[magduf]

You're confusing mass production with adoption of similar artistic concepts.

Buildings have light switches and ceiling heights standardized because of building codes which mandate these things, and the fact that many components are mass produced so they're extremely cheap (like light switches) and are used everywhere.

Vehicles have things in the same places because regulations require it, it makes sense to do so (mirrors aren't much help if they're behind your head), and because drivers expect similarity.

Laptops are similar because that's really the only way to make them work. A keyboard over the monitor isn't usable: your arms would block your view. This is just silly.

Websites may be built with the same components (python, PHP, web servers, HTML, CSS, etc.), but that in no way means they need to look the same. It's entirely possible to make them look very different, and you only have to go to the Wayback machine and look at how sites used to look 15+ years ago, and compare to modern ones, to see this.

Basically, most of it is a cargo-cult mentality: sites update to "newer" designs that are less useful because it makes them look "fresh" and "modern" even though they waste a ton of whitespace and make the site slower and less useful. Sites used to be much better in the mid-2000s.

[rpeden]

It's not just a cargo-cult mentality.

There's a decent amount of research that has found that putting things in non-standard locations impairs most users' ability to accomplish what they want on your website.

The Nielsen Norman Group is a decent place to find some of this research. There's some decent information here about how conventional layouts tend to be more effective:

https://www.nngroup.com/articles/horizontal-attention-leans-...

and how placing logos in the center instead of on the left often prevents users from accomplishing their goals:

https://www.nngroup.com/articles/centered-logos/

and the reason all of this matters is that the computer skills of a typical person using your website are probably far, far worse than you think they are:

https://www.nngroup.com/articles/computer-skill-levels/

so if you deviate from the "standard" layout too much, you'll hurt your website's business value because people won't be able to use it effectively.

I think there's still lots of room for creativity within a standard layout. But too much creativity might result in a site that looks better from an aesthetic standpoint but is less effective at actually delivering business value.

[pintxo]

> Buildings have light switches and ceiling heights standardized because of building codes which mandate these things, and the fact that many components are mass produced so they're extremely cheap (like light switches) and are used everywhere.

You mean similar to included youtube videos, Facebook like buttons, Google analytics, ...

> Vehicles have things in the same places because regulations require it, it makes sense to do so (mirrors aren't much help if they're behind your head), and because drivers expect similarity.

GDPR cookie settings? Imprint? Privacy statements?

> Laptops are similar because that's really the only way to make them work. A keyboard over the monitor isn't usable: your arms would block your view. This is just silly.

It would be silly to place the menu on the bottom, as you would always need to scroll down the whole page to see what options you have.

> Websites may be built with the same components (python, PHP, web servers, HTML, CSS, etc.), but that in no way means they need to look the same. It's entirely possible to make them look very different, and you only have to go to the Wayback machine and look at how sites used to look 15+ years ago, and compare to modern ones, to see this.

Houses may be built with the same components (bricks, wood, concrete, glas, metal), but that in no way means they need to look the same. It's entirely possible to make them look very different, and you only have to go out in the world and look at all the different implementations.

Sorry ;-)

[r00dY]

Menu on the bottom is obviously pointless, but sticky menu on the bottom would be so cool, most of the mobile apps moved to tabbar.

Why it's not used extensively in web? (it would make it at least a tiny bit less boring).

On iPhone when you scroll down the page, bottom Safari bar collapses and your sticky menu would be still sticky at the bottom. But clicking any button on the sticky bottom menu wouldn't invoke click on the menu. It would make iPhone Safari menu uncollapse -> huge confusion -> potentially less engagement -> retention.

Web today is full of similar constraints. That's why there are usually one or two options that make sense and everything looks the same.

[HereBeBeasties]

Because on mobile the primary interaction point is your right (or left if you're a leftie) thumb, which naturally falls over a tab bar at the bottom of a screen. Whereas on the Web the primary click zone where your cursor focus is, is at the top of the browser window where the navigation elements are.

There's also the obvious stuff about top-to-bottom reading order making such a layout making the navigation much more obvious when you first look at a site (which is probably want you want for most things, although not all, which is reflected in things like blogs often putting ancillary navigation elements in the page footer).

I don't think this is a broken thing that needs fixing.

I'll also note that it has taken mobile UX designers YEARS to start moving navigation elements to the bottom of their apps, driven by larger and larger screens making elements at the top of the screen quite uncomfortable for one handed use. It's required that to become sufficiently cumbersome that the trade-off in immediate discoverability becomes worth it. And you'll note that the most popular mobile apps such as WhatsApp still have all their navigation elements at the top, regardless.

[lotyrin]

It's not mass production that causes this though, it's the constraints (the things that make the process "design" instead of "art") that come from how they are used -- and while some of those have solidified into technical or legal constraints, many are simply matters of convention.

Similarly, any user experience study of a website isn't going to find that you should make your menu some diagonal shimmering nonsense, or try to convey information on the side of a spinning cube or any of that sort of thing.

[r00dY]

Thank you!

Design != Art !!!

If designer is more interested in fulfilling his artistic/creative ambitions than business value for client (making things great looking, stable and reasonably easy to implement -> in budget), he shouldn't even touch digital today. It's too complex and filled with too many constraints.

