By Matthew Hile on 1/17/2007 1:17 PM

Support groups on the web are a huge phenomena. However, what is the best way to organize the user experience. For asynchronous groups there are two common options a non-threaded time oriented bulletin board were users' entries get included in a single list and a threaded discussion form where a root entry (any number of root entries) is made that people then comment upon.

The non-threaded form seems like it would be the easiest for “mom and pop” to understand but for groups of any size it would be a difficult cognitive task to keep up with the various threads of conversations. Sort of like arranging all of the individual statements at a party into chronological order then trying to figure out how they were grouped with one another. For small groups this would be possible because I think they would be more likely to have a single conversation. You mentioned that you keep groups to around 50 participants. I would guess that many of those have a very few individuals who participate actively and that in effect they have become a small group.

The threaded discussions make sense to me and do reduce the cognitive load of keeping conversations coherent. But I have never really liked the narrow view you get of them when they are collapsed into the original post. While I have seen different ways to represent the richness of the comment space none has been particularly satisfying to my tastes. You exchange load problem of the first option with a new one, the need for the user to search for and identify which conversations of interest. The threaded interface seems like a difficult navigational hurdle for non geek users.

I have been considering merging the two approaches perhaps giving the user a moment by moment choice as to the way to look at it or by offering two views of the same data. That would be easy enough to implement but the trick would be to devise a model/approach that would make sense to mom and pop.

By Matthew Hile on 12/8/2006 8:22 AM

Jon Udell is joining Microsoft.

Why should I/we care? I started working with computers over two decades ago and at the time knew very little about them. I was a clinical psychologist and had only worked with statistical programs (using a mainframe and punch cards). No computer classes in college or graduate school. No personal computer (this was after all at the very beginning of the PC). So how did I learn? I started reading Byte Magazine (as well of lots of other stuff). Each month I would read the articles in Byte, which I often did not understand, and slowly I began to understand to terms, concepts, and approaches that under gird computers.

Jon wrote for Byte and I have been reading his articles ever since. Over the years he has taught me a great deal and has clearly and powerfully presented new ideas in a way that helped me see how they could be used to support my users. When Jon makes a recommendation I listen. For example he recently mentioned the podcasts from the Long Now Foundation, Seminars On Long-term Thinking, I started listening and have become hooked. They are terrific (more about them in some other posts).

At any rate, I have been reading and trusting Jon for years and if he is as excited as his self-interview suggests, then I am really looking forward to what he is about to do.

Congratulations Jon!

By Matthew Hile on 11/17/2006 6:50 AM

Clay Shirky's talk, Making Digital Durable: What Time Does to Categories (Audio/Discussion), is a part of the fascinating Seminars About Long-Term Thinking (SALT) hosted by Stewart Brand of the Long Now Foundation whose fame began in the 60s with the creation of the Whole Earth Catalog which Steve Jobs, the co-founder of Apple Computer, called the conceptual forerunner of the web search engine.

Shirky talked about the difficulty of preserving our information over the long term. Much of his focus was on how one could find information that had been stored. He described the difficulties of structured catalogs. Comparing them to lost languages for which we have no Rosetta stone. As a solution he suggested "degeneracy" that is classifitory redundancy so that the same thing is said in many different ways. Degeneracy is embodied in the "new" social tagging systems such as flicker, and del.icio.us. "The only group that can categorize everything is everybody" and by enabling this sort of categorization we have an ever emergent understanding of how people think about these items and can therefore find they. Perhaps more importantly, these systems also alow us to understand how our understanding changes over time.

The notions of tagging have long been used to help locate information. In searching the scientific literature these are called "key terms" which are associated, by the author or others, with a particular article. When searching for some information I would rarely choose to start with such classifications as the Dewey Decimal system. Rather I enter some key terms that would seem to me to fit the concept I want to understand and use what I find there to expand my search to gather in the web of knowledge that surrounds it. What tags adds to this information is that other readers add sharable terms which will expand and enrich that web of understanding making our information links more fruitful.

Shirky's talk is a great listen and a good jumping off point to expand your own thinking about how to find what you want and how to understand the meaning of what you find.

By Matthew Hile on 11/1/2006 8:28 AM

In the most recent version of DNN 3.3.5 and 4.3.5 clicking a file link will fail in Firefox if the file name has any spaces in it. The work around is to eliminated those spaces until a fix is provided.

By Matthew Hile on 11/1/2006 8:13 AM

Error: You do not have permissions to view the requested file

In earlier versions of DNN uploading a file would allow one to easily make it available to portal users. With the more recent version 3.3.5 (and I assume the 4.3.5 version as well) they have added an option to store files in the file system (secure and unsecured) or in the database (secure).

By default the directory with the previously uploaded files which was formerly open is now marked as secure. This breaks the accessibility that we had come to expect. While I am all for security breaking this is not a great choice.

To change the behavior back to what you expect open the file manager go the the Security Settings at the bottom of the page and allow All Users to view these files.

One important note about the file system secured files. These files are only secure when DNN accesses them. The user can enter the full URL directly into the browser and access the, evidently not so secure, files.

By Matthew Hile on 10/19/2006 5:04 AM

In its default installation DNN (up to and including 3.2.2) breaks the expected user experience with regards to links. By default the style sheet sets links text-decoration: none;. What this means is that the user has only a slight shift in color to identify a link. Moreover, unlike the browser's typical default, visited links do not look any different than unvisited links. I for one rely on being easily able to identify a link because it is underlined and whether or not I have visited it based on its color. Jakob Nielsen's Alertbox article entitled Guidelines for Visualizing Links provides great information on why this is important.

So how do you fix the problem? There are two options.

• For a single site go to the Style Sheet Editor which you will find on the Admin > Site Settings page and add "text-decoration: underline;" (without the quotes) to the a:link and a:visited styles.
• If you want to fix all of the sites change the a:link entries in the ..\Portals\_default\default.css file.
  

Making these changes helps your users better understand and use your site.

UPDATE: This code only works in Firefix and not IE (tested in version 7). After much searching, experimentation, and trial and error work I found that dropping the trailing semicolon allowed this approach to work in both browsers.

By Matthew Hile on 9/21/2006 1:08 PM

I like Joel Spolsky's take on the software world. He also is interested in good user design. In a recent review of Sprint's new Power Vision Network phone he has an hilarious (or sad if you were the designer of this disaster) review of the free phone they sent him. The quote I liked the best

13. Remove the battery and put it in again. That should get you back to the main menu.

To get the full effect and read about a really poor user experience read his full post.

 

By Matthew Hile on 9/19/2006 8:53 AM

OK so I get a technical magazine each month in the mail. There is an article which attracts my interest and a reference to a “InstantDoc” id so I can quickly get the supporting materials and the code from the web. Cool!

Step 1 - So I log onto the web site. First it tells me that I need to be registered. So I search my records to see if I have already registered and find nothing. Try to create a new registered user and it appears that I am already registered (OK so this was my bad for not recording my information in my secure ID/PW file.) So I try to get an emailed reminder and find that my registration has my old email address that no longer works. So an email message to the publisher asking them to change my email address. I never got a reply but checked again in a few days and could then send my current email address the reminder.

Step 2 – So I log onto the web site with my ID/PW and request the “InstantDoc” and up it pops. Nope. Now it tells me that I must associate my user name with my customer number. Where is my customer number? On the package that encased the publication which is, naturally, long in the trash. That seems like a reasonable place to put this number so I wait for the next issue to arrive…

Step 3 – My next monthly issue arrives and I save my customer label and head back to the web site. I enter my ID/PW and try to associate my customer number so that I can get my subscriber access to the “InstantDoc.” In response I get a message telling me that I need to contact customer support and get them to associate my ID and customer number. Off goes an email and the next day I get a friendly response telling me that the two pieces of information have been associated and am good to go.

Step 4 – Imagine my excitement. I log on to the site with my ID/PW, enter the magic “InstantDoc” number and…well I get to a page that has an abstract about the article but where is the additional information? I click the “subscriber only” link – nope that just give me the opportunity to subscribe to the publication (there is also another “Subscribe to …” link just in case I missed the first opportunity). How about the “Main article” link? After all the blurb with the “InstantDoc” did promise “an expanded version,” but no the same information on the web as in the print version. But wait there is a sidebar link surely that will give me the additional information? Again disappointment, it took me back to the initial abstract page.

I tried other links. I could get a listing of everything that author had written, a listing of similar articles, order reprints in various formats. I could Digg the article, add it to Del.icio.us, email it, print it, add a readers’ comment…lots of stuff but not what I was looking for - the expanded article and a download of the sample code.

Step 5 - Wait a minute. For safe browsing I use the Firefox extension NoScript I try turning scripting on and voila … another advertisement with the opportunity to subscribe to the print version AND for an additional $5.95 a month I can get a “Online Pass.”

Perhaps the information I wanted is there somewhere. Perhaps it isn’t. Perhaps if I were to spend the additional $s I could get access to the additional information. Perhaps I couldn’t. Perhaps there is something in the fine print in the publication or on the web site that would explain all of this. Or perhaps there isn’t. At this point I have been trying for a month to get the advertised “sample scripts” and it is just not worth that sort of time.

I get a lot of IT related publications at the Institute and this is not the first time that one of them has become more trouble than it was worth. Print publications are struggling trying to figure out how to survive in the era of free and bountiful internet based resources. I hope that they succeed because I actually do like and learn by reading what they publish. But in their struggles they are becoming more and more irritating making me much less likely to subscribe. So in their attempts to make their way in this new world they are driving off those of us who are supporting them.

By Matthew Hile on 7/28/2006 9:25 AM

According to a 2005 Yahoo survey it may not make any difference if people know the answer to this question.

27% of online users consume third-party content on these pages without knowing that RSS is the enabling technology.

So one clear area of focus is how provide folks with the cues that they can use to add this active content to their web browsing experience. Having it built into the browser, as is true for the Live Bookmarks in the current version of Firefox and will be true for the new version of Internet Explorer, will be a great jump up. Having it built into one's email client, as will be found in the new Outlook 2007 will also help.

Another help would be to change the term we use from RSS to feed or subscription. These may help "real users" feel more comfortable with the process and increase their use.

By Matthew Hile on 7/27/2006 6:29 AM

To continue my exploration for a good "real user" description of RSS I turned to Wkiipedia. Their definition includes these statements:

...the news media is utilizing RSS by bypassing traditional news sources. Consumers and journalists are now able to have news constantly fed to them instead of searching for it.

A program known as a feed reader or aggregator can check a list of feeds on behalf of a user and display any updated articles that it finds. ...

These highlight a very important advantage of RSS over the more traditional emailed newsletters. An RSS subscription does not require you to give out your email address. Thus you will get no spam.