vanderwal.net

For Many AJAX is Not Degrading, But it Must

A little over two months ago Chad Dickerson posted one of the most insightful things on his site, Web 0.1 head-to-head: 37Signals' Backpackit vs. Gmail in Lynx. You are saying Lynx? Yes! The point is what 37Signals turns out degrades wonderfully and it is still usable. It could work on your mobile device or on a six year old low end computer in Eritrea in a coffee house or internet cafe (I have known two people who have just done that in the last year and found Gmail did not work nor did MSN, but Yahoo did beautifully).

Degrading is a Good Thing

Part of my problem with much of the push towards AJAX (it is a good, no great thing that XMLHTTPRequest is finally catching on). But, it must degrade well. It must still be accessible. It must be usable. If not, it is a cool useless piece of rubbish for some or many people. I have been living through this with airline sites (Continental), commerce sites (Amazon - now slightly improved), actually you name it and they adopted some where in this past year. In most cases it did not work in all browsers (many times only in my browser of last resort, which by that time I am completely peeved).

When Amazon had its wish list break on my mobile device (I (and I have found a relatively large amount of others this past couple years doing the same thing) use it to remember what books I want when I am in brick bookstores and I will check book prices as well as often add books to my wish list directly) I went nuts. The page had a ghastly sized JavaScript, which did some nice things on desktops and laptops but made the page far too large to download on a mobile device (well over 250 kb). In the past few weeks things seemed to have reversed themselves as the page degrades much better.

Is There Hope?

Chad's write-up was a nice place to start pointing, as well as pointing out the millions of dollars lost over the course of time (Continental admitted they had a problem and had waived the additional phone booking fee as well as said their calls were up considerably since the web redesign that broke things for many). Besides Chad and 37Signals I have found Donna Mauer's Designing usable rich internet applications as a starting point. I also finally picked up DOM Scripting: Web Design with JavaScript and the Document Object Model by Jeremy Keith, which focusses on getting JavaScript (and that means AJAX too) to degrade. It is a great book for designers, developers, and those managing these people.

I have an awful lot of hope, but it pains me as most of us learned these lessons five to seven years ago. Things are much better now with web standards in browsers, but one last hurdle is DOM standardization and that deeply impacts JavaScript/DOMScripting.

Two Wonderful Phone Service Calls - How Odd

This past week I had two wonderful experiences. The first was with Adobe customer service and sales support. I have been trying to do a "cross-platform upgrade" from my Windows Photoshop to the current Mac version. I have been trying this for a couple years. The phone services has been miserable and often would take over 45 minutes for them to pick up.

Not this week. The phone was picked up by customer service by Adobe in less than 3 minutes (I seriously thought I had dialed the wrong number). They fixed some problems with my account information, but I needed to talk to another department to get the upgrade I wanted completed. They transfered me to the sales group, which had all the information I had just changed and we started working through what I wanted (ultimately the Adobe CS Premium Suite). We found I could not directly do that going the cross-platform route, but I could upgrade to Photoshop CS and once that was installed and authenticated do the upgrade to the package I desired. All of this was less than the full price, not the optimal price I had been hoping trying to get.

What impressed me was the competence, speed, politeness and the working through their arcane rules to get me what I desired. In all I was done in 20 minutes.

The second also involved the phone and a voice automated solution. But after trying to exchange airline flight miles for a magazine subscription on the web, which seemed not work in any browser I had to verify my frequent flier number and extra authentication code. I first called the help number for the magazine people, which took me to a human, who seemed very confused with the information his computer was providing him (he had at least eight addresses and name variations for me (there was a more button to see the rest). He asked how I heard about the program and said I was on my preferred airline site and was linked to theirs to get magazines. That seemed confusing as he asked how long ago I got the software. A couple minutes later I could not give him answers that fit his check boxes and our conversation ended.

