Guidelines For An Uncomplicated Web
March 01, 2005 | 
39 Comments
Summary: Offering up some friendly advice to those who, like me, would like to see a less complicated Web.
I used to say that I hate computers.
With a few exceptions, there are many more things I’d rather be doing than sitting in front of a computer. I’ve said this a thousand times if I said it once — computers should work for people, not the other way around. Unfortunately, this doesn’t seem to be the case much of the time.
I used to blame the computer. Lately I’ve come to realize most of my, er, issues with technology seem to be rooted in poor design, poor planning or a simple lack of understanding about how people interact with technology. So I guess I should lay the blame on people.
Usually this is due to over-complication.
Why’d we have to go and make things so complicated…
For some reason people tend to complicate things. I’m guilty of this and it’s something I battle against every day, both on- and offline. I’m a collector. And all about “add-a-trick” you know? I’ve been trying, for the last year or so, to adopt a “less is more” attitude toward life.
It’s hard, but I’m getting better and I do think it’s having a positive effect. The less complicated you can make your life, your work, your relationships, the easier things are to manage, use and have fun with.
A shift towards the less complicated?
I don’t think I’m the only one who is seeing this. I’m seeing a shift toward the less complicated, in many ways. Look at applications like Flickr. Its beauty lies in its clarity. It’s not simple, by any stretch, but neither is it complicated. I learned most of what I needed to know about it in about 10 minutes. Now that is a learning curve I can handle.
Basecamp also comes to mind. Why? It’s the first of many project management tools I’ve tried that actually made my work easier. I can think of many applications I used in the past that were simply too much—I spent just as much time managing the application as I did using it. This was usually do to “features” I never used and couldn’t disable.
Why would you use a program (or Web site, or process) that is so hard to use it actually makes your work harder? Ask yourself that question about things you use every day. I bet there is something you can do with out.
I recently ditched my feedreader. There was simply too much to keep track of. I prefer Blo.gs which simply shows me who was updated when and lets me choose what I feel like reading.
“Features” can be a dirty word
At Children’s I was introduced to Office 2003, the only part of which I used on a regular basis was Outlook. They added all sorts of cool features and they “improved” the UI. All of which ended up making things much harder for me.
Another example of how “features” can have an adverse effect on the user-experience: Macromedia Dreamweaver (at least on OSX) has so many “features” is almost too slow for me to use. It didn’t used to be like that, in fact, while I never have used it much, I seem to remember it being much faster and easier a few versions ago.
(And boy do I miss Homesite.)
Features don’t always add much of value to most software and Web applications, designers and developers need to be really careful about adding features. My favorite applications and Web sites do a few things very well. The more you add to something, the more opportunity you have for something to break.
A few things to remember…
Here are some guidelines to remember when designing, architecting, building and writing for the Web:
- Less is usually more. Less code, less words, less visual elements—all usually better.
- The best solution is often the easiest one.
- If you’ve got a choice between simple and complex, choose simple.
- It doesn’t have to be simple, but it has to be clear.
- Be careful with clever. Straightforward is almost always preferable to clever.
- Use plain language.
- Find the core of the problem, or project and nail that first. Then stop.
- Be really good before you try and be better than the rest.
- Originality isn’t always a good thing. It’s possible to be too different.
- Don’t design for everyone.
- When it comes to features, go for the win-win.
- Make sure your users actually want and need any features you plan on adding. Weigh anything extra against your goals and focus on added value. If it’s not there, or your not sure, don’t do it.
- Before you launch, see what you can safely strip away.
- If you feel compelled to add something extra, make it as unobtrusive as possible.
- If it doesn’t work, or isn’t used, remove it.
- Only publish what you can manage (thanks Gerry).
- Know your audience.
There are no rules on the Web, only guidelines, there are exceptions to every item on this list. A personal site, for example, shouldn’t necessarily concern itself with all of this.
