John Gage is famous in Silicon Valley for coining the phrase "The Network is the Computer" way back in the 90's. Sun Microsystems took the phrase as it's product management mantra a couple years later. This was a bit of a leap for most people in Sili-Valley, leading to a number of jokes at Sun's expense (my favorite was a t-shirt that read "No wait.. the network is the network, the computer is the computer. Sorry for the confusion.") Gage may be eventually proved right (though Sun might have been decades ahead of the market on this one.)
Software as a Service (SaaS) has been a viable mechanism for delivering solutions to customers for at least a decade. The number of Application Service Providers (ASPs) are growing at least as fast as Independent Software Vendors (ISVs) focused on desktop systems and if we are to believe the trade media, corporate IT managers are clamoring for more Cloud-Based solutions. ISV growth is focused on mobile platforms (almost exclusively Apple iOS and Google Android.)
But maybe there's a future for the mobile web. Maybe the idea of buying and explicitly downloading a an app will go the way of the floppy disk. There are several very good economic reasons for corporate users to move to the mobile web; if ASPs can deliver half-way decent experiences via the mobile web, they have a good shot at taking a chunk of the mobile and desktop app market away from existing ISVs.
And who knows... maybe the next desktop computer you buy will run a web-oriented operating system like Google's Chrome OS or Mozilla's Firefox OS.
So what is a Cloud-Based Operating System anyway?
Ask a hundred product managers what a "cloud based" operating system is, and you'll probably get one hundred different answers. There are a lot of people who use the term to describe customized Linux-based OS distributions optimized to run a browser (and little else.) Google's Chrome OS is the exemplar of this class of distros, but it's certainly not the only one. Other people will tell you the term "Cloud OS" means a particular web application running in a browser that lets you perform typical tasks like editing documents, sorting pictures and even sending documents off to be printed.
But I think I agree with John Gage on this one; "the" computer is no longer the one on your desktop (or in your hand.) If you think of "the computer" as that thing that performs data manipulation tasks on your behalf, it's now spread out across your desktop, your phone, your tablet, your home router and any number of servers across the network. So if "the computer" is now spread across so many different systems, then "the operating system" is too.
Nomenclature fails us in this regard; in the mind of a typical user, terms like "computer" and "operating system" have evolved beyond their technical definitions. "Computer" is synonymous with "Desktop Personal Computer" or "Laptop Computer" these days. You rarely hear people refer to their smartphone or game console as "a computer." And what do most people know about operating systems? Not that they're privileged code bases which manage system resources, but that "an OS is the thing that draws windows on the screen and determines which apps you can run."
From a technical perspective, it seems wrong to call a collection of application software running on a constellation of computers an "Operating System," but from the user's point of view, that might be the most salient feature of future systems. In the future, users may stop looking for the Windows or MacOS X logos on shrink-wrap boxes and start looking for the "Salesforce Compatible Data Source" logo on various web pages.
In the future, when people talk about "the operating system," i believe they'll be talking about network APIs that let systems from different organizations safely share a user's data for the purpose of doing something useful for that user.
No, Seriously, What do I install on my PC?
In the future, I think there will be less "installing" going on. I believe you'll browse to an application's web page, click the "I Agree" button on the EULA and User Interface components will magically appear on your devices. Where we now have App Stores, I think we'll have independent reputation app databases in the future. As part of the process of "installing" a future Cloud-Based OS app, you can configure your user interfaces to check one or several of these reputation databases to ensure they're not malware (or that they don't use bad language, or that they don't show naked people, or ...)
People have long since stopped installing operating systems on hardware (with the exception of power users and Linux fanatics.) PC-based video games are now increasingly distributed online (see Valve Steam or search Amazon for "software download.") Even with bandwidth, security and server costs, online distribution is much cheaper than putting DVD-ROMs in physical boxes. The only reason to distribute your software through a physical retail chain is to appeal to those people who still prefer to buy "things" in computer stores.
In the future, we may see the thin client market evolve into a "slightly thicker client" market. Thin-client class hardware might be fitted with an embedded Linux operating system which boots into a browser. The browser's home page would be set to Google Apps, Facebook, Salesforce or a "Web Desktop" like Glide or G.ho.st. Maybe Microsoft or Oracle will leverage their relationships with enterprise customers to build a "Enterprise" Web Desktop.
Or maybe Community ISPs can become relevant again by selling cheap thick clients that point at Web Desktops served by their systems. Ditto for wireless operators.
What do you install on your PC of the future? Nothing. It's all over the net.
Where does my data go? Where does it rest?
So hopefully you've been asking yourself, "Hey!? Who's got my data!?" Imagine me waving my hands as my eyes glaze over and I solemnly intone, "It's in the Cloud! Beyond the confines of physical reality!"
I would hope that at this point, you would be thinking of how many seemingly decent ASPs can't seem to properly hash your password. There are true risks to putting all of your data out in the cloud. Luckily for web application operators, it's easier to quantify beneficial cost savings than detrimental security risks.
There is no simple answer to "how much risk can I tolerate?" Enterprises will (no doubt) insist that sensitive corporate information be stored on their own servers or those owned by trusted third parties. Individual consumers, who may be using the web to share data with friends and family, may be more risk tolerant.
Any viable web of the future must be flexible enough to support the high-assurance requirements for enterprise customers and the cost-sensitive nature of the consumer market.
Value Added Interfaces
If you go to any typical online application today, the way you get data into the application is by direct user input or uploading files from your PC. And yes, I do mean PC. Even though it's possible to upload photographs and videos from tablets and smartphones, uploading other data files is a less than compelling user experience.
In the future, I believe we will see "Value Added Interfaces." These will be RESTful Web APIs applications use to share data on a user's behalf. For example, imagine you had a mailing list in a spreadsheet on Google Drive. Ideally, you should be able to give a Google Drive URL to a network based printing house who will do a mail merge for you. The printing house's software queries the Google Drive URL for a CSV formatted address book and "does the right thing."
I believe it will be possible to effectively secure these interfaces; even sensitive data will be available to authorized consumers. In the future, payroll companies like ADP may provide an interface directly to Intuit so your personal tax information is fed directly into the Intuit app. Or Exxon will provide details of fuel purchases directly into your corporate accounting system.
If we want to enable a "all my internet of things data is in the cloud" future, we'll need to move past our reliance on file uploads. They are a legacy of the PC era.
What will I see when I log in?
I have to admit, I don't like the term "Web Desktop." It enforces a PC era model on the cloud-based mobile data future. I don't like the idea of a grid of applications 'cause many times it's useful to think of documents or tasks or modes of thought. Twitter is talking about expanding the types of "cards" they use to summarize information in tweets. Maybe a stack of twitter-esque cards would be a suitable interface? Maybe a "dashboard" summarizing information and tasks of interest to you?
I don't know what you'll see on your screen when you log-in in 2018. Whatever it is though, it will likely be rendered using JavaScript and HTML5.
Software as a Service (SaaS) has been a viable mechanism for delivering solutions to customers for at least a decade. The number of Application Service Providers (ASPs) are growing at least as fast as Independent Software Vendors (ISVs) focused on desktop systems and if we are to believe the trade media, corporate IT managers are clamoring for more Cloud-Based solutions. ISV growth is focused on mobile platforms (almost exclusively Apple iOS and Google Android.)
But maybe there's a future for the mobile web. Maybe the idea of buying and explicitly downloading a an app will go the way of the floppy disk. There are several very good economic reasons for corporate users to move to the mobile web; if ASPs can deliver half-way decent experiences via the mobile web, they have a good shot at taking a chunk of the mobile and desktop app market away from existing ISVs.
And who knows... maybe the next desktop computer you buy will run a web-oriented operating system like Google's Chrome OS or Mozilla's Firefox OS.
So what is a Cloud-Based Operating System anyway?
Ask a hundred product managers what a "cloud based" operating system is, and you'll probably get one hundred different answers. There are a lot of people who use the term to describe customized Linux-based OS distributions optimized to run a browser (and little else.) Google's Chrome OS is the exemplar of this class of distros, but it's certainly not the only one. Other people will tell you the term "Cloud OS" means a particular web application running in a browser that lets you perform typical tasks like editing documents, sorting pictures and even sending documents off to be printed.
But I think I agree with John Gage on this one; "the" computer is no longer the one on your desktop (or in your hand.) If you think of "the computer" as that thing that performs data manipulation tasks on your behalf, it's now spread out across your desktop, your phone, your tablet, your home router and any number of servers across the network. So if "the computer" is now spread across so many different systems, then "the operating system" is too.
Nomenclature fails us in this regard; in the mind of a typical user, terms like "computer" and "operating system" have evolved beyond their technical definitions. "Computer" is synonymous with "Desktop Personal Computer" or "Laptop Computer" these days. You rarely hear people refer to their smartphone or game console as "a computer." And what do most people know about operating systems? Not that they're privileged code bases which manage system resources, but that "an OS is the thing that draws windows on the screen and determines which apps you can run."
From a technical perspective, it seems wrong to call a collection of application software running on a constellation of computers an "Operating System," but from the user's point of view, that might be the most salient feature of future systems. In the future, users may stop looking for the Windows or MacOS X logos on shrink-wrap boxes and start looking for the "Salesforce Compatible Data Source" logo on various web pages.
In the future, when people talk about "the operating system," i believe they'll be talking about network APIs that let systems from different organizations safely share a user's data for the purpose of doing something useful for that user.
No, Seriously, What do I install on my PC?
In the future, I think there will be less "installing" going on. I believe you'll browse to an application's web page, click the "I Agree" button on the EULA and User Interface components will magically appear on your devices. Where we now have App Stores, I think we'll have independent reputation app databases in the future. As part of the process of "installing" a future Cloud-Based OS app, you can configure your user interfaces to check one or several of these reputation databases to ensure they're not malware (or that they don't use bad language, or that they don't show naked people, or ...)
People have long since stopped installing operating systems on hardware (with the exception of power users and Linux fanatics.) PC-based video games are now increasingly distributed online (see Valve Steam or search Amazon for "software download.") Even with bandwidth, security and server costs, online distribution is much cheaper than putting DVD-ROMs in physical boxes. The only reason to distribute your software through a physical retail chain is to appeal to those people who still prefer to buy "things" in computer stores.
In the future, we may see the thin client market evolve into a "slightly thicker client" market. Thin-client class hardware might be fitted with an embedded Linux operating system which boots into a browser. The browser's home page would be set to Google Apps, Facebook, Salesforce or a "Web Desktop" like Glide or G.ho.st. Maybe Microsoft or Oracle will leverage their relationships with enterprise customers to build a "Enterprise" Web Desktop.
Or maybe Community ISPs can become relevant again by selling cheap thick clients that point at Web Desktops served by their systems. Ditto for wireless operators.
What do you install on your PC of the future? Nothing. It's all over the net.
Where does my data go? Where does it rest?
So hopefully you've been asking yourself, "Hey!? Who's got my data!?" Imagine me waving my hands as my eyes glaze over and I solemnly intone, "It's in the Cloud! Beyond the confines of physical reality!"
I would hope that at this point, you would be thinking of how many seemingly decent ASPs can't seem to properly hash your password. There are true risks to putting all of your data out in the cloud. Luckily for web application operators, it's easier to quantify beneficial cost savings than detrimental security risks.
There is no simple answer to "how much risk can I tolerate?" Enterprises will (no doubt) insist that sensitive corporate information be stored on their own servers or those owned by trusted third parties. Individual consumers, who may be using the web to share data with friends and family, may be more risk tolerant.
Any viable web of the future must be flexible enough to support the high-assurance requirements for enterprise customers and the cost-sensitive nature of the consumer market.
Value Added Interfaces
If you go to any typical online application today, the way you get data into the application is by direct user input or uploading files from your PC. And yes, I do mean PC. Even though it's possible to upload photographs and videos from tablets and smartphones, uploading other data files is a less than compelling user experience.
In the future, I believe we will see "Value Added Interfaces." These will be RESTful Web APIs applications use to share data on a user's behalf. For example, imagine you had a mailing list in a spreadsheet on Google Drive. Ideally, you should be able to give a Google Drive URL to a network based printing house who will do a mail merge for you. The printing house's software queries the Google Drive URL for a CSV formatted address book and "does the right thing."
I believe it will be possible to effectively secure these interfaces; even sensitive data will be available to authorized consumers. In the future, payroll companies like ADP may provide an interface directly to Intuit so your personal tax information is fed directly into the Intuit app. Or Exxon will provide details of fuel purchases directly into your corporate accounting system.
If we want to enable a "all my internet of things data is in the cloud" future, we'll need to move past our reliance on file uploads. They are a legacy of the PC era.
What will I see when I log in?
I have to admit, I don't like the term "Web Desktop." It enforces a PC era model on the cloud-based mobile data future. I don't like the idea of a grid of applications 'cause many times it's useful to think of documents or tasks or modes of thought. Twitter is talking about expanding the types of "cards" they use to summarize information in tweets. Maybe a stack of twitter-esque cards would be a suitable interface? Maybe a "dashboard" summarizing information and tasks of interest to you?
I don't know what you'll see on your screen when you log-in in 2018. Whatever it is though, it will likely be rendered using JavaScript and HTML5.
No comments:
Post a Comment