Current Directions in HCI I
From CS160 User Interfaces Sp09
Lecture on Apr 29, 2009
- Jeffrey Nichols, Tessa Lau. "Mobilization by Demonstration: Using Traces to Re-author Existing Web Sites," In Proceedings of IUI'2008. Maspalomas, Gran Canaria, Spain. January 13-16. pp. 149-158. PDF
- Jeffrey S. Pierce, Jeffrey Nichols. "An Infrastructure for Building Applications That Span Multiple Personal Devices," in Proceedings of UIST'2008. Monterey, CA. October 19-22. pp. 101-110. PDF
Denise Ngai - Apr 28, 2009 11:57:42 am
Considering the overwhelmingly increasing number of mobile web devices, it is no question that some type of technology such as Highlight would emerge. I believe this is a genuinely good idea, especially after seeing first-hand how certain websites look on a mobile web device (iPhone & Blackberry). A lot of websites become hard to read or hard to navigate when viewed through a mobile web device. Highlight can possibly fix this problem, allowing users to access their favorite (or not favorite) websites comfortably and maneuver their way around the sites with ease.
Highlight's ability to lower the barrier to make web applications is indeed a benefit; however, I still see this technology being used mainly by those who are more computer/internet savvy. The customization Highlight offers is also a good idea, but again, I still see mainly computer savvy individuals taking advantage of this feature.
Overall, Highlight is a good idea, but I feel like it needs to be catered more to the regular individual, not those that are comfortable with using computers and software. Furthermore, from my experience, when viewing sites through a mobile web device, I am mostly bothered by a website that I rarely visit. For example, if I just happen to stumble upon a website I have never been to or if I type in a URL to a website I've never been to, because I am unfamiliar with the site, the ease of navigating the site on my mobile web device would be of greater concern than that of a site that I normally visit since I have already grown accustomed to it.
Saung Li - Apr 28, 2009 07:03:07 pm
The increase in popularity of mobile devices raises the importance of making web interfaces suitable for them, but most websites are designed for desktops. Highlight is an innovative way to address this problem. It allows users to make their own versions of the websites by having the program trace through their common tasks. Though this seems to work very well with tasks like shopping for an item on amazon.com, this might be difficult for complicated tasks. Websites that require a lot of tasks might require the user to trace a lot of steps in order to create a mobile version of the site, and this may be tedious if the user needs to do this for a lot of complicated sites. Allowing for the user to easily modify the mobile version of the site is essential (more interaction through the storyboard sounds like a great idea). Highlight sounds like it may be hard to use for users that are not too familiar with computers, though it is probably expected that most mobile phone users who want to surf the web are familiar with doing so on a computer desktop. Highlight does not address the problem of users just wanting to surf the web with their mobile devices, thus encountering sites that don't work well on them and decreasing their satisfaction. Users would still have to have used the sites first on their computer to understand how they work. The program, though, is still a cost-effective way for users to make task-centered sites fit with their mobile phones. The rise of mobile devices also calls for the need to be able to remotely control them. It would be useful, for example, to access files from my home computer on my mobile phone. This requires efficient communication among the devices, and the infrastructure introduced in the reading used the model of instant messaging, which works pretty well, and allows for services to send commands and information between devices. It would be nice if the interface was made so that those not too familiar with computers/programming could use it, though this topic itself is quite complicated. Allowing devices to communicate with each other can make way for "ubiquitous" computing mentioned in the previous readings, as less effort is required by the user to do tasks using the multiple devices. People may soon not realize how many devices they are using if they can effortlessly access content of all of them and issue commands to them from one device.
