Sat

Dec 8
2007

Nat Torkington

Nat Torkington

Nick Carr discovers Software Above the Level of a Single Device

"Software above the level of a single device" is Tim's phrase for software that you interact with across multiple devices. The iTunes Music Store was the classic instance of that: a store on the servers, an organization tool on your laptop, a player in your iPod. In fact, that's the example Tim used in the canonical What is Web 2.0 paper from 2004, and he's blogged about it before.

I chuckled when I read Nick Carr on the Kindle where he says:

It shows that we're rapidly approaching the time when centrally stored and managed software and data are seamlessly integrated into consumer appliances - all sorts of appliances. The problem with "Web 2.0," as a concept, is that it constrains innovation by perpetuating the assumption that the web is accessed through computing devices, whether PCs or smartphones or game consoles. As broadband, storage, and computing get ever cheaper, that assumption will be rendered obsolete. [...] The next great wave in internet innovation, in other words, won't be about creating sites on the World Wide Web; it will be about figuring out creative ways to deploy the capabilities of the World Wide Computer through both traditional and new physical products, with, from the user's point of view, "no computer or special software required."

Close, but no banana. There are plenty of problems with Web 2.0, but none of them are that nobody's thinking about devices. I had an absolute barnburner of a conversation with Matt Webb in 2005 about this at an Amsterdam Foo Camp. Matt talked about it a little in his ETech keynote (slides) this year.

In Amsterdam, Matt made me rethink the world in terms of general purpose computers and specific devices where the focus isn't on flexibility but on suitability to a purpose. E.g., very few people grumble that their mobile phone can't connect to a scanner, doesn't have 320G of storage, and fails to have a VGA connector. They grumble that it drops calls, doesn't do predictive text, and the screen gets covered in ear sweat. For each a general purpose computer does, it's worth looking to see whether there's a single-purpose device that can be made to focus on that task. Then you can customize the device and its interactions to give a great experience for that purpose.

Matt was particularly thinking of social software, and he's living true to his words now—building the Olinda, the BBC social radio. Social software is one aspect of the Internet's functionality, and as Nick pointed out, there's a strong future for single-purpose devices that connect to the Internet. If he's looking for a phrase to use to encapsulate that, he need look no further than "software above the level of a single device".


tags: web 2.0  | comments: 11   | Sphere It
submit:

 
Previous  |  Next

0 TrackBacks

TrackBack URL for this entry: http://blogs.oreilly.com/cgi-bin/mt/mt-t.cgi/6105

Comments: 11

  Nick Carr [12.08.07 06:27 AM]

I'm not sure what "software above the level of a single device" means, but assuming it means something like what I was talking about (the incorporation of centralized computing services into consumer goods), then I think actually Tim and I are in agreement, as he wrote that it's "one of the principles that is not properly understood."

  jay [12.08.07 07:07 AM]

I've commented on a few of Tim's posts, and as you point out he has a way of leading us to the future. However I believe "software above the level..." is a bit misleading.

Although it the discrepancy may not be visible to the general public, more technically minded people might point out that it's not rarely if ever one piece of software running above multiple devices.

I believe the real sense that Tim's trying to convey is "Purpose above the level of a single device".

As you illustrate, it's not one device doing everything, it's every necessary device doing something well.

I realized that may be taking the term "software" a little too literally. However, I think it's an important distinction to capture the true intent and desires.

As a customer I want _continuity of purpose_ across (or above) multiple devices, even if there's more then one software component required. And I believe rendering that desire explicit may help better guide us all.

  Tim O'Reilly [12.08.07 08:28 AM]

Nick,

If you'd read my What is Web 2.0 paper closely, you'd definitely know what this phrase means, since it's one of the key principles -- and in fact one of the core ideas of web 2.0 -- that one of the consequences of the internet as platform is that applications become ambient. I've talked about the idea that you describe in your blog post as a new one from the very beginning of my evangelism of Web 2.0.

In the original paper, I talked mainly about the iPod and iTunes, even though people don't think of these as "web" applications. I was making the point that the way that an application designed to span multiple devices (the handheld, the PC, and the internet cloud) doesn't need to have all the functions in one place. The actual term "software above the level of a single device" comes from Dave Stutz. I wrote about that recently on Radar, reminding people to re-read and think about Dave's post: http://radar.oreilly.com/archives/2007/11/software_above_single_device.html

In my talks, I've been highlighting for years as key Web 2.0 applications things like Norwich Union's "pay as you drive insurance" which builds insurance plans on automatically submitted GPS data

I think Nat's point was that in making a post that takes Web 2.0 as a straw man, and says "The problem with "Web 2.0," as a concept, is that it constrains innovation by perpetuating the assumption that the web is accessed through computing devices..." , you're creating a straw man that you then proceed to attack.

