Web Services and Devices
We are witnessing massive innovation in and around web services. It seems that rarely a day goes by without some new web service being launched.
And the same is true with devices of all sorts. We have Microsoft’s Origami device being discussed in the blogs this weekend before it’s even announced by Redmond. We have Amazon’s new music device to consider, and the Samsung Z5 which features software designed by the same guy who did the iPod’s software. And of course we have new cellphones, PDAs, game players, and digital cameras coming to market in ever increasing frequency.
Yet there is something missing from all this innovation and that is the integration between these new web services and these new devices. Sure you can upload the photos you shot with your cool Sanyo Xacti video/still camera to Flickr, but its not seamless. You can watch YouTube videos on your PSP, but it takes some work. You can put the songs you bought on iTunes in your iPod, but not your new Z5.
The problems are both technical and political. Web services are open for the most part, but some of the most popular ones, like iTunes (which isn’t even a web service) are not. Devices are almost always closed.
Much of the discussion regarding Amazon’s digital music strategy as well as the debate over the Z5 revolve around whether consumers require an integrated/turnkey solution. Can consumers figure out how to mix and match web services and devices or must they be served up a complete solution by a single vendor?
I think we all hope that a truly open environment will prevail where devices and services can play together and best of breed will win out on both fronts. But that is not where we are today and we have a lot of ground to cover in order to see that vision realized. And the powers that be don’t want it to happen. The businesses whose margins depend on their continued control over a distribution channel – carriers, media companies, and consumer electronic manufacturers - will find it hard to embrace innovation that potentially undermines their margins.
But if the experience of the personal computer and the interenet is a guide, open architectures where many players compete in a decentralized ecosystem do accelerate innovation. When web services integrate with devices in a open architecture, we believe that the consumers' interest in innovation and integration will trump the vendors' interest in preserving control over limited proprietary channels.
If you agree and are working on any aspect of this opportunity, we’d like to hear from you.
February 27, 2006 02:45 PM, By Fred Wilson
Tags: ce devices webservices
Comments (10)
QUOTED: "Where are the web services that for example synchronize my address with my desktop, smartphone, home appliances, other applications like shipping sites?"
Out of interest our group is steadily building the ubiquitous functionality you talk of, all via web services. My vision is to be able to access my business critical information (ie address book, documents, sales pipeline) from any device without having to think about it. I want my "stuff" with me, wherever I am and on whatever device is handy.
With this in mind, we've gone about building a platform based on multiple web services all working together to deliver meaningful data. Today this is available over a web interface. Within the next few months we'll have a thin desktop client, a thin MS 2005 client and a java client. All integrated into your local Outlook, Exchange, Quicken etc and all working without the user having to manage settings and different environments.
As Fred stated, there is no single, open environment for devices. We'd pretty much need to code a client for each environment. For example, we haven't yet been able to figure out how to get our business applications accessible on an iPod. The calendar integrates just fine, but synching through documents, sales reports etc is just going to be a lot of work.
Posted by Daniel Barnett , February 28, 2006 03:24 AM
A seamless experience across all services, devices and networks is a driving goal behind the recent “convergence” initiatives, most notably IMS. With wireless services, service creation, provisioning and deployment are much harder and more expensive than in the wired Internet. As a consequence, carriers need business plans for services that address a large number of customers to justify the expense. One of our visions is to make the economics of service creation and deployment scale down below the 1K customer range. Imagine local vendors being able to create wireless web services tailored to their specific clientele and have them delivered and integrated naturally over Sprint’s, Verizon’s, etc., networks. While there is much hype around IMS, it’s potential to open up service creation and delivery in wireless networks has real potential.
Posted by Dave Famolari , February 28, 2006 08:46 AM
Regarding Zack's statement above:
"The ability to drag and drop web services onto a dashboard that is pervasive from laptop to devices is key. It shouldn't matter about the device it should be more the dashboard that gives your ubiquitous access to your information or songs."
This is what we are working on at Grimaldi Productions. Our flagship product is called Mentations. You can download a public preview version to see our progress (right-click on things to see available options).
Posted by Brian Schneeberg , February 28, 2006 09:22 AM
Fred, seems like your flushing a lot of interesting ventures out of the bushes w/ this post. Now we need a similar universal platform for media. I want my HBO subscription, my Sirius subscription, and whatever other media services I subscribe to, equally available cross platform w/o having to kludge a one-off solution--business critical applications may be where the price insensitive buyer is today, but leisure applications are where the people are who turned the iPod into the runaway hit that it is.
An instructive analog is MIDI--a universal standard that allowed the then-new digital musical devices to interface w/ one another that was manufacturer agnostic. The engineers managed to sneak the standard past competing manufacturers and, voila, the business of digital music devices bloomed. But in consumer facing digital media today we're still stuck w/ format wars (Blu-Ray vs. HD-DVD, for example) or bespoke platforms (iTunes), that, for all their buzz (and in the case of iTunes, success) are ultimately a hinderence not a boon to the growth of the biz
Posted by Jason Chervokas , February 28, 2006 02:05 PM
Your definition of "web service" is all over the map here Fred, and mostly incorrect.
Flickr and YouTube are websites, not web services. To get a website on any form-factor is simple ... just add a browser. Size is your major limitation here.
A "web service" is a software engine that is accessed via a web protocol (originally, web services == SOAP, but thankfully that's expanded to all integration protocols). A web service exposes its content and functionality to other devices over the internet via a predefined interface, namely an API.
Simple example ... RSS is a web service ... your blog is a web site. Your blog is only viewable via a browser, while your content is viewable on any device that "speaks" RSS ... be that other websites, or subscriber's iPods.
Web services are actually not the problem, quite the contrary, they are the first step to making content and functionality available everywhere ... because they do not rely on a browser to expose functionality previously limited to being distributed by (one) website.
This difference (between sites & servces) is important & is often misunderstood by web 2.0 users - because sites now provide so many cool "sevices".
The challenge in executing on your utopia is immense ... adding a web service to a device is like teaching a person a new language ... there's service-specific translation & collaboration expertise required to "adopt" any service in any device. From a support perspective, all problems are amplified because your product now constitutes features and products beyond your control ... bottom-line, any-service interoperability is a VERY EXPENSIVE product.
What's are the solutions? Simple --> standards or leverage.
Industry-wide standards ... or industry-leading leverage.
Standards like RSS (... check out SSE) are fascinating but not VC-fodder.
So, in this space, as a VC you're left needing to pick the next iPOD ... good luck!
Posted by David G , March 1, 2006 12:19 AM
The guys at RIM have figured this out.
http://www.eweek.com/article2/0,1759,1932933,00.asp
We're also doing some work in this area, but with more of an RSS focus.
Posted by Charlie Wood , March 2, 2006 11:40 AM
David G, I think Fred is talking about web 'services' in the abstract sense of functionality provided through the web, independant of the access mechanism (e.g. DHTML/REST/RPC). Many would also disagree that Flickr is not a web-service (in the technical sense of a programmatic HTTP-based interface)—in fact most of their web pages and Flash employ their API; the site is therefore an interface layer built upon the web-service. YouTube also functions as a web-service for video content through their own API (which unfortunately, does not extend to uploading). However this is all academic.
Fred, I'm glad to see that others recognise the issues surrounding facilitating the movement of personal assets amongst services and devices. I dropped the USV feed in a purge a while ago beacuse I didn't feel it was relevant to me, but I'm happy to discover I'm wrong and reinstate it.
One of the key elements restricting the adoption of web-services is findability and interchangeability (enabling data-portability), and this as you observe, results in applications that are tied into specific functions (and APIs). You rightly question whether consumers want the flexibility, and to look at Apple you could easily say no. But I don't think there's a right or wrong answer. As time and technology lowers the barriers, no doubt it will become entirely unacceptable for anyone to be locked into any single path or solution (presently, it's just an inconvenience).
I've been working on a simple specification (for media-sharing services) that promotes [web-service] API discoverability to try and avoid the issue, and ultimately enable easier consumer adoption of applications utilising web services. In the short to mid term, I believe that the desktop (Operating System, whether PC or portable) environment provides the best platform for aggregating a user's assets/identity—from devices and web-services. Not least because visualisation and management [will] need richer interfaces following the increases in production and consumption of medias. This aligns with the concept of the digital-hub (where a single device takes a key role), and fulfils the ideal of Digital Lifestyle Aggregation—but wherein the web services perform only ancillary roles (mainly of transport and distribution). Certainly the future web will offer all of this, but for now it's not quite there…until then I'm sure we'll be seeing more 'desktop' solutions to these issues emerge (and indeed I'm working on one such direction).
Posted by Jacob , March 2, 2006 12:09 PM
Jacob - why not just use wsdl? If we're going the standards route, it really helps if everyone is on the same page - what we don't need right now is more standards.
Posted by David G , March 3, 2006 04:15 PM
fred,
you invested in fusionone with Flatiron which was seeking to do something of the sort. How is that investment doing ?
Posted by mike , March 8, 2006 03:04 PM
The ability to drag and drop web services onto a dashboard that is pervasive from laptop to devices is key. It shouldn't matter about the device it should be more the dashboard that gives your ubiquitous access to your information or songs.
Where are the web services that for example synchronize my address with my desktop, smartphone, home appliances, other applications like shipping sites? We are far away unfortunately but moving incrementally.
Posted by Zack Perry , February 27, 2006 08:54 PM