Interpretation of the referencing rules

that made me wonder whether I fundamentally am misunderstanding that rule.

Perhaps :)

I received a statement from an SE staff member saying that if a source of copied material is online, then adding a link to it is required. Although my response to your original comment might have seemed "unusually sharp", I'm following the interpretation I've been told by SE, so I am trying to be clear and unambiguous when I state the referencing rule. There's no point in giving false hope of a compromise etc. when there isn't one.

Due to the amount of copied information which comes onto our EE.SE site, I expect we are one of the SE sites most affected by that referencing rule. It is an imperfect rule and I've already been in contact with SE about how best to raise a "laundry list" of issues and corner cases which I have seen and which are not fully addressed in that current rule, for SE to provide guidance on or perhaps for them to update the rule. (It's not an easy or quick process to build that list for SE and I thought it was less urgent (though not less important) than responding to other meta questions, flags and other mod work here, but I am doing it in the background. It needs to try to be done all in one go, as much as possible, since all of the various issues should be considered by them together.)

But the main point you raise in your question here isn't a corner case - it is clearly stated in the referencing rule, and that's why I was direct in my response to you. As you quoted from that rule:

If you copy (or closely rephrase/reword) content that you did not create into something you post on Electrical Engineering Stack Exchange (e.g., from another site or elsewhere on Electrical Engineering Stack Exchange), make sure you do all of the following:

• Provide a link to the original page or answer

{... more requirements ...}

The datasheet image is copied content, it was not created by the OP, and it was posted onto EE.SE, so the requirement to provide a link applies (since the source is online).

To be clear, I'm not saying that only a link should be given - we should always try to reference to the best of our ability, e.g. give title, source website name etc. too - but a link is also required, if the source is online.

You said:

a datasheet is not another post on SE, nor a site

The rule says if the content comes "from another site" - and the datasheet will have been copied from a (web)site, in fact you kindly provided a link to one possible source site. Are you saying that the requirement to provide a link to a datasheet doesn't apply because it didn't come from a website? If so, then I don't understand that point.

It would be unreasonable to expect SE to produce an exhaustive list of all possible types of content, down to the level of them knowing what a PDF datasheet is, as compared to a datasheet displayed in an HTML iframe, or how an application note is different from a datasheet, or from a reference design and treating a datasheet differently somehow. See next point below.

is my interpretation of the rules that do not explicitly refer to datasheets so far off?

SE don't explicitly mention datasheets, agreed, but an interpretation that the referencing rule doesn't apply to datasheets is incorrect.

I checked with SE a short while ago for a very similar situation to the one you have raised. A user claimed that giving a datasheet document number was enough as a reference for a screenshot, without them also providing a link, despite the datasheet being available online so a link could also have been given. I disagreed with that user and escalated that specific situation to SE for advice. Their response confirmed my interpretation: If the source is online then its link must also be given.

As it says in the referencing rule:

Always give proper credit to the author and site where you found the text, including a direct link to it.

Yes, it says "text" but it really means content, as confirmed by the response I got from SE about a user refusing to add a link for a datasheet screenshot (which is content, but not plain text). So in the case of the question you're referring to, adding the datasheet link to the existing device information was the correct thing to do, so thanks for doing that.

You raise the point (and I'm paraphrasing my understanding of your point here) - that people wouldn't get any more information from the whole datasheet, than they have already got from the datasheet snippet copied into the specific question you're referring to. Therefore there's nothing to gain from requiring a link to the datasheet source.

That might sometimes be true. But we have had many examples where, after we remind an OP to provide a link to the source of something they have copied into a question, further details from that source provide additional context which helps people to answer their question. I've just seen another example of that happening today: The original question had no source link for a schematic adapted from elsewhere, making it difficult to provide a good answer (as well as not complying with the referencing rule). A source link was then added and a good answer could then be written, due to additional details from the original source.

Also, if an OP doesn't include a link for something they copy into their question, let's say a datasheet, we can end up with many people all having to spend their time searching for that datasheet link, before they feel confident to consider answering the question. What a waste of time that duplication of effort can be! So another advantage of an OP linking to the copied material in the question is that potential answer-writers can easily read it. Therefore requiring people to link to the source of copied information has that extra benefit for a technical site like ours, of making more context available for potential answer-writers (which wouldn't necessarily be relevant on other SE sites where answers may not be affected by additional details in the source).

FYI the main SE staff member who did the last revision to that referencing rule has unfortunately since left the company, which means I can't ask them anything directly. That's why I've had to find someone else in SE who could advise a way for me to ask that list of questions & concerns.

One final point: There are aspects of the referencing rule that SE might decide are not fully being complied with (IMHO improvements are needed to what many people do e.g. including providing titles of linked material, as you mentioned) even when people do also add the source link. One consequence of raising these issues to SE is that they will inevitably look closer at what we're currently doing. That could mean extra, or stricter, requirements might be given and us mods would then have to enforce those.

I think moderation could be a bit sensible and just apply common sense on case to case basis:

• Is this picture likely to be copyright infringement, plagiarism etc?

  If it comes from a published work or if it's some sort of artistic photo (unlikely to be posted here) or if it is a picture of a person (even less likely to be posted), then yes the source should be stated. Not necessarily available online or for free. IEC standards in particular are not.

  Common sense is already applied whenever someone posts their schematic asking for feedback or posts a photo of a component asking for identification. The same common sense could be applied to most pictures though.

• Datasheets are generally protected by copyright but freely distributed. The silicon vendors mostly cares about ensuring authenticity - they want to sell components, not datasheets. Therefore no sane silicon vendor would reach out here and claim copyright to pictures taken from datasheets on this site. Rather they'll be grateful for the free support and exposure. And making their customers pissed off isn't in their interest either.

• Really old parts do not necessarily even have online datasheets, in case the datasheets were never scanned when the Internet era begun.

And this would be why we have human moderators who are expected to at least have a bit of electrical engineering knowledge - not to apply policies by the letter but to be sensible from case to case.

However, there is a good reason to reference a certain specific datasheet or other technical document: namely revisions. In particular, datasheets & manuals to more advanced components like microcontrollers might differ a lot depending on revision. It has happened to me that I referred to a certain chapter of a microcontroller manual and then the OP couldn't find it, because we had mismatching revisions.