Oh yeah. User-agent stylesheets and a list of links. The pinnacle of human-centered design. /s The author either doesn't understand design or has a very low-bar for tactically good user-centered experiences, or both.
I can list an endless list of bad things on this website:
- Bad information architecture.
- Poor contextualization, narrative, and direction of a message (just because it's Berkshire Hathaway, doesn't mean that I automatically know what I'm trying to find in their website)
- Poor document structure that renders at the whole width of wide screens and doesn't wrap for small screens.
- Bad typography and typographic scale (can a visually impaired person read the small text that shows dates and footnotes?)
- Bad use of color. Everything is blue, the links are purple. Usually in plain stylesheets blue is reserved for links and purple for clicked links.
- External links don't open in a new tab.
- No branding. Can a user trust that this is an official website?
- No navigational elements (navbar, breadcrumbs, etc) that allow a user to return to the main page (which is supposed to be the "master class of human-centered design). Every click is a dead-end.
- Multiple SEO and accessibility issues:
- Does not have a <meta name="viewport"> tag with width or
initial-scaleNo `<meta name="viewport">` tag found. Document
does not have a meta description
- Text is illegible because there's no viewport meta tag
optimized for mobile screens.
- Tap targets are too small because there's no viewport meta
tag optimized for mobile screens
- The page does not contain a heading, skip link, or
landmark region
> No navigational elements (navbar, breadcrumbs, etc) that allow a user to return to the main page (which is supposed to be the "master class of human-centered design). Every click is a dead-end.
This is the web, it works like people expect web pages to work. No custom behaviors, links work like links, the back button does what the back button should. That is absolutely human friendly. If external links were supposed to open on a new page then the browsers would default to that.
> Multiple SEO and accessibility issues
Whether a site has SEO issues is irrelevant to usability of the site. It's not like they are looking for organic traffic. If I'm looking for BRK's homepage, it's easy enough to find which is all that matters from a SEO perspective for a site like this.
The accessibility issues are a concern if they actually existed. This kind of site is ideal for screen readers and while it doesn't automatically adjust for mobile, it doesn't get in the way of zoom or reader view which makes the site just fine for people who struggle with small fonts.
> The page does not contain a heading, skip link, or landmark region
You don't seem to understand why the skip links are there. In this case, there is no heading full of links you need to skip over, if you open the page and start tabbing, you are already on the first bit of content. It is better from a usability perspective than a skip link.
You shouldnt conflate browser defaults with good usability.
Sure its generally speaking a good rule of thumb to follow but saying the default is the right thing to do because its a default is a logical fallacy. Its appealing to authority for no actual reason.
Why is the default the default? Is it because of user studies or is it because a developer said thats the easiest thing to implement in the time i have available.
I think the high-bar that you desire is less user friendly and more designer-job-friendly.
- Here the information you're probably after is right at the top: letters from the CEO, and annual statements. If you're visiting the site you probably know what you're after.
- Websites optimised for mobile usually wrap text and remove zooming because the designer is perfect at deciding at what size the text looks prettiest, and doesn't want the user to spoil this. This is the worst thing that's happened to web design, probably ever. It's worse than IE6 at it's ugliest.
- I am not visually impaired, so can't comment on this.
- Links are underlined consistently making it clear what is a link.
- I can middle click, or open in new tab. I don't want new tabs popping up just because the designer wants me to stay on their site.
- Does branding ever help this? They have the .com, why would a pretty logo make things safer? They're quite easy to screenshot.
I agree that a link back to the homepage would be useful.
Overall a return to information dense websites like this one would be a wonderful thing.
Berkshire Hathaway has highly visible brands like Berkshire Hathaway Home Services (https://www.bhhs.com/) and Berkshire Hathaway Energy (https://www.brkenergy.com/. You see their brands very frequently on yard signs and advertisements.
Many people can land on this website first. Not everyone is looking for the CEO letters and annual statements. That's basically assuming that anyone who interacts with BH has to be an investor which is definitely not the case.
