Help and Program Flow
Lecture on Mar 20, 2008
- The Need for Web Design Standards. UseIt.com Jakob Nielsen.
- Writing for the Web Usablity.gov U.S. Department of Health & Human Services.
- Designing Apple Help Interface Design & Document Design Kevin Knabe.
Eric Cheung - Mar 19, 2008 02:49:48 pm
It seems like most of the guidelines are suggested in the 2nd (usability.gov) reading are easily done using wikis. Namely bulleted lists, headings, and numbered lists. I suppose that's one reason why we use the wiki so much in this class. The Apple help reading was helpful to a point, but I think it kind of glossed over a larger problem within in-product help. The main reason I search on the Internet rather than use in-product help is that I don't really know how to phrase my questions. It's easier for me to Google for some forum posts with the problem phrased in layman's terms rather than trying to guess how the problem is formulated within the given help files.
Diane Ko - Mar 19, 2008 03:51:14 pm
One of the things I thought of when reading The Need for Web Design Standards was the way that email systems work. For instance, in the calmail system and the squirrel mail system for instructional accounts, the buttons for delete, move, send, etc. are all in different places and often times when I'm using squirrel mail (because I don't use that as often) I have to search around for the appropriate button. The same is true whenever I go to different sites where I need to log in. It's becoming more standard now to have logins in the upper right hand corner, but there are still many sites that have log in areas in different places, which can get rather bothersome. So many other things have become standard for applications that it seems like web design should have some standards that we can use.
Benjamin Lau - Mar 19, 2008 08:08:48 pm
I found Nielsen's remarks on web design to be very helpful. I suppose I probably knew this at some subconscious level, but 1 min and 49 seconds really isn't very much time. It's like an essay, if you want to catch the reader, you'll have to do it immediately or risk losing him. Also the formal discussion about what kinds of headings are good (eg imperatives, action phrases, questions, etc) was helpful. I thought the rationale given for using HTML for the help system in Macs was kind of funny-- it basically came down to the designers being afraid of having to force application writers to create 2 separate help systems for Windows and Macs for the same application. So more of a programmer interface consideration than a user interface consideration.
JessicaFitzgerald - Mar 19, 2008 07:49:36 pm
I really liked the comment about when they were user testing the Mac operating system, people had a hard time finding the help button as a question mark on the right side of the main menu. As a result of this confusion, they moved the help to the left side where users had expected it to be and looked for it. I thought this was a good example of how user testing can improve a design. Even something as small as this may help many users and save them a good amount of time and frustration when they already need help. With respect to the idea of adding automation as much as possible, and that automating it for them once will help them learn how to do other tasks. I feel that if I was put into that situation, and I knew the help would automate a task for me, the next time I had to do that task or a similar task, I would just go to the help to have it automated for me instead of having to figure out how to do it myself from my experience from the previous time. I feel that automation is a good thing to have in a help menu, but to teach users how to perform the task effectively next time, they should also have instructions so that next time help is not needed, instead of just having the automation.
Maxwell Pretzlav - Mar 19, 2008 10:38:54 pm
I really liked how the usability.gov website was a self-contained example of almost everything it was explaining. It was a perfect example of how well its suggestions work -- it was easy to read, well organized, concise, to the point, and easy to skim. As someone with experience and opinions on web design I found the usability and Nielsen writings really insightful and well put-together, and applicable to many things, not just web design. I would like to read more about Nielsen's opinions on web-design standards and what he suggests we do about the 'confusion' cases. I found the apple help article interesting mainly in its demonstration of how user testing can help improve a design — and an interesting look at the Apple of the past; A friend of mine who works there says they no longer do any user testing. Maybe that's why Leopard's help viewer is awful. I also liked how the user action/task model they described when helping users perform sequential actions is very similar to the gulf of execution/evaluation model we have been studying.
Gerard Sunga - Mar 19, 2008 10:47:00 pm
This group of readings were particularly interesting. I found the first article particularly interesting (especially the description of the web as "an anthill built by ants on LSD"). Although I feel it's somewhat restrictive in terms of creativity for what can be good designs, it seems to be a good idea for the majority of websites (especially the mess known as Myspace and Xanga), providing a manageable interface for the majority of users. As for the second article, as the previous poster noted, it was good to see that they followed their own advice. As for the advice itself, it seems to be common sense (and sadly, many don't follow this paradigm) and it's nice to see decent tips in a single source. One thing that struck me in the Mac article was the similarity between advocated paradigms and their implementation in many of the help files in Windows applications (i.e. central access, easy accessibility of the help menu, etc.)
Gary Miguel - Mar 19, 2008 11:25:29 pm
Nielsen's idea of standards for web-site design is very interesting. I think it probably has a lot of merit: the goal of a website shouldn't be to reinvent the wheel in terms of the layout. If every e-commerce site used the same template, then people wouldn't have to figure out how to use each one separately, and they could get to buying the stuff more quickly. However, I don't think this will happen anytime in the near future. I think most companies think they can gain an advantage over competitors by having a unique design.
The article on Mac OS's help system showed a nice concrete example of the iterative design cycle in action, as well as giving a few pointers about what a PC OS's help system should and shouldn't do.
Katy Tsai - Mar 20, 2008 12:09:41 am
Nielsen points out that because the confusing design elements are often the bigger issues like navigation and search, usability is reduced. While I agree that these are important elements that should have some sort of standardization, I also feel that many sites use these elements to distinguish themselves from one site to the next. The web is convoluted with content now, that each site designer must use design elements and different layouts to create a unique experience. Thus I think some of the issues with inconsistency stem from a direct desire to be different and unique.
I do agree with Nielsen that standards and usability are important to user satisfaction, but I think there’s definitely a lot of compromise required surrounding that issue when it comes to designers trying to provide users with unique experiences. This is where compromise really comes into play. While users are stimulated by visual elements and graphics, I completely agree that frustration over functionalities is a huge reason behind visitors leaving the site. And while articles like “Writing for the Web” communicate key design standards that contribute to usability as well as visually, there still seems to be several limitations when it comes to a designer really taking ownership over a site.
