The question Is Microsoft going to produce an operating system based on Linux? asks whether Microsoft plans to develop a Linux-based operating system. It references an article from source that appears to be more or less reputable.

In this article, a Microsoft spokesperson is quoted verbatim, and the quote states explicitly that Microsoft are developing an operating system which "is a custom Linux kernel complemented by the kinds of advances that we have created in Windows itself". The accepted answer quotes a Microsoft website which confirms the statement by the Microsoft spokesperson.

Apparently the question was not whether the Microsoft spokesperson actually made that statement, but whether Microsoft was working on such an operating system.

I was surprised to find that this question is considered on-topic by people who understand skeptics.SE much better than I do, but I'll follow the suggestion to discuss this on skeptics.meta so that I can eventually comprehend the rationale behind this.

To quote the FAQ:

This site is about applying scientific skepticism. We only accept answers based on independently verifiable applications of the scientific method ("facts").

So there are three requirements that acceptable answers must meet (independence, verifiability, based on scientific method). Let's go through them one by one.

Independent evidence

The question quotes a Microsoft spokesperson. The accepted answer uses a Microsoft website as its reference. This is no independent evidence. Any statement by Microsoft about their future plans cannot, by definition, be independent. It's extremely difficult (if not impossible) to imagine what form such independent evidence would take. Would a Microsoft whistle-blower from the "Linux-based development department" be considered independent evidence? Do we need something like an article in which the authors conclude e.g. from observing the job advertisements that Microsoft has posted in the last few months that Microsoft has shifted their attention to the development of Linux-based operating systems?

Evidence like these two far-fetched examples are unlikely to emerge. But what type of independent evidence would be needed, then?


Even if there was independent evidence, that evidence would be difficult to verify. The business plans of Microsoft are not publicly available. How can we know whether the claims by that independent source (e.g. the whistle-blower, the article authors) are true? Can we ever?

Scientific method

This question doesn't appear to be something that can be answered by using the scientific method. It fails already when it comes to falsifiability. The hypothesis at hand is that Microsoft is going to produce a Linux-based operating system. This is a projection into the future – even if there was a way to test whether Microsoft was working on such an operating system now, there is no way to be sure that these plans can't change.

In fact, I'm rather lenient when it comes to this part of what is on-topic for skeptics.SE. There are many good and acceptable answers on skeptics.SE which are not, strictly speaking, based on the scientific method. Accepted answers to the plenty "Did XY say Z?" questions typically consist of someone providing a suitable reference for Z, and everyone is apparently happy with this type of answers even though it doesn't have much to do with the application of the scientific method.


What all this boils down to is, I think, whether this is a type of question for which answers can exist that meet the skeptics.SE standards. I don't think that such answers are possible – and as it is, neither of the two top-voted answers produce independent evidence, neither presents verifiable facts other than statements by the involved company, and neither answer is based on the scientific method. I assume that questions that can't produce acceptable answers can't be on topic.

Note that I focused on this particular claim about Microsoft, but I don't think that the scope of this discussion is restricted only to it. The questions that I raised above would apply to any other post that doubts whether a news piece about future plans of an individual, a company or an organization are true.


(1) Can someone explain to me how an acceptable answer can be provided that is based on

  1. independently
  2. verifiable
  3. applications of the scientific method

for this type of question?

(2) Are questions on-topic for which acceptable questions are virtually impossible to provide?

  • Literally, we can't predict the future, but in this case I imagine that the question meant "is Microsoft openly planning to build a Linux OS?" which is factual.
    – Sklivvz
    Commented Apr 20, 2018 at 13:25
  • I thought there was an implicit (or explicit) duty on SE to use it for that which cannot be easily researched. This question is literally "Does water contain hydrogen and oxygen" Answer: "Yes.". Commented Apr 20, 2018 at 20:51

1 Answer 1


Your question is long question and has several parts to this. I haven't addressed them all, but I will say: in many ways, I think you are right.

I don't think the Microsoft question is a great one, but I do think it passes the rules, and I try (with varying degrees of success) to keep my opinion out it when closing questions.

You are right that if the question was "Will Microsoft one day invent the HoverCar?" we would probably close it as it is unanswerable with evidence, and would boil down to speculation.

However, I interpreted this question as "Are Microsoft currently working on inventing [something something Linux]?" which is acceptable. Asking whether they will succeed wouldn't be.

You are right that the level of evidence posted in the question was relatively high. When that happens, it means the quality of the evidence in the answers must be even higher, to be adding anything new.

It makes answering the question much more difficult, but it isn't normally considered a Close reason.

That may warrant a comment asking the OP why they are skeptical of the sources, so we can check the question is is good faith, isn't based on a simple misunderstanding, and find out what sort of evidence it would take to persuade them.

Maybe they think the people were misquoted or impostors or something, so we could provide some evidence to support or refute that. (You seem to suggest that could never happen, but I wasn't convinced by that part.)

You are right that we accept history questions here. (What I mean by that is factual questions about whether something happened or whether someone said something. We don't accept the discussions of motivations and influences that form a large part of understanding history.)

There can be a little friction there, because history, necessarily, has different standards of evidence than, say, physics, and it is difficult to determine whether "scientific" is the appropriate term. [Economics questions often seem to cause a little friction for similar reasons.]

I don't think I have addressed every point you make, but I am hoping this is enough to explain my decision not to close the question. As always, 5 close votes will get it closed without a mod intervening - and when I see that happen, I try to learn from it, to see what the community thinks is off-topic.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .