A Skeptic’s Take

Whatever your view of the cloud, a little perspective goes a long way.

Cloud computing? “I think it’s a lot of hype,” he says during a recent interview in the company’s midtown Manhattan office. “Time sharing made a lot of sense when computers cost $3 million in 1970, but now they’re commodity items that anybody can buy.”

I went straight from college into the timesharing business. It was a license to print money until PCs arrived in the mid 80s.

Yes, but with the likes of IBM, Oracle and Microsoft joining Amazon in the public cloud game, won’t that kind of processing capacity be available cheaply and on demand? “You can buy one of these servers for $500 and have it under your desk,” Goodnight responds.

Note that is much cheaper than the typical $2-5,000 PCs cost in the 80s. Of course hardware isn’t the main cost anymore. The big money is in the application software, and the really big money in the professional services required to make application software work.

Jim Goodnight is a CEO with more than 30 years in the same job, he recognizes hype when he sees it. The hype is that cloud is somehow new and magically cheap. It really is timesharing packaged under a new exciting label. The true cloud play has to do with price – the old rent versus purchase choice. If you live in New York City, parking costs more than the price of the car – rental may be a better option. If you live in the suburbs parking is free, rental is not as good an option. When you consider your cloud options, be a skeptic with Jim Goodnight and evaluate the true costs. For example data backup and the need for offsite storage / retrieval of backups may make cloud backup services a better option. ERP with a 7-10 year lifespan and the need to control your data may make on premises deployment a better option. At HarrisData, we are happy to provide both cloud and on premises options to our customers. The customers, not the hype, determines which option is best in each instance.

Posted in Cloud computing | Tagged | Comments Off on A Skeptic’s Take

Cloud Computing is not a Pricing Model

Try searching for “cloud computing is cheaper” and you’ll find no shortage of opinions that support the notion that by going to the cloud, you will will save millions of dollars. Cloud Computing is full of hype and marketing over-reach. The most common overreach, in my opinion, relates to pricing. According to the true believers, there is a magic to Cloud Computing that instantly lowers your costs. The web site Cloud Computing Defined is typical, claiming that “… because users pay only for the resources that they use, cloud computing is cheaper.” This is not to say that cloud computing cannot be cheaper – many cloud computing resources are even free. But Cloud Computing is not inherently cheaper – because Cloud Computing is not a pricing model.

In September 2011, the National Institute of Standards and Technology (NIST) published a paper titled “The NIST Definition of Cloud Computing” including the following definition:

Cloud computing is a model for enabling ubiquitous, convenient, on-demand network access to a shared pool of configurable computing resources (e.g., networks, servers, storage, applications, and services) that can be rapidly provisioned and released with minimal management effort or service provider interaction.

The NIST goes on to discuss five essential elements of Cloud Computing, and four distinct deployment models. However, search the entire NIST document for price or pricing and you’ll be disappointed. Pricing is simply not part of the definition. In fact, the “cost” of Cloud Computing can be cheaper, the same, or more expensive than alternatives, depending on the pricing models used and the specific circumstances.

Many erroneous claims that Cloud Computing is cheaper come from the perception that subscription-based and/or pay-per-use pricing models are inherently cheaper than an up-front purchase. A quick example shows the folly of assuming that one pricing model is inherently superior to another. You need a car for transportation. Which is cheaper, renting, leasing, or buying the car? Renting at $50/day sounds cheaper than leasing at $500 per month, and both sound a lot cheaper than buying a $25,000 car. Plus, when you rent a car, you don’t have to worry about servicing your car, and you get all of the latest cool features (because you get the new model year when it’s available). The Rental Car companies get much lower prices from the manufacturers than consumers, and can share the fleet to keep costs down. It’s almost CaaS (Car-as-a-Service)! If you find yourself away from home for a week or two, you rent a car. So why doesn’t everybody rent a car all of the time? Because even if you only rent on weekdays, you spend $50 * 250 days/year = $12,500/year. If you can by a car for $25,000, and you expect to use it for more than two years, you wouldn’t rent it (especially if prices go up over time!).

Economics 101 starts by teaching us that different people have different preferences. Pricing on the cloud is no different. As Frank Scavo notes in his blog Cutting Through the Fog of Cloud Computing Definitions, “how the customer pays for the service has no bearing as to whether the service is cloud computing.” A perpetual software license (a license purchase) can be delivered over the cloud as effectively as a pay-per-use model, and in some cases may be preferred. With ERP software lasting 7 years or more at most organizations, a perpetual software license model delivered through the cloud may offer all of the advantages of the cloud, with a pricing model that better meets the requirements of the business.

The Latin phrase caveat emptor loosely translates into “let the buyer beware,” and applies to anyone who accepts the marketing hype without inspection and automatically assumes that Cloud Computing is cheaper. Just because the label says Cloud doesn’t mean the solution is inherently cheaper – look at the price and compare to your other alternatives. Often, Cloud Computing solutions are cheaper – sometimes dramatically so. Just make certain you know what you are buying before you sign on the dotted line.

Posted in Cloud computing | Comments Off on Cloud Computing is not a Pricing Model

Cutting Through Vendor Noise

In a tough economy, software vendors pull out all stops to get your business. As Trevor Perry explains this often results in so much noise that the vendors hurt themselves as much as they hurt your ears. His article is full of useful tips for cutting through the noise to locate what value the vendor has to offer. This is a must read article for everyone including, perhaps especially including, the vendors themselves.

A quick example is Perry’s approach to technobabble as expressed in vendor claims about Service Oriented Architectures (SOA):

Service Oriented Architecture (SOA) also suffers from this kind of confusion. For example, the “A” in SOA stands for “Architecture,” yet you can find vendors selling SOA “software”—none of which is actually software for architects. The concept of Service Orientation is one that will become an inherent trait of all future IT and applications, regardless of its name and misrepresentation by vendors, because it’s the next logical evolution in IT development. You may have SOA already, but you may not understand what it is—or that you’re actually doing it—because of vendor noise.

His recommendation is straightforward:

Your response should be this: Find the detailed definitions of the buzzwords and terminology (preferably from the vendor site), research the concepts being proffered, then re-read the marketing.

There is much, much more in the article including problems with marketing speak, branding, celebrities etc. In each case Perry includes a recommendation for cutting the noise to get to the point. And what is the point?

The beneficiary of the right choice in product selection should be your company.

This puts the responsibility for effective communication of how a product benefits your company squarely on the vendor, while giving you techniques to help the vendor reach the needed clarity. Once you select your vendor you can use these techniques to focus the implementation on what benefits your company as well.

Posted in Uncategorized | Tagged , | 1 Comment

Operating for the Benefit of Wall Street is Bad for Customers

The lights have gone on for Josh Greenbaum at ematters as he confronts the reality of Oracle at their annual user conference.

The problem with the Oracle of today is that the focus of a group of the some of the best technology minds in the industry has been hijacked to fulfill a vision that is skewed more towards fulfilling the promise of a decade-old merger and acquisition strategy than it is towards making customers both successful and happy.

What Josh ran into was the standard big software monopoly rent seeking approach to customer relationships. In this strategy the vendor acquires their way toward ever greater footprint then exploits the customers by increased maintenance fees with no concern for customer ROI. Competing products are acquired to capture more users in existing product areas, while more products are acquired to fill out wall street analyst function checklists. Acquired products are then ruthlessly stripped of costs to maintain margins for Wall Street’s satisfaction. Oracle’s results are typical:

Oracle’s applications strategy is the antithesis of integrated , in reality a hairball of products and underlying technologies, data models and deployment models, that could have only been “engineered” by an M&A strategist. Any real engineer would be crucified for pretending this software strategy makes sense for the customer looking for an integrated applications environment.

Because in the end Oracle’s roll-up the best of breed strategy has never been about better TCO for the customers. It’s been about optimizing the sales opportunity for Oracle’s incredibly effective sales machine, while bringing smaller, inefficient software companies under the razor-sharp cost-cutting eye of Safra Catz.

The next step in the big software monopoly rent seeking model is to give lip service to integration and usability while using the monopoly to increase sales of programming services. Customer service in the sense of actually supporting the customers’ business requirements takes a back seat to revenue opportunities. Oracle is right on target here as well:

Oracle’s customers have been pulling out the soldering irons and user manuals in order to realize their vendor’s integrated software stack vision.

There is no magic bullet, no easy-to-configure wizard, for the majority of the integration that Oracle customers require to run their businesses on Oracle software. Nope, it’s all about custom development, using expensive development resources.

Josh concludes that while Wall Street may like Oracle’s strategy, it may actively harm customers.

Right now, Oracle’s case to its customers on the value of engineered systems looks too much like the case it’s making to Wall Street. Until that changes – if it can change – Oracle is headed down a path that at best lacks customer-centricity and at worst is genuinely customer hostile.

