Kevin Cheng  

It Bugs Me, It Bugs Me Not

July 15th, 2005 by Kevin Cheng :: see related comic

There are weeks when I write about a topic that I simply feel should not need to be written. In those weeks, I might decry some user experience design faux pas which seems so blatant that I feel like as big an idiot for even mentioning it. It’s like saying, “it’s bad for business to lock your doors such that people can’t get in your store.” Doesn’t need to be said, right? Anyone writing about that would get laughed out the door because … who does that?

Yet here I am, talking about people locking the store front doors on their web sites.

##Designing With Barriers
The New York Times, LA Times, Atlanta Journal, and more online newspapers than I could possibly enumerate, are the most cited examples of designing with barriers. Each of these sites requires visitors to provide information before they’re given the ability to read the majority of the site’s content.

Why? Because the sites feel they have a right to collect this information to help them discern what areas can be monetized. They’re providing free content, after all and that’s expensive! Didn’t you know? Many other companies ask for far more intrusive information than newspapers do so you should feel lucky. Look at cell phone services, they ask for outrageous information! Who cares that cell phone subscribers are _subscribers_ and essentially have a credit account with the company, the newspaper companies feel it’s apples to apples!

I’m not unrealistic about business. I don’t believe that the internet is a revolution that makes everything and anything free to all online. Businesses providing valuable content and services have every right to make money. They’re just going about it the wrong way - and not just newspaper services.

For example, how many online photo album sites out there require you to register just to _look_ at your friend’s wedding pictures?

##Removing the Barriers
In the traditional software land, downloadable software often comes in “Trial” versions. I’ve used a few such programs before, most notably [Trillian][1]. I was able to use it exactly as advertised - to instant message across all four IM platforms. Not long after, I purchased the full version (which was worth every penny). The majority of utility applications I can think of come in similar packages: [WinAmp][5], [AppRocket][6], [TopDesk][7] are all programs I’ve tried or still use, some of which I purchased, some I found to be not as useful and uninstalled, others I was happy with the trial version.

The question is, how did we lose this model on web sites? Let’s look at the two examples I gave. First, the photo sharing sites. Certainly, the easiest improvement is to not require visitors to register just to view pictures but can we take it one step further?

What if, as a new visitor, you were able to actually upload photos, organize them into albums, tag them, label them, etc - all without registration? All of a sudden, if I was the new visitor, I’ll have spent half an hour playing with the web site’s capabilities _without logging in or registering_. Then, if I actually wanted to _keep_ this information, __that’s__ when I have to register. By then, not only have I had enough experience to make an informed decision, I’m invested in the site already - invested through time.

As for the newspaper sites, perhaps they should look at [Weblogs, Inc.][2] and [Gawker Media][3] sites. All the content is freely accessible, regularly updated, and some of the sites are incredibly popular and yes, they generate revenue. Newspaper sites claim that location based advertising just doesn’t work anymore and this may be true to an extent but they fail to see that the context of the content itself is where they should advertise from. The one piece of irrefutable data everyone has is, “I know what content is on this particular page, so if you’re looking at it, you might be interested in these things.” Giving visitors a brick wall every time they try to visit is not the answer.

##If You Must …
Sometimes, requiring some sort of account really is unavoidable. If it must be done, make it as painless as possible. Creating a 3 step process involving a long form to fill out and two e-mail confirmation links is not a painless process.

A painless process is one that mimics the simplicity of logging in: What’s your e-mail address? What password do you want? Thank you, here’s the content. It’s that easy.

(While I’m on the subject, can we stop requiring users to choose usernames in addition to e-mail addresses? It’s like trying to remember two passwords.)

By all means, if you want more demographic information, ask for it - but do it _later_. Has no designer ever heard the term “A bird in hand is better than two in the bush”? Get them registered, have them taking advantage of your content and services, and you can ask them whenever you want. Ask too much to start, and you get nothing but some more data on your access logs to tell you yet another person exited your site at the registration page.

##Combating Useless Logins
Many people already know of and use this service already but I felt it was worth mentioning as a parting gift for those who don’t know. If you do encounter a useless and annoying login/registration screen, you might try [BugMeNot][4].

