|
|
My work blog. Social Computing with Agile Software Delivery. See also my personal blog.
-
Last week's failure of the HMRC to undertake basic steps to protect
the data of some 25m-odd British citizens is a classic case of how IT
by itself is of no value at all. All the retinal scanning in the world is of no use if it's not actually employed.
We're the tail end of a cultural revolution right now -- moving
into a true digital age, where online is woven into society. What
happened above was stone age thinking leading to a blundering
oversight. The issue fundamentally comes down to perception -- how can
losing a couple of small pieces of metal and plastic potentially cost the taxpayer £300m? The risks associated with sending several hundred kg of printed material of a similar nature would've been abundantly clear. Lack
of basic IT literacy is what caused this debacle. Governments all over
the world lumber in a taxpayer-funded gravy train, using old value
systems and old methods of working that just don't take into account
modern day needs and risks. It's time some of the fat was cut.
And it's not from the bottom -- it's from the grey suited middle
management encumbents that have no real incentives to improve. With the
money saved, get some decent IT literacy into the departments.
Or better yet, hire some digital natives -- almost anyone under
25 fits this: they'll be full of ideas and be fluent in today's new
digital channel.
|
-
The dangerous mix
Why do we
need the Treating Customers Fairly (TCF) principles? Many financial services products
are a dangerous mix of elements: they’re generally uninteresting, complex to
understand, and yet carry dire consequences if not used properly. A bit like an
army mission then: 99% of the time stupefyingly boring along with 1% sheer
terror. And weirder still – some you pay for in the hope that you don’t need them!
Consumers
need protection from this. You buy a new fridge – it’s pretty easy to compare
one against another, whereas you may not even know you really need critical
illness insurance. The FSA is responsible for making sure consumers are
protected, and with the recent subprime crisis, it’s quite clear that even with
them there are some delinquents in the FSA nest. £500k mortage? That’ll be fine
on a salary of £30k, no problems! Without the FSA at all, you can imagine the
mess we’d be in.
TCF has to be taken very
seriously now.
December
2008 looms, and with it a TCF mandate from the FSA requiring demonstrable proof
that customers really are being treated fairly, with management information and
everything. The FSA has offered extensive literature on the framework, but in
addition, Conchango has some specific recommendations for your digital customer
experience.
I read a customer review
about it. They’re great when it comes to claims, and they love them for it.
The way
customers find out about products and services has fundamentally changed. The majority now trust more what their
peers say than any amount of product literature. Their experiences are
shared online like dirty laundry, for all the world to see. Sometimes then, customers’ first contact with your product can
be from these raw, subjective experiences. And for financial services
products, servicing experience is what people will be most interested in: it’s
hidden from view, yet is often the whole point of the product! Every company
has a bed of roses onramp to their product – this is assumed. But when you accidentally
go into overdraft, or you drop your ipod into the drain at the bus stop, what
did they do then? You need to deploy
agents online and directly contribute to these conversations, correcting errors
and making amends. This is one on one: broadcast has no effect here.
They answered all my
questions, in terms I understand (I shouted from the rooftops about it).
The foundation of TCF is understanding and
satisfying the customer’s genuine needs. Not
wants. So to avoid having the FSA waggling their finger at you as a product
provider, it’s a good idea to figure out how to cost-effectively educate your target
audience. Modern day technology makes knowing your customer much easier than
before.
- Use plain English: being complex to understand,
let customers ask questions in plain language with plain responses. Let
the entire customer history influence the responses.
- Give a little,
get a little: let
customers build up ‘what if’ scenarios based on small increments of
information they share. I’m a male, what will my premiums be like? What if
I’m a male dentist smoker? What if my date of birth is 1964? No boring
long forms to traipse through – it becomes an insightful, exploratory
exercise. Financial products are complex and are great candidates for
rich, visual applications that today’s web browsers can show easily.
- Allow them to
share: share selected
details, but anonymously. Then foster communities that can benefit from
this information, and learn together. You’ll find passion in these
communities: reward the evangelists with special privileges. Bring what the
community shares into the product design process.
- Integrate the channels: when a question is asked, this
should be treated like gold: keep it. Have a ‘call me about this question’
button, next to one labelled ‘chat to me online about this question’. While
they’re waiting, look the question up in your knowledge base. And if there
isn’t someone available after a moment, archive it, and give the option of
a callback, email, or SMS.