I then tried the airline's number to complete the magazine transaction if the web did not work or a person needed assistance. It was a voice interactive system (I loathe these). It asked some simple questions and I responded and it understood. It authenticated me very easily and quickly by me reading my needed information. I authenticated by telling it my address and it understood (this is a beautiful task given my street name and city). Next it started reading the magazine offerings and said I could interrupt and just tell it what I wanted. I interrupted (this never works with the local phone company for information) and it understood what I wanted. My transaction was complete and in under 5 minutes. I was completely impressed, which happens very rarely.

Mobile Search is Not Interested in Mobile

One of the things that has been bugging me for a while is mobile search. I mostly use Google mobile search on my mobiles for search. It is not the interface, but the results that get me.

Mobile search should return mobile site results. I gave Google a huge clue as to my usage, "I am on a mobile device", which they have yet to find as a helpful part of their algorithm. If I search for information I on my mobile I should be able to get the search results pointing to mobile ready content. If not by default, let me set this as a preference (not that I want to with Google as they have this wonderful way of poorly allowing me to manage my identity (there is no way to manage your own identity on Google).

I would love to have mobile search engine give me mobile sites. Why? Many sites have moved to flooding their pages with rich interfaces (AJAX and Flash) for no value added to the customer. This turns a 25kb or even a (formerly large page) 60kb into a 200kb or even a 450kb page. Much of this added interface is of little value other than it is cute or cool on a desktop, but on a mobile device it make the page and the information on it inaccessible.

Myself and many people I talk with who use mobile search often have not tucked the information we want into our bookmarks or sent it to ourselves for easy access. I know what site had the information I am seeking or what site I would like to have inform me while I have a little downtime away from home or the office.

Delicious Lesson and Social Network Ecosystems

Joshua Porter brings up a wonderful point he is calling the "Delicious Lesson". The Del.icio.us Lesson is incredibly important, as it is one thing that many tools and implementations of the social web do not get. The person must get value for their interaction in the service or it will fade.

I see so much focus on the technology, the interaction components, the network effect, etc. But, the driver for these services that are successful is that they have a direct primary value for the person choosing to use them.

A Little Effort for Greater Personal Payback

Jeff Hawkins (the inventor of the Palm device and pen-based writing language (Graffiti) Palm used) talks about the most important point for people to adopt and learn Graffiti was it gave the person value. Jeff points out that learning Graffiti took a little bit of time, but people could see value of learning Graffiti as it made for a quicker input of information. There was personal value that did not take a lot of initial effort to learn, which returned a much greater value.

Social Network Ecosystems

In social networks and personal interaction with web applications and their associated communities there is a ecosystem. The social networks have value chains. I have been playing with this idea for a few months (mostly with in the intersection of the Personal InfoCloud and the Local InfoCloud. I have been using it on some personal projects and it is weaving its way into my consulting practice (but with focus on the full ecosystem and values).

The personal interaction with the system/application/service and value derived is a viable measure, particularly when there are two points of value for the person. The first personal value is derived from the service returning direct value to the person for their interaction. In del.icio.us it is making one&039;s own bookmarks/favorites more easily refindable, the ability to expand one's own bookmarks/favorites beyond the functional restrictions of the browser, and having access to the bookmarks/favorites from any browser anywhere one has web access. The second value is the network value, which can be a feeling of digital philanthropy (doing it out of goodness), personal attention (being an authority, coolness, building points for alphaness, etc.), a driver for monetary reward (recognition increases clicks to an site with ad revenues, builds attention for a business, etc.), etc.

Attention is Value

Attention for those providing development, like in the widget communities for Yahoo! Widgets and Apple Dashboard is very important. In the podcast of the Niall Kennedy and Om Malik interview with Kevin Burton regarding APIs this issue gets brought up (beginning at 15:19 into the podcast). The example discusses Konfabulator (the original product name for Yahoo! Widgets) and Apple Dashboard. Kevin Burton (I believe) states Apple Dashboard has larger exposure than Konfabulator does at the moment and Apple could offer the developers more attention to get more people writing widgets for them. Here the initial value for the developers is attention, as is pointed out in the podcast. The developers are passionate about what they do (personal value in a platform for their expression of their development prowess), but the secondary value received is attention. In a limited pool of developers (not only for widgets, but APIs, and other open development arenas) value to the developer is attention, which can lead to monetary value. This value to the developer is going to be a driver for which service they provide their services (a secondary driver is ease of development). The value to the network is more widgets equal more cool things for the service, but the primary reason the developers are there, is often the value to themselves.