Timothy Edgar - Mar 20, 2008 01:32:53 am
The first two readings seemed quite obvious. I suppose the point of the first article was to explain the necessity to follow conventions and standards with the second article giving standards and conventions. However they weren't standards but general high-level suggestions (use blank space!). It was interesting, but not too terrible useful. I did find Jakob's law interesting where the motivation of a design rules are quite apparent. There are so many sites out there and we don't necessarily only go to one site at a time. I suppose this fact does differentiate the web design from a desktop application design as a user is most likely going to spend less time on an individual website due to the speed at which users surf the internet. The last reading reiterated a lot of principles we've seen through the semester, although I did like their use of making things automatic. A small gain at the sacrifice of learning does make learning a new application into bit-size pieces.
Edward Chen - Mar 20, 2008 01:38:38 am
I thought the most interesting reading out of the three for this lecture was the "Writing for the Web" one from Usability.gov. Many of our past readings have talked about being clear or emphasizing a point without actually doing that in practice themselves. What I loved about that reading was they actually practiced the techniques that they talked on the page itself. Essentially, it was an interesting way to illustrate their point because you saw it in action and saw how effective it was. Going through a lot of points of that reading, the only site that clicked in my mind in satisfying the criteria presented was Wikipedia. It followed many of the points that the reading talked including linked table of contents at the top the page, extensive use of headers, chunks of text, tables, bulleted lists, numbered lists, and pictures to break the monotony of solid text.
I found the first reading not quite so applicable to current major web sites only because 4 years in internet time can end having huge differences. A lot of the confusion that the reading pointed have been largely become convention or even standard in sites. I know that I certainly expect the search bar to be close to the top right of most commerce sites, and the lack of such makes it completely difficult for me to find what I'm looking for.
Henry Su - Mar 20, 2008 02:45:26 am
This was actually the first time that web design standards came to my attention. Previously, I kind of assumed that every web designer has an incentive to make their site unique, but now, it seems pretty obvious that if the user has trouble navigating and using the website, he/she will probably try to refrain from using the site in the future, and go to a competitor's site.
The usability.gov article also presented some information that may seem contradictory at first. For example, writing teachers mostly emphasize paragraph structure, and grammatically correct, complete sentences. However, this is not necessarily the best way to set up a web page, since users don't read web pages like they do a book. Thus, short sentences and fragments in small paragraphs or lists are actually preferable.
The Apple Help document elaborated on one major point: guide users only where necessary. That is, if most users can figure out the workings of the application through exploration, don't spend resources writing up a help file for that. Instead, concentrate effort on writing useful troubleshooting tips, which can be useful when the application is not behaving like it should.
Gordon Mei - Mar 20, 2008 02:59:06 am
The usability.gov site outlines guidelines for avoiding many mistakes I see with documentation out there, particularly those of more technical depth. In an effort to be comprehensive, some page creators end up overlooking many of the techniques that make the page built for bite-sized consumption to prevent an overwhelming flood of text that drives many users away.
For example, perhaps a user comes along this site:
...looking for ways to set up webspace. Assuming they don't know to search for "public_html", they're going to have to go through some unnecessary intensive perusal to find what they seek. As Nielsen stated, visitors left websites after an average of 1:49 minutes, as most of the time, by Jakob's Law, is spent on other sites (bouncing from Google hit to Google hit until a quick digestible solution comes up). The casual user will be tempted to skip a site like this, lest he or she can't find better presented help elsewhere. There are some useful headings, although they're sized so poorly that they're buried among the content, and consequently there's no shift in the eye's focus there first. The headers could also be adapted to catch what users frequently query as their keywords. The haphazard sequence of information does not seem to list what users need most first (how could modem connections be one of the higher items), and finally, there's a lack of brevity in both text and tables/layout. Many of these instructions can be explained far more concisely, and the use of spacing and alignment make this page difficult to read naturally and quickly.
"Designing Apple Help" points out the observation that HTML was used to make it easier to author documentation. This was a wise move, as you want to have make this process quick and easy to write very presentable help with an easy-to-learn standard across many platforms. If you look at applications like Photoshop, operating system help like that in Windows, or HTML readme files with many small freeware out there, many developers have chosen to write their own help in this manner. Many websites today use MediaWiki to create easy-to-author help sections (either by fellow staffers, or by community users). The ease of rapidly learning the formatting facilitates improved documentation behavior.
Much of Nielsen's article comes as no surprise, as aside from the quantitative measures, we've qualitatively seen these patterns in what users expect, and what common design approaches work on the web. This is not to say that designers can't push and stretch beyond the same strict set of layouts, as long as the creative layouts they implement actually work. For instance, the placement of the shopping cart link at the upper right corner may be the convention, but I've seen an AJAX-powered site using a shopping cart "bar" on the bottom of the page, where you could drag product images into the bar (into your cart) to add it to that shopping cart. It felt intuitive, and worked at least as well as the conventional method.
Alex Choy - Mar 20, 2008 03:43:43 am
Nielsen says that websites should comply with design standards. However, sticking to a standard can be bad and can limit creativity. Many sites (when they were new) had a unique layout for that time and as long as the site has a good idea, it will probably succeed. Yahoo's main page used to be very plain and simple. Currently, it contains many different sections (news, search, top searches, marketplace, etc.), which I did not like at first, but have become accustomed to it. I agree with Eric that many of the suggestions in the second reading can be easily done using a wiki. Another way to write visually is to use tables in the webpage layout, bulleting/numbering the information inside of each table entry. Also, the article mentions using blank space effectively. I feel that when viewing some websites, information can be cluttered into too narrow a space and a good balance between content and whitespace is necessary. The Knabe reading on Apple Help was interesting. However, I tend to get help with an application by searching on the web. In many cases, searching through an application's help browser did not return what I was looking for. As such, while the design goals mentioned are all sound, I believe that the issue of help content should be addressed more.
Ilya Landa - Mar 20, 2008 03:57:19 am
Finally, a useful and hot at all boring reading; may be, the key is a lively HTML format instead of 20 pages of PDF. The description of standards and conventions for creating websites that are easily traversable and readable was interesting to read. However, it was not eye-opening. Even without it, I knew that sites should look similar to each other, that the site logo is in the upper left corner, and the search box is likely to be in the top right corner. The second reading wasn’t too revolutionary either. Even though, CS people rarely stoop so low as to refer to the in-product help, I do know what the “help” browser looks like, and how the topics are or should be arranged. I also agree with several people’s comments that looking up information on-line usually yields better results than looking through the help file; still, one definite exception exists to that rule – looking for an item with a known name. In that case, help systems immediately give the correct result. And, on a personal note – I absolutely love automated buttons in the help files. They often save me time by circumventing long walks through menu links.