The clock’s
ticking – but luckily there’s still time to innovate. Let me know if you have
any other ideas!
|
-
(I'm in geek mode at the moment because I've been setting up an automated test framework for the project I'm managing, which has seen me learn C#, which is fun. So I guess I'm not a typical PM. But I digress...) Context: you want to parse some HTML using XPATH in C#. All roads appear to point to HtmlAgilityKit. I specifically was trying to do is make it really easy for testers to check the contents of dropdown lists so I write a Custom ValidationRule for Microsoft WebTest called 'ValidateDropdown' which could also easily be called 'ValidateSelect' but Select is such a uselessly generic term. Problem: HTML defines select option like this: <select ... ><option value="[value]">[InnerHtml]
This means the 'InnerHtml' is not associated with the option tag. XHTML however defines it like this: <select ... ><option value="[value]">[InnerHtml]</option> HtmlAgilityToolkit was driving me nuts because it was actually doing strictly the right thing -- ignoring the InnerHTML. The fix is easy -- see link to forum Enjoy!
|
-
James Gardner, Head of Innovation and Research at Lloyds TSB was recently wondering: it is users themselves, everywhere else, that are deciding not only how things are personalised, but also what content should be shown in the personalised context. We can see where social finance is going when examining Mint, Wesabe and others of their ilk. So,
sitting around the table listening about "personalisation", I couldn't
help but wonder if we're somehow stuck in a financial services time
warp. Or at least, why the institutional part of the market is.
.. to which I replied: I think it's a mix of the two -- that personalisation done right draws from both
- software-initiated (implicit personalisation), and
- user-initiated (explicit personalisation).
Amazon is the poster child of software-initiated customisation, aka
'collaborative filtering' aka 'recommendation-engine driven
personalisation'. All the features and items are derived entirely
implicitly from the user's actions. Last.fm is another great example --
working secretly in the background as you listen to music to figure out
what you might like. They're both screaming successes so I'd not want
to deemphasise this aspect.
Netvibes could, correspondingly be seen as the poster child of
user-initiated customisation. Netvibes has spawned well, an entire
class of web sites now called 'start pages'. This concept is Also a
great success, particularly in the intranet space.
Both work well, for solving certain things, and I think they're mutually supporting.
What's particularly fascinating in the social computing space is how
it spans both too -- recommendations can be based on your own past
behaviour ('the page you made' in Amazon) while also based similar
patterns with other users. User-initiated customisation at the social
level is a bit more tenuous but is starting to take shape shape in the
form of a kind of 'group evolution' -- Yahoo Pipes! for instance allows
cloning of pipes so you can make a better pipe. You've customised your
experience, but it's based on what someone else did.
So why isn't implicit personalisation more popular then? When the
rubber hits the road, it involves some pretty tricky problems that
require some great business intelligence, which is often hard in an
enterprise. First base being single view of customer (SVC) of course,
or least what I coin 'incrementally better view of customer' ('IBVC').
As an aside into what's involved for recommendation engines, check
out a precis of Google New's recommendation engine at on my personal
blog.
|
-
Dapper: convert any web page into a web service
"Dapper allows you to easily build an API for any website. Initially, if the content source you’re interested in is not Dapped already, Dapper will take you through a visual and intuitive process that takes just minutes. When finished, you will have at your disposal an API with which you can interface to get the content you are interested in."
Basically a screen scraper. Obviously this only works with reasonably static web sites. As the trend has sites become more interactive and use other rich formats such as Flash etc, this tool system will become less and less effective. This app has caused a sensation -- been bookmarked like a billion times on Delicious today.
I've had a quick play -- here's the source for a Google Desktop Gadget that presents a London Tube Service Update [XML].
Legal implications
Stripping out branding and repurposing content will definitely violate that little 'legal' tab at the bottom of most web pages.
|
-
Jamie Thomson, SSIS blogger extraordinaire asked me this today: Take a look
at the tags on our blogsite: http://blogs.conchango.com/tags/default.aspx You will
see that there are alot of very similar tags on there and a rationalisation
process might be beneficial in increasing the usability of the site - making it
easier for people to find the information they are interested in. It would also
help to reduce clutter. e.g. There
are tags for "Analysis Services", "Analysis Services 2005",
"AS", "AS2005", "SQL Server Analysis Services",
"SSAS". 6 different tags and they all relate to exactly
the same thing. It would make sense (to me anyway) to group them all as a single
tag "Analysis Services" On the
other hand, I'm aware that dictating how people should and should not
categorise their posts might not be a very (here we go) Web 2.0 way of
thinking.
