Not long after ads started appearing on the Web, there was research showing that Web browser users quickly learned to visually filter them out.
So, when Material Design came out, and it wasn't even distinguishing the extents of UI elements (e.g., transient UI object with same background as what it partially overlapped, with no border), it violated much of what we knew about HCI (i.e., in the interests of the user or task), and it looked like a brochure (i.e., in the advertiser's interests) more than anything else... Occam's Razor needed only to mutter the words "advertising company".
I know that some percentage of the people who have to work atop Material Design are doing good HCI despite it, but they're fighting against sabotage, and we all are dumber and less effective for it.
What exactly is wrong with Material Design? It seems to fall in the category of "good enough", where the differences between "good enough" designs are rounding errors when compared to the benefits of standardization, consistency, and user familiarity.
Parent comment doesn't say it, but I infer one of the things that they are talking about concerns how some secondary interactable elements look indistinct from text, and only hint at being interactable if you have a cursor and hover over it. So beside a flashy CTA "Upgrade plan" might ba a "no thanks, let me cancel" interactable that looks like text
If you are a software engineer with no sense of design, do make it like Google. Material Design allows anybody to put something together that looks half decent, is somewhat consistent and familiar with users.
Funny I like the MUI React widgets for applications that are used by people who use those applications all day for work. For consumer-oriented mobile they just seem to be meant for maximum feasible misunderstanding and then some.
The good designers, yes. I'm sure there are plenty of other designers who think their opinion is better than the likely thousands of person-years of UX research that the major OS vendors (combined) have invested into their interface guidelines.
For each of the major platforms, following the guidelines will at least make your app consistent with the other apps on the user's device, which is a decent-enough baseline. Going wild and re-inventing how a drop-down should look or how scrolling should work is probably just going to annoy the user.
What's the alternative? I feel like I read a bunch of vague critiques without anything specific.
For what it's worth, I think Google's products are far superior to it's competitors, and I've never had issues with a site looking too Google, though I've been burned by some complex SPAs that were slow and buggy.
I could not disagree more with this piece. I actually quite like Material Design. Gmail does feel like the red headed stepchild mostly because it predates Material Design by a decade, and there are still many power user settings that rely on old UI (similar to the problem Microsoft has with Windows 11 nowadays).
We can also pass off Apple GUIs as clunky and bloated by simultaneously opening every possible modal and sidebar in a manner that isn't representative of real world use, then taking a screenshot.
Every article that criticizes some general UI thing, like the general state of current user interfaces, how everything was better 10/20/100 years ago, how the start menu or settings in some operating systems are bad, ... really should be forced to provide some actual examples and analyze them in some detail. All we get in this article is a screenshot of Gmail, resized to a small size so that we don't even have a chance to decipher anything on it, and the repeated assurance of the author that this does in fact represent an unusable UI.
But even I, as someone who doesn't use Gmail, can quickly understand that interface on the screenshot after zooming in a bit. Maybe it looks a bit chaotic, but there seem to be some menus opened just for the sake of argument. Maybe this UI is incredibly powerful? Maybe they didn't dumb down the interface, which is something that is also criticized here a lot. It's hard to tell from a screenshot alone.
I feel like the article is completely unconvincing when they give GDrive as an example of bad design without going into specifics, then give Apple as an example of good design, then fail to comparatively examine how iCloud offers a better UX
> Every time I use Google Drive or the G Suite admin console, I feel lost. Neither experience nor intuition helps—I feel like an old man seeing a computer for the first time.
I’ll do you one better: try making a chart in Google Sheets.
Incorrect. Making a chart in Sheets is mind-numbingly easy.
The problem comes (as you describe in a later post) if you don't just want a chart, but instead want a particular chart for a particular purpose. Not some weirdo crazy purpose, either; there are many ways to get into trouble trying to make minor adaptations.
Same thing with Material. It's great for making a bland clone of a thousand other interfaces, preferably one with only a handful of relevant interactables ("find the beige car" is hard when they're all beige). If you need to distinguish things that are different because they have a reason for being different -- say, labels and buttons -- well then you're SOL.
Readit News
So, when Material Design came out, and it wasn't even distinguishing the extents of UI elements (e.g., transient UI object with same background as what it partially overlapped, with no border), it violated much of what we knew about HCI (i.e., in the interests of the user or task), and it looked like a brochure (i.e., in the advertiser's interests) more than anything else... Occam's Razor needed only to mutter the words "advertising company".
I know that some percentage of the people who have to work atop Material Design are doing good HCI despite it, but they're fighting against sabotage, and we all are dumber and less effective for it.
For each of the major platforms, following the guidelines will at least make your app consistent with the other apps on the user's device, which is a decent-enough baseline. Going wild and re-inventing how a drop-down should look or how scrolling should work is probably just going to annoy the user.
For what it's worth, I think Google's products are far superior to it's competitors, and I've never had issues with a site looking too Google, though I've been burned by some complex SPAs that were slow and buggy.
But even I, as someone who doesn't use Gmail, can quickly understand that interface on the screenshot after zooming in a bit. Maybe it looks a bit chaotic, but there seem to be some menus opened just for the sake of argument. Maybe this UI is incredibly powerful? Maybe they didn't dumb down the interface, which is something that is also criticized here a lot. It's hard to tell from a screenshot alone.
I’ll do you one better: try making a chart in Google Sheets.
The problem comes (as you describe in a later post) if you don't just want a chart, but instead want a particular chart for a particular purpose. Not some weirdo crazy purpose, either; there are many ways to get into trouble trying to make minor adaptations.
Same thing with Material. It's great for making a bland clone of a thousand other interfaces, preferably one with only a handful of relevant interactables ("find the beige car" is hard when they're all beige). If you need to distinguish things that are different because they have a reason for being different -- say, labels and buttons -- well then you're SOL.
What's your preferred way? Matplotlib? Seaborn? R?
For example, adding data labels to a scatterplot. Completely insane how they want you to do it.
Or removing the Y axis from a bar chart. Have to make the text size 11 and white. Why?
There's a lot of little things like this that are just mind-numbing.
I like native PowerPoint charts (not Excel) and R's ggplot2 personally.