Yunfei Zong - Mar 20, 2008 05:43:11 am
The first article makes some good points on the necessity of standards. However, the author goes too far in the amount of standardization he asks for. Nobody writes out a page from scratch unless they are doing some sort of supertrendy artsy website; they use some sort of reference, a frame layout, or take inspiration from another website. But it's not like they just steal Intel's website and change the text. The author seems to want all search boxes in one place, all the company logos in one place, etc, in essence having all websites be the same except for the content and color scheme. At what cost does creativity go out the window?
The second article was also quite refreshing in the numerous examples they give to emphasize their design instructions, as well for its brevity and lack of a 100mb download. However, I do hope we get some readings that go into discussion about the color layout aspect of web design. Personally, when building a page, I'll know exactly what I want in a webpage in terms of content, but will rarely have a clue when deciding a decent color scheme to back it up.
Jeremy Syn - Mar 20, 2008 06:09:19 am
I noticed that in the second reading it says that successful websites are those that have the information that users want and that they are able to find it quickly and easily. I felt that this relates to the Nielsons article in that if we use standards for our web designs, then a lot of this can be done easily. With standards, users will know exactly where things are placed and so can quickly and easily search for the piece of information that they are looking for. I also noted that in the second article, it says that we should write clearly. This is where good writing skills come in handy. We should only present the most important pieces of information to the users in a short, concise way so that they wouldn't be required to do too much reading.
Jeff Bowman - Mar 20, 2008 08:00:37 am
The Nielsen article was most engaging for me, partly because I disagreed with it most. While certain conventions have emerged from the use of the web across nearly 20 years, I feel like it's overly limiting to expect eighty percent of websites to follow standards. Nielsen's arguments are valid, for why standards have a place in website design; however, his focus on ultimate usability seems to stifle technologies and practices that make sense in context. At the end of it, it seems we would be left with designs all matching Google Sites or the average Drupal installation, with layouts like the Berkeley Visualization Lab and berkeley.edu having different (and evidently non-standard) layouts. Frankly, I want to know what he thinks about Google, because it seems to break conventions in very usable ways.
Granted, his article is great and useful; I have his Designing Web Usability book on my bookshelf. I just don't agree with his conclusion. Instead, practices such as the web-writing suggestions in the other two readings seem to be the biggest problem on the web today. And interestingly, the usability.gov web writing practices seem to echo much of the Wikipedia Style Guide, with which I was more familiar before. I'm sure this is part of the reason that Wikipedia is so popular.
Ultimately, it's true that information could be pulled from the web most easily if all websites were the same: They would all have the same interface, and the concept of "web usability" would be passé because, "well, there are the guidelines, make it look like that". And maybe someday X/HTML will be obsolete, and we will move all our data into the "Web 3.0" semantic data model where everything is connected through markup, and the web can all look and feel the same [exactly as the user wants it]. However, one site at a time, there may always be a more optimal way to display data, and Nielsen's well-known affinity for standard practices seem to stifle innovation.
How not to write/design for the web
I may disagree with Nielsen, but not in defense of these sites.
- HavenWorks.com - A 10-year-old news blog that takes itself seriously.
- HRODC.com - An international technical training school.
- FastLaner.com - Find three things wrong with this web design. Then find the other ninety.
Jonathan Wu Liu - Mar 20, 2008 09:43:20 am
I agree that some standards need to be followed such as information about the company, investor relations, etc. However, as for page design (eg. having the logo at top left, having a search bar), I think standards should not always be followed. Different designs can be equally user-friendly. Humans follow similar principles, such as the fact that the eye wanders to the top-left. So each site can determine what is most important and place that object in the top-left accordingly. It can also be argued that in our fast-paced society, we are enthralled with anything new. So, a different design that breaks standards can actually keep users there longer than 1 minute 49 seconds.
As for help, I think it is always needed. However, I hate the system of a centralized help structure. The primary reason is that it takes a long time to load the external program. Internal help (ie. not a centralized help) would seem to be faster, and also would lead to more credibility because the help seems actually native to the program.
Lita Cho - Mar 20, 2008 09:28:57 am
I didn't know that our government had a usability website. That is pretty awesome. I found "Writing for the Web" article to be more useful and wish that people would treat their writing on the web like an English paper. Sometimes, I see blogs that have just terrible writing. If the blog was their personal journal, then I could understand, but a lot of gamer blogs I've read seem like they didn't even proofread for grammatical errors.
I particularly like the idea with Nielsen's idea of standardizing the web in a usability stand point mainly because the first website I find, should give me the information I need. For example, I was searching for rules to a card game a few years back. When I went to the sites that came up on top, I saw this website: . Looking back, this was the information I needed and I didn't need to go elsewhere. However, I didn't like the interface so I didn't read it and found some other site. However, if the website was standardize, then I wouldn't need to have this problem where not only I have to find the right information, but one where the information is presented in a way that is aesthetically pleasing.
Robert Glickman - Mar 20, 2008 10:33:39 am
"An anthill built by ants on LSD." The man is a genius. Anyway, this exploration in usability was very interesting and very fun to read. It provided an informative and well-structured look at website usability. I especially enjoyed reading through the usability.gov page, which I found to be particularly good at explaining itself because it used the usability principles it preached and was effective in doing so. Also, the Apple Help was an interesting look at the origins of many help tools, but since I have never been a Mac user, I recognized many of the principles which one can find in Microsoft help, as well. The only thing I would have liked to know more about is usability as related to accessibility for people with disabilities (those who are color blind, deaf, blind, have tunnel vision, etc).
Randy Pang - Mar 20, 2008 10:14:18 am
Hooray! Concise articles with examples in non-pdf format!