If given the choice, my advice (for what it’s worth) is to go for as uncomplicated as possible, whenever possible. By taking the least complicated route you’ll make life easier for your customers, readers and users—and probably do yourself a big favor in the process.
Filed under: IA and Usability
Comments
1. Dan Bowling said:
You bring up a good point. As designers, we often get in the mindset to do the most that we can, believing it will be the best that we can. Sometimes bringing back simplicity is the key, and the real designers are the ones who know what to leave out, not what to put in.
Bravo!
Posted on March 1, 2005 01:11 PM | #
2. Graham said:
There was a solid trend in web design to put in as much crap as the developer could, just to show off their skills. Drop-down/Fly-out menus, Splash(Flash) intros, animated crap everywhere…..these elements have cluttered web sites for years now. It’s really nice to take a step back and and focus on making users happy with the site’s functionality instead of impressing the client’s CEO with a goofy Flash intro.
Posted on March 1, 2005 01:52 PM | #
3. Garrett said:
Amen. Now if only everybody will buy into this approach and start focusing on the important parts.
I really like the 80/20 rule here. You can please 80% of your audience with 20% of your features.
We need to stop enabling the fringe cases at the expense of their core audience.
If you chase two rabbits, you won’t catch either.
Posted on March 1, 2005 01:54 PM | #
4. Justin Perkins said:
Good point indeed, as a developer I find myself always adding “features” to my sites (partly because *I can* and partly because I think somebody would acutally use said feature) and as it turns out I have to spend more time explaining the “feature” than the potential time-savings I was expecting.
I would love to simplify my life in general as well, I like this attitude.
You kind words towards Basecamp, and Jason’s mention of it yesterday has finally motivated to create the freebie account. I never really understood *what* Basecamp actually was (a web app.) until now…very cool. I actually had the idea for a web-based project management app. three or so years ago, just after I finished my own Senior Project (for my BS in CSc). I rushed out and bought the domain name “eprojectlog.com”, then never did anything with it. Oh well, that’s what I get for dropping the ball because I was spread too thin.
Posted on March 1, 2005 01:54 PM | #
5. Stu Schaff said:
In development, I always think “just because you can doesn’t mean you should” (or, if you prefer, “keep it simple, stupid”). Great guidelines, Keith.
Posted on March 1, 2005 02:02 PM | #
6. Justin Perkins said:
I prefer the friendly kiss, Keep It Short, Simple. A stupid kiss only makes complicates matters.
Posted on March 1, 2005 02:18 PM | #
7. Sean Voisen said:
“The less complicated you can make your life, your work, your relationships, the easier things are to manage, use and have fun with.”
Amen.
Posted on March 1, 2005 02:38 PM | #
8. John Zeratsky said:
Once again, 37signals is helping to lead the way here. Not just with Basecamp, but with the Less Software principle that Basecamp embodies. We would all do well for follow in their footsteps. Thanks, Keith, for these excellent guidelines.
But remember: Each advance (or mini-revolution) that sweeps the web design community requires a refined round of client education. Clients love features, so it’s important to keep them on the straight and narrow. Fortunately, they will often see that the uncomplicated route is best, once they get a chance to use the site/app/whatever they are paying for.
Posted on March 1, 2005 02:39 PM | #
9. John Zeratsky said:
By the way, I am curious – What feedreader did you use that was too complicated?
Posted on March 1, 2005 02:41 PM | #
10. Keith said:
John – Client education is always an issue isn’t it? ;) As far as my feedreeder, that wasn’t the problem at all. The software is fine. My problem is keeping track of the sheer number of feeds I’m interested in. It’s not really a software/application issue – more an information issue.
When I see 436 unread feeds I just want to mark them all as read and be done with it.
Posted on March 1, 2005 02:48 PM | #
11. John Zeratsky said:
Now I see what you mean about the feeds. Reading sites you enjoy shouldn’t be a chore, right?