An important point to consider, Oracle is not alone in applying the standard big software monopoly rent seeking strategy. Oracle’s results are not atypical. Any software vendor applying this strategy will get similar results, Wall Street’s favor may vary, but the customers suffer in all cases.

Posted in Software Support | Tagged , | Comments Off on Operating for the Benefit of Wall Street is Bad for Customers

Who is in Charge of Cloud Security?

Cloud service providers believe their customers are responsible for security, and license accordingly. That is the message cited by David Rosenbaum in his article on security risk and liability in the cloud.

“Cloud,” says Bruce Lynne, managing partner of Financial Executives Consulting Group, “is just a fancy word for outsourcing.” And, as smart CFOs know, when a company outsources, it sheds work, not responsibility.

Anyone contemplating cloud computing should read this article to get at key licensing issues which tech industry hype glosses over. Rosenbaum identifies a key point about cloud – it is a financial decision and requires intense effort by the CFO and company lawyers to be successfully implemented. The big issue is security, the related liability from security failures, and what can be done about managing the business and financial risks of the cloud.

But while the policy may be familiar, the ramifications could be huge. That’s because the cloud — which enables companies to outsource everything from e-mail to ERP and then access it all through a browser — is inherently insecure. The same ease of access that makes it appealing also makes it vulnerable. Yet many non-tech-savvy buyers of cloud services are not adequately aware of the security issues, says James Reavis, director of the nonprofit Cloud Security Alliance.

The recommendations include negotiations advice, things to learn about your vendor, and encouragement to be bold about getting the security your firm needs. Security needs may trump some common cloud architectures. For example the need to “find out who’s in the cloud with you. If the provider has an insecure customer, that makes you less secure” argues against common multi-tenant approaches to cloud applications. This advise also applies to “private cloud” offerings (anything delivered via a browser even if it is implemented on premises). In short, by selecting a vendor that you can reliably partner with, you gain an opportunity to manage cloud security risks.

Posted in Cloud computing | Tagged , | Comments Off on Who is in Charge of Cloud Security?

Implications of Changes in Investment

Charles Stewart dives into the numbers behind the economy to focus on gross private domestic investment. The outlook and policy options he presents are bleak – almost anything the government does will lead to reduced employment. However he does offer a path to better outcomes for individual firms:

Total factor productivity has risen 3.2% in each of the last 2 years, the highest rate since the BLS started keeping tabs in 1987. The average long-term rate has been around 1%. We know that it is investment, not labor, that is mainly responsible, because labor productivity increased by only half, to 3.6%, while total factor productivity tripled. We can produce the same output today with 7% less labor than two years ago. Investment in equipment and software is reducing employment.

Stewart’s insight has several implications for domestic manufacturers looking to improve their chances in today’s economy. First is a hard lesson — if you are not getting the same output from 7% less labor, you are falling behind your competitors. This benchmark comes from national statistics, so half of all manufacturers have plenty of room for improved results. Even those getting by on a 7% labor reduction may be falling behind, that is the average not the top performance.

To catch up to and potentially leap ahead of your competitors requires investment. This investment must be carefully targeted. Adding staff exposes manufacturers to uncertain regulatory costs. Adding inventory is trickier, unless it moves it is dead weight on the balance sheet and prevents investments with real returns. The key is investment in equipment and software to increase total productivity of the firm. This is not an either / or investment, modern CnC machines and ERP software work together to generate greater returns than either alone. Work with vendors to fit these investments within your balance sheet.

When you set investment priorities, focus on the business value, and on enabling strategies that take advantage of opportunities in the current environment. Identify a business opportunity tied to a successful business strategy for today, then invest in a solution to that problem. A troubled economy is the wrong time to get misled by industry hype about technical nuances of competing products. If the vendor isn’t focused on direct business value, no value will be realized. Potential success strategies include processing faster and more flexibly than foreign competitors by combining lean production techniques with automated product configuration. Investments include better ERP plus modern factory equipment. Results include dramatically reduced order processing time, fewer production errors, reduced inventory, and reduced labor costs. If you can invest before your competitors do you may achieve growth in spite of the poor economy.

Posted in Uncategorized | Tagged , | Comments Off on Implications of Changes in Investment

ERP Lessons Learned

Frank Scavo posts an excellent review of the history of and lessons learned from ERP systems over the past 20 years. Both Frank’s post, and his keynote video are well worth your time. Some key points from the post:

ERP is not primarily a planning system, it’s a transaction processing system. Its benefits are primarily in standardizing and automating