Jeffrey Patzer - Apr 28, 2009 11:50:49 pm
Mobilization by Demonstration: So what I really liked about the potentials for this application was its ability to tailor sites to a mobile interface. The problems it addresses are really smart. Trying to simplify things to work better on a smaller screen and a lower processor capability is awesome. It is safe to assume that users will begin to adopt more mobile 'smart' phones as time goes on and the ability to start making websites more mobile user friendly is a good direction. Especially since users are accustomed to a certain level of interactivity based off their experience with a typical computer, the ability to make sites more usable on the mobile level could have very big implications when convincing a user to buy a mobile device. Despite all the potential benefits that this platform seems to offer there are a few drawbacks. Besides the technological capabilities listed in the paper, I think the biggest problem is that people are lazy. More specifically if they have to spend painstaking time figuring out what might be a mobile set of actions they would want, they won't even bother despite the advantages they might yield in the end. I think a more automatic scheme is needed before people would truly adopt it.
Infrastructure for Extending…: What I found particularly interesting about this article was the thought that went into how to bring user's work to the cloud. I only have experience in so far as google docs with keeping things on the cloud and allowing them to be collaborative. This paper offered a really inventive way to allow your applications on your computer to interface the work you are doing to a central location that keeps other devices updated. This is an extremely powerful when you consider it. I know for a fact that I have to very much limit myself to a few devices due to the limitations suggested by this article. The ability to keep things in a central location and have those things updated in a very efficient manner would be very convenient. It sounds like the implementation is currently a little too difficult to use for normal users. Noting the times it took for tasks to be completed, it appeared like the learning curve was pretty steep. I think that decreasing the learning time will prove key to the quick implementation and adoption of use by people.
Ling Chen - Apr 29, 2009 12:52:19 am
As we become more and more mobile, I guess it doesn't really surprise me that something like the Highlight will start emerging and becoming more and more useful. Although, like someone mentioned earlier, this will probably be for people who are already familiar with the web and the technology. It won't be long until all the old web designers have to think about how to make their websites better fit for mobile devices. Maybe one day it will be that, when a user surf a site on a mobile device, the site will automatically know that, and adjust itself to the mobile version. (Just a thought.) This way, even people who might not be familiar with Highlight could surf the net with ease. However, before that happens, Highlight would be a good transition. And I imagine the CoScripter system with all these repository of traces will become very useful in future research and development of mobile sites.
The idea of Personal Information Environments is also very interesting. Being able to access information across multiple devices will no doubt bring a lot of conveniences for the users. I like how the infrastructure is targeted at helping the developers, by allowing them to provide functionality across devices without writing code to manage them. In addition, it's built on something we are already familiar with: the IM architectures. Everything is going mobile nowadays... I guess it's up to the designers and developers to make the experience of accessing information across different devices easy and seamless. I do sometimes worry though, about having everything stored in one big place...but maybe that's inevitable as we advance further and further.
Shoeb Omar - Apr 29, 2009 12:51:20 am
I think the Highlight idea is extremely interesting and an awesome idea to solve some of the problems with mobile site navigation and use. However, the main problem is that you need to sit through the somewhat tedious tasks of making the custom highlight site version of the site you want to navigate, and you have to do this on your desktop first. If you're navigating on your phone willy nilly and found something annoying to navigate, there's not much you can do there and then. I think a good next step would be to socialize the concept, let people make and post their own highlight apps for various websites. Then make a phone application that incorporates the highlight concept into their phone's browser. A little popup that came up on an annoyingly hard to navigate site that said "Want a different version" and popped up alternate highlight sites other people had created would be awesome.
Dwij Garg - Apr 29, 2009 01:05:13 am
I think that Highlight is a great idea for the direction that technology and web browsing is headed today. Since the invention of the Blackberry and more importantly the iPhone, most people have been using their smart phones to access some of their daily websites. Since many of these websites have a lot of CSS content, they take a long time to load on smart phones, whose browsers are not entirely capable of speeds available on the desktop. Thus, in this day and age, I think that it is important for at least websites with a large everyday audience to have mobile versions as well. The best part about Highlight is that it allows users to make any website mobile. It basically enables them to hold a significant piece of technology in their hands and use it to their advantage to solve everyday problems that would make life a lot easier. I feel that technologies such as Highlight is what we need today for progression.