[1]: “Trillian Instant Messenger”
[2]: “Weblogs, Inc.”
[3]: “Gawker Media”
[4]: “BugMeNot!”
[5]: “WinAmp”
[6]: “AppRocket”
[7]: “TopDesk”

21 Responses to “It Bugs Me, It Bugs Me Not”
bugmenot wrote:

Funny how you require both Name and Mail to post replies to Ok/Cancel…

Kevin Cheng wrote:

I assume you mean “why can’t we just ask for one or the other”. Well …

- There’s a lot of comment spam and/or trolling in blog or blog-like sites so just asking for name seems to invite low quality comments (we feel through this, our reader comments have been of high quality and really more valuable than the article itself).
- We don’t ever show nor link the email address so only asking for email isn’t an option.
- If you want, I can ask you for your name and address before you can read the article, which is really what I’m talking about. =)

james wrote:

Good article. You make an excellent point about making the system for registering, when it is unavoidable, as painless as possible. Too bad more sites don’t follow that route. One thing though that I find more annoying is the collection of anonymous demographics. Check out and surf around for a while. Eventually you will be asked for gender, date of birth and zip code; ocassionally they have asked for more anonymous data. This is really annoying since they’re interrupting the use of the site just to collect the most mundane of data.

Erik S wrote:

Besides what was just said above, there’s a big difference between registering for a site and posting comments on a blog. Kevin’s main point was forcing the user to remember two pieces of information (user/pass) as opposed to one (pass).

It doesn’t matter what you put in the Name field on this or most other blogs, and many times it automatically populates the fields through cookies anyway.

Chris McEvoy wrote:

While we are talking about “people locking the store front doors on their web sites”, when are you guys going to fix your store link which just seems to take people to the home page.

Chris McEvoy wrote:

And while I am on the subject of dead-end navigation. Did you intend to remove the “View Related Comic” that used to appear on the articles?

If I deep link to an article on OK/Cancel I used to be able to click on the “Related Comic” link so that I could see the strip that the article related to.

Is there a reason for this change?

Kevin Cheng wrote:

Whoops. Both of those are unintended, Chris. Store is now fixed (how embarassing =) ) and the Related Comic link not being there is a bug that we just haven’t gotten around to fixing.

Timmy wrote:

Sorry for staying off topic, but regarding bugs in the OK/Cancel store, I notice that all the links in the shirt descriptions are broken. (I mean the [rap song] link next to the first description, and the [this comic] and [follow up comic] links in the third.)

They take me to a page that says “The file you are looking for doesn’t seem to be here.” I *guess* clicking on a link is similar to “looking for something”…

Brian wrote:

I recently filled out a long form to create a user account on a forum of a newspaper site. After giving them 2 pages of information I was rewarded with an error telling me that Hotmail, Yahoo, Gmail, or other free email accounts are not accepted to create a user account.

I felt like this could have been stated up front instead of wasting my time and telling me after I had entered the information.

Paul Brown wrote:

Brian makes a good point; even if you must ask for information have the courtesy to make clear what you are asking for. How many times have you been presented with a registration form, filled it in and been told that your UK postcode is not an acceptable ZIP code, your pasword is too short/ long / doesn’t contain numbers etc, and one of the fields that you left blank is compulsory? Even if I am willing to invest the time in completing the form once a kickback like that will usually put me off unless I really must have the content.

Tom Woods wrote:

It’s always easy to spot the symptom, but much harder to spot the cause, and the cause is what you have to address to fix an HCI problem like this.

I think Kevin was headed in the right direction when he spoke of “monetizing” and providing content for free, but he missed the real problem, and it’s not just limited to newspapers. Many businesses make the mistake of trying to extend their brick-and-mortar business model onto the web instead of embracing a new (read: correct) business model appropriate for new kinds of clients.

Newspapers in the physical world pay for the content they povide by collecting a small fee for each paper, which typically covers the cost of distribution, and significant fees for advertising, which covers the cost of production. They get to charge more for the advertising if they have a higher circulation, and having a paying customer vs. a non-paying customer is solid proof to an advertiser that their message will actually reach the community. This is exactly the model extended to the web except without the distribution cost. They still sell advertising based on “circulation” but now they need you to register as proof to their advertisers that you are a reader. No doubt this is why Brian encountered the “no free email services” ban.

