Thursday, September 27, 2007

Holdout client applications

If the web is so great, why are there still client application hold-outs?

Look at the popular PC applications of recent years: Skype, IM, Office 2007, iTunes, games. Why have they followed the client approach?

ApplicationWhy client?In two years
Skype, IM Skype and IM require real-time communication between two parties. That's something that the HTTP request-response model of the web will never solve Browser plug-ins are incorporating real-time technologies like XMPP - this will remove the need for separate client applications
Office 2007 Office 2007 has exploited its massive installed base, a far better user interface from the previous version, and certain browser weaknesses in page layout and editing. Google Docs, Zoho and Wikis are already encroaching on Office's territory. These businesses will expand massively in the next couple of years, forcing less dominant Microsoft to lower prices.
iTunes iTunes relies on an offline model, using downloads to iPods. It also relies on DRM, which browsers are not equipped to handle. Both of these dependencies are crumbling (iPods are getting Wifi access, and DRM is fading) - I wouldn't be surprised if iTunes was replaced by iTunes.com soon.
Games Most games rely on advanced graphics and animations, which browsers are not designed to support I can't see browsers competing in the next few years, except perhaps for simple games like Tetris

Conclusion: except for games requiring powerful graphics engines, the web will continue to replace today's common client applications.

Saturday, September 22, 2007

Hardware and Software

Throughout their histories, Microsoft and Apple have had different strategies. Microsoft sells only software (WIndows, Office) but Apple sells hardware too (iMac, iPod, iPhone).

For a long time, Microsoft had the better strategy. It could focus on one thing - software - and build a dominating position in the space where most of the technology innovation happened. It could then dominate an ecosystem of hardware partners to help grow the PC market at double-digit growth.

Times have changed, leaving Apple's approach looking better (at least on the client side):

  • New form factors arose - e.g. the smartphone - that require different types of software, focused on battery power and communications
  • Hardware innovation became as important as software innovation, for example the iPhone touch screen, camera phone, satellite navigation, and Wii controller, leaving Microsoft struggling to catch up
  • The establishment of common standards (e.g. HTML, ODF, USB) and common user interface paradigms removed barriers to change for the end user
  • Open source competed with much proprietary software, but not with proprietary hardware
  • PC operating systems became commoditised; what you really need is a good browser, and a few basic productivity apps

The biggest change is surely the internet - the client OS is becoming the browser. That means on the client side, hardware is just as important than software.

News in: Microsoft reconsiders its software only approach.

An integrated business model of making both device and software could make sense, executive tells investors at tech conference. Microsoft said on Tuesday that it is "not unreasonable" for the company to introduce a mobile phone combined with features of its Zune digital music player to compete with Apple's iPhone....

If that's true, Microsoft will have to turn the company on its head (again). And what about making their own laptops and PCs? Somehow I can't see it...

Friday, September 14, 2007

gPhone based on Google Gears

The only way I can understand current gPhone rumours is if the gPhone is based on Google Gears.

The gPhone is Google's allegedly forthcoming mobile software platform. It's very unclear what it'll actually do, but it sounds like an API for Google and third party providers, presumably for applications such as maps, emails, photos, and videos.

But isn't Google an internet company?

Trouble is, this doesn't sound very like Google. Didn't Eric Schmidt say "don't bet against the internet"? What are they doing building an API for the client? Can't they continue their strategy of developing stripped-down websites in HTML for small mobile browsers?

Two major problems with mobile browsers are the lack of bandwidth and the intermittant connections. That's where Google Gears comes in - I think it was designed with phones specifically in mind.

That 8Gb phone memory is just a cache!

For example, your phone's photo application could be a link to www.google.com/photos. If you're in a Wifi or HSDPA zone, the site will be pulled up over the net (and synched with your phone). If the connection speed is slower, then the local cache from your phone memory will display instead, using Google Gears.

Similarly for email, maps, and even your phonebook contacts - to access them, you visit an internet site, but if the connection speed is too slow, the local cache is displayed instead via Google Gears.

Now it makes sense

Using Google Gears, gPhone application development uses just URI, HTML, and javascript. All gPhone apps sit on the internet, and data is stored locally when your mobile browser accesses the site.

Google is still betting on the internet - in fact it's making it work in an environment where network speed is uncertain.

Wednesday, September 12, 2007

Web Art

Walking through Greenwich market in London recently, I passed lots of stalls selling jewellery, cards, fabrics, and various pieces of art.

It occurred to me that "web art" - digital multimedia produced by an artist and published online - is still in its infancy (with the exception, perhaps, of photography).

Where is the art exchange selling beautiful SVG clockfaces for my desktop or wristwatch, or abstract paintings for my digital photoframe? Where can I purchase thoughtful and arty online cards, or "moody" background videos for plasma screens?

This must be partly down to artists preferring traditional physical materials - and to be honest, given current display and design technology, I don't blame them. But in a couple of years, the market will be there.

Webifying art

Art could be the next area to be turned upside down by the internet. If every piece of virtual art has a URI, this means perfect copying - if someone creates the next Mona Lisa online, everyone can see the "original" perfectly in their browser, and download it.

It's similar to the music industry, - business models will have to change. What will it mean to "own" virtual art? Will anyone pay for it? How will the artists' rights be protected? In music, some suggest that the web will eliminate the record company - but art has never had the equivalent of record companies, and it still may have serious issues.

Of course, there will still be offline art. But it's already clear that the internet will change the world of art, just as it has for music.

Saturday, September 08, 2007

Browser-based wristwatches

Digital watches have never been the most fashionable accessory. But I think that will soon change, as display technology has developed in leaps and bounds.

Imagine if your watch contained a screen like the new iPhone, showing an arty clockface that could be touched to reveal more (e.g. date/time, news headlines, weather, your latest calendar appointments).

You could set this all up by registering at a website on your PC and customizing the look and feel - for example, setting the background to be a family photo.

The technology is already all there - touch screen, 3G data, browser displaying HTML / SVG. Finally, the digital watch can overcome its geekiness.

Sunday, September 02, 2007

Does Apple get the internet?

It might seem a silly question, but does Apple really understand the internet?

After all, they're best known for producing hardware and personal operating systems (iPhone, iMac, iPod, Mac OS) and media applications (QuickTime and iTunes). These are all client-side; apple.com is just an old-fashioned download site, and mac.com is their only attempt at a modern web application, languishing as the 800th most popular site.

Apple has had no success in converting iTunes into a community site, where people could share recommendations, music gossip and events, or post their own music. And with broadband now prevalant, why not store your music at iTunes.com, rather than on your c: drive - that way, you could access it from any computer, or directly from your iPhone.

Finally, if Apple is not careful, sites like Photobucket, Picasa and the forthcoming photoshop.com will steal its tradition in graphics.

Proving they understand the web

Even Apple isn't immune to Silicon Valley start-ups, especially those competing with their key music and graphics applications.

Apple have shown enormous flexibility in the last few years, transitioning to Intel processors and moving to touch screens from their famous clickwheel. They'll have to demonstrate it again by moving to web applications.

I've written some success measures for Apple, to demonstrate how far they've got to go in making use of the web:

  • Replace iTunes with iTunes.com, a browser-based social application
  • Enable direct iPhone access to iTunes.com
  • Create the best photo editing site

Are Apple ready to create the next great online applications for graphics and music? If not, they'll be limited to selling internet devices.

Saturday, September 01, 2007

Projects are social networks

Project management is natural for the web. That's because it's all about continually creating, maintaining and communicating project information - i.e. "collaboration" - so that the project remains on track. No one's found a better collaboration system than the web.

Project management systems have been web-based for several years now - I've had experience with Planview and several internally developed applications. But they suffer commmon flaws - data is hard to find, hard to update, and too complicated (especially around time management, approvals and project workflow).

What's missing is Web 2.0 - i.e. the use of simple, social web technologies. This will bring less focus on formal workflow, and more focus on straightforward collaboration. What could be easier than a Wiki-based project homepage, or a social network containing the project team?

Below are some ideas for improvement that I haven't seen in any existing project management system.

RequirementWeb Technology
communicate project informationWiki
maintain project planWiki (using SVG / VML for Gantt view, hyperlinks for dependencies)
track status changesWiki versioning
staff notificationsRSS feeds
time managementMicroformat integration with calendar
system integration (e.g. with a financial or CAD tool)RSS mashups or open web APIs

One unexplored opportunity is microformats. A project management system that expressed project plans using microformats would express the who, when and where in a machine-readable format, so all sorts of possibilities open up - linking to mapping software, people's calendars, or a corporate directory, for example.

Another relatively unexplored opportunity for project management software is RSS (or Atom). Staff would subscribe to receive notifications when key pieces of information change (for example, project risks or milestone changes).

And the work breakdown structure (a.k.a. plan) is just a widget that uses microformats to integrate with people's calendars and maps.

Projects are social networks

What I like about these ideas is that they are simple ways to directly support project managers using today's technology.

Enterprise Project Management systems started out as monolithic client-server applications. What they're turning into is social networks, because that's what a project team is!