Talk:Spacetime/Archive 16

I believe the section to be ready for transfer to article space.

D.H's suggested revisions correct what I felt to be deficiencies in the narrative that I had developed with help from all of the rest of you, but which I wasn't quite able to pinpoint. At this point, I believe that the history is ready to transfer to article space.

I will begin the process of transfer immediately after this post.

Thanks to everybody who helped out! Stigmatella aurantiaca (talk) 02:30, 17 July 2017 (UTC)

@Stigmatella aurantiaca: I'm reading through and it so far is quite natural to read, interesting, and logical. I am confused about this sentence: In 1900, he recognized that Lorentz's "local time" is actually what is indicated by moving clocks by proposing an explicitly operational definition of clock synchronization assuming constant light speed. What does “explicitly operational definition of clock synchronization”mean? Can you flesh that out within the sentence? Greg L (talk) 04:39, 17 July 2017 (UTC)

P.S. Very good! I went all the way through it and made only a few minor edits for clarity. I found it very easy to understand with the exception of what I wrote in the preceding paragraph. Greg L (talk) 04:52, 17 July 2017 (UTC)

"fleshing out" is on my to-do list. Should get to it in a day or so. Stigmatella aurantiaca (talk) 11:08, 17 July 2017 (UTC)

Geoffrey left a trail of typos and careless edits, not just here but in Special relativity. I didn't go around following him closely to proofread his work, although I obviously should have. I had assumed that a working physicist would be more careful than he actually was. So I'm taking care of those items first. Mea culpa. Stigmatella aurantiaca (talk) 18:04, 17 July 2017 (UTC)

Your culpa runneth over with goofs to fix.

With regard to “…an explicitly operational definition of clock synchronization assuming constant light speed,” an alternative way to address insider-speak like that would be to give it a more general treatment, e.g. In 1900, he recognized that Lorentz's "local time" was actually more akin to what Einstein later posited five years later.

Alternatively, a more detailed but plain-speak treatment. For instance (though I have no idea it is technically correct): In 1900, he recognized that Lorentz's "local time" is actually as Einstein would describe five years later, where ‘time’ (when events occur on a clock) must necessarily vary for different observers in a universe where the speed of light has one fixed value regardless of the two observers’ relative velocities to an observed light source. Greg L (talk) 18:50, 17 July 2017 (UTC)

Clearer? Stigmatella aurantiaca (talk) 06:51, 18 July 2017 (UTC)

Very. I enjoyed the learning process, so it’s good stuff. Greg L (talk) 17:18, 18 July 2017 (UTC)

You're welcome. Galison's book is a great read, by the way. Stigmatella aurantiaca (talk) 18:00, 18 July 2017 (UTC)

Translations of Poincare frequently have him using the word "convenience", which was used in D.H's text. I suspect it may be a case of one or another of Poincare's translators settling on the wrong English word to express some subtle nuance of the French. "Convention" makes more sense in most instances. Stigmatella aurantiaca (talk) 18:00, 18 July 2017 (UTC)

I know someone who speaks fluent French. She’s a French citizen, born in France, and married to an American. Would you like to share the precise passage with me and I’ll have her look at it? Greg L (talk) 19:07, 18 July 2017 (UTC)

I'll need to track down the original French for these passages that I know only in translation. This may take a while. Stigmatella aurantiaca (talk) 22:39, 18 July 2017 (UTC)

By the way, she's a language genius. She and her husband play Scrabble in any language so long as it is one of the Romance languages. She and her husband (my boss at the time, though we really had more of a Darren Stevens / Larry Tate relationship) came to me requesting a ruling: she wanted to play Scrabble in Korean, which wasn’t originally a Romance language but only recently has received government-endorsed “official” translations into the English alphabet. I ruled in his favor (I’m no fool). Whatever verdict you obtain regarding her native French, well, you can take it to the bank. Greg L (talk) 00:52, 19 July 2017 (UTC)

Bold for intra-page hyperlinks

@Stigmatella aurantiaca:

Here's a thought. I’ve never seen intra-page hyperlinks denoted by bolding before. Is that something relatively new to Wikipedia, or is it a home-brewed thing?