Del.icio.us is Back

Yeah!! Del.icio.us is back. After many hours, if not a day of being down due to residual effects from a power outage. I bet Joshua is looking forward to somebody else managing the servers.

I has been a bad week for the popular stuff on the web with TypePad outage problems in the past week as well (yes, that meant Personal InfoCloud was down).

Does it bug me? Not so much. Del.icio.us being down meant I was not cross posting with Yahoo! MyWeb rather than to both places. If this site is down I am not too happy as my work e-mail is on the same server and I have been living in e-mail lately. But, I think with TypePad and Del.icio.us and the like with their outages I have appreciation for what it takes to keep that up and running. I also know the problems inherent in scaling those type of services. At some point the killer ease of use applications become more about killer sysadmins and server/datastore optimization skills. That is where one learns to grow up.

Along those lines, I am quite happy to see Technorati get their server situation sorted out and they are now running at usable speed again (it was a seemingly long time coming).

Web 2.0 Dead?

It was bound to happen sooner or later, but it was a little sooner than expected. Richard McMannus explains why Web 2.0 jumped the shark as an follow-up to his Web 2.0 is dead. R.I.P. post. This pronouncement has an impact as he is co-writing a book on Web 2.0 for O'Reilly Books (with Joshua Porter) and writes Web 2.0 Explorer on ZD Net. In Richard's explanation he gives the prime reason is to get away from the hype and cynicism.

Tim O'Reilly describes Web 2.0 in rather long detail. But in the more than a year that the term has been around it has not been used in any specific specific sense and it quickly turned into a buzzword with little meaning. There are some profoundly different things taking place on the web, when we compare it to the web five years ago. These things seem to be best described by their terms and pointing to what has changed and where we are going now. Richard writes that he will still largely be writing about the same things, but will not be using the Web 2.0 moniker.

The Rich Interface

During the past six to nine months one could easily see that the term Web 2.0 getting flattened into hype and mis-understanding. Many articles were written about new technologies that were changing the landscape, but neither were the technologies new, nor were they doing much of anything different than sites were doing or trying to do in the previous three to five years. AJAX was not new, it was a new name for xmlhttprequest (which most web developer worth much of anything knew about, but knew there was little browser adoption outside of Microsoft IE). Jesse James Garrett provided a much easier means of calling the long term, mostly to talk more easily about what Flickr and Google (in Gmail and Google maps) had been doing in the past year using it as part of their rich interfaces. The rich interfaces were absolutely nothing new as Flash had been providing the exact rich interface capability for years. The problem was much of the design world had not worked through its documentation and design specifications for a rich interface using Flash, but they jumped all over AJAX with out ever working through solutions to the problems of state, (re)addressing information, breaking the back button, addressable steps in a process, etc. Web browsers growing up and becoming consistent and more processing power and memory on the machines under the browsers have enabled the rich interface more than anything that gets credit for being new.

Web as Platform

The web as a platform is a great step forward, but it is anything but new, just ask the folks at Salesforce. But it has been embraced as a replacement for the desktop . The downside is most people do not have continuous access (or anything near it) and many do not want it. People have set workflows that cross many devices, contexts, and information uses. Thinking the web is the only way is just as short-sited as closed desktop applications. The web as a platform is insanely helpful, but it should not be the only platform. We have to work towards cross-platfoms and cross-device use development as an end not just the web.

Forbidden Term

Very quickly this year the Web 2.0 term was forbidden from usage from many conferences and large meetings I went to. It was forbidden as by that point it had lost its meaning and using the more direct terms, like social networking, social bookmarks, rich interface for mail, web as an application platform, etc. It was also noted that people should not say the new web, with out explaining why they thought it was new. There needs to be clarity in understanding so we can communicate, and Web 2.0 did not provide that as it was an umbrella term that was used as a buzz word to replace specific changes people did not understand.

Without a Term How to We Understand

There have been a handful of people who have been writing on the Web 2.0 changes and landscape and using the term well and describing the components that are being used in new ways. Richard was one and his writing partner Joshua is another. The group that is aggregated at Web 2.0 Workgroup are most of the rest.

With out the term Web 2.0 it will be tough, but it was more a marker of a confluence of many different things that shifted than a bright line in one or two areas. Understanding what has changed will make sense, which is a large part or what Joshua has been doing and a small handful of others. When the confluence is the streams and rivers of technology, social interaction (as Bruce Sterling calls it "technosocial"), interface, web services, application that provide uses that are needed, cultural and social changes along the lines of privacy (this could swing back massively), cultural changes with more people having comfort with social interactions using technology, trust, etc. take place there will be problems describing it. There will also be only a rare few that can cross the chasms and grasp, make sense of the subtle as well as vast changes, and explain them intelligently and simply to others. As the majority of writing has proven it is a very rare few indeed that have the background and wit to handle this challenge.

Now that I am at the end of the brain dump (some of it long festering), I think I am a wee bit sad to see the term losing traction. But, I don't have to think had to remember one of the vast many of poor articles that every journal has had somebody write.

In full disclosure I spoke on the BayCHI Web 2.0 Panel held at Parc in August. Been writing and speaking on digital information use across devices and platforms for three or four years and the underlying information architecture that is needed to support it. In this past year I frame the need for it as a change from the "I go get web" to the "Come to me web" (not quite equivalent to the push/pull analogy, but I will explain this later for those that have not heard the presentations or me just ramble about it). I felt it important to frame what I change I was talking about rather than rely on the Web 2.0 moniker.

BBC Knocks Audio Annotating Out of the Park

Tom Coates shows off the BBC's Annotatable Audio Project. Tom gave me a preview early on Saturday and I was ecstatic. You see, what the tool does is provide an interface to annotate and segment audio on the web. Yes, podcasts can be easily segmented and annotated. This has been my biggest complaint with podcasts over the past year, okay since they started getting big (that is big for an early early adopter). I complained to people I knew at Odeo about the problem and they said they were working on it. I mentioned this to podcast enthusiasts at Yahoo! about nine months to a year ago and they said if they did podcasting that would be one of the first things in it as it was a big complaint. Did they? No, they made a product not too indistinguishable from every other product out there? Where is the innovation?.

Why is this Huge

The reason I am so excited about this is voice/audio is not easily scannable, like type. I can not easily skip ahead in a 30 or 45 minute podcasts to find that which I am interested in. Many friends will forward me links to a podcasts stating I have to hear what somebody says. Finding that segment usually means listening to much of the whole podcast.

The other downside is if I hear something stellar in a podcast my mind will mull over that item for a little bit. This means the minute to five minutes that follow in the podcast are lost on me. This is not a problem with written materials as I can skim back through the content and pick-up where my mind drifted (it is usually in these moments of drifting that I find the best solutions to things that have bugging me - the Model of Attraction came out of one of these).

A couple other items of note about this product. It is great interface design as it is interactive helps the person using the product know exactly what they are doing. The second is the segmentation is a great asset. With segmentation I can easily see writing a script to grab items of interest (27 seconds for here and 36 seconds from there, etc.) and having an automated audio stream built for me. Not only do I have a personalized audio stream, but since the originals are annotated and I can keep track of where the information is extracted from I can easily point others to the spot so they (or I) at some later point can go back and listen to more so to get better context (personally I don't think people are against attribution, it is just that we have made it so hard to do so in the past).

Voice and Audio is a Common Problem

The last couple time I have travelled in the USA I have run across people quite similar to me. None of us like voice. We are not particularly fond of the phone, for much the same reasons as I have problems with podcasts. Too much information gets lost. In phone conversations I am often saying, "I am sorry can you repeat that", in part because I did not hear, but the something that was stated just triggered a good though process for me and I missed what came after that moment. (What would be a great application is Tivo for the phone.) I continually am running new ideas and thoughts through what I believe and see how they may change it. It is the examined life - I enjoy living.

So what Tom and his cohorts did was make podcasts and audio more usable. It makes it searchable. One thing that would be a very nice addition is to have those annotating the information each have their own distinct layer. Just like with folksonomies, the broad folksonomy where each individual and each annotation on a distinct element provides a richer understanding and richer layer. (Such things would be really nice in Wikipedia so that I could remove the people who I do not think add any value to entries (in not polite terms - those who I know are wrong and are polluting the value of Wikipedia, which is far too much noise for me on the entries I would love to point to), or conversely to use a "white hat" approach and subscribe to the annotations of people and the distinct tags or terms they use in annotations. I have many people whose opinions and view I value, but on rare occasions it is everything a person has to say.

Filtering information in a world of too much information to keep track of is a necessity. Filtering is a must. It is about time we got here.

Thank you Tom. I hope your new team can innovate as much as you were allowed at the BBC, which has been the most innovative large enterprise going.

Mobile Location Needs

I have written up some of my frustrations with mobile local needs in a posting Location? Location? Where am I? over at the Personal InfoCloud. (Comments are on and it fits the Local InfoCloud theme that is also there.)

Microformats hCard and hCalendar Used for Web 2.0 Conference Speakers

Tantek has posted new microformat favelets (bookmarklets you put in your browser's toolbar). The microformat favelets available are: Copy hCards; Copy hCalendars; Subscribe to hCalendars; feed Copy hCalendars (beta); Subscribe to hCalendars feed (beta). Look at Tantek's Web 2.0 Speakers hCard and hCalendar blog post to understand the power behind this.

Microformats are one of the ways that sites can make their information more usable and reusable to people who have an interest. If you have a store and are providing the address you have a few options to make it easy for people, but a simple option seems to be using the microformat hCard (other options include vCard and links to the common mapping programs with "driving directions").

There will be more to come on microformats in the near future here.

Interview regarding Digital Identity and the Personal InfoCloud

Today I appear over on Under the Radar: That cloud kinda looks like you! in an interview by Scott Hirsch who is a partner at MIG5. The interview focusses on digital identity, which is integral to the Personal InfoCloud and interacting with other people and services in the digital world. Increasingly it seems digital identity is tethered to physical world identity for access to buildings, accessing our computers, medical services, etc. This has some problems around privacy that must be addressed and there must be trust in the services that interact with out digital identity.

Web 2.0 Mash-ups and the Model of Attraction

I posted a write-up on Mash-ups and the Model of Attraction, which explains the Housing Maps through the Model of Attraction lens.

Read and comment over at Personal InfoCloud.

User Experience Design in the Come to Me Web?

A question came up with Rashmi in the week prior to the BayCHI Web 2.0 event that I thought would definitely come up at the panel in the Q&A session, but most of the questions related to the application and technology side of things.

As content can be repurposed in and pulled into various tools with drastically different presentations than the sites they sit within. There seems to be a logical question as to the value of the user experience of the initial site. We are spending a lot of time, effort, and resources building optimal user experience, but with more and more of the content being consumed in interfaces that do not use the user experience should we spend less time and resources on perfecting it?

One answer is no, things are fine the way they are as the people that still consume the information in the traditional web manner (is it too early to call it traditional web manner?) are a narrower audience than the whole of the people consuming the information. The design of the site would have to add value, or provide additional service to continue enticing people back. I have been talking about the Perceptual Receptor in the Model of Attraction for a few years and the sensory components of design, look, and appeal should be targeted to the expected users so it fits their expectations and they are attracted to the content they are seeking in a manner that is appealing to them.

The converse to this is we are spending too much time on the ephemeral in relation to the benefit. With increasing consumption of the information done though RSS/ATOM feed readers and aggregators on the desktop, mobile, or web (as in Bloglines or My Yahoo) interfaces, which nearly all strip the presentational layers and just deliver the straight content with the option for the person to click and get to the site we developed. Information is also pulled together in other aggregators as summaries on various websites and versions e-mailed around. The control of the user experience has drifted away from the initial designer and is in the hands of the tools aggregating (some provide presentational layers from the content owners to show through on the aggregators), or the people consuming the information that choose their own presentation layer or just strip it for other uses.

With content presentations in the hands of the people consuming and not the crafting designer how does branding come through? How does the richer integrated interface we spent months designing, testing, and carefully tweaking? Branding with logos may be easier than the consistent interface we desire as the person consuming the content has a different idea of consistent interface, which is the interface they are consuming all of the information in. People have visual patterns they follow in an application and that interface helps them scan quickly for the information they desire.

Where the content creator puts their content out for aggregation in XML related feeds, they have made a decision at some level that having their content in the hands of more people who want it is more important than a unified user experience. Consumption of the media has a greater impact than fewer people consuming a preferred experience. All of the resources we put into the refined user experience is largely for the user's benefit, or at least that is what we say, but it is also for the business benefit for consistent branding and imprinting. The newer consumption models focus on the person and their getting the information and media they want in the easiest and their preferred manner for that person.

Is there an answer? One single answer, most likely not. But, I personally don't think we and crafting designer have a great say at this point. As tools people use mature, we may get more control, but optimally the person consuming is the one in control as they want to be and should in the "come to me web".

Presented Personal InfoCloud and Folksonomy to MIM Class

Last night I presented my Designing for the Personal InfoCloud - Including a Focus on Folksonomy to the User Interaction with Information Systems class in the Master of Information Managment program at University of Maryland.

I had a blast doing this, maybe more than last year. This year's class is quite sharp and had great questions during the class and after. I have seen two classes in the MIM program and helped a student project last spring and I think the University of Maryland is on to something. The program is a professional masters program for information professionals in management positions.

Designing for the Personal InfoCloud presentation at WebVisions 2005 Wrap-up

I have posted my presentation from yesterday's session at WebVisions, in Portland, Oregon. The files, Designing for the Personal InfoCloud are in PDF format and weigh in at 1.3MB.

I really had a blast at the conference and wish I could have been there the whole day. I will have to say from the perspective of a speaker it is a fantastically run conference. Brad Smith of Hot Pepper Studios did a knock out job pulling this conference together. It should be on the must attend list for web developers. I was impressed with the speakers, the turn out, and how well everything was run. Bravo!

WebVisions is held in one of my favorite cities, Portland, Oregon, which has some of the best architecture and public planning of any North American city. I have more than 300 photos I have taken in 48 hours and will be posting many at Flickr in the next couple of days.

Passion and the Day-to-day

This has been an up and down month so far with health, work, technology, and time. In general 2005 has been a rough year for respiratory issues already for me as I am nearly 3x the normal problems for a full year. These issues zaps energy and fogs the brain (something I really loathe).

The day-job is muddled in past problems, issues that have been plaguing people and have been solved years ago, but where I am resources and bureaucracy keep the long past current. Outside of the day-job I am working with the now and future, which I have really been loving. I have been working on responding back to many questions that have come in through e-mail about possible work and helping people through problems grasping and implementing efficiencies for current web development, folksonomies, and Personal InfoCloud related items.

I have also been working on my presentation for WebVisions, which involves completing it, tearing it apart and nearly starting over. To date I have nearly 25 hours working on this presentation, mostly integrating new material and editing out past content. This is in contrast to day-job presentations, which take me about an hour to build.

In a sense I am still time traveling on my daily commute. The gap is about four to six years of time travel in each 40 minute to hour commute. This is really wearing on me and it is long past time to move on, but I have not had the time to put forward to nail down the essentials for moving my passion to my day job (time and family needs that have filled this year).

So today, I was quite uplifted as my subscription issue of August 2005 MIT Technology Review arrived. The cover topic is Social Machines and I am quoted and have a sidebar box. That was up lifting as it relates to my "real work". This is right up there with Wired's Bruce Sterling article on folksonomy and Thomas Vander Wal.

Now the real work continues. If you are in Portland for Web Visions or just there in general later this week, please drop me a note and I will provide my contact info. If you are not in Portland and would like me to come to you and discuss and help along these topics please contact me also.