In fact, that was Nat's original backchannel message, before writing this post:


"Nick Carr discovers "software above the level of a single device" ... but doesn't realize it's been discovered before.

"The problem with Web 2.0 is ..." ... that [someone] can make up a definition and then accuse it of being lacking. Or, in this case, skip the pesky "make up a definition" and just go straight to the accusation of the strawman without even identifying the strawman in question.

This may not entirely be fair, because I think what you were really saying is that the "Web" part of the Web 2.0 term is misleading. And I certainly agree with that. That's been a long-discussed topic. Web 2.0 isn't really the web at all. And of course, we talk about p2p file sharing and 3D worlds and Google Earth and distributed computation and internet telephony as parts of Web 2.0 even though they aren't web applications either. Web has come to mean "internet" or even just "network" (as the term includes the cellphone and other devices), subsuming the old sense of internet as "network of networks."

So as to the substance, we do in fact agree. I've been talking a lot lately about ambient computing, and agree that things in that direction are becoming more and more interesting.

  Tim O'Reilly [12.08.07 08:34 AM]

jay --

I don't think that the term "software above the level of a single device" is misleading at all. I'm trying to point towards a trend to design integrated applications that indeed reside on, and depend on, multiple connected devices.

People who are stuck in the PC era tend to design for the network as if it's an add-on. People who really understand "Web 2.0" or whatever you want to call the new paradigm know that you need to design systems that inherently assume multiple devices in much the same way that a PC programmer assumes access to multiple components of the PC.

It's been my premise from the first that we're turning the internet into an operating system, and designing applications for that operating system means thinking of the system as a whole.

  Nick Carr [12.08.07 11:49 AM]

Thanks, Tim.

What I'm talking about is the common conception of Web 2.0 (which is independent of what you or anyone else may have written about the subject), which continues to reflect the "site" or "destination" metaphor, if in a new social guise, and hence perpetuates the idea that the means of access to the web is a computing device. (Indeed, I would argue that your reliance on the geeky word "device" also, if inadvertently, perpetuates this conception.) The next wave of internet innovation will, I believe, focus on funneling centralized computing, data, and software through new or traditional consumer appliances. I certainly don't suggest that I "discovered" this idea - that would be absurd - bit I do believe that entrepreneurs will need to free themselves of the common conception of "Web 2.0" if they're to exploit this rich opportunity.

Nick

  Tim O'Reilly [12.09.07 10:36 AM]

Nick --

As to new consumer appliances, have you looked at Chumby? (O'Reilly Alphatech Ventures is an investor.)

  Tom Coates [12.09.07 11:21 PM]

I really hate dissing people in public, but Nick's rhetoric in his piece is absolutely predictable and utterly frustrating to me. It must take a particularly complex and intricate architecture of self-deception to continue to write the kind of pieces he writes - normally epic pieces dismissing the rhetoric of whatever is coming out of the technologist environment as being absurd, utopian, clumsy or representing some naive and debased form of idealism. For some reason he has managed to find an audience for this stuff, which absolutely stuns me and has led me to observe—as I do again now—that naysaying and decrying new developments as a debasement of everything that has come before is an easy way to look serious and intelligent without any of the labour of actually having to think!

To say that Web 2.0 rhetoric constrains innovation by keeping people focused on the browser is laughable as it only requires you to look at any of the examples that anyone cites when they talk about Web 2.0. You don't actually have to have read Tim's piece, or have gone to any conferences or seen anything I've talked about in public over the last three years to see that it's a specious sentiment, but if you had I'd be stunned if you could still make it.

I look at Flickr and I see phone clients (updaters and downloaders), I see Aperture plugins, I see data expressed beyond the single site and manifesting all over the web. I look at Last.fm and I see senses extruding into other software (iTunes) and manifesting in desktop applications. I see the Apple TV sitting under people's televisions connecting to the same data sources as iTunes and the portable phone-based iTunes music store. I see widgets everywhere, each expressing data beyond the browser, whether on a computer or a phone.