If home brew, I propose that 1) to better clue readers to the availability of these convenient hyperlinks, 2) and to better embrace the principle of least astonishment, and 3) to avoid such links being essentially an Easter egg, we try something like the following:

We were discussing the genesis of spacetime in early July when…

Would instead be lke this:

We were discussing the genesis of spacetime (click here to skip to that section, above) in early July when…

Now that I think about it, I might have encountered this situation before in a previous article (it might have been our Kilogram article) and ended up with this sort of solution. It works quite well and best serves our readership because it’s perfectly clear and unambiguous. Greg L (talk) 22:57, 19 July 2017 (UTC)

Bold intra-page hyperlinks are already very common in mathematics articles for linking to a numbered formula. For example: Acceleration (special relativity). Various of the numbered formulas are referenced within the text by a bold formula number. I am merely extending the use slightly. Stigmatella aurantiaca (talk) 23:49, 19 July 2017 (UTC)

Since it is a convention that isn't exactly ubiquitous on Wikipedia, would you mind if I changed the non-math ones to what I am proposing above? I think it will be helpful to non-expert users of Wikipedia. Greg L (talk) 01:46, 20 July 2017 (UTC)

The links are already very prominently blue because of the boldness, and it is not necessary for readers to click on them to understand the article. You will notice that there is usually a gentle hint of some sort associated with the links that the reader is encouraged to explore their use: "Click here for a brief section summary", "invariant interval (discussed shortly)", "(see Fig. 1‑1)", "Fig. 2‑9 illustrates that...", "As we have discussed in the previous section on four-momentum", etc.

There is also the issue that many of the links don't necessarily work the first time for phone users, and even more importantly, phone users don't have a good way to return to where they were, especially since they do not have the benefit of a table of contents. The "back" < button often leaves them stranded after a javascript-assisted leap.

The lack of TOC for phone users is an exceptionally irritating point to me, since it makes navigation through the article extremely difficult. I would like to be able to have a "click here to return to TOC" or some sort, but it would only function for Desktop users, who don't need such a feature.

What the MediaWiki "jeniouses" should have done was provide an initially collapsed TOC for phone users. <sarcasm> But no, that was too hard for them to implement. </sarcasm>

So no, don't add the extra verbiage, since except for "Click here for a brief section summary" where I have explicitly provided a "back" mechanism, I don't want to rub phone user's faces into the fact that desktop users have available to them a feature that they may be very hesitant to use.

Re the math intra-wiki link templates, they are incompatible with "collapsing" formulas that compress on a narrow screen, otherwise I would have used them in a couple of points in the Spacetime article. Instead, I rearranged the text so that discussion about a formula was always immediately adjacent to the formulas being discussed, which were hence not necessarily located in the positions where their display may have been most natural. I also completely omitted discussion of one formula (of lesser importance, admittedly) where I couldn't resolve the placement issue. Stigmatella aurantiaca (talk) 05:37, 20 July 2017 (UTC)

I tried putting a note at the beginning that the Section Summaries can be read as a stand-alone "Introduction to Spacetime", but I had to undo because of anomalies that phone users experience if they make the Section Summaries their first point of entry. Stigmatella aurantiaca (talk) 13:04, 20 July 2017 (UTC)

OK. It’s not a big deal anyway. I should think though that what the mathematics editors are up to insofar as using bold to indicate intra-article hyperlinks is one thing; using the convention in this more general fashion is an uncommon interface element that doesn’t appear in WP:LINKS and runs an increased risk of suffering frustrating drive-by shootings in the future. I think you are setting yourself up for needless hassle in the future.

With regard to your mention of a lack of an explicitly provided "back" mechanism, that issue doesn’t appear to be relevant since providing an intra-article hyperlink imbedded in the phrase is functionally identical to imbedding the link a few words adjacent in a parenthetical.