Stephanie Shih - Apr 29, 2009 12:15:04 am
As society becomes increasingly mobile, the need for an app like Highlight becomes necessary. The only drawback is that everything must be organized on the desktop first, which means that any new sites found while browsing on ones mobile device is still as unwieldy to navigate as ever. It is useful, however, for the sites that the user will more commonly browse. Nevertheless, unless someone can come up with a way that reorganizes a website optimally and automatically, it is unlikely to be used by more than those who are already comfortable with such technology.
Sean Hansen - Apr 28, 2009 10:04:20 pm
The multidevice sharing development kit is a nice idea, but I wonder if it will soon become obsolete. I figure as portable electronics take on more functions and become less varied in function (because everything will do everything) that people will pare down their volume of devices and stick with one or two that handle all their needs. Then again, I also thought it would be PDA's and not cell phones that became the ubiquitous jack-of-all-trades, so there's my prognostication skills for you.
William Cho - Apr 28, 2009 07:06:29 pm
Some of the statistics presented in the articles were interesting to think about. On average, the people surveyed in "User Experiences Across Multiple Personal Devices" used 5+ computing devices daily. I think for a second and see that I use a cellphone, laptop, and ... that's about it really, daily, regularly. I suppose this article isn't as relevant to someone like me, who mainly uses his laptop for everything. Nevertheless, the authors have tackled a real and current problem, the problem of an application interacting across multiple devices smoothly, and their infrastructure seems solid. The article "Using Traces to Re-author Existing Web Sites" has the promising concept of users slimming down an application for use on mobile devices. I don't have much experience in using web applications on mobile devices, but even I can see how useful this can be for those people. However, I agree with some people above that average users will probably have a hard time using this to tailor web applications for themselves, with the way it is now... although the statistic that 30% of users are not satisfied with their mobile device web experience is ... compelling.
Cuong Ngo - Apr 29, 2009 02:02:54 am
I found both readings pretty interesting. The first article, titled "Mobilization by Demonstration: Using Traces to Re-author Existing Web Sites," discussed what Highlight is and how it works. Basically, it is a system that "enables end users to re-author mobile web applications from existing web sites simply by demonstrating how to complete their task in a desktop browser." More specifically, "Highlight Designer works by recording the actions a user takes in the browser, and converting these actions into a mobile application description." I find this tool pretty handy and useful. Users who are always on the go can now streamline their work with Highlight. They can perform their daily tasks without worrying about where they are. As its title suggests, the second article discussed the implementation as to how we can improve user experiences across personal devices. It's a pretty tedious job for mobile users to share information across their devices and keep them in sync, not to mention the process doesn't always go smoothly. That said, we need an infrastructure that's robust enough to get the job done with ease. The infrastructure described in the article does just that. It "provides simple mechanisms for applications to send information, events, or commands between a user’s devices."
Timofey Titov - Apr 29, 2009 01:58:19 am
Highlight is an interesting way of dealing with porting of applications. The number of platforms and personal devices are increasing, so there is a need to have functionality available on all of them. At the very least Highlight can be used when developers are working on a port. I'm sure that letting users produce actual applications in such a way will result in new and original ideas. One could use the applications created to analyze how and what users use the most.
The main problem behind coupling of multiple devices is synchronization. This is also a problem from the developers point of view. The proposed infrastructure is similar to "Software as a Service" model which makes life easy on users' and developers' side. The developer benefits from an ease of use of such infrastructure by not worrying about each hardware specifics. The user gets the flexibility of a network of personal devices that he or she can actually figure out how to use in conjunction.
Victor Lum - Apr 29, 2009 02:10:11 am
I don't really use the internet on a mobile device, so I can't say how difficult it is, but this Highlight idea sounds like it would make it a lot easier. Users can just go to a website and use it like they would normally, and Highlight keeps track of everything the user does, showing only what the users wants on their mobile device. This can tailor each user's version of a mobile site to his specific needs, only showing that user the parts of the site he actually interacts with.
