it - it isn't a proper title. Anyway, most major publications aren't capitalising it any more. Popcornduff (talk) 18:15, 7 June 2015 (UTC)
I'm far from being a linguistic expert, but a unique entity should qualify as a proper noun, which in turn should be capitalized. — Dsimic (talk | contribs) 05:09, 8 June 2015 (UTC)

HTTPS use[edit]

How come we are prevented from putting in HTTPS links for sites that support both properly, but not enforce HTTPS?

A recent edit on the FaceTime article pointed to this project page, and I would like to understand why we are putting users at risk by redirecting them to non-secure resources. #!/bin/DokReggar -talk 09:17, 22 January 2016 (UTC)

Hello, DokReggar
The edit in the FaceTime article uses a protocol-relative link. It means, I, who visit Wikipedia over HTTPS, see that link over HTTPS too.
As for why, did you notice the footnote? There was a discussion at "Wikipedia:Village pump (policy)/Archive 111 § As WP uses HTTPS, should (some) external links, too?" and this page only reflects that consensus.
Best regards,
Codename Lisa (talk) 09:27, 22 January 2016 (UTC)
Oh, well, I knew I already read something on that topic, but did not manage to find it. Thanks, Codename Lisa.
That being said, Wikipedia is a major website, and user security should be considered. I do not understand why sites properly supporting HTTPS, but not mandating it, should not be pointed to using secure links, particularly after all the revelations about mass-scale surveillance. A secure link has no downside over a secure one, as long as the website supports it, which I checked before editing. Wikipedia can (and should, in my opinion) take a stand towards a secure Internet for its users.
Thanks for the feedback anyway :-). #!/bin/DokReggar -talk 10:44, 22 January 2016 (UTC)
In case of, it does not offer anything unique to each visitor, so security is practically not a concern. Participants in that discussion were more concerned about the feasibility of loading the site over congested networks. So, the PURL clause helps both group: People who are overly concerned about security will get HTTPS, while people who have to navigate network congestion will get HTTP. It is a win-win. I myself get HTTPS.
Best regards,
Codename Lisa (talk) 12:09, 22 January 2016 (UTC)
The URL does not display things unique to the visitor, but there are nevertheless cookies involved, which may hence be a confidentiality breach. But security does also concerns itself with integrity, and it has been shown multiple times that large-scale injection of exploits is possible (and done) through insecure connections. Using a secure connection prevents a third-party from tampering with the connection, hence securing their browser and operating system.
I will not (unless requested) bother you or take more of your time on that topic; I accept the current consensus, even though I strongly disagree with it. #!/bin/DokReggar -talk 13:08, 22 January 2016 (UTC)

Command names at the start of sentence[edit]

Should the name of a CLI command be capitalised if it begins a sentence? Fizzwhiz (talk) 15:17, 27 February 2016 (UTC)

If the natural case is lowercase, no, especially if the command is case-sensitive. If you feel uncomfortable still, wrap them in <code>...</code>.
Best regards,
Codename Lisa (talk) 10:52, 28 February 2016 (UTC)

Rewording "shell-specific sequence"[edit]

I don't consider describing the shell syntaxes as FOO-like a good idea. Shell language should be categorized into families, but the MOS should not do any overly clever comparisons to describe these families.

Instead of:

If a shell-specific sequence is required for proper explanation, provide an example for the ALGOL-like shells (Bourne shell, Korn shell, and Bash) as well as one for the C-like syntax of C shell and tcsh.

The MOS should say:

If a shell-specific sequence is required for proper explanation, provide an example for the POSIX family of shells (Bourne shell, Korn shell, and Bash) as well as one for the C shell syntax of C shell and tcsh.

--Artoria2e5 emits crap 21:58, 27 May 2016 (UTC)

Hi. The suggested "POSIX family of shells" already encompasses C shell and tcsh. Then, the suggested sentence compares two non-homogeneous groups. That's the recipe for trouble. —Best regards, Codename Lisa (talk) 10:27, 29 May 2016 (UTC)

Kernel Linux: "name advocated by"[edit]

@Dsimic: I don't understand well this Special:diff/725707657/725707915.

With this phrase

GNU/Linux is a name advocated by the Free Software Foundation (FSF)

...but also the word "Linux" is also "advocated" by someone else. It's a registered trademark of the Linux Mark Institute to talk about the Linux kernel project. So, for completion, I don't think that there is anything strange to keep this information for the balance of the POV. --Valerio Bozzolan (talk) 10:41, 17 June 2016 (UTC)