Regarding the Nielson article, I feel that design standards are useful to keep in mind when thinking about usability, but ultimately I feel your biggest concern should be on making your site usable rather than standard. Standardization is one approach to usability, being more familiar with something is helpful, but we have innovated plenty of non-standard user interfaces that have been far more usable than the 'standard' counterparts (by Nielson's logic we might as well stayed with the command line because that was the standard before Xerox came out with the GUI). People Think Different for a reason.
And while I do agree that Nielson standardization mentality isn't neccesarily the healthiest, and certainly isn't the most innovative -- I actually agree with Nielson in the Flash article that Jeff Bowman linked to (which "stifles innovation"). Although I believe Flash has for better or worse become more useful on the web (think video players, mp3 streaming, games, Scribd iPaper, etc.), I felt the sites that Nielson was talking about seemed like those flash sites from the 90's that used Flash for the sake of using Flash and were definitely grave abuses of Flash (flash intros, gratuitous animation, etc.). I also see his beef with the lack of a mapping between browser actions (say your back button) and flash actions (in that, there is no mapping, the flash interface is completely detached). To some extent I also agree with his arguement against non-standard GUI controls (I've seen some ugly scroll bars and buttons), though not always (For example, you can have bigger or more visually descriptive buttons + more feedback). Outside of Nielson's points, Flash is also fairly slow (you might not notice it on your desktop, but what about a mobile? Though to be fair, it's probably not supported on your mobile anyways. Opera's rendering engine for Flash is also terrible, despite it being blazing fast in every other aspect.) and proprietary (seems bad for a lone company to dictate what is increasingly becoming part of the 'standard'). So no, Flash might not be bad 99% of the time, it might even be great and the perfect tool for the job, but please, please use it with care.
Harendra Guturu - Mar 20, 2008 11:15:30 am
I don't agree with the idea of web design standardization. Good design is such a subjective concept that should be left open to creative designers and their customers. If we were to force standards on every website, the internet would be so drab and boring. There are clearly instances in which some web designers don't know how to make a proper interface, but this is not a valid reason to penalize the designers that do come up with innovative interfaces that are a pleasure to view as well as to use. Also, I don't think people are that "dumb" that they need the same interface at every site. Internet usage has been growing at an exponential rate even without standards, so who is to say the user even needs or wants standardization. It should be up to the user to find a site that serves their needs appropriately and it should be up to the site to find a way to attract its users.
Chris Myers - Mar 20, 2008 11:20:59 am
"The conclusion was clear: the Web's strength comes from narrowly targeted sites that provide users with highly specialized information that they need or care about passionately."
I think it's interesting that Nielsen says to focus on repeat visitors, since the 'info dippers' aren't worth your time anyway. They will be using a search engine anyway to extract what they want and don't need an interface.
I hate corporate websites. They are nearly always guilty of excessive embellishment from some bullshit web design team that sold some template to marketing. I've notices the most usable sites are ones that sell a lot of products. Which make sense, as Nielsen points out - ecommerce is based on repeat business and a good, consistent interface is important for selling stuff otherwise the person will buy somewhere else.
Johnny Tran - Mar 20, 2008 11:35:09 am
I am going to admit that this is only tangentially related to the readings, but it does concern web usability.
In the Nielsen article, the first link in the text points to another of his articles called "When Search Engines Become Answer Engines." It goes into how the Internet has evolved from a collection of disparate, mostly isolated websites into a gigantic searchable web of pages, and many users, when confronted with a new task, would start by going to a search engine for the vast majority of the time.
This sounds like a big leap for usability, and is certainly a big boon for users. Nielsen does rightly point out that it puts an additional strain on websites, as they must cope with the additional load and remain usable even though most users will probably glance at a single page for several seconds. However, Nielsen then links to another of his articles titled "Search Engines as Leeches on the Web" where he describes search engines as a bad thing. He then details several rather questionable tips on how to deal with problem.
- Email newsletters were one solution. While they sound good in principle, and they are certainly a way to maintain a relationship with a user, does anyone really need more email? I would rather visit the site to get the information I want, instead of it being delivered to me on a weekly basis. A similar suggestion was RSS. I would still prefer the website to contain a superset of all the information available by the site, however.
- Another solution involved asking the user for what they want, and then telling them when they can have it. This implies that the information was moved off the page, and then provided only when the user requests it. This is a giant loss for usability.
- Yet another solution was adding a hardware component, like the iPod is with iTunes. Not only will this not work for all sites, but one of the greatest things about the web is the low startup cost for a website, and now that advantage has been taken to back alley and shot.
I do agree with Nielsen when he makes his point that the main idea is to make users come back. But really, can't we do that in a less-sneaky fashion, such as providing high-quality information and building user satisfaction?
P.S. Let's all contemplate what Nielsen has to say about preventing false clicks for pay-per-click advertisement: "Use negative keywords to keep your ad from being seen by users who are searching for free or cheap services. For example, add '-free, -cheap' to your keyword list. For most businesses, it's a good idea to add '-sex, -porn, -nude' as well." I never thought advertisement needed help being less visible.
Yang Wang - Mar 20, 2008 11:47:40 am
First of all, these were some good readings. I disagree with one who said the writer of web format standardize wants to make every web page has logo, search functions, and help functions etc. to be placed on the same place. I think he simply said some standard formats help people to read and navigate through the web page. I'd say though, there are some different types of web sites. It doesn't make much sense to see a wiki page has same format as a search engine or news websites. Most online wiki pages all follow a same format in sense of where search bar is where log in and comment option is. I liked the other two articles well, usability page gives some decent ideas about headlining and how to make web pages clear to users. About the apple help though, maybe it is just me, but I very rarely find any integrated "help" functions help me at all.
Michelle Au - Mar 20, 2008 12:06:06 pm
It's interesting to note that all three websites in the readings had very similar structures. The logos are in the upper left corner with links back to the home page, there are navigation and search tools in the upper right corner, text is broken up into smaller sections with headings, and bullets and pictures are used to help break up text. Despite these similarities in basic information structure, there are still enough differences to give each site its own look. While I see that sites need to distinguish themselves from one another, as a user, I would want all sites to have a similar basic structure in terms of navigation and search, the two features that are most important to me when I browse sites. It is frustrating when I browse a new site and have to spend time trying out different links and scroll through pages to figure out the navigation system. At least for me, it is more important that sites distinguish themselves by content rather than appearance.