Carolc - Apr 29, 2009 02:06:52 am
The first paper, Mobilizing by Interaction: Using Traces to Re-author Existing Web Sites, explains how using an interaction trace through an existing web site, Highlight allows for design of the mobile web interface of the web site. Something that struck me as really nice was how Highlight leverages users' existing knowledge of enterprise web applications. It reminds me of how many GWAPs take advantage of users' small contributions to learn the solutions to big problems. It's also very convenient that end users themselves are equipped to created mobile web applications based on their own customized tasks.
Regarding the second paper, An Infrastructure for Extending Applications' User Experiences Across Multiple Personal Devices, I believe that there is a strong user need for a seamless experience when moving between multiple devices. Omniscience and history are attractive to have in all of a user's personal devices. Since the authors focus on personal devices, they benefit from the fact that ownership is stable and doesn't change over days, but months and years. Their infrastructure facilitates communication between a user's personal devices of information, events, and commands. Just like in development of Highlight in the first paper, the authors conducted usability studies that confirmed usability by new users within reasonable time periods. The difference was that for this infrastructure the users were developers, and the implementation took several hours for each developer.
Chunwei Lai - Apr 29, 2009 02:47:22 am
I've only personally used mobile devices for browsing the Internet on a friend's iPhone a couple of times. The experience itself wasn't too bad and was within my expectations of a mobile device. Many popular site have a mobile device version of their site which I believe is more or less the norm (something like a low-bandwidth and high-bandwidth version for computers). Technology like Highlight can be useful, if it is adopted but it seems these days every company wants to set their own standard as the golden standard. Some of the limitations due to interactive elements also adds additional restrictions to browsing and it would be interesting to see some standards be defined clearly for mobile devices. Flash in a way has been strong in this market (at least on devices supporting Flash).
Derek Liu - Apr 29, 2009 02:26:08 am
I really liked the idea of Highlight. I can't think of how many times I've navigated through a website that was unoptimized for the Iphone and how much of a pain it was. Highlight allows a user to create a more mobile friendly webpage for viewing. Indeed a user would have to be more tech saavy in order to use this system, however I feel that an application like this could be widely used by many web developers to make their websites mobile friendly. If this app was a little more user friendly, I feel that many developers could make use of it and navigating on my mobile device would be come much easier. Also, in response to Ling Chen's comment "Maybe one day it will be that, when a user surf a site on a mobile device, the site will automatically know that, and adjust itself to the mobile version" some websites already do recognize when you're using an iphone or blackberry and adjust accordingly =).
Yin-Zen "Johnny" Hwang - Apr 29, 2009 03:02:40 am
highlight helps solve a really important problem about web browsing. it's also a good paper demonstrating what our coding project really should have been like, along with the writeups. it all ties together in a paper like that. these readings really make me wish that i've read them before designing my project, because now it looks really really elementary =(.
Alexei Baboulevitch - Apr 29, 2009 02:59:30 am
"Highlight" has the potential to be a really useful plugin for Firefox, but there are several problems with the concept. First, users reauthoring (or "customizing") interfaces often leads to less efficient interfaces, since users are generally not interface designers. This was mentioned in an article we read previously, and I tend to agree. Furthermore, webpages tend to have elements that don't often appear and that users aren't aware of until they need them. This means that neither of the component adding methods would be effective. Some elements can't be added at all (such as Java), which significantly limits the scope of the project. I think that a better solution would be to return to the form/function divide of HTML/CSS and add a separate standard for advanced interactivity, but this might do as a short-term solution.
The Jabber interface seems like a really good idea. I think a peer-to-peer IM system, if one exists, might be a better option, since it would simplify the user's experience and remove the central point of failure. A scripting language would be very helpful, since it would allow users to write extensions on the fly instead of having to compile them.
Phiroath Chan - Apr 29, 2009 03:04:27 am
With the emergence of 3g, mobile devices can further enhance the user's experience of web browsing. Highlight overall is a good idea since being able to replicate desktop web browsing on such a smaller scale like in a mobile device is a great idea. Replicating desktop web browsing is a tough task indeed, but with Highlight there is no longer a need to fully replicate desktop web browsing in its entirety. People might visit the same web page, but can totally use them differently. With a task-oriented emphasis on the design of the mobile web browsing i think users can actually enjoy the web page like they would on a desktop even though not every feature of the web site is not implemented on the mobile browser. If the user doesn't use it, then why implement that feature of the web page? A problem i see with Highlight is that if a friend were to borrow my phone and goes on the web, but uses a webpage diffenetly than i do, then there would be a problem with consistency. My point is since Highlight focuses so much on customization of web pages that one web page on one user's mobile device might not suit the needs of another user that uses that same mobile device.
Chang Su - Apr 29, 2009 03:10:04 am
I must admit that I'm a caveman by an EECS major's standards. Today my phone is still a camera-less, Internet-less, very much anything-less Nokia 3120. I guess personally I never saw the need to check my emails while waiting in line, on the bus, or in a toilet cubicle. But objectively speaking, I do agree that being able to use a mobile device to check out the websites that one frequents would be nifty. As such, I am more interested in the first paper, which if I understand correctly, allows users to customize a website, view it as it would appear on their mobile devices, and finally actually accessing the website as they customized from the said devices. However, I do have my reservations as all of the most frequently visited websites, including eBay, The Weather Channel, Yelp, Amazon, etc., already have professioally developed applications for the iPhone. Spending time and efforts to customize a website therefore does not seem to have much market. Well, just MHO.
Alexander Cho - Apr 29, 2009 03:43:39 am
The Highlighter article talked about a pretty interesting concept. It would be very nice to bypass some of the tedious work one has to do to execute common tasks on the computer and internet. I found the paper structure, explanations, and diagrams quite clear and well organized and a great example of presenting a UI. I think this technology is a very smart and a great idea especially in this fast paced world where people are trying to trim off time in almost anything. However I could see how this interface would be quite error prone if the users were not experts and the interface automatically chose certain commands without the user knowing, but with enough testing and modification I'm sure that this could be workable and great. The article about extension of multiple personal devices was also very organized and brought up a concept I never heard or thought about too much. It presents some ideas that would make things quite convenient to collaborate work on different machines. It would be great to see these technologies in its final usable forms.
Moonway Lin - Apr 29, 2009 03:22:13 am
Personally, I am not a smartphone user, and I do not like the inconvenience of browsing a webpage on a very tiny and limited screen. I would much rather browse the Internet on a full-sized laptop or desktop screen. However, the Highlight technology does show a lot of potential in making mobile web-browsing easier. If there were a database of user-submitted Highlight Designs for each website, people can then choose the ones they like the best. But the disadvantage to Highlight is that all of this work must be done in advance; one cannot browse to a new site while on the road and decide that he wants to use Highlight on it. This is a potential drawback. It's also noteworthy that many of the most highly-trafficked websites, such as Google, MySpace, YouTube, et cetera already have dedicated mobile versions, and if mobile web use increases in the years to come, the trend for web companies to produce separate versions will become more popular. Regarding the second article, I think having the infrastructure to support many different types of mobile technology will become increasingly essential in the 21st century as more and more people connect using portable devices.
Alan Young - Apr 29, 2009 02:00:32 am
"Mobilization by demonstration: Using Traces to Re-author existing web sites" introduces the Highlight system, which a mobile version of a web service. Since most web sites are developed for desktops, a lot of changes must be made to ensure its usability and usefulness on mobile devices. In order to do so, analysis must be made or a trace must be made on the interactions between content and user for mobile applications. Highlight Designer is a firefox plugin which allows developers to create mobile web apps. An example of such an app is a mobile app to use to buy a single item on amazon.com. Highlight Designer includes a storyboard for the developer to see all possible interactions. I think Highlight Designer seems like a great tool that addresses a real pain for developers who are trying to expand their web services to the mobile platform and make it more ubiquitous and useful. By being able to trace through the different interactions that users can take with content on a web page, anyone can create a similarly satisfactory usability experience on the mobile device form factor.
"An Infrastructure for Extending Applications' User Experiences Across Multiple Personal Devices" points out that since so much work on content in the web space is now done from many devices, ranging from laptops to smartphones, there must be an infrastructure in place for developers to design multi-device services. The paper goes into the infrastructure implementation details and later into the applications. The example include a Shared List application, such as a grocery list, that can be modified from any personal device. The paper also provided results from experiments on users who were given laptops and desktops with Eclipse and the service skeletons to create applications and the results showed that the proposed infrastructure in question is very feasible and ready for further long term studies with early adopters.
Sean Kim - Apr 29, 2009 04:25:01 am
After reading the first article, I can recognize that designer for interfaces also need to think the possibility for current interfaces to be extended to other form such as pc to mobile phone. The highlight can be the good example for representing the good interaction within a bridge between the two media. From this article, I can see how an interface system is changed based on the platform where the application is implemented. Similar to the first article, personal information environment concept makes boundary of interface design so vague. therefor, I think there doesn't exist any specific and particular fixed environment in where interfaces implemented on.
Raymond Young - Apr 29, 2009 04:59:57 am
My interest is growing in mobile phone apps, so/and I thought that Highlight was a really useful system. The only issue I could see was even addressed in the article itself, and that's the part about the system not being very flexible or forgiving to mistakes. It would be cool to be able to edit the mobile site in real time by viewing the original page, and then selecting different elements of the page as the app extracts the code necessary to recreate that chunk/section of the page. Then you go back to the mobile view once you've fixed your mistake about something important/useful you left out. Not all interactions on a website, even one as one-pointed as amazon, are going to use the same interactions every time. But it was cool to see how well the idea was developed given the very specific target user group. It was also neat to see a report on something in its mid-development. I enjoyed the specific target user group again, because this infrastructure felt more "by developers for developers" and it seems that the way it was built would make the most sense to people in the computer industry.
Kevin Huey - Apr 29, 2009 05:23:07 am
I can't help but think that showing these two reports are also sneaky methods for showing us students the type of analysis you're looking for in our final project reports. Very sneaky.
The reports were pretty interesting, and Highlight has major potential based on the reading. Pretty much everyone has commented on it, so I don't feel the need to. Instead though, I think the idea behind these implementations was to make web-browsing on mobile phones much easier and convenient to the average user. All these tests definitely cater toward giving the user a more streamlined experience. And they probably had that result. My question is if these things would convert more mobile-web users, such as what the 2nd reading kind of tried to do when conducting user tests on non-mobile-web users. Maybe the greater convenience converts more people. Personally, I still find phones to be a hassle. It's like most interfaces are not set up for web-browsing.
Eric Hernandez - Apr 29, 2009 07:53:47 am
Highlight is an interesting idea, where, as far as I can tell, nothing has ever really come close to doing this. Taking a trace of a user task and building a set of mobile phone pagelets to accomplish the same task seems pretty innovative. In this way, all of the important tasks on a website that requires a large display become small sequences of pagelets that fit on a mobile screen. The fact that even non-programmers can do it is a little disheartening though, less use for us!
The second paper, which aims to create one user experience over multiple devices, seems to largely solve a common problem. A single server that essentially does all of the necessary data sending and receiving to all client devices is not a new idea, but this particular implementation was interesting. Allowing developers to easily exchange this information in a way that is relevant to their applications on different devices is what makes the difference. One problem however, is that the odds of a developer designing a suite of applications for all of the devices that a user just so happens to use is slim. Luckily, this framework is in Java, which presumably, further generalizes to more devices. Now a developer can develop their application for desktop/laptop, and also for Java Mobile; these two applications would likely be enough to run on the vast majority of any modern mobile or non-mobile devices.
Prahalika Reddy - Apr 29, 2009 08:22:22 am
The readings today were really interesting. I think it's necessary that better forms of mobile web browsing be available since mobile usage is increasing a lot. My personal experience with mobile web browsing is limited to checking email and browsing through a couple of sites, and even with that, I was never completely satisfied with the experience. The Highlight technology seems like a really nice solution to customizing mobile apps for each person. It seems pretty cool that a person can browse through websites like they normally do and have that sequence of actions be stored for later reference. It not only makes mobile web browsing easier, but also more efficient since it's specific to each person. Overall, the reading had some very interesting ideas.
Anatol Tsang - Apr 29, 2009 09:42:36 am
I've never really thought about streamlining websites to make them more convenient for mobile users to view (I guess this is because I've never owned a mobile phone that had internet capabilities). I guess another possible (society cooperative) solution is to have website makers streamline their websites in general (since clutter on the internet is a problem in general, according to the CS 61A webpage anyway). Highlight seems to be a very useful tool- I would use it to make my mobile phone access webpages faster. However, the webpages Highlight creates are not very flashy. Therefore, if accessing web pages on a mobile phone gets faster, then Highlight has to strike a balance between keeping some visual elements (to keep users interested) and getting rid of the useless ones (to make the application go faster).
Maintaining some sort of synchronicity between different devices one owns is something that I would like to do for my own devices (when I get them...). It would make things very convenient. Perhaps, there would be some built-in backup functionality (like backing up files for a project between a home computer and a work computer). Having a server to provide this functionality was not something I expected (although in retrospect, this seems almost obvious). I don't know what to say about the different functionalities- all I would imagine doing is sending and receiving files and/or synchronization to be my top priorities.
Sum Sum Wong - Apr 29, 2009 10:05:10 am
Of the two readings, I found the first one more interesing and I really like the idea of Highlight. I do have a mobile device that can get me online and it is undeniable that some websites designed for pc and laptops do not work well in mobile devices. Although many popular websites do provide mobile version nowadays, but the interface of the mobile version is not always very satisfying. A good example would be eBay. The mobile eBay cut off (or hid to a very secret place that I am not able to find) many useful tools to make the interface cleaner, but it seriously damaged the usability of the website. Highlight technique would be helpful in this case since it let users to decide what they want to keep and what they want to leave behind, so that the version of website user is browsing could exactly fit their needs.
Andrew Chen - Apr 29, 2009 09:55:34 am
Both of these papers were very interesting to me. For the Highlight paper: knowing little about the code behind webpages, I did not realize that it was so modular as to be able to be taken apart piece by piece by an application, without a programmer. Although buttons and text fields could obviously be taken piece by piece, I was surprised that even their effects and functionality could be brought along.
David Burban - Apr 29, 2009 09:36:28 am
I don't think that web sties are going to be too pleased with "Mobilization by Demonstration: Using Traces to Re-author Existing Web Sites" research paper, as it gives users the ability to strip away unwanted advertising. While I see that the research was implemented as a plugin for Firefox, it was not actually tested on a true mobile device, where I believe that sideways scrolling is more painful that just dragging a horizontal scroll bar (i.e. on the amazon search results page).
The "An Infrastructure for Building Applications That Span Multiple Personal Devices" outlines a client-server infrastructure based on Jabber to allow synchronization between devices such as desktops, laptops, and smart phones. However, it seems that the infrastructure is only good at sharing lists (or arrays). While this may be good for simple software that just uses lists, it seems that it would not scale very well when the applications need to share graphics data. Also, synchronization might be an issue when data is modified on one device which ends up not having internet which conflicts with data on the server. This can lead the the user unknowingly working with obsolete data.
Szu-Chun Mao - Apr 29, 2009 10:54:25 am
Both papers provide ideas that can make our life easier. I particularly like the Mobilization by Demonstration: Using Traces to Re-author Existing Web Sites. It addresses the important issue of surfing the net on our smart phone. Though, many popular sites already have mobile version interface; however, they are not optimized. Highlight is a cool approach to this problem by tracing users common tasks. This paper sets a good example what this course is all about. Using traces to re-author existing web sites is a user-centered design that allows end users to create task-based mobile web application. This paper also provides the analysis of user study and evaluation. After reading these two papers, I get a better idea how our final report should be produced.