So, we use a tool called 'Community Server'.
How to encourage people to use the same tags One
thing I thought CS does well is that as you’re typing it brings up a list of
existing tags. Except it seems only to do it for tags you yourself have
entered, which is weird.
The
concept is: if the system suggests tags that look a bit like what you’re typing, you’ll
select one of the existing ones, so you get resonance. This is called ‘user
paths’ (dry, boring name) or my much more fun made up description ‘sheeptrailling’. Which can
help reduce aliasing a bit. (Del.icio.us does this very well to encourage tag reuse when you're creating entries.)
The
fundamental idea as you say about the concept of a ‘folksonomy’ (user specified labels) vs taxonomy
(predefined hierarchy by centralised authority) is that how people think of things is subjective. A
fascinating concrete example of this was an exercise in the US with some
museums – curators pompously tagged stuff and then got visitors to tag as well.
80% of the tags the visitors made were different to the curator’s tags. Centralised authority only works to an extent. So we need to allow people to enter their own tags. Aliasing: how to improve this a bit
So Jamie imagine you’re motivated by a desire to clean it up, and reduce confusion a
bit. Fair call. I’ve not seen it, but I think it’d be fun to use the RDF idea
of aliasing – someone can nominate that one tag is actually another. So you
could, either in a posting, or as a separate task, ‘clean it up’ but specifying in some way. Let's say you entered this as a tag:
AS2005
is “Analysis Services 2005”
etc. Then when you click on one of the tags, everything
tagged with the other appears too.
I
wouldn’t remove the tags however, as multifaceted interfaces work well to accommodate
different user mental models. Hope this helps. JRH.
|
-
Retail Week has a special on Web 2.0 to be launched tomorrow (14/3). To be up front, Conchango has sponsored the event, but Retail Week wrote and produced the material. If I were take a single statement away from the supplement, it would be: Web 2.0 is necessary, but luckily aspects are lower risk and cheaper than you'd think; BUT get Web 1.0 right first.
Joanna Perry, supplement editor talks in her editorial that retails can and are adding Web 2.0 tools to their sites, that it can be done easily and cost-effecitvely, and that there is clear measurable value from doing it. Perry then highlights the value of RSS as a specific value-add increment. Without doubt RSS can't be said to be 'in the mainstream' yet but with Windows Vista gadgets this will all change. Here are some interesting snippets from other parts of the supplementy with my follow-up commentary: Customers behave differently now Mike Pierce from Paul Mason Consulting underscores the power shift that comes with web 2.0, through what I call the 'Web 2.0 backchannel of communication": "Traditionally, retailers have had a selling process that customers have had to conform to. Now consumers have a personal buying process that retails have to conform to". Part of this personal buying process is the trend of cross-industry aggregate views -- customers are going to 'relatively independent' portals to compare retailer goods head to head. What's important not to miss however is that this head-to-head comparison is no longer just about price -- post sales service performance and other customers' experience with the product are being factored into these portals.
There's also an interesting discussion on Bottlerocket -- a shop with an active community in New York and without advertising how they're donig well. Get Web 1.0 right first Conchango's own Paul Dawson underscores an important message -- don't do Web 2.0 until you've done Web 1.0 right. Part of this is getting your basic user experience and design principles right. On my other blog I highlighted on Newspapers 2.0 about USA Today's controversial 'social redesign', the uproar from users, and why it wasn't Web 2.0's fault. Other useful bits; Building communities online, how communication is being affected, Retail in SecondLife -- virtual worlds and finally a hit of how web 2.0 can affect internal communications (this is sometimes called 'Enterprise 2.0'). I really enjoyed it overall. How can you get a copy of the material? Good question. Retail Week is subscriber only. But Contact Conchango and we can arrange for free copies to be sent out to you.
OverallI found the supplement to be a helpful perspective from the retailer's point of view, a compliment to the conversations I've been having with other communities (mostly horizontal -- such as web 2.0 community best practices, RIA best practices, etc). Other snippets: - Sainsbury's online has a great discussion forum -- however conversations were kickstarted by a faceless employee called 'Sainsburys', sometimes 5 months ago, and despite dozens and dozens of replies, the shadowy 'Sainsburys' doesn't then follow up the engagement... To 'Sainsburys' -- who are you? Why hide behind the mask? Why not 'Bob Nelson from Sainsburys'. And a bit of follow-up to reader comments would be well respected.
- Tesco.com's RSS feed: their front page has 'today at Tesco' RSS feed for specials and timely deals. Nice start! I wonder what the uptake is and, how are they doing with analytics on number of subscribers? And, let's see a full 'RSS view' of the site -- once you leave the front page it looks a bit barren.
|
-
Project Red Stripe » The Economist seeks fortune tellers
"In a nutshell, The Economist Group has picked a small team of employees, moved them to London, and allowed them to come up with whatever they want, even something entirely unrelated to The Economist Group – as long as it is innovative and online (for details go here)."
Jeff Jarvis highlighted this -- saying these guys have $US200k and 6 months to figure it all out. They came to London because apparently this is where the newspaper innovation is.
Something serious is afoot. Lots of things are changing. People are really treating Web 2.0 seriously in newspaper publishing and they're all clamouring to make sure they're not left behind. What's particularly interesting is is follow-on quote: “What makes a week in London
so damned exciting for me — professionally and intellectually invigorating — is
the competitive race for innovation I see all around here. It has been a great
week.”
I'm genuinely interested in this comment -- I thought all the action was over in California (Jarvis is from the US). What happened at USA Today then?
But USA Today, opening in fanfare with it's 'social redesign' is interesting: what's your guess. The new version had "lots of social network features like comments, reader rankings, blogs, photos, and better search. Reader reaction? 92% of readers don't like USA Today's Social Redesign". Is Web 2.0 a white elephant? Or did they just do Web 2.0 wrong? Or was it something else?
Dodge goes on to offer a view that it's likely the layout redesign was the failure, not the mechanisms introduced. It's just that they happened to now have a way of venting their concerns.
This is a good thing -- at least, USA Today now has an opportunity to acknowledge the feedback, and respond.
This is Web 2.0 in action by the way. Web 2.0 has enabled customers to whine at each other quickly, permanently, and most worryingly for customers, searchably (if there were such a word); USA Today is, as it should be, capturing some of this, and if they're smart, will respond in a way appropriate to the medium: honestly, and promptly. People may whine, but they should've got their darn user testing right in the first place. Blame the design process, not web 2.0! Hats off to USA Today for leaving their editorial throttle right off -- those comments are there forever now, and so will be their response.
|
-
Zopa was covered in Channel 4 news recently. Headlines:
Default rates are very low because there is a personal relationship established -- being more personal, people are less, not never, less inclined to default.
Zopa's growth places them at 135,000 members,
(embedding still doesn't work here :( -- see my personal blog for video for now... )
Also profiled was FairFinance
|
-
Twingly Screensaver Beta
"Twingly screensaver is visualizing the global blog activity in real time. Forget RSS readers where you see only what you're interested in. With Twingly screensaver you get a 24/7 stream of all (viewer discretion advised) blog activity, straight to your screen. More information about Primelabs can be found at primelabs.com"
Twingly shows a world map that you can drag around and in real time as blogs are posted the titles scroll by and an indicator of regional volume builds up. You can click on any post and view the blog summary; one more click and you're on their site. See my personal blog for the video while we get video embedding sorted on this blog...
|
-
So Read/Write Web wrote a piece about the inevitable term 'Web 3.0' being:
- Web 2.0 (Content, Commerce, Community), +
- Personalization and Vertical Search.
Well, I'm not so sure about that. Personalisation has been around a long time; I'm an ATG Dynamo Java programmer by background and its Personalization Server has been very popular for oh I dunno, the last 7 years?
What's more interesting is the Vertical Search business. I've not been super convinced of this, which is where I stand when I don't really understand it.
But BarCamp London 2007 and other work recently has connected some dots. One is actually, brushing the dust off my Computer Science course on Artificial Intelligence -- the basic tenet was, constrain the context of use, and you can build in more assumptions about what might be interesting. So, I'm not going to call this Web 3.0 but the concept is interesting anyway so let's explore it further.
To start, let's look at what RDF is, and what an ontology is.
RDF RDF stands for 'Resource Description Framework' and is used in the world of the Semantic Web. One of the most useful things to consider about RDF in its various formats is that it's very good for this kind of thing: [thing] [relates to in some way] [some other thing]
Building up semantic relationships is called building an 'ontology'. There are 2 primary ways of doing this in the world of the Semantic Web -- there's 'Notation3' which is Very Human Readable and then there's the XML version which is just Human Readable. Here's an example Notation3 in action: :Spot is-a :MoneyTransfer.
Not so bad? Clearly there this example brings up more questions than answers; for those who want to open the bonnet on N3, check out this useful primer on Notation3 and RDF. Suffice to say, the point is you could build up a semantic vocabulary in your industry of focus, and then trawl the web / blogosphere etc. Your semantic view would be able to do much more clever things by having this insight. The system could infer relationships. In the example above, it means that if you did a search for 'Money Transfer' and you found an article talking about Spots, and your search engine focused on resources primarily focusing on financial services resources, you could show articles with the word 'spot' as well. Clearly this would be naive to do this one thing but hopefully you get the idea. If you want to learn more about ontologies, or even explore building your own (imagine a company-specific ontology?), then check out Protege, a free open source ontology builder that is pretty wizzy and allows you to build ontologies in an ontology building language called OWL which extends RDF to do More Clever Things.
So I'll finish with a glib catchphrase: Web 3.0. It's all in the context.
|
-
Howard asked me today what my top 10 Web 2.0 apps were, and why they were killer. A good question. Here they are anyway.
|
Web 2.0 site
|
Killer feature (primary
reason: UGC, RIA)
|
|
Digg.com
|
UGC. Blows Slashdot away for interesting geek news by
democratizing the moderation process. Inspired open source tool ‘pligg.com';
I did a proof of concept of pligg at a tacky half day job of a site called Britnews. Yes it crashes a bit. But hey, half a day, cool eh? :)
|
|
Flickr
|
UGC. True web 2.0 in action - flickr isn't just about sharing
photos, it's about people's opinions of the photos. In flickr you can create
your own groups, which is really where the ‘network effect' comes in. Check
out ‘Hit Miss or Maybe' - staggeringly addictive group where you rate others'
photos in exchange for still others to rate yours. Warning - it is a serious
time sink.
Also that Flickr has APIs which make for some killer mashups.
|
|
Tripadvisor
|
UGC. Trusted source of real experiences with hotels. It really
works; you go to quality places. I've used it lots of times and been inspired
to contribute.
|
|
Ask Metafilter
|
UGC. A rabidly active ‘community weblog' about ‘everything'. This
is where I go to answer questions I cannot find answers to. It's the
grandfather of Google Answers etc.
This site embodies the spirit of web 2.0 despite being founded in
1999. See one such question of mine to get a sense of the kind of thing that is still very hard to find.
|
|
Kayak
|
RIA. Flight portal ‘Live' price controls - sliders that hone in to the best
flight dynamically. www.kayak.co.uk see also skyscanner for cool 'view price fluctuations over the month' view. Kayak seems to have more airlines.
|
|
Reevoo.com
|
UGC. Reviews on products with support for validating the reviews
against real product purchasers - the idea is, the opinions are legitmate
purchasers (through partnerships with online vendors).
|
|
Nestoria
|
RIA. Google maps mashup with houses for sale. I hated
findaproperty and other sites because it didn't have the street number. Cuts
through the rubbish about ‘10 minutes to station' when it's clearly 2km
away.
|
|
Wordpress
|
UGC (user-generated code J). My favourite
blogging tool. There's an enormous community of extensions including great
spam support, etc etc. Massive, and very easy to use. All for free.
|
|
Technorati
|
UGC. One of the key members of the blogosphere. Set itself up to
be the source of changes to blogs by being recipients to the RPC ping service
pioneered by Dave Winer and www.weblogs.com
(I had julian.weblogs.com but sadly it all died).
|
|
Mybloglog
|
UGC. Instant blog community on your own site. Drives traffic and
builds a sense of community on a blog almost instantly.
|
Honourable mentions
- Reviewcentre.com
I
go here regularly so I have to put it up there but its execution is fairly
lame. For peoples' reviews of products. Kind of like epinions.