Hsiu-Fan Wang - Mar 20, 2008 11:51:44 am
Nielson's website looks like a throwback to the 90's. If we are to say that usability and design go hand in hand, it seems that Nielson is a bit outdated.
I remember reading a rebuttal to that article awhile back where the author complained that by simply agreeing to some uniform set of standards it would retard the internet's growth and allow things to settle at a local maximum, whereas the unruly nature of development was akin to a parallel state space search that would eventually find some global maximum.
The usability.gov reading was good. And nice and skimmable. Many people have commented how the only times they remember seeing header hierarchies is on Wikipedia and on that site, and I wonder if it may perhaps be more an issue of them simply not reading sufficiently long pages online? I feel like most articles have some semblance of a hierarchy that is brought out. Also, by nature I would argue many pages do not really need this level of structure (blogs for example tend to be short bursts of content where a table of contents could end up spanning as much vertical space as the entry itself)
As for help... I wonder how applicable this is in the present day. I haven't seen a help program in ages, and the only reason I use CHM is because it serves as a fairly good packaging format for HTML documentation (the PHP documentation is available as CHM for example).
Jiahan Jiang - Mar 20, 2008 12:06:02 pm
I enjoyed the readings for the most part; some of the suggestions such as keeping texts small and using headers and sections. I didn't agree with the parts about standards, conventions, and confusions because just because an interface has certain elements that belong to "confusions" but more efficient and innovating, then there shouldn't be a reason to eliminate them. The user might take some time to get used to the new setup, but as long as the learning curve is relatively smooth, once they are used to something that is not conventional but better, it would become more efficient. (In short, sometimes it's okay to do things that are not how they've always been done if the changes are for the better. Who knows, they may become the new "standard" after a while.)
Mike Ross - Mar 20, 2008 11:53:24 am
The first article was amusing. I'm all for creating websites that adhere to a common format. I had a friend once who reversed the order of a navigation menu "just to be different from all the other websites." Alarm bells went off in my head, and I'm glad to see I'm not alone. In most cases, though, I'm not extremely picky, I consider it a win if a website even has a search bar. As long as it's somewhere on the main page, even if it's tucked away in some corner, finding it takse about a two seconds at max.
Regarding help pages, I'm all for html documentation. Once you've written your documents in html, you then have the bonus of users being able to navigate them with tools they're already familiar with (i.e. their browser), thus practically cutting your work in half. This way, it automatically scales to the user as well. Power users will have features like session saving and iterative search that you would otherwise have to implement yourself, and basic users won't have to deal with a bevy of new extra features in the way.
Bo Niu - Mar 20, 2008 01:05:01 pm
The most important idea that i see from today's reading is that following convention is always a good thing, it's actually easy to follow web design conventions since we are constantly using the internet as users. The elements that were discussed in the second readings basically listed all the important and obvious features of those conventions. We can easily see examples of these features being used in various websites such as the wiki that I'm using right now.
Brian Trong Tran - Mar 20, 2008 12:47:25 pm
I agree that there should totally be a standard in the way that web sites are designed. I agree with the article that many people that use web sites oftentimes find themselves lost and navigate away from the page even though the information that they are looking for is there. It would be really nice if all web pages featured a search bar and site map. Navigation would be much easier and users would not have to go through all these different web sites. It was also interesting to read the web design and see parallels with other aspects of user interface design for applications like use of white space.
Glen Wong - Mar 20, 2008 01:24:53 pm
I honestly cannot agree with Nielsen's view of things. Yes, I agree that there is a lot of variation on the web in terms of interfaces and layout. No, I don't think this is necessarily a bad thing. Neilsen's suggestions seem to go towards a web that is highly standardized and not creative. I definitely do think a lot of sites have serious usability issues and elements in the interface are so over embellished that they are hard to use, but that doesn't mean the solution is to throw out a set of strict rules on how things should be done. I personally take great satisfaction in using websites that are creative in their user interface but still highly usable. That aside, I felt like the second reading was good. I agreed with all of their points. There wasn't really anything that I felt was web specific, as these points are valid for other medium such as printed reports, presentations, etc. The last reading was interesting because I personally have found myself highly frustrated with help. In fact, I'm at the point where the help menu is the last resort, the first place I usually go to for help is Google search. The folk doing Windows could certainly learn a thing or two from the Apple help guys.
Adam Singer - Mar 20, 2008 01:37:47 pm
Upon reading the Nielsen Article, I found myself screaming "yes! yes! yes!" to myself in my mind. The web is a largely standards-abusing entity and we really do need to standardize many design elements. However; one of the culprits in restraining true web standards from coming to being are not the web designers themselves, but the browser manufacturers (yes, I'm looking at you, Microsoft). For years, Microsoft, Netscape, and other companies' attitudes have turned a blind eye toward web standards. As a result, web designers have been restrained from making their designs look more like familiar mediums (such as print or television). Luckily for users in 2008, most of these browser manufacturers have realized the need for web standards and have been hastily revising their browsers to conform to the standards put forth by the W3C ages ago. The "big three" browsers, as I call them, Internet Explorer, Firefox, and Safari, can now handily pass the Acid2 CSS test. That means that the current (or upcoming, for IE) versions of these browsers actually conform to web standards. This conformity will allow web designers to create more standardized designs without hurling themselves through a window.
In addition to this, many browsers now enforce some of the standards Nielsen pointed out in his findings. Changing the color of visited links is now a standard that all browsers conform to. Some web programming frameworks, such as Microsoft's ASP.NET, allow for one-line addition of things like horizontal breadcrumbs and automatic site mapping. Even things like navigation are becoming more and more standardized in development frameworks.
Now, I'm not saying that all of the onus is on browsers and frameworks, but non-web developers are often at an advantage these days since many operating systems come with development tools that force the designer to conform to user interaction standards. An excellent example of this is Apple's Interface Builder. There isn't any similar tool for web developers that can actually force one to conform to predetermined design standards. It will be interesting to see in the coming years if web designs will continue converging to a small set of standards, or if we will see a resurgence of ridiculously confusing designs that we saw in the late 90's.
