The Cloud Versus SaaS: Compiere’s Don Klaiss Weighs In

“Traditional SaaS offerings inherently restrict the customer’s ability to customize the solution, require co-mingling of sensitive data from various customers, and force all customers to comply with upgrade and downtime schedules dictated by the vendor. SaaS customers are required to sacrifice freedom and control for the convenience of the hosted SaaS model. Compiere’s Cloud Edition offers the convenience of traditional SaaS, but with far greater freedom, control, and cost savings.

“Cloud computing is the next-generation of software as a service, in which a complete software environment is licensed as a subscription from a software vendor and low-cost, secure, and dependable IT hardware infrastructure is ‘rented’ from a utility-computing provider on demand. The customer has complete control over its own secure and private IT environment at a very low cost and without the hassle of procuring and managing its own data center. It can quickly scale IT resources up or down as computing needs change. And [the customer] has complete freedom to customize the solution as it sees fit and complete control over upgrade cycles and all other aspects of its IT environment.”

What do you think about this definition or the Compiere-Amazon pairing? Leave your comments here.

Comments

[Arthur Kruk]

It's really just another positive step forward in utility computing. Enterprise software environments that were too complex for cookie cutter Saas offerings had to maintain their own infrastructure or go the ASP route. Applying the cloud model leaves applications management in the hands of the licensee, while eliminating the big commitment of scarce capital. Sounds like a win-win for midmarket arcompanies that need customization. And if those customizations fit Compiere's enhancement protocol, chalk up another win for the open source model.

[Peter Cohen, SaaS Marketing Strategy Advisors]

Most folks weighing on what to call this market (and there are plenty of them) don't position "cloud computing" in opposition to "SaaS" or as a "next generation" of SaaS. Instead, many explain SaaS as a subset of cloud computing. I've also seen references to "cloud computing applications" as distinct from "cloud computing services."

From the point of the people I work with who are responsible for marketing these applications, we'll all be better off once this dispute over what to call these things is settled.

As for the description offered of "traditional SaaS offerings," I'd say that would that would more aptly describe a "bad SaaS offering." The more successful solutions do, in fact, allow configuration to suit individual companies' needs, they do segregate and protect sensitive data, they publish upgrade schedules, and their uptime is often better than applications maintained in-house.

[Trevor Miles]

Hi Bruce

The two most interesting points are without a doubt the use of Amazon EC2 and the price point. While I agree with the bulk of your post, there are 2 areas where I must disagree:

You make the statement at the end of your article that Compiere will need to get “brand-name” companies to sign up in order to be taken seriously as a disruptive technology. I am willing to bet anyone $50 that none of SFDC’s first 100 customers, perhaps even none of their first 1,000 customers, was a brand-name. If Compiere can get significant traction in the SMB space, the brand-names will pay attention.

I think the much bigger barrier to broad adoption of Compiere will be the insecurity around support and maintenance with open source based solutions, not core functionality or price point. I have a personal bias against open source for these reasons so I would be interested to hear what others say about this issue.

I think open source and cloud computing are at odds. As Don Klaiss states, "And [the customer] has complete freedom to customize the solution as it sees fit and complete control over upgrade cycles and all other aspects of its IT environment." Much of the attraction of cloud computing is the lower TCO, much of that achieved by reduced headcount in IT and reduced cost of maintaining highly customized or custom-built applications. In the current economic climate specifically, but in the long term too, I am believe the TCO argument will prevail.

There are several other open source ERP companies, most of which are based in Europe. A comprehensive comparison of several of these was made by Smile in France, which is in French. (http://www.smile.fr/publications/livres-blancs/erp-open-source__1). I presume this comparison does not include Compiere Cloud. None of the ERP systems they compared has, to the best of my knowledge, gained any significant traction. One reason I believe to be the issues around support and maintenance. The other is one you bring up, namely direct sales. My experience of open source companies is that they are replete with technology people, with very little focus on sales and marketing, and somewhat short on people with deep operations knowledge and experience.

Regards
Trevor Miles, Kinaxis

[Bryan Larkin]

I've read your First Thing Monday piece and the above post two times each.

So, this differs from traditional ASP environment how? Oten the ASP leases access to a single instance of their software. And often that software is sitting on hardware that is a platform located at - and owned by - a hosting company.

Is the difference the hardware is cheap or that Amazon owns it? Or is it the "thin client" aspect? Or is it something else.

According to Wikipedia, "Common features associated with ASPs include:

* ASP fully owns and operates the software application(s)
* ASP owns, operates and maintains the servers that support the software
* ASP makes information available to customers via the Internet or a "thin client"
* ASP bills on a "per-use" basis or on a monthly/annual fee

I see one goal of the next wave of application delivery being to have a two-dimensional application that allows the multi-tenant capabilities where appropriate to keep costs low with customizability (perhaps in modular instances) for the aspects that need to be unique. The infrastructure is standardized and supported - whether it comes from the same company that provides the s/w or a different company - like with EC2. The one concern I have about the EC2/Compiere solution is that there are two necks to choke on this one.