- Zopa
P2P moneylending marketplace. Cool idea. Although I'm a
member not had a lot of cash to invest recently but keen to use it when that
opportunity arises.
- Pandora.
Music with opinions. Caches MP3s on your machine so you don't need a 100%
reliable connection. I regularly find lots of stuff it likes. Slick gui
(unnecessarily ‘slidey' however) - it's only part web 2.0 as the core of their
‘dna' is brute force. There are others like iLike etc which are ok but Pandora still brings the great hits to my ears.
|
-
The spreadsheet market has been stagnant for a long, long time. Excel at its core has been fundamentally the same for the last 10 years. OpenOffice's spreadsheet tool just blindly copied it, and Google Sheets is just a dumbed down, otherwise uninteresting wannabe. DabbleDB however is a fresh look at how spreadsheets should work.
I've long seen Excel as a kind of 'database prototype' space -- people
muck with semi-structured information -- when those structures stabilise they would then naturally go into a database -- but they don't. Why?
Because most databases are just too difficult for people to understand, even really slick ones like FileMaker (which allows you to create multichoice fields without a master/detail relationship....), and, critically, the support for migrating from loose spreadsheet structures to more strict database / table structures is hugely lacking. Dabble fits this gap beautifully, swinging a serious punch at Excel and all its old school clones. 
How DabbleDB works DabbleDB is incredibly good at taking textual data and manipulating it to represent rich formats (in tech jargon -- refactoring data). Two great bits in the demo; the following are drop-dead easy:
- Moving a repeating field into a separate 'category' (aka database table)
- Representing information with time values in a calendar view (Excel eat your heart out!)
- All views can be exposed as RSS feeds.
- All data can be edited collaboratively, with a log of all changes (borrowing from wikis)
I must apologise if I sound a little evangelical, but as I mentioned in previous posts, I really dig elegant interfaces, things well made, things which have clearly had love and attention. DabbleDB is the kind of app that makes you sigh and think, thank God there is still some innovation in the world -- and more interestingly, hope that there is yet a way to challenge Microsoft Office without spending billions. (If DabbleDB cost more than $500k to make, something's wrong.) Where DabbleDB should go - Extend its 'rich type' concept to support more types -- e.g. allow just 'time only' fields without needing a date.
- There's mention of a mashup (aka SOAP) interface -- this will mean that you could expose and relate your data into other systems, such as google maps (no brainer there). In fact, why not have a rich data type which is a geolocation in DabbleDB. Or have types be pluggable via other mashups. Now there's power.
- Tie it up with a wiki for content. Then you could build bespoke content management systems easily. Then expose it to javascript so you can write your own situational apps (IBMese for 'apps created quickly for specific purposes written by people with no programming expertise -- a kind of 'visual script' concept).
Undo! Like a true office app, DabbleDB has undo. This is definitely a significant delineator between the 'web page app' concept and the 'rich internet app'. Undo is so important to preserve integrity, but so rarely implemented. And it's not like technically it's that hard! Programmers have got lazy, and business owners have forgotten the value customers place on it. The small print Just a word of warning with those interested in looking further -- they have a 1 month free trial -- for these specific terms in their T&C list you have to agree to: "You may terminate your account at any time upon ten (10) days prior notice to us" -- written notice? You kidding??
"We will do our best to preserve and protect your data, but make no
guarantees that it won’t get lost, corrupted, or inadvertantly leaked" -- Ouch! Go and have a Dabble
|
-
For a lot of people, Web 2.0 is a buzzword, i.e. used to "impress people with the pretence of knowledge" [Wikipedia]. There are snake oil salesmen in every bazaar, but there really is a wow factor at the core of Web 2.0 -- Web 2.0 is what I call a transformational principle, or more simply, a transformer, in that it requires a step change in the thinking about how business gets done. For Web 2.0, it's the cumulative change that has steadily occurred in the last 5 years. This is why it's so important for businesses to sit up and listen.
There are tonnes of articles out there that try to cover all the possible ways in which applications could possibly be labelled Web 2.0 -- but to me, the most interesting unique and importantly highest impact aspects of Web 2.0 also happen to be two passions of mine: rich user experience and democratic empowerment. In this entry I'll wax lyrical a little about the first, rich user experience and its companion, the Rich Internet Application (or RIA).
I've always loved interface design. Heck, you can't help it if you learn to program on a Mac which I did back in the late 80s. Bruce Tognazzini (Apple Human Interface god) was my inspiration with his two books Tog on Interface and Tog on Software Design, followed very closely by Jakob Nielsen and Donald Norman. This led me to the company I cofounded back in New Zealand in the late 1990s, Musick Point Software having a User-Centred Design service offering. I spent a lot of my time dreaming about the craft of interface design -- how to create software that gave you that warm, fuzzy feeling you get when using something has been a labour of love, where a lot of thought, effort and care has been invested. So I know interface design well, and I know about the importance of user-centred software design.
Highly usable interfaces can and have been designed inside web browsers. But fundamentally, they've had almighty limitations due to their original intent as text browsers. For an end user, they've won in many ways (e.g. what installation?), but at its core, the web browser's user interface toolkit (HTML forms) has been a spectacular retrograde step. Change a button's state as you type? Woah baby, that's unthinkable. Forget the fact that you've been able to do it in desktop applications for oh, at least 15 years! This whole idea of a 'web application' was coined, fundamentally to describe and work around these incredible limitations. And, because of the other advantages people gained, things went reasonably well. Yes it took friggin' ages to build a form because you had to get a designer to draw a picture of a button and then add a bunch of javascript that worked in 47 different browsers just to simulate highlight, but hey, that's all there was. And it was nice getting designers involved, they'd been asked with multimedia CD ROMs but not other desktop apps.
But then I saw the Oddpost email client back in March 2002. I'll just quote what I wrote back then as there's really not much else to say about it:
To describe it as an online mail reader, which technologically is exactly what it is, would be missing the point.
Oddpost is the first practical online web application with a desktop-quality user experience.
You click to launch; very quickly a real gui appears, with real menus,
drag and drop, inline editing, multiple selections, all super quick.
It's all real! It couldn't be much better if it were a local app. It's
a superb achievement.
For me this is so key: they have not innovated in functionality; nor
even have they innovated in user experience! All they have done is set a precedent for the user experience quality of web-based tools. Quite simply no one has seen this kind of quality before. This is the Internet Explorer for Windows Killer App, if in fact it needed it.[you must remember in 2002 there was no Firefox - JRH]
Web guis for web-based tools have been too long in the dark ages!
Everything about using traditional HTML for user interface widgets has
been fundamentally flawed due to its history. Don't get me wrong, there
are some great workarounds, and it's set incredible precedents but HTML
still is only good at what it was designed for: content presentation.
My feeling is that this is going to set precedents everywhere. The ante
has been upped, and this has a profound impact on any other browser and
platform. And I bet mail.yahoo.com and www.hotmail.com are quite
interested as well. Can this app be ported to other platforms and
browsers? If not, why? Remember it doesn't have to use the same
technology; just has to present the same user experience; can this be
done in Gecko? Why not? That's the big question people will be asking
now. Welcome to Internet User Experience 2.0.
As I predicted, Oddpost was subsequently bought by Yahoo, and is now Yahoo's rich email client (in good Web 2.0 form, still beta), which looks very similar, except is more portable as, again as I predicted, they ported it to support Gecko (the core engine used by Firefox). But the point is, it made me realise that we were in 2002 at the dawn of an age where this retrograde step in user experience was being restored back to its full desktop application glory: most importantly, the glory of 'live' interaction, where you drag or type stuff and get instant feedback. We're there now; this is what I mean when I say Web 2.0 is the manifestation of a rich user experience -- and what a rich internet application (RIA) is: it's simply what users had 15 years ago! It's a shame it's taken so long, but it's very exciting to rekindle the fire of rich user experience creativity that is now open to the world on the web.
In future articles, I'll open up the bonnet to describe how RIAs work a little: how Ajax / Javascript compares with Flash, Flex, and some thoughts on Microsoft's upcoming Flash/Flex killer, 'Windows Presentation Foundation / Everywhere' or WPF/E (catchy innit).
Further reading
No mucking around -- read the seminal texts on What Is Web 2.0 by Tim O'Reilly (who coined the term for a conference they were designing):
|
-
(can't embed objects on this server -- post moved to my personal blog).
|
|
|
|