Note: I would like to briefly respond to a peer's comments above, referencing Nielsen's article about flash as "stifling innovation". I'm sorry, but I'm sick of flash-based websites. Sure, the pictures are pretty, but they are a usability nightmare. I definitely concede that flash has a place and time (especially these days, for displaying video and audio content in a standard way), but most websites that use flash, and especially (gasp) make me CHOOSE between a flash and non-flash based version of the site, are so ridiculously over the top I simply navigate away.
As a web developer myself, I am always in favor of web standards and always cringe when a client asks me to do something in Flash. I'm usually able to talk them out of it.
Megan Marquardt - Mar 20, 2008 01:32:55 pm
I thought the article from usability.gov was pretty ironic, since it kept explaining the need for white space and tables and bullets, breaking up the text into chunks to make it easy to read, but I had some difficulty reading it since it was too broken up. There was only one sentence per heading in a large portion of the page, so I felt like it didn't flow at all. I'm sure they meant that in practice there should be more substantial portions under each headings when the website is very information oriented, but I thought it was overdone on that page itself.
In reference to the Nielson's article, I agreed that there should be strict rules regarding placement and conventions for websites to follow, and I don't think this limits creativity. Some earlier comments disagreed with Nielson's need for convention, but I fully agree with it. It doesn't limit creativity, there is still plenty of room for creativity within the confines of a structure. It's the same reason the majority of the public doesn't like and doesn't understand modern art, because there is a lack of convention and structure. Websites are seen by the entire world, and should be easily understood by the majority of the public, so introducing standards that the sites should follow is absolutely necessary for a website to be successful if it contains any sort of ocmplexity.
Pavel Borokhov - Mar 20, 2008 01:53:02 pm
These were interesting readings that seemed to have a bit more relevance than usual to the course. In large part, I agree with what Nielsen is saying, though I feel like in a lot of cases people would highly resist such moves due to potential "stifling" of "creativity" and freedom in creating website designs. Of course, I also feel like this only applies to websites of a purely informational nature, and for example, the guidelines for websites of photographers displaying their work would be very minimal and broad, since there is actually something to be said about the design having an actual function in the website itself. One thing I will mention is that the somewhat standard nature of Mac OS X application design - especially in the era of Mac OS 10.2 and 10.3 - is something that really sets Mac OS apart from Windows to me and makes it that much more a pleasure to use. It's nice that to change the preferences of every single application (that was written for Mac OS specifically, at least...) in the exact same place and don't have to go hunting for it.
The .gov reading was interesting (if not somewhat obvious), but I would take issue with some of their claims. For example, I don't think that having one-sentence paragraphs is a very good idea, because most paragraphs (both on the page, and elsewhere in written material) are not that short, and this leads to a feeling of imbalance on the page. It just feels jarring. This can be seen especially well on Wiki pages with proper headings, where a certain section will have only one paragraph, and that paragraph is only a line long. My gut reaction to seeing something like this is to either expand the paragraph so that it has more content, or to merge it with existing content, not to say "oh this looks great!" another interesting thing is that the page doesn't even validate against W3C standards and lacks a proper heading structure of its own, because it has no <h1> elements, which are kinda requisite for having <h2> elements (which are used, properly, for the headings on the page. though one has to wonder why the subheadings are not <h3>).
Lastly, it was interesting to see the ideas raised in the Apple Help System reading, and how some of the things mentioned in it were dropped in the OS 9 to OS X transition, to be brought back eventually in later releases. For example, they talk about how "Apple Guide was designed to address the problem of layer switching. When users access help in Apple Guide, they stay in the active application." Yet it was only in Leopard (10.5) that the Help Viewer ceased to be a "separate application" (at least to the user, it's still actually a separate app) and became a floating panel window attached to the application from which it was called. Similarly, the ability to connect to the internet for additional help items didn't appear until I believe 10.3 or 10.4 (though I can't find documentation of this on wiki right now...). It's good to see that old research is being "reimplemented" at least...
Kai Man Jim - Mar 20, 2008 02:06:49 pm
I totally agree that we should have websites in a standard form. It is because I sometimes have problem with browsing website to search for information when there are too many links and resources on the page. It just make me confuse when I first look at it. If websites are standardize based on one well organized website(namely the popular websites that people usually go for information), then we can go to any website and know where is the location for certain links for the information that we are looking for. This way can save us lots of time to search and browse. From that, it gives me an idea that maybe we should make our new features in the project more standardize with the existing cell phone feature because it is the way mostly people would do to experiment the new cell phone features. Don't need to be too fancy, but have to be user friendly enough.
Michael So - Mar 20, 2008 02:16:21 pm
The readings this week encourage following standards and conventions. The reason being for the benefit of the user. It makes sense because standards and conventions are recognized easily and the majority understand them. Something unconventional will confuse the user and impose a learning curve upon the user. This is not good for a practical level because the user will not be able to accomplish his or her goals or tasks efficiently. All the techniques and suggestions and guidelines marked out in the readings all contribute towards making a user-friendly interface where the user can efficiently and comfortably accomplish whatever the user wants to accomplish.
Andrew Wan - Mar 20, 2008 02:12:25 pm
The readings were useful. I find it interesting that sites have essentially established their own general layout conventions despite a total lack of regulation. As others have noted, the usability.gov site also serves as an excellent example of it's own message.
I think wiki and other trends in web markup all reflect a move towards more deliberate structure in website design. The widespread usage of bullet points, numbered lists, and (increasingly) additional markup languages like css provide cleaner, more accessible websites (when done properly). It seems that the choice/availability of interface elements has become mostly standard across the web, while their layout on-screen has changed.
Bruno Mehech - Mar 20, 2008 02:21:07 pm
The Writing for The Web article presented some interesting guidelines that I had never thought about and never noticed, though now that i think about it seem to be followed frequently and when they aren't followed I usually don't stay very long. Though the guidelines for keeping sections short and usin fragments doesn't always need to be followed in my opinion, one such example is wikipedia which can have very long sections, but if its information the reader is interested in they will stay and read it. The Apple Help article seemed to me to be a very good example of the design process in action. They went thruogh the cycle several times each time looking at user responses and fixing what they found wrong for the next release, eventually coming to a pretty good solution. I don't see how this article relates to the other two however.