Client education is definitely always an… issue, for sure. Although the design community (particularly the group of us into blogs) generates a lot of good ideas and good discussion, none of it matters unless we can translate that into meeting goals for our clients. Simplicity, blogs, standards, Ajax and all that other stuff are cool and all, but we need our clients to see why it matters.
Posted on March 1, 2005 02:54 PM | #
12. Thomas Baekdal said:
Although I do agree with you message, there is a conflict between making somthing simple to make - and making it simple to use.
A good example is this is the Tada List (www.tadalist.com). Here it is very simple to add items to your list, but the code behind is somewhat complicated compared to regular form handling.
So it is quite often harder to make something simple to use. But it is worth the extra effort.
Posted on March 1, 2005 03:01 PM | #
13. R.B. Boyer said:
This seems very much like the UNIX methodology, where you write little programs that are all really good at the one thing that they are written to do. Then you just hook them all together to do something wonderful.
Posted on March 1, 2005 04:37 PM | #
14. Mike said:
I agree whole-heartedly. The simpler the better. I really like that list you came up with as well, that will definitely make it into my workflow/testing/proposal process.
Thomas has a good point as well. I’ve found quite a few times the simpler you make something to use, the harder or more complex the back-end is. It is totally worth the extra effort.
Hopefully there will be a shift towards this ideal. I know a lot more people would appreciate it in the end.
Posted on March 1, 2005 09:50 PM | #
15. seth said:
I miss Homesite too…I’m using Dreamweaver in code-only mode now, and it’s ok but still not as good.
I tried Eclipse as well, and was scared of it.
:/
Posted on March 2, 2005 12:00 AM | #
16. paul haine said:
Though I love Dreamweaver, it’s very much a development tool that’s struggling to move with the times. The Design view is pretty useless when you’re designing CSS websites, with elements overlapping each other - I stick to using Code view, and then previewing in a browser now. I stick to using it because I like its code-completion when I’m writing, but I could get by with Notepad if I had to.
Posted on March 2, 2005 12:43 AM | #
17. Mahi said:
Agree with the post in general.
Kindly disagree with a few of the comments. In particular the “Clients love features”. I go back to the sales job I had in high school during training when the boss said “show features, sell benefits”. What good is the super high tech revolutionary spring if the client doesn’t know it will make them run faster and jump higher?
When presenting to clients I constantly get “we would like to have ____ on our site” at which point I ask/tell them to show me/themselves the real BENEFITS of having that. If they can prove it to themselves, I gladly put it in for them. Most of the time they find that they neither want/need/like these features which leads to happier clients and more importantly a more user/customer friendly site.
As for HTML editors, I also use Dreamweaver to lay the HTML groundwork, but use Firefox and the Web Developer Extension for all css related testing. Talk about a simple product… Firefox and the extension both embody this idea.
Cheers!
Posted on March 2, 2005 02:39 AM | #
18. Andy Budd said:
Unfortunately features sell products. If I had a pound for every time a client has asked for a “cool” feature like a new ticker, I’d be a rich man. I seem to spend a good portion of my life arguing again the addition of unnecessary features because the MD likes sports and thinks it would be great to have sports news on the home page of their internet.
It can be very difficult to explain that every non essential addition vies for attention and makes using the application significantly more difficult.
I agree with some of the comments that making things appear simple, can actually involve a lot of hard work. Once the application is finished it looks so simple you can’t believe it took so long to build. Because much of this work is intangible clients may feel that they are not getting value for money. This is because people tend to latch on to more tangible things like extra features.
From a marketing standpoint, it would be very difficult to sell an upgrade to Office boasting 50% less functionality, even if it was 100% easier to use.
Posted on March 2, 2005 03:15 AM | #
19. Richard Kendall said:
Dreamweaver MX 2004 on OS X is painfully slow, it must take 25% longer to do simple updating, sometimes a text editor would be far quicker. Good broad points about technology, too many features don’t necessarily give more helpful options. Too many distractions on the Web make for a lot of potential time wasted in front of a screen, when those of us in the industry may have spent a full working day already glued to a monitor.