To fix this, you would have to re-engineer how the papers sell their advertising, and I bet that is a huge problem to tackle, requiring a change in the way the papers relate to (and contract with) their large ad customers.

Garbage wrote:

And the point of the email address is what exactly? You can enter any old rubbish that looks like an email address and it works. It really can’t be justified as a way to stop spammers. And even if you went off and did an MX lookup on the site I could still enter somethind valid and you wouldn’t know if it was right or not. Compulsory asking for email is just pointless.

Tom Woods wrote:

I’m not defending the practice, merely pointing out why it has happened and that it is not as simple as just not doing it anymore. The fact that an email address (and therefore whether the reader is “real”)can be faked is a primary criticism of the whole model.

It’s not pointless from the perspective of the newspaper, where they have to honor a contract that speaks to numbers of subscribers, or change around internal business process that links this to every report and business practice. It’s expensive to do that kind of internal shift and even if you can and want to pay for it, it takes time. They are just doing what is easy and expedient for them. Their advertisors are willing to accept the “just good enough” argument that a real person had to at least set up the address and therefore a real person sees the add.

I doubt this is a case of stupidity on the part of each paper. Eventually, one of them will solve the problem and all of them will follow, and then the registration thing will go away.

Eric wrote:

Regarding photo-sharing sites, I’ll agree that users want to get on with it and get their photos, et al, uploaded.

But it might be a good idea to tell them before they get started that, at the end prior to their “site” can be published for the world to see, they’ll have to eventually provide some info.

Not only is that just plain polite, but not doing so can impact a provider’s word-of-mouth reputation.

Rabbit wrote:

Wow. I’m thoroughly surprised that despite the bit in the article and at least one of the comments regarding photo sites, no one mentioned Flickr.

IMO Flickr gets it right on SO many accounts.

Ian Stalvies wrote:

One additional point - not only are most of these sites annoying visitors, but in most cases (in my opinion) they are wasting their time - as noted above, people throw in whatever they can, as quickly as they can, to get past the annoying page in question.

When confronted with a form demanding personal knowledge, I select randomly … which means to them, I’m a gardening assistant earning $500,000+ per year in an engineering firm. Not sure how they manage to use that data to “monetise” me ;o)

Pixeldrift wrote:

A great example of a good way to do things is the “Guest” feature on Ebay. It allows you to get to accomplish tasks without needing to log in, but then will transfer that information to your personalized records once you do. It clearly explains what you are able to do and what added benefits are available if you choose to log in.

Most commerce sites usually let you add items to a cart or begin training a recomendations engine to your tastes, and then require you to register in order to save that information and make purchases.

Users are much more likely to register once they have already invested in the site by creating something and have a basis from which to determine whether registration is worth the effort, a preview to assess value.

Jon wrote:

Another “bug” which has been driving me crazy on your site is the distinction between comments about the comic and comments about the article; why do they take us to two seperate comment areas? :(

Kevin Cheng wrote:

Jon, lots of reasons.

- they’re two separate things.
- the comic is sometimes directly correlated to the article but other times only tangentially so.
- if there’s two articles, it’s not clear where you leave a comment about the comic.
- some feel initimidated to leave comments on the article (where there’s usually fairly in dpeth discussion), the comic is a lower barrier place to leave simpler feedback like “wow, that guy is hairy”.

Jon wrote:

Have you user tested that? None of that was clear to me…

Comment Area Reasons wrote:

yeah right or a comment like
“wow that bigot : a person obstinately or intolerantly devoted to his or her own opinions and prejudices is really heavy into his ingrained assumptions about HCI- Human - Computer - Interaction!”

Leave a Reply

OK/Cancel is a comic strip collaboration co-written and co-illustrated by Kevin Cheng and Tom Chi. Our subject matter focuses on interfaces, good and bad and the people behind the industry of building interfaces - usability specialists, interaction designers, human-computer interaction (HCI) experts, industrial designers, etc. (Who Links Here) ?