Thanks for the link — that subject has been my #1 complaint with Wikipedia for a long time.
I'm glad they're finally doing something about it, even though it seems like I still won't be happy with it — judging by the wikis where the update is active, it's still too wide.
The rationale in the article is interesting, but IMO wrong — the idea that you need to balance scannability and line length suggests that line length is the main factor in improving scannability, when it seems obvious that what wikis have been missing for a while is better, probably sticky tables of contents for quick navigation (which something like Wikiwand figured out a while back).
Making users scroll 25% less is a poor solution to improve scannability, and degrades readability tremendously.
There are issues tied to the variability of layouts on Wikipedia that are more legitimate (how to deal with boxes, etc.) and would involve deeper design changes, but they would probably be worth it considering how important the basic reading experience is to wikipedia.
This is what I don't get. Why not just make your browser window narrower? Wikipedia is one of increasingly smaller set of sites which have the decency of NOT WASTING MY FRIGGING SCREEN SPACE by putting text into a narrowest of columns in the middle of my screen and leaving the sides empty.
I get that line scannability is an issue for some people, but that's why _you_ can adjust the way _you_ read the article on _your_ device, without affecting _anyone else_. Web browsers, and desktop environments in general had this magic window-resizing ability for decades.
At least there are cool browser extensions like CustomCSS, which allow me to fix sites I visit often. I guess I will be adding Wikipedia there soon... :/
Because I don't want to spend all day resizing my browser when I switch tabs?
I want websites that are made to be read to optimize for legibility and comfort, not for "not wasting screen space."
As I type this comment on HN, the right hand side of the text box where I type my comment is entirely empty.
Should the text box be full-width so it fills the screen and doesn't "waste space"? Hell no. It would be a pain in the ass to read my comment as I type it.
I don't know about you, but I'm very seldom (never?) reading more than one column of text at a time. Therefore, as I'm reading, anything else on my screen is "wasted".
Doesn't matter if it's white space or anything else.
It seems like we can't avoid "wasting space."
To see the absurdity of that metric, why don't you just make your window as small as possible all the time so you're never wasting any space? Why don't you reduce the font size to 1 pt so you don't waste space with fonts that are comically large?
I must admit that logic kind of escapes me, since it misses the most basic purpose behind reading on websites.
Now, I'm not unfamiliar with the regular HN comments about how everything is "dumbed down" and it's all about "information density" and cramming everything in one screen should be the be all and end all of design.
Unfortunately that opinion also seems to emanate from people who can't substantiate their preference with anything else than appeals to "power user" aesthetics, with very few actual use cases justifying it.
These are mostly ego-based arguments I tend to dismiss when discussing design decisions.
What does a screenful of text allow you to do that you can't do with a better proportioned column of text?
Unless you're the Flash, in my experience, absolutely nothing.
As far as saying the experience of reading properly should be left to the user/reader, I simply disagree. I'd rather benefit from the skills of designers and typographers presenting the content for me in the best manner than relying on hacks like resizing my window to get a decent experience.
I don't expect to go to a restaurant and be thrown ingredients for me to customize them _my_ way, I want my food cooked deliciously. And I want my content formatted properly and elegantly.
P.S. Note that your argument is entirely symmetric: if you don't want to "waste space" on a fixed width website... why don't you resize your window?
This is not at all about "dumbing down" or "information density", this is about choice and ability to consume content at one's own terms.
If a website leaves the text width to the reader (or their user agent), everybody wins, because everybody can read the text at their preferred width.
If, on the other side, a website tries to prescribe how wide the text should be, that choice is gone, and some readers inevitably lose.
So, no, my argument is not entirely symmetric - one approach gives everybody the same choice, the other one removes it.
To answer your "why don't you resize your window?", in the approach you seem to prefer, I do not have that option, because the website took that away from me, and no amount of resizing my window will change the text width the website dictates.
But as I mentioned, I already do go out of my way to configure my user agent to work for me, by means of a browser extension. Unless you want to take that choice away from me as well...?
(I will purposefully ignore your attempts at psychoanalyzing people over the internet based on a few bytes worth of text they wrote.)
Everybody can always read at their preferred width provided they do what's necessary to get there. Resize window. Modify CSS.
You can always do that.
My point is that the smart default is to minimize having to make any change. The question is what the smart default is.
Defaulting to full width text means that most of the time, on a large desktop monitor, I need to resize my window to adapt to the text (different font, font size, layout, etc.), so it doesn't meet that standard.
Using a maximum width solves the issue, unless you want, for some reason, to display text wider than is commonly considered comfortable (you can always resize your window to shrink it further if you need to).
Can you show me an example where you'd like to resize text beyond 10-ish words on screen?
I'm very curious about concrete use cases and what the result looks like (and why it's superior), other than the case indicated in the original article (i.e. logging, not long-form text).
I'm not interested in "taking choices away" from anyone. I'm interested in not wasting my time building custom CSS to avoid having bad reading experiences on websites I spend a lot of time on.
Maybe it comes down to what you think Wikipedia is as a product. I don't see it as an API that spits out unformatted content. I see it as an encyclopedia, and as such, expect it to provide the best reading experience for me out of the box, like I would expect from a printed encyclopedia.
So far, Wikipedia is at best mediocre when it comes to basic typesetting. I just wish it were better.
I’ve seen the “stop limiting my column width” argument presented on HN on many occasions and each time it seems to be an argument of optimization: let the user decide how to better process the information. That works well for people who argue that most things can and should be handled via a terminal window, but the average user doesn’t think that way. An average user expects curation.
Right. I think there's a case to be made for a distinction between letting the user decide, and making the user decide. Customisation should be possible, but not necessary, especially for the most common use cases.
I guess we'll just have to agree to disagree. You prefer narrower lines, I like my lines as wide as my browser window allows. For me, it's more comfortable reading text with long lines and having to scroll less, than having shorter lines and having to scroll often.
> it seems like I still won't be happy with it — judging by the wikis where the update is active, it's still too wide
I'm not sure what they looked like before, but I also visited a few of the localized wikis and ended up getting a horizontal scrollbar. I know that the theme on English Wikipedia is very old and not exactly responsive, but this is not something that happens there; I get no horizontal scrollbar. Everything is sized to fit within my window, except on certain articles containing wide tables or (unfortunately) multi-column reference sections, and even then, it's limited to those elements, which overflow, and the main article contents end up sized correctly. So I have to think that whatever changes they're introducing might actually be causing new problems...
I'm glad they're finally doing something about it, even though it seems like I still won't be happy with it — judging by the wikis where the update is active, it's still too wide.
The rationale in the article is interesting, but IMO wrong — the idea that you need to balance scannability and line length suggests that line length is the main factor in improving scannability, when it seems obvious that what wikis have been missing for a while is better, probably sticky tables of contents for quick navigation (which something like Wikiwand figured out a while back).
Making users scroll 25% less is a poor solution to improve scannability, and degrades readability tremendously.
There are issues tied to the variability of layouts on Wikipedia that are more legitimate (how to deal with boxes, etc.) and would involve deeper design changes, but they would probably be worth it considering how important the basic reading experience is to wikipedia.