Matthew Can - Apr 29, 2009 10:46:06 am
Highlight seems like a very interesting tool to me. I use the internet on my phone all the time, but quite honestly, it's often a painful experience. Most of the websites out there are not meant for tiny screens. I'm especially interested because the paper showed the example with a shopping site, Amazon.com. It would be great to see this applied to a food shopping site like campusfood.com. Ordering food from a mobile phone seems like a task users would perform often, and it's something I would do often if it were easy to do on a mobile browser.
The problem statement in the second paper is quite true. One of the reasons I bought a smart phone is because I wanted access to my e-mail from my phone. However, there is still no easy way to remotely access my Word documents and PDFs. As a developer, I seriously hope research like this gets commercialized.
Siddharth Shah - Apr 29, 2009 09:51:26 am
I see you Jeff Nichols!! :D
Highlight is SICK. I need it so bad, you have no idea. My phone browser is pretty good in general, but on some websites with too much *STUFF* it just freaks out and shows a blank white page... Highlight is the cure. Storyboard is the perfect "metaphor" or w/e for this app. I don't really understand HOW you did all this stuff (I don't know XPath, Co-Scripter, or slop), but it looks good :)
Nice idea, I like PIE :) I like the whole "Send a URL" thing, even though that's not what it's called in the context menu. The history thing is also a really thoughtful idea, and it can make this application so much more useful than without it; the multiple devices features is also a nice touch.
Salman Rahman - Apr 29, 2009 10:52:20 am
These papers were both pretty interesting and insightful. I thought the idea of Highlight was important because customizing websites for mobile device users is something that is necessary. There are already some websites that will automatically detect whether one is using a mobile device web browser and automatically pull up a mobile-device version of their site (i.e. facebook) but it would be pretty cool to allow a user to customize it. The one issue that might arise with this technology, however, is that novice users might be unable to take advantage of this. It would be mostly the expert users who would be customizing.
Bernardo de Seabra - Apr 29, 2009 10:59:45 am
The paper "Mobilization by demonstration: Using Traces to Re-author existing web sites" presents a very interesting concept names "Highlight" that can be very useful for today's world where computing is shifting towards mobile phones (iphone, blackberry, etc.). Since the majority of websites were created for the desktop environment navigating these on a cell phone can be proven to be hard. The paper introduces the idea of creating a storyboard of the tasks containing all possible interactions that will enable the developer to create a mobile version of the website. In other words, sub-dividing the task into smaller steps where each step fits on a single screen. These mobile website will not only increase accessibility but also usability for the users. The paper "An infrastructure for Extending Applications' User Experiences Across Multiple Personal Devices" defends that there has to be an infrastructure in place to support developers in designing services that are compatible across multiple devices. The paper presents details of the implementation of such infrastructure and also presents a few of its applications.
Kevin Nakahara - Apr 29, 2009 11:27:11 am
I think a concept like Highlight is definately becoming more and more necessary as mobile devices continue to proliferate at a rapid pace. Once the mobile device market settles and a few clear products are left on the market, then something like Highlight could also probably be tailored to each device, rather than having to adapt to the bevy of selections on the market today. However, eventually, I also believe that mobile devices may very well supplant desktops as the predominant form of personal computing, at which point a reverse Highlight might be necessary, since developers could very well be focusing their new efforts onto mobile devices.
Chao Michael Zhang - Apr 29, 2009 03:24:15 pm
The Mobilization by Demonstration paper reminds us that while websites have historical been designed with pretty much only the desktop form factor, new devices are causing a paradigm shift, forcing developers to take into consideration how easily the websites will be able to be used by people with things like mobile devices. In addition, tradition inputs are experiencing a paradigm shift as well, with touch screen, multi-touch, and other input methods.