I’ll change just one of those links as an example to consider. I hope you ruminate on my example link (near the bottom of the section, here) for a good while before acting. I think you’ll soon conclude that it reads exceedingly close to what you had before, is typographically elegant, is fully compliant with WP:LINKS (and what our users are accustomed to), makes the experience no different for users on mobil devices, and—very importantly—fully and unambiguously communicates to all users what they can expect if they click the link, which is a fundamental imperative of all good man-machine interface guidelines. Greg L (talk) 16:23, 20 July 2017 (UTC)

With regard to your mention of a lack of an explicitly provided "back" mechanism, that issue doesn’t appear to be relevant It's far, far worse than that. Your linking to a top-level section works fine on a phone, because the top level section exists even when the section has not been expanded, and the "back" arrow works fine. But when you experiment with wikilinks to destinations at deeper levels, you will find that the implementation of Wikipedia mobile has a lot of flaky elements that leave you absolutely exasperated, and the back arrow can take you to completely unexpected destinations. Stigmatella aurantiaca (talk) 16:40, 20 July 2017 (UTC)

Agreed. But we’re still talking past each other. My point is only as follows:

This example of trying to understand our target readership (discussed above)

…is functionally identical (mobil users included) to this:

This example of trying to understand our target readership (discussed above)

The blue color in both methods makes it clear that there’s a link. The only differences between the two examples are that 1) the second example fully complies with WP:LINKS, and 2) the second example makes it perfectly clear the reader will be taken to an intra-article location and won’t be taken to another article.

I think you believe that what the math people have been doing (using bold links to denote an intra-article jump to the formula) is a useful interface element that should be used on this article in hopes it might serve as a paradigm and be more widely adopted across the 6,821,895 articles on this wide project. However, the vast majority of our readers’ eyes glaze over on abstruse math sections. Using bold to denote an intra-article link isn’t even covered in WP:LINKS (neither prescribed nor proscribed) because what consenting mathematicians have been doing is a practice that’s largely flown below the radar. As a result, it’s a user-interface technique that is unfamiliar to the vast majority of our readership. Greg L (talk) 17:17, 20 July 2017 (UTC)

Sections too long

Sections longer than, say, half a screen, are possibly intimidating. An easy way around this is to create sub- (subsub-) sections. If you worry that the TOC becomes too long, it can be made to diplay only down to a chosen depth. YohanN7 (talk) 08:57, 18 July 2017 (UTC)

I'm worried that the fonts of level 4 and deeper subsection headers are indistinguishable. I sort of got around this for • Active, passive, and inertial mass, • Pressure as a gravitational source, and • Gravitomagnetism by adding dots, but it was a kludge solution that I really didn't like. Is there some more elegant way of customizing subheader fonts? Stigmatella aurantiaca (talk) 11:38, 18 July 2017 (UTC)

Don't know. Maybe have a look here. YohanN7 (talk) 12:04, 18 July 2017 (UTC)

TOC could stand a bit of abbreviation. I'll try limiting its depth. Stigmatella aurantiaca (talk) 15:30, 18 July 2017 (UTC)

Trying to figure out visual cues that would work to break up a wall of text, not necessarily subsections. Stigmatella aurantiaca (talk) 15:44, 18 July 2017 (UTC)

Sometimes very simple page layout elements can be employed to break up long stretches of text to alleviate the mind’s energy devoted to eye tracking. Sidebars are a common way in print; though I think it’s possible, I haven’t personally seen them used on Wikipedia.

Bullets, enumerated lists, and cquotes are other examples that break up visual monotony, can easily be used on Wikipoedia, and all of which require minimal-to-zero truncation of text material.

In the History section, as an example of what I’m talking about, I converted an imbedded quote in an already tortuously long paragraph into a cquote, which also effectively bifurcated that long paragraph. All in all, that single giant paragraph became three elements, which I find to be much easier to read because the mind doesn’t have to struggle so hard to direct the eye.

Before and after.

Greg L (talk) 17:40, 18 July 2017 (UTC)

Thanks! Stigmatella aurantiaca (talk) 17:48, 18 July 2017 (UTC)

The {{TOC limit|3}} moved the lede to below the table of contents (funkiness here). I repositioned the tag (∆ edit) to restore the wiki-standard convention of having the lede before the TOC. Greg L (talk) 21:44, 18 July 2017 (UTC)

whoops! Stigmatella aurantiaca (talk) 00:16, 19 July 2017 (UTC)

This version doesn't look too shabby. Then for sectioning without sectioning, one could just play around with standard emphasis.

Subsubsubsubsubsubheader

This topic can be roughly be broken uo into (one), (two), ...

Subsubsubsubsubsubsubsubheader one

Yada yada yada.

Subsubsubsubsubsubsubsubheader two

More yada.