Posted on March 2, 2005 03:39 AM | #
20. Gordon said:
Great article - I ponder the seeming trend for simplicity recently as well (here if you are interested).
To pick up on Richard’s comment - I probably use a text editor to make about 30% of the changes to any of the sites I work on. So whilst there is definitely a requirement for software to be focussed on the usage experience, there can sometimes be a requirement for education as well (Richard, not a dig at you, I’m as guilty of it as well!).
Posted on March 2, 2005 05:13 AM | #
21. Small Paul said:
Amen to simplicity.
http://weblogs.media.mit.edu/SIMPLICITY/
Posted on March 2, 2005 07:37 AM | #
22. Keith said:
I eluded to this in the article but didn’t exactly state it implicitly but “simple” isn’t a word I really jive with. The reason why is something that appears simple, might not really be simple, as a few of you have stated.
I like “uncomplicated” or “clear.” You can have something that is complex yet clear. Still, if there is a solution that is simple, chances are it’ll have a better chance to be uncomplicated–if you know what I mean.
As far as clients and stakeholders loving features. I’m sure this is the case at times. Luckily I’ve not see too much of this myself. But then again I’ve only been “outside” for a few months.
I’m not sure how well this would work on the consulting, freelance, agency side, especially considering the bottom line goals of the developer, but as an in-house Web professional you can usually show the lack of value in such features.
A sports ticker on the homepage of an intranet would be an easy one for an in-house person I imagine. A call to the HR people might just do it.
But I get the point. It’s really too bad. We try to help, we try to save time, money and effort and increase ROI, but sometimes the people we work for just won’t let us…
Posted on March 2, 2005 09:17 AM | #
23. Nick Finck said:
Keith, it looks like there is questions about the future of Blo.gs …apparently it is for sale as of just the other day.
Posted on March 2, 2005 09:47 AM | #
24. JonathanB said:
Thanks Keith, best web development article I’ve read in a while.
For those of you that miss Homesite, I recently discovered TSW Webcoder. A product very similar with (from what I can tell so far) as many useful feature as Homesite had.
Once you register, the program is free (although very worthy of a donation).
http://www.tsware.net/
(I’m not affiliated with them, just a satisfied user)
Posted on March 2, 2005 10:17 AM | #
25. jake said:
Simplicity always wins out. But when it comes to features, I’m a huge fan of add-ons and check boxes to turn things on. Keep the thing super simple, but then allow me to add in things I actually want.
Posted on March 2, 2005 01:09 PM | #
26. Guy said:
But, Keith, can’t you just get HomeSite itself from Macromedia? I use an older version, but it dovetails nicely with Bradbury’s TopStyle.
Posted on March 2, 2005 01:35 PM | #
27. Keith said:
Guy - Ahh, no. I’m on a Mac. I actually bugged Nick B. about this last year at SXSW to no avail and no surprise. I’m slowly getting used to BBEdit, which many people swear by.
I’ve been doing most of my coding with the text editor…
Posted on March 2, 2005 01:46 PM | #
28. Guy said:
Ahhhhh. So, MM does not do HomeSite for Mac? Pshaw!
On the other hand, you are way less prone to malware assault than those of us in the Windows majority.
Posted on March 2, 2005 02:28 PM | #
29. BryanJ said:
Speaking of less is more, staying simple, and Basecamp, you should definitely read the Basecamp Manifesto.
Posted on March 2, 2005 09:31 PM | #
30. b said:
Thank you for articulating those points. How often it is that I need to remind myself that, especially,”Be really good before you try and be better than the rest.” Cheers.
Posted on March 3, 2005 05:53 AM | #
31. Ryan Brooks said:
Hmmm…
At risk of being stoned, I’ll have to say that I disagree with the overall concept of the KISS principle, and the vast majority of the responses to your post here.