Hello! See, the term "GNU/Linux" is considered a point of view here on Wikipedia, because it's used by a minority that pretty much boils down to the Free Software Foundation and its associated projects. As a result, and regarding the diff you've linked above, it's better not to blur the MOS:LINUX guideline with additional explanations that actually don't contribute to the guideline itself. — Dsimic (talk | contribs) 10:49, 17 June 2016 (UTC)
I know that in GNU/Linux is considered a POV to talk about the operating system, bus you have to consider that Linux is neutral only to talk about a kernel, but for the same reasons you said, it is itself a POV if used to talk about a complete operating system. --Valerio Bozzolan (talk) 12:04, 17 June 2016 (UTC)
Well, this is the English Wikipedia. — Dsimic (talk | contribs) 11:24, 17 June 2016 (UTC)
Please don't change the subject: the title of this section is clear and I've already explained my edit. Could you be so kind to explain why you don't want to solve the POV about the "advocated" phrase? --Valerio Bozzolan (talk) 12:04, 17 June 2016 (UTC)
I'm not changing the subject, and "GNU/Linux" is clearly advocated by the FSF. Moreover, I've already answered your questions. — Dsimic (talk | contribs) 12:10, 17 June 2016 (UTC)
 Unrelated Please take a small wikibreath... Read carefully what I have done. This is not the place to discuss the GNU/Linux term, I'm not discussing that, I've not done an edit about that here (IMHO: even if it's still an historical very sad bug). As I've already said, in the evidenced sentence there is an heavy POV. We should say that in the term GNU/Linux is *not* avoided for the reason in that sentence, because on the other hand the word "Linux" it's *also* strongly "advocated" by someone else who want that only attributed to a kernel, and I'm not talking about GNU-developers, very-boiling-FSF-lovers, Linux-haters, Ubuntu-geeeks-haters, etc. I'm talking about that "Linux ®" is a trademark that without an explicit permission is only to attribute to a kernel project and not to something else. Again, don't talk about the MOV, that it was not touched there. So, to be complete, I've added both the backgrounds. --Valerio Bozzolan (talk) 13:24, 18 June 2016 (UTC)
@Dsimic:: Hi. You have reverted this addition by Valerio Bozzolan:

"GNU/Linux" is a name advocated by the Free Software Foundation (FSF) that refers to a complete operating system and "Linux" is a registered trademark of the Linux Mark Institute that refers to the kernel project.

Mind if you tell me what's wrong with it? Thanks.
Best regards,
Codename Lisa (talk) 12:04, 18 June 2016 (UTC)
As I explained here, that the Linux kernel is a thing does not mean that the Linux operating system does not exist or that Linux solely refers to the kernel. I agree with Dsimic's edit, as the change made it seem as if Linux were solely the name of a kernel and any instance of Linux refers to the kernel; this is not the case, as reliable sources (Linux Foundation, Torvalds, IEEE and others) show. Despite what the edit suggested, Linux does not solely refer to "a kernel project". - Aoidh (talk) 01:18, 19 June 2016 (UTC)
Hi. Judging by what you say, the problematic part is "that refers to the kernel project", not the whole thing. Even then, we can still change it to "originally intended for the kernel project". But I myself am in favor of Dsimic's removal on the grounds of brevity, staying on topic and avoiding redundancy.
Best regards,
Codename Lisa (talk) 09:16, 19 June 2016 (UTC)
@Codename Lisa and Aoidh: I'm also in favour of brevity, but after neutrality. It's true that Torvalds developed a kernel? It's true that Torvalds used the already existing GNU operating system in top of his kernel? *It's true that Torvalds use to call that whole thing as his kernel*? If so, and please don't discuss here about the Linus Torvalds work, we have to be neutral: GNU/Linux is "advocated" by the FSF? Well, that is not the reason because of Wikipedia don't choose that form: because we can't prefer the also criticized Torvalds POV where he is giving something-with-his-name-plus-"x" on top of 7-8 years of other people work (a.k.a. GNU development). We want brevity? We are OK with the deletion of the Dsmic's edit? We have also to say that Wikipedia don't choose the Linus Torvalds POV, but choose "Linux" *only* because of the English language. Wikipedia can't prefere a POV instead of another POV. --Valerio Bozzolan (talk) 09:46, 20 June 2016 (UTC)
Neutrality of point of view belongs to the article. This is MOS we are talking about. It should tell the editor what to do and what not to do and shouldn't waste their time giving factlets like '"Linux" is a registered trademark of the Linux Mark Institute that refers to the kernel project'. Having said that, I might as well add that this MOS isn't playing unfair with the matter at hand either; it is already neutral. Your change does not improve that aspect. Fleet Command (talk) 11:55, 20 June 2016 (UTC)
@FleetCommand: Neutrality can be also be reached with less words:
Use "Linux" (e.g. "Linux distribution") instead of "GNU/Linux" to refer to the family of operating systems that make use of the Linux kernel, but keep the disambiguation: "Linux" itself it's not an operating system. The term "GNU/Linux" may still be written as part of the proper names of individual operating systems.
GNU/Linux naming controversy does exist, but Wikipedia prefers the name that is most commonly used (as determined by its prevalence in reliable English-language sources). The discussions are archived in the Talk:Linux/Name.
--Valerio Bozzolan (talk) 13:28, 20 June 2016 (UTC)
Notes: a lot of operating systems are not based on a kernel: it's better to say that they use a kernel; as said, the FSF viewpoint is not too much important over the Linus Torvalds viewpoint: it's better to avoid that argument to avoid flames and/or POV favouritisms. The small part in the code it's what we should discuss in the see also discussion to be even more neutral. --Valerio Bozzolan (talk) 13:28, 20 June 2016 (UTC)
@Codename Lisa: Sorry for my delayed response. I've reverted that edit simply because it doesn't contribute to the MOS:LINUX guideline, and instead serves just as a "trampoline" for yet another "Linux is a kernel, not an operating system" crusade, so to speak. As Aoidh and FleetCommand already pointed out, our MOS entries should be brief and to the point. — Dsimic (talk | contribs) 21:31, 20 June 2016 (UTC)
Dsimic, with your statement on record, we are now officially four people in favor being brief and to the point. (CL also said "brevity, staying on topic and avoiding redundancy".) Fleet Command (talk) 11:13, 21 June 2016 (UTC)