Sure, why not? I just reverted your self-revert! Stigmatella aurantiaca (talk) 07:39, 19 July 2017 (UTC)

How about this version, where I've lumped all of the legacy sections under "Technical topics"? Stigmatella aurantiaca (talk) 07:52, 19 July 2017 (UTC)

Probably a good idea. The separate topics don't really each warrant a top level section. YohanN7 (talk) 08:20, 19 July 2017 (UTC)

That's what I thought, too. So I reverted my self-revert. Stigmatella aurantiaca (talk) 08:29, 19 July 2017 (UTC)

My two cents? It’s a really bad idea. It certainly looks better (there’s no doubting that bit) but the tradeoff in functionality is overly steep.

As wikipedians, we tend to get used to what is what and where things are located after working on a single article for a long while. However, the experience for long-time visitors, who are accustomed to Wikipedia conventions but are new to this article will be VERY different. I can guarantee you that many readers will have experiences like this:

“ Damn! I knew I was reading a section on Lorentz transformations but “Lorentz” doesn’t even appear in the table of contents!! ”

Such a reader was reading here but struggles to return to the section because the TOC doesn’t even mention “Lorentz.”

We don’t make a city smaller and easier to drive through by tearing pages out of its telephone book. I think you two just suffered from an industrial-grade case of groupthink.

It takes, what, a half-second longer to scroll through a two-level TOC with a modern trackpad or mouse? Greg L (talk) 15:54, 19 July 2017 (UTC)

I've set limit=3. Let's see how this works. I'm at work right now, so can't check its appearance/functionality on a tablet. Could somebody with a tablet and someone with a minitab check how the full site and the mobile site look on a tablet and minitab? Changes won't make any difference to a phone. Wikipedia mobile on a phone size screen doesn't show TOC at all. Stigmatella aurantiaca (talk) 17:29, 19 July 2017 (UTC)

TOC left|limit=2 looked fine on minitablets in Desktop view.

TOC left|limit=3 looks bad on minitablets in Desktop view.

TOC left|limit=2 looked fine on full size tablets in Desktop view.

TOC left|limit=3 looks marginal on full size tablets in Desktop view.

Stigmatella aurantiaca (talk) 21:26, 19 July 2017 (UTC)

On an iPad running iOS 10.3.2, four hierarchical levels show in Safari in desktop view and they aren’t numbered. On a laptop running MacOS 10.12.5, two levels show (no more than 3.1, for instance) in both Chrome and Safari. Greg L (talk) 21:37, 19 July 2017 (UTC)

And I thought I had near-OCD levels of perfectionism. Two levels on a desktop is fine. Four levels for some tablets isn’t a deal breaker because just a flick of the finger is plenty fast. Greg L (talk) 21:40, 19 July 2017 (UTC)

I was using an online simulator, trying out different virtual tablets and minitablets. The simulator is not always accurate, so I trust your report more than I trust the simulator. The question was how the LEFT option looks. I can tell that the online simulator goofs up in how it handles TOC in Mobile view. The only accurate way to judge definitively is with actual tablets.

It looks like we have to go with a regular display, if we want limit=3.

LEFT with limit=2 looked nice and clean, though, on all displays. Stigmatella aurantiaca (talk) 21:57, 19 July 2017 (UTC)

"On an iPad running iOS 10.3.2, four hierarchical levels show in Safari in desktop view and they aren’t numbered." - AAARGH!!!! Stigmatella aurantiaca (talk) 22:00, 19 July 2017 (UTC)

I conceived of and designed the {{val}} and {{xt}} templates we enjoy today and take for granted; I sweated atomic-level details like this during their development. But the TOC is established business that isn’t going to be changing. These huge TOC differences are all due to vast differences on the consumption end of things. Optimizing the experience for desktop at two levels showing (e.g. “3.1” via TOC3) is best, IMHO, even though other platforms are compromised in different ways. Look at the bright side: There are no “errors” in the TOC regardless of what platform is being used; the only difference is hierarchical depth, which is just a grayscale judgement call anyway. Greg L (talk) 22:08, 19 July 2017 (UTC)

Actually, that was iPad running iOS 10.3.2 in mobile view. One way or another, my reaction is still AAARGH!!! Stigmatella aurantiaca (talk) 21:28, 21 July 2017 (UTC)