I’m a software developer first, and a website designer/developer second; I’ll touch on both concepts here.
I’ll pick up with Andy on this one. Features DO sell products. Let’s look at the content management giants. Stellant and Vingette sell for one million plus for a soup-to-nuts system, but they do a lot more than content management. They can search a network and manage digital assets, integrate with HR and financial software and more. And the reason for this? Most likely because they’re competeing with the other products. When one company adds a feature, the other isn’t far behind, while trying to come up with new features to stay ahead of the curve. The problem becomes when you’re outright TRYING to keep the interface simple and usability directed at a business user and not a technical user.
Jakes concept of ‘addons’ and ‘checkboxes to turn things on’ is great, but eventually it will get to the point when some systems are far too robust to have a simple ‘install’ button. Not to mention that some software and web applications out there have a massive install and customization time ( 6 months +)
On the flip site, a complex website isn’t too hard to stay away from, but this almost ALWAYS changes with size. When you start developing a site that could be 1000 - 5000 pages or more, things can always get complicated. Not just for the developer but the user as well. You could easily have a navigation structure that’s six, eight, or 10 levels deep. Breadcrumbs and listed navigation? Or subdomain sites entirely? It’s easy to confuse the user, and the designers best efforts to keep it simple are most likely going to end up being moot.
As technology in the web industry gets more attention (Stuff like Ajax and Livesearch and so forth), clients want those features. They want dynamic sorting of tabular data and complex layouts, and it’s harder to educate the client.
-Ryan
Posted on March 3, 2005 05:56 AM | #
32. jimmyd said:
I still use Homesite 5.2 for PC. I just can’t bring myself to use anything else.
Simplification seems to be the next big thing…I keep hearing from friends who are liquidating many of their toys, dvds, books, etc in favor of the simpler lifestyle. With less to manage, there’s more time for social activity and more focused learning.
Didn’t Cliff Stoll predict this movement a long time ago?
Posted on March 3, 2005 07:19 AM | #
33. Keith said:
Ryan – In some ways I recognize and agree with what you are saying. Clients will want these features and in some rare cases these features will make sense.
You bring up CMS and Stellant and Vingette. You know, it’s very few that actually need all of those features. Hey, if they can sell them, more power to them, but if I were a consultant with one of many companies lookin at those systems I may advise against it.
I think it’s a rare sitution that warrants a CMS of that caliber. But then again I have a beef against tradtional CMS implementation.
I know from personal experience how much of a waste a CMS can be to an organization. Usually it’s because of three things.
1) Overcomplication – they are too hard for the people who are supposed to use them to use.
2) Too many features – even on a small CMS the feature bloat can be overwhelming.
3) Lack of flexiblity – they have features that sound good, but they’re not flexible for proper implementation.
More features rarely mean “better” when it comes down to it. At least not that I’ve seen. But there are exceptions and they might be just what your talking about. For some people, in some situations you’ll want a feature rich application.
If you focus and don’t design for everyone you’ll be fine. The problem, when it comes to Web interfaces, is that you often have to accomodate a large, non-expert audience.
This can apply to software too.
It’s like my Office 2003 example. I was in a training class where we got to learn about all the great new XML-based features that’ll help us with collaboration, etc.
Everyone walked out of that sesssion thinking, “I’ll never use any of that stuff. I just want to check my e-mail.”
There are times when you need to add features, etc. and, again if you can sell this stuff, what can I say, go for it.
However, the huge caveat I’ll add to that, and it applies to design, development and more, is that what a client really needs and what they think they need are often two different things.
Usually they want more than they actually need. At some point, if we really have their best interests at heart, we need to figure out how to talk them out of “features” that they don’t need and that might become a burden to their users.
Posted on March 3, 2005 09:25 AM | #
34. John Zeratsky said:
Wow, nicely said.