But if same designer wants to make decent money, well...

[zbtaylor1]

It's not a cargo-cult mentality. It's the convergence on a set of best practices "illuminated" by analysis of user-behavior data, as well as the application of basic graphic design principles (like the use of white space to create a visual hierarchy).

[ken]

> Laptops are similar because that's really the only way to make them work. A keyboard over the monitor isn't usable: your arms would block your view. This is just silly.

With tablets like the iPad, the monitor is the keyboard, and your arm does block your view. Maybe it is 'silly' but a whole lot of people buy and use them.

There's been lots of other laptop designs, and they seemed much less silly to me than tablets. Compaq used to put the trackball on the right side of the case (even lefties I know mouse right-handed). HP made a laptop with a little pop-out mouse, which didn't even require a surface to place it on. IBM made a keyboard wider than the case, which unfolded when the lid was opened.

Everyone I know who tried or owned these laptops loved them. Why did they die out?

[Izkata]

We were actually talking about something else in the office today that seems similar: Apparently around 2009-2011, several companies were coming out with dual-monitor laptops that folded out when open... Which then died out and disappeared.

[jsgo]

Are we talking like Razer's concept from a couple years ago at one of the tech shows (CES? I don't remember, honestly) or something like Acer's Iconia dual screen tablet/notebook?

[jazzyjackson]

I had the Thinkpad W700 with the pop out side screen, not to mention the wacom tablet built into the right of the touchpad. It was a great parlor trick, and genuinely useful for running a terminal or a music player to the side of your main workspace.

[0] https://www.engadget.com/2009/01/26/lenovo-w700ds-laptop-rev...

[gfosco]

Microsoft also had a foldable dual-screen design in the works called Courier. Apparently they're trying to revive it: https://m.windowscentral.com/microsoft-bring-back-courier-20...

[Izkata]

Acer Iconia was actually one of the models we found online, I don't know what Razer concept you're referring to.

[jsgo]

Just had to look it up: https://www.razer.com/project-valerie

They demoed it and apparently the unit they demoed was stolen when they were leaving the show.

Not going to lie, I’d love an Acer Iconia style portable at this point. Wouldn’t use it for desktop publishing or probably anything with heavy typing, but just seems like a fun form factor.

[Animats]

Bob Lutz, the car guy (senior positions at BMW, GM, Ford and Chrysler), once commented on this. He noted that his designers liked Bang and Olafson design, with black on black buttons and unusual form factors. This was all wrong for a car interior.

[onion2k]

make the site slower

Websites are far faster than they've ever been in the past, even with the bloat of frameworks and fonts. Most sites load in < 5s these days. That was not the case 15 years ago.

[stonewareslord]

I have to disagree with this. Websites loaded slower because our connection was much slower. You’re comparing network speeds to page load times, which are different.

With today's speeds, the older internet is so much faster than “modern” sites. Some sites even have a loading bar (ex: gmail). Compared to sites with minimal/no Javascript, they’re much slower. 5s is not a good load time for a webpage especially with today's internet speeds.

[onion2k]

5s is not a good load time for a webpage especially with today's internet speeds.

I completely agree. I work to a perf budget of 200ms to first paint on the things I build. 5s is a huge amount of time to wait. It's the upper limit of what I think is acceptable.

[freehunter]

What you're describing is not a web site, but an application delivered through the web browser. Sites in the past were smaller and faster because they were just web sites. Today we have a whole host of applications delivered through the browser, with the Internet as the hard drive they're loaded from.

It's not a valid comparison at all. Compare the time it takes to download and install Thunderbird versus the time it takes to log into Gmail, and there you'll have a valid comparison. Compare the time it takes to install Office versus the time it takes to log into Excel Online and there you'll have a valid comparison. And in comparison, it's a hell of a lot faster.

[paulryanrogers]

Gmail in 2018 feels like I'm downloading Thunderbird every time it opens. Ideally I'd only need to load it all once and then it'll be cached for faster load next time. But maybe that's too much to ask when changes must be deployed multiple times a day.

And yet the Gmail experience is not significantly better than non-JS alternatives, at least for m my use cases.

[JoeAltmaier]

Irrelevant. When more resources become available(like bandwidth) we use them. The fact is pages are loading faster than 15 years ago.

Yes they could be faster, but most are still useful at a 5 second load time.

[TeMPOraL]

Which ones?

In what is most likely a parallel universe that I inhabit, websites have been getting consistently slower for a while. No amount of edge black magic can compensate for the growing bloat of fonts, frameworks, multiple-dozen-megabytes background videos, and half a thousand requests to ad networks.

[com2kid]

> Websites are far faster than they've ever been in the past, even with the bloat of frameworks and fonts. Most sites load in < 5s these days. That was not the case 15 years ago.

I had a cable modem 15 years ago. Pages loaded faster, even though I had 1/10th the bandwidth that I do now.

[paulryanrogers]

Could be that web servers don't have much more bandwidth per client. Or that they're actually cold starting VMs or lambda functions instead of running constantly. Still, I agree the overall feel of the web had gotten shower and jankier over time.

[r00dY]

True.

A while ago most people didn't realise to what extend loading times affect engagement.

If they knew, flash would be dead much quicker.