The whole bloody point of most of this stuff, as well argued by many people more intelligent and able than I, is that there are useful and repurposable bits of information being created and manipulated by powerful new interfaces online, pretty much all of which are essentially device independent. The value is in the data, and the product extends in every direction as far as the network touches. That will obviously include more physical appliances as they become more connectable and connected.

While it might conceivably be reasonable to claim that such an understanding of Web 2.0 might not be in the popular consciousness, I find it ridiculous that someone who has spent the last couple of years decrying the term (and social software and wikipedia and pretty much everything else that's gradually become mainstream) could have missed it along the way.

My experience has been that in the UK there's considerable resistance and suspicion among the business classes to ideas from the technology sector. Engineers and technologists are perceived as 'back room boffins' with a limited grasp of reality and no sense of business. Nick's rhetoric is absolutely typical of this type of person, who are probably his audience. They are—in my opinion—one of the most substantial blights on the UK technology scene and one of the most significant reasons why very few decent new start-ups have emerged out of London since the beginning of the web.

Apologies for getting so angry about this, but this kind of ignorance and grumbling is a blight I've experienced a lot in the UK. It's one of the reasons I've left.

  steve [12.10.07 12:22 AM]

That kind of criticism can be levelled at just about any industry pundit, including Radar posters. What they're seeing as something worth commenting about is what is obvious to somebody else.

While I may snipe from the sidelines, I have to conclude that I would do no better myself. I too have things that are new to me, yet obvious to other people. You would also be annoyed by me.

Yes, I find Nick Carr to be obvious and annoying. I also find some Radar posters to be obvious and annoying. Even Tim sometimes comes into that category with a few of his posts. I even think the same of my own posts when I read them later.

So smeg it, just scan and move on. Read it when it's interesting, ignore it when it's not. Like Fidonet, don't be annoying and don't be easily annoyed. Haven't found better advice than that.

  jay [12.10.07 09:03 AM]

Tim,
When I hear "Software...", especially in the context of iTunes I visualize iTunes on the computer managing an embedded device like an iPod. In my mind this represents sort of a "master-slave" relationship, where the master (iTunes) runs on a PC and manages UI/Data on the PC and then pushes the user's intent to the embedded device(s).

I say software is the wrong word because I believe it depicts this type of expectation, i.e. that it's sufficient for a product to explicitly manage multiple "environments" and that the control can be centralized.

I believe the focus should be on "purpose beyond a single device" because it shouldn't require one "master" piece of software. Or necessitate a pre-ordained accepted exchange (i.e. anything but iTunes is a hack).

Consider the Facebook library application where you list the books you've read (or want to read). I believe this should be integrated with Amazon's wishlist.

In a "Software above the level..." world I believe the interpretation will result in the app taking a push or "slurp" mentality.

In the "Purpose above the level..." world, I visualized a much more dynamic exchange where two SW products accept that they need to openly communicate, not explicitly to each other but with any potential SW product.

In this world, my booklists exchange information and I can add a book in either one and the other can be made aware of this exchange. There's not even an "only two" restriction and another application might participate in this data exchange.

It's the desire to "keep a booklist" that matters, not which piece of SW I keep it on. I believe that's the form of the future.

  Vlad [12.10.07 02:43 PM]

Just to add another to this list that 'lives above the level of a single device'. Nike+, the Nike Apple collaboration. Very impressive application that lives on the web but is driven by your running experience in the physical world. As a user it is experienced in iTunes, at Nikeplus.com and on your iPod. It is also global, US, Europe and Japan.

  Nick Carr [12.13.07 05:31 AM]

Tom,

I'm happy to give you an opportunity to vent your frustrations, and I'm sorry about your bad experiences in your home country. But even in your comment here you display the same narrow conception of the net that I described as a constraint on innovation, particularly when you write:

>>I look at Flickr and I see phone clients (updaters and downloaders), I see Aperture plugins, I see data expressed beyond the single site and manifesting all over the web. I look at Last.fm and I see senses extruding into other software (iTunes) and manifesting in desktop applications. I see the Apple TV sitting under people's televisions connecting to the same data sources as iTunes and the portable phone-based iTunes music store. I see widgets everywhere, each expressing data beyond the browser, whether on a computer or a phone.

Post A Comment:

 (please be patient, comments may take awhile to post)






Type the characters you see in the picture above.

RECOMMENDED FOR YOU