Posted on March 3, 2005 10:19 AM | #
35. Ryan Brooks said:
Keith,
As always, I commend your thought-out response.
However, I assure you, I’m not comparing Stellant nor Vingette to being great examples of a ‘solid’ application. Both those CMS systems, though large and robust, are built to manage sites to the scale of IBM, or Microsoft. They’d integrate into outlook, or manage permissions across a network, or even accept satellite data.
I agree with you, not everyone will ever, ever, ever need all the features that the giants boast. The upside is that they are there, for those that need to use them. However, when it comes to my own CMS system, those are the giants that we’re striving to compete with. We’re not even there. The next version will most likely be in competition with CrownPeak, or SiteRefresh.
For most of our clients, our current system works nicely, and it’s quite small. ( I wrote about version one here: ryanbrooks.net/?p=29) It doesn’t manage digital assets in the way a stellant system does, nor does it provide root control to the server to add DNS records. But it manages permissions, and is a good start.
Now, I’m hoping that you didn’t misunderstand me… I am not saying that more features is better; All I am saying is that features are good so long as they’re properly implimented. In fact, the hard part is keeping things simple while at the same time adding more functionality. In my mind, an implemented solution for a cms system, or a email system, or even a house, should address the needs of the client; that’s it. When you buy a house that’s already built, you go in, you look at it. You might need to paint it or rip out a wall. The same thing could be said for a CMS system, or Outlook 2003.
But in my mind, what you’re saying is full of “What if” - what if we don’t need two bathrooms? Or what if we don’t need task management in outlook? What if we need to search our files over a network?
On the flip side, what I’m saying is: What if we do?
Designing for everyone, be it a house or software is next to impossible. No matter what you’re doing, designing an interface, or building a house, or building an enterprise class CMS, it’s almost always easier to add things in then take them out (look at software, you have all manner of things to worry about, like backwards compatibility).
On a final note, I want to touch on this:
However, the huge caveat I’ll add to that, and it applies to design, development and more, is that what a client really needs and what they think they need are often two different things.
Thank you for that, it is an entirely different approach that I’ll have to give some thought to.
-Ryan
Posted on March 3, 2005 03:54 PM | #
36. nick said:
My only contribution will be a quote, which I think summarizes everything that was said and meant by the participants in the discussion so far. I think it is just enough.
Posted on March 4, 2005 11:19 AM | #
37. Amos said:
I totally agree. One of my favorite books on web usability is Steve Krug’s ‘Don’t make me think’ (www.sensible.com). This book has some good points on keeping things simple.
I wasn’t familiar with Basecamp and will give it a try. Anything must be better than MS Project.
Do you or anyone know of a good and simple online tool for team collaboration? I work with remote team members and would love to simplify how we collaborate…This might be a good discussion topic.
Posted on March 4, 2005 03:55 PM | #
38. Jennifer said:
Nice set of rules. Remember, your site visitors will determine whether or not your features are “useful” regardless if they are usable.
Design for them what they will use to facilitate accomplishing their goals on your website, not features that you think your visitors will want to use because they’re there.
In regards to what Ryan said about how features sell products - sure, if they’re useful and they help get the job done without issue. Keep in mind that any product which is feature driven is either a toy, or an expensive failure.
Posted on March 11, 2005 04:31 PM | #
39. Miguel Alvarez said:
Hi.
I think that the “simplicity” balance is broken
every time a ground-breaking technology comes
along (i.e. flash sites or even RSS). It then
takes time for webmasters and developers to
“get over it” and start caring about what really
matters: getting the message across clearly.
The problem is that new technologies are getting
out faster and faster and the webmasters are
just jumping from one big technological advance
to the next…
Do you think there’s a way to really stop this
and make the web the nice, simple place it used
to be a few years ago?
All the best!
Miguel Alvarez
http://www.webdesignersuccess.com/
Posted on March 21, 2005 05:10 PM | #
Comments are now closed