[Doug Hadden]

This argument about customization vs. standardization has been going on for some time. Companies believe that they have internal processes that generate competitive advantage.

Many observers suggest that companies reduce customization as much as possible in standard enterprise software in order to reduce long term administrative costs. The problem of adapting to the customer requirement was a serious inhibitor in the early ASP days. SalesForce has managed to get a reasonable mix of configuration and customization by focusing on a single domain. We are entering into a new era of focused applications for specific vertical or horizontal markets that can be configured and more easily maintained.

Any new ERP suite, whether SaaS or not, that attempts to satisfy a broad market of multiple verticals or horizontal functionality (ERP, SCM, CRM) is doomed to providing customization as the main avenue of adaptability. This was once considered a competitive advantage - "complete freedom to customize". Not so any longer. The SME market is not prepared to accept 4 to 8 month system upgrade cycles or having to hire certified IT staff. (Larger businesses are starting to revolt too.) This is one of the attractions to SaaS, as long as the need to adapt can be contained mainly to configuration.

[Ming Ahmad]

Bruce,

As only a technology novice (more focused on the process and performance lens), I did not read an overview on the security measures through the open source, cloud computing initiative. It will be interesting to know the structure (in business/layman's terms!) on how to leverage the advantages and to mitigate the risk exposure.

[Dan Druker]

Bruce,

I thought this was an interesting enough spin on cloud computing that I posted about it on my SaaS 2.0 blog at http://intacct.blogspot.com

Running applications on EC2 addresses hardware virtualization, which can be better than installing hardware in your own data center, but to say that this is the next generation of SaaS says to me that someone doesn't really understand what SaaS is.

With SaaS, the vendor takes on the costs and risks of purchasing, maintaining and operating all of the infrastructure required to run the applications. This includes things like Oracle licenses, SOA platforms, security software, and a lot more. The SaaS vendor also operates the system, handling security, tuning, backup, upgrades, maintenance, disaster recovery, etc. SaaS vendors also spend lots of money and resources on operating procedures, SLAs and SAS 70 audits to guarantee things like like uptime, performance, security and quality of service.

There is also a lot of magic in multi-tenancy at the application level - with major associated benefits and economies in scaling and operations. Single tenant applications running in a virtual machine in EC2 by definition don't get this - above the hardware level you've still got the old single instance, single tenancy, single maintenance model.

If you are running your applications on Amazon EC2, by definition you are doing all of the above stuff yourself. Sure you get virtualized hardware and that's good, but buying and maintaining infrastructure is up to you, as is installing, maintaining and operating your applications. There are no SLAs, that's also up to you, and you're also on the hook for backup, disaster recovery, performance tuning, etc.

So the idea of running business apps on EC2 instead of on your own hardware is interesting, but to say it's the next generation of SaaS is an awfully big stretch.

[Subraya Mallya]

I completely agree with Dan. It is one thing to assume Amazon EC2 as the IT hardware plugin into your IT footprint but to assume that it is just half a person each doing backup and configuration is a bit too much of a stretch. There is no way a small vendor can get caught between two vendor to resolve SLA conflicts and finger pointing on the cause of a unscheduled downtime. Also getting the two vendor to clear SOX and SAS-70 audits is completely unrealistic.

Amazon EC2 works well for a homegrown legacy application whose maintenance costs can be shaved off with virtualized environments or if it comes in the form of solution wrapped into a SaaS service.

I can however see a few things that might be possible with Amazon EC2 and SaaS vendors
1. Companies can use Amazon S3 as a escrow for the code, documents that are housed in a SaaS solution.
2. Customers can use S3 can also be used a way to get a local copy of the backups of their slice of data in a SaaS multi-tenant environment and have SaaS vendors replicate it into their S3 accounts.

[Bill Wagner]

The distinction between the cloud and SaaS is fairly straight forward in my mind: The cloud is a computing environment, whereas SaaS is an application environment. If that's true, then a reasonable definition of each would be the following:

Cloud computing is effectively IT as a service (ITaas). It's a hosted, utility based computing platform accessed via the internet for use by multiple, simultaneous businesses as an application development & delivery platform.

SaaS is a hosted, utility based software application accessed via the internet for use by multiple, simultaneous businesses.

Therefore, Cloud and SaaS can (and do) exist independent of each other. The cloud is not SaaS 2.0 … it’s an enabling paradigm for things to come.

The fundamental problem may lie in the term "cloud" ... it doesn’t distinguish between the concept of a cloud model which has existed for a long time, and the recent advent of commercial cloud computing offerings such as EC2. The latter is what is exciting and potentially game-changing in my mind because of what it has the potential to enable.

[Pankaj]

in total agreement with peter cohen. SaaS is a subset of cloud computing. If by "cloud computing" we mean "platform-as-a-service" then the freedom provided by PaaS is too much to be of use to anyone. SaaS solutions nowadays do bring a nice balance of power, flexibility and customizability.