You could do something as simple as making the links blue instead of purple and it would already be an infinitely better design because people assume purple links have been visited.
I doubt very much that anybody can launch a website / web-app today with only user-agent stylesheets and be successful. Even for an internal product that would be seen as bad/lazy/unoptimal.
Although there are also a fair number of academics who invest more effort in web design than I do, this type of design is fairly common in math. I haven't heard anyone within the field criticize it as "bad" or "lazy".
My students are part of the audience, but not all.
For example, if I give a talk at a conference, then someone who was in the audience might plausibly look up my website to get a sense of my research interests.
For some reason I have always seen academic websites like yours as a different concept. Most academic websites look like yours and I think that’s fine.
I was thinking more of something like a CMS control panel or some data-entry tool, or a consumer website like an e-commerce.
> I doubt very much that anybody can launch a website / web-app today with only user-agent stylesheets and be successful. Even for an internal product that would be seen as bad/lazy/unoptimal.
Yeah that was a profoundly underwhelming article, especially for an "expert". You would think Maria and Matt would set a higher standard for contributors.
It's an interesting idea, and I totally agree with it, but the article seems like a long shower thought. I'd love to see more evidence that a list of links is actually the right solution for Berkshire Hathaway customers. I bet it is, but the author is making a big assumption that what exists works for their customers because Berkshire Hathaway is successful.
Some of what you're describing is very important for users, like accessibilities issues, but not every general solution makes sense for every audience. That's the essence of user-centered design: a ruthless focus on optimizing for particular customer needs. You're not wrong that the site could use an information hierarchy pass—items like "A Message From Warren E. Buffett" are clearly more important than "Berkshire Activewear," for example—but when you're designing for an audience that knows exactly what they want, the best job you can do is to get them there with no frills. Real design is about solving problems for users, and an ugly list of links can do that if the links are right for that user base. In fact, I'd hope the designer spent the bulk of their time figuring out exactly what the right links are. That's 80% of the solution in this case. Better visual hierarchy, better typography and color, navigational elements to help users go back or discover different aspects of the site–the rest are accessories to the design.
So, you're not wrong in your design criticism, but it's important to know what decisions will make the most impact for the users you're designing for and which are less important. For Berkshire Hathaway's target customer, this website may actually meet their needs perfectly fine.
Edit: One note: the author of this article does NOT provide any evidence that this solution actually is the right one for that customer base. I wish they did, because I suspect it's actually pretty close to what will do the best job for their users, but the author assumes it's right based primarily on Berkshire Hathaway's success.
> - Poor document structure that renders at the whole width of wide screens ...
Does everyone just use their browser full screen now? I never want super-wide text, but I can always narrow my browser window or switch to portrait mode if I'm using a phone or tablet.
I just checked hn, and it lets text get too wide to comfortably read, but stops somewhere before full screen width.
The claim made in the article is that ignoring UX and SEO conventions tells you all about the company. That is the “human centered” message behind the site design.
A website with a bunch of responsive elements, meta tags, and fonts spent a lot of time thinking about how they are going to convert clicks.
I can list an endless list of bad things on this website:
- Bad information architecture.
- Poor contextualization, narrative, and direction of a message (just because it's Berkshire Hathaway, doesn't mean that I automatically know what I'm trying to find in their website)
- Poor document structure that renders at the whole width of wide screens and doesn't wrap for small screens.
- Bad typography and typographic scale (can a visually impaired person read the small text that shows dates and footnotes?)
- Bad use of color. Everything is blue, the links are purple. Usually in plain stylesheets blue is reserved for links and purple for clicked links.
- External links don't open in a new tab.
- No branding. Can a user trust that this is an official website?
- No navigational elements (navbar, breadcrumbs, etc) that allow a user to return to the main page (which is supposed to be the "master class of human-centered design). Every click is a dead-end.
- Multiple SEO and accessibility issues: