Everything is moving into the cloud. Well not exactly everything, but you kinda know what I mean. Very many IT products are morphing into services and vanishing into an amorphous existence on the Internet, known as “The Cloud.” Let’s take stock. First of all, for the sake of perspective, I’ll list some well known capabilities that either were born in the cloud or quickly headed into the cloud as part of the early web. We might not think of these things as being “as a service” (aaS), but there’s no reason not to think of them that way:
- Search/Navigation (aaS): The search capabilities of Google, Yahoo!, Microsoft, Ask, etc. provide the directories of the web as a service, either as a search capability or as structured navigation. the business model is advertising-led.
- Information (aaS): The web embodies a vast information pool that opens up to some degree with the use of any search engine. However, there are also sites that specialize in providing pools of organized information (Wikipedia, Wikileaks, etc.) or organized links (Netvibes, Google News, etc.) and particularly socially-networked links (Del.icio.us, Digg, Reddit, etc.). The revenue streams are primarily through advertising.
- Market Making (aaS): These are sites that enable the sale of goods by the owner of the goods. They make the market. It includes auction sites like eBay and Taobao, which take a percentage of every sale. It also includes sites like Craig’s list and Freecycle, neither of which involve a fee of any kind.
- Payment (aaS): Paypal pretty much took the air out of this market with its brilliantly viral service - although Google is rumored to have a competitive capability in gestation. The banks missed this opportunity entirely. Which is bizarre because the “payment transaction” is their business.
- etail i.e. Retail (aaS): eTail, particularly the success of Amazon, is what launched the web as a business “el dorado” and the amount of retail on the web continues to grow year-on-year. This could be viewed as a beyond-your-imagination extension of what was previously the consumer catalog business.
Let’s now consider more recent developments in the cloud, starting in the obvious place: running hosted applications:
- Application Software (aaS): This is what most people would classify as SaaS, a trend that has been lead by Salesforce.com, which includes Oracle online (in terms of business apps) and both Zoho and Google Apps in terms of office productivity apps. Salesforce.com now has a whole software ecosystem of other related apps. Ultimately most business apps will be available in this way.
- Security (aaS): Sophos, Kaspersky, and others offer Security as a service. The actual mix of what you get varies. Spam filtering as a service was where it all began and it has been growing from there.
- Backup and Storage (aaS): On-line back-up and storage as a service has been available for quite a while. I remember meeting with LiveVault (whom I believe had the first service) about three or four years ago. More recently Mozy has been attracting attention as a from-home back-up service. Indeed, it attracted a great deal of attention, when EMC acquired it.
- Communications (aaS): You could argue that communications has always been sold as a service, with the telephone being the access device. In any event, there are now a number of web-hosted communications services. Aside from Skype, there’s Grand Central (owned by Google), Ribbit and Jajah, and probably a few others that haven’t made it onto my radar.
- Collaboration (aaS): Webex was the first collaboration as a service web site, unless you include instant messaging capability, in which case it may have been ICQ (but you could classify that under communications as a service). More relevant are Dim Dim (a free Webex service) and Second Life, the virtual world which some companies have used as a venue for virtual meetings.
- Software Lifecycle (aaS): Bungee, a company I wrote about recently, provides a software development environment and set of life-cycle tools as a hosted service. They also host the applications you create, which you pay for on a usage basis.
- Mashups (aaS): StrikeIron, another company I have written about, provides mashups as a service - normally on a paid basis. There is a variety of functionality that is offered. The Programmable Web also provides links to mashups, but most of these are free to connected to.
- Virtual Infrastructure (aaS): Finally, Desktone has a virtual PC business model that involves providing virtual PC infrastructure as a service. It involves thin clients on the desktop with cloud-hosted servers running the PCs.
Finally, lets consider the Open Source movement as a separate part of the cloud. The reason for separating it out is that really it’s all about program code and the cloud itself is a downstream thing:
- Open Source Web Engines (aaS): The software ecosystem first became of age (imho) with Firefox and its large population of plug-ins. Now we can add Wordpress, Drupal and Joomla, all of them web engines that also provide a wealth of plug-ins that truly qualify as an ecosystem. These products, and many others, are usually available ready loaded on any web hosting site. Associated web sites give you access to the plug-ins.
- Program Code (aaS): There are hundreds of thousands of projects on SourceForge, all containing free source code (free as long as you abide by the associated license) that you can download and use. It’s a shared resource that is used by many developers to cut development times, and it may not be intended as an open code library, but it is undoubtedly used as one by some programmers.
If we look at the whole picture here (I doubt I’ve included everything), it’s clear that the IT industry has become very “cloudy”. Also, irrespective of the success of any particular company I’ve mentioned, it’s hard to imagine that these different categories of cloud activity will not all continue to grow. I don’t even think that anyone disputes that all the movement is into the cloud. What is not clear is what it will look like when the full migration has occurred.
Note: Also relevant to this article is Vertica: Revolution in the Database World? which discusses a new product that delivers data warehouse as a service.
It is exciting that so many as a Service(aaS) companies are emerging, however, I am curious if all these companies can be successful. There are so many challenges that face aaS companies: customizing their solution, billing, and monetizing their service; how can all these companies overcome these challenges? Can they all overcome these challenges and be successful?
I have heard of a few resources to support aaS companies, such as eVapt (www.eVapt.com); do you know of other resources that will help truly help “everything” become aaS?
Biggest challenge for aaS company is getting economies of scale.
There are some hosting operations such as Jamcracker, that specialize in aaS type operations, but I’v enot looked at the field in depth.
IaaS or Content as a Service is interesting. I believe the business model shouldn’t just be through advertisements — companies should look into premium content as well.
Well, to tell the truth,I simply emphasized the free aspects of this. Iaas for $ is already a thriving business. Much of what StrikeIron provides is access to information for a fee.
SaaS, MSPs, and other ‘function rentals’ are all growing in popularity - classically most business focus on ‘buying’ and ’selling’ items/services. On the ’sell’ side almost everyone would agree that salesforce.com has won and that even Seibel, who’s well know reputation of ‘under delivering’ on it’s functionality promises fueled sf.com’s success, is now OnDemand as they say. On the ‘buy’ side various purchasing SaaS plays are predominant with Ariba offering either customer adoption methodology.
However, I think when looked at from a true business/finance perspective many client’s may wish to ask whether this ‘function being rented’ - CRM, Email Messaging, Information Security, ERP, etc, etc - will be needed temporarily by the business or more likely for the next 10 to 20 years. If so then one might imagine that the recurring cost model of renting the function will be less cost efficient than bringing this function in house.
[...] determining factor in connectivity in the United States. Robin Bloor’s recent post on havemacwillblog.com clearly illustrates this change to what he refers to as, [...]
This post resonates well with what I have been seeing with regard to the shifting meaning of the phrase “Digital Divide”. I’ve blogged about this at:
http://theanthrogeek.wordpress.com/2008/07/05/the-digital-sub-divide/
What I found was a recent conflation of the old use of the phrase (to imply ANY internet access) as opposed to current uses of the term (to imply HIGH SPEED access). Although this sounds like a minor distinction that most geeks figured out a while ago, National Public Radio got it wrong last week.
Leave A Reply