Zhou Li - Mar 20, 2008 02:25:51 pm
I understand the ideas presented in the "Web Design Standards" reading, namely, having standardized design elements for all web sites on the Internet would make users' lives easier. Since they would know exactly what to expect from each web element therefore require no extra time to figure out how to navigate or use elements on a new web site. However, I think if all web sites on the Internet follow the same layouts or standards for each element they use, then web surfing might become much more boring, because users will be seeing the same elements over and over again. Having a distinct or innovative web interface design is what makes a new web site pops out from the rest of existing sites. If all new sites just accept and follow the same standards, then nobody would know if a different approach to a web element is more effective or easier to use. So in my opinion, web interface designs should still have their uniqueness, because as long as the interface is well designed and tested, users shouldn't have much trouble finding the information they need.
Joe Cancilla - Mar 20, 2008 02:30:16 pm
One of the greatest things I like about help pages nowadays, is the "how useful was this help" rating system that most sites incorporate. Although this isn't of direct use to the person trying to find help, it does allow that person to vent their frustrations when they find that the help that they found is insufficient. I recently had an issue with a mac machine that simply would not complete its update. This page on Automatic Software Update in Mac OS X seemed like it should help according to the title, but it didn't address the failure I was getting. I felt better after filling out the survey.
I don't believe that error messages were addressed in the Knabe article, but I think that if there is a wealth of information documented on a particular error, the error message should have a link to that documentation. I think it would be good if programmers got in the habit of including links to documentation in error messages.
Reid Hironaga - Mar 20, 2008 02:46:43 pm
The readings describing standards for web design basically boiled down to standardization efforts and guidelines for focusing reader attention to optimize usability. However, I don't see how they could possibly imagine the entire internet, with its diversity in goals and resources, would possibly come close to complying with any sort of standards with this level of strictness. The freedom of the internet is part of the appeal people see in it, and the standards would only limit user imagination. Any of the standards are debatable, and situations can be found where the particular approaches would be sub-optimal. For instance, short paragraphs and use of tables would be useful in a summarizing page, but for an in depth analysis of research or vast amounts of data, it may be better to allow for larger structures of text and graphical or interactive forms of data display.
Designing Apple help was interesting because it went over the many details that went into designing the help. Writing minimally seems to be a recurring theme throughout good design, and I suppose this is an important part of good communication that designers strive to attain. Visibility is also important, and this is similar to the concept of visibility and mapping defined by Don Norman.
Zhihui Zhang - Mar 20, 2008 02:52:44 pm
I agree that having standards for website design is a good idea. In the idea of recognition over recall, users expect to things to operate in certain ways. For example, websites that use fancy ajax to dynamically add or remove page elements will often confuse users because web pages are expected to be static. My question is what if you already have a site that breaks these conventions. Users are already accustomed to your existing interface (although it's bad) so how do you gradually update your interface to conform to these standards.
Daniel Markovich - Mar 20, 2008 02:43:41 pm
I found all the articles interesting and helpful, but I especially liked Writing for the Web . The section called "How do you write clearly" was particularly helpful and I feel that most of the concepts presented in the article can be applied to not just web sites but user interfaces as a whole, with a few tweaks of course. One of the big flaws in our project groups low-fi prototype had to do with bad messages and unclear text, and I believe that following set guidelines as these throughout the project will reduce or eliminate these problems.
Although I realize this is a User Interface course and not a specific focus on mobile interfaces, I would still find it helpful if some of our readings and concepts were focused on mobile devices in particular. Articles such as these but specifically for mobile interfaces would give us a great deal of insight into implementing our applications and even redesigning parts of them if needed. But as many other people said, I felt these articles were some of the more interesting and applicable to the course so far.
Jonathan Chow - Mar 20, 2008 03:06:19 pm
I found the Nielson article interesting in pointing out that almost 1/4 of the websites are confusing and don't follow a convention. In many ways, it seems almost like a style issue for website now that try to make themselves stand out from the others. Especially with all the fancy web 2.0 things that you can add to your website, many people seem to think that having flashy buttons and scrolling menus is better than just placing all the text in a visible area. The second article by the US Dept of Health and Human Services seems to be common sense. Those are the kinds of things that you are advised to do when writing articles for a newspaper/newsletter. Most internet websites are like that anyway (they are just sources for online news). So it makes sense to keep the wording concise and to add pictures whenever possible (although, I guess you have more liberty to do this online than in a newspaper). I was also interested in reading how Apple decided to design it's interfaces. In many of the other articles that we read, we saw why Apple had better interfaces, but the process of how they went about choosing those interfaces was not made clear.
Siyu Song - Mar 20, 2008 03:12:21 pm
My first reaction to The Need for Web Design Standards was initially very negative. I did not like the idea homogenizing the web to have a uniformed look. I do understand its more about usability than enforcing designs, and ultimately it is about avoiding confusion for the user. However I feel like designers who get negative feedback about their interfaces would be able to change their interfaces to cause less confusion for users, I think this is a more organic way to iteratively better a design as opposed to having standards for people to follow. In this way a designer could come up with useful novel features or designs that he/she would not have thought about if there was a standard to follow.
Raymond Planthold - Mar 20, 2008 02:59:19 pm
I liked what Nielsen had to say, though I wish he had provided some concrete examples. The remarks that sites' "sign-in process[es]" confused people were especially ambiguous. I also think that the focus on placement is somewhat less important than process; it doesn't really matter if breadcrumb navigation is the 3rd item from the top or the 5th, as long as it is used. In my experience, people who are less familiar with computers and the Internet tend to find things on pages by reading the whole thing left-to-right, top-to-bottom anyway, instead of skimming or jumping to likely locations. Putting things near the top does get them found faster, but exactly where they are at the top has less of an impact. Consistent behavior is a much more powerful metric, such as the example of uniformly labeling the "site map."
I found the AppleGuide help system quite interesting. I'm familiar with help systems that provide "do it for me" buttons, but the AppleGuide seems so much more useful in the long run, since it demonstrates how to access the desired functionality without opening Help next time. On the other hand, someone who didn't know how to open AppleCD but discovered the pictured entry in the Help system might go to the Help page every time they want to open AppleCD -- not because they can't find the normal way, but because they think it is the normal way. Again, my experience has shown me that novice users regard different ways of starting a program as separate programs. When told to open Firefox, they ask whether to use the desktop icon, QuickLaunch icon, or even the Start Menu item, and they really want you to tell them which -- if you say it doesn't matter, they stall.
Khoa Phung - Mar 20, 2008 03:16:04 pm
Roseanne Wincek - Mar 20, 2008 02:19:55 pm
I thought there were some gems in the Nielsen article. My favorite being, "Unfortunately, much of the Web is like an anthill built by ants on LSD". But I think that the most insightful thing he said, however, was that people spend most of their online time at other sites. Just wondering, is the author the Jakob of Jakob's Law? I think it's really important for designers to give users the opportunity to use and apply what they have used elsewhere on the designer's site. Obviously, standardization will improve users online experience and make them more productive.
As for help programs, with the advent of the internet, I never actually use them. I feel the hardest thing about them is generally trying to verbalize the problem that I'm having (large articulatory distance, anyone?). In that way, Google has become a better help source than any application. I love being able to use plain english and find forums with people asking the same question. However, it is extremely frustrating when you can find multiple people asking the same question you are but you can't find anyone actually answering the question.
Ravi Dharawat - Mar 20, 2008 03:10:18 pm
I agree with the first reading, but I think it practically impossible to implement without creating some new language of the web as a standard that enforces certain design principles, and I think it would be too expensive to do, in terms of both time in money, considering the payout is of questionable value. I wonder if the second article takes into consideration many of the things that can be done using AJAX, which certainly allow for a little more flexibility. This is regards to their remarks on organization. It is certainly very important to keep text chunks and sentences as simple as possible, and I believe the reason for this is that the web is an interface for information retrieval, and so should make the access of information as simple as possible. It seems to me that these are the very guidelines for technical writing. The one question I had about the third article is where they derived their design goals from? What were the criteria for these design goals?
Andry Jong - Mar 20, 2008 02:59:19 pm
From all the things in the three readings we have this week, I like the fact that Jakob Nielsen put the "Jakob's Law" in it. It is simple, and it is a common knowledge, but we all know that it is true that "[internet] users spend most of their time on other websites". This Law reminds me of a class that I took last semester in which the professor told us that one of many reasons why a search engine is more preferable than the other is not just based on the search queries it returns, but also the time it takes to return those queries. Most people will find search engine which takes more than 1 second to load to be "slow".
This is why, I think, speed is the key for making website design (or any design, maybe?); and standardized design help user to speed things up when they try to look for something in a website when they see it for the first time, since they already know where things should be. Also, organizing the site in some order might help user to speed search through the website, too. Joining these two ideas together, we'll be able to make a popular website (or maybe not).
Anyways, jumping to the third reading, I always find help files to be confusing. Even with Knabe's 7 design goals when designing Apple Help, I still think it is not much better. I don't know how to make help files to be more interesting or helpful, however. So, I guess Knabe's design goals are good enough.
Jeffrey Wang - Mar 20, 2008 03:20:31 pm
In the third reading, I understand that it is helpful to have all help content in one area. However, I also realized that the help content is never useful. I never find the topic that gives me trouble in the help section. Even if I do, the trouble shooting is not very helpful and only states the obvious. Lastly, I find it hard to navigate through the content. I prefer just looking up the solution in google.
Tam La - Mar 20, 2008 03:15:22 pm
It was interesting to read about the guideline for the wed design. I think it very useful to have a standard for web design. A lot time I find myself having trouble navigating some web page because the design is really bad. Every time you go to a new web page you have to figure where to click and learn a new interface set up. For example sometime I want to find store hours it take a lot of effort just to navigate to that information. Sometime it even hard to find the store locater on a web site. By having a standard on where those information is on a web page it would reduce the user searching time.
Scott Crawford - Mar 20, 2008 03:24:21 pm
Web design is something I've never gotten to do much of, but have always been interested in. One thing I'm not sure I saw in the reading was about how writing web code that works on all browsers is actually still an issue (sadly). It is a horrible UI flaw when something you expect to work, doesn't because of poorly written html. That also brings to my mind the design of the web design languages as a UI of interest, since for a long time it was very unstandardized and made writing good web sites quite painful. Now that's starting to get better, so I imagine that future web designs will reflect that and be more usable in turn. I would imagine that the failings of web pages comes more from older pages or were designed by people who first learned how to do so in an older version of the paradigm, and so the quality of web pages as a whole should improve asymptotically as the methods for designing pages improves.
Brian Taylor - Mar 20, 2008 03:23:51 pm
Thank you for the short readings after that lengthly midterm. Overall, I found the readings fairly interesting and insightful, and found it easy to relate to since nearly everyone peruses the internet everyday (and for hours on end everyday). The usability.gov page was relatively nice because, beyond the fact that it was concise, it gave specific instructions, for the most part, on what are good ways to organize text. As a user, I and pretty sure almost everyone prefers to peruse headings and short globs of text quickly than attempt to scan through a well organized essay to obtain information. Bulleted lists are an efficient man's best friend.
Benjamin Sussman - Mar 20, 2008 03:41:15 pm
I also appreciated the shorter readings. I glad we are leaning towards some more practical applications of our knowledge recently. In my opinion, mobile and web applications (and operating systems on a more meta-level) are going to be the most important developments in UI in the next decade or so. Desktop applications have been so beat to death that there seems to be only one track to success. The web is on its way to this place, but because of the broad range of users which you will be targeting on the web it becomes a more complicated problem.
Max Preston - 2008 03:04:39 pm
Having standard formats is certainly extremely important, not just for web browsing, but for computers in general. If different manufacturer's hardware didn't have the same standard formats for connecting with each other, either computers wouldn't work at all or every single part of the computer would have to be manufactured by the same company. Admittedly, this can still be a problem since, for example, CPU sockets vary by manufacturer, which means that you usually need to buy a new motherboard if you want a new CPU. File formats can also thought of in the same way. If, for example, you make a new image format that is far superior to anything that is currently being used, it would be totally useless if no one made software that follows the same format for viewing or editing it. If popular tools like Adobe Photoshop didn't support this format, then it probably wouldn't acquire mainstream popularity. Likewise, following common conventions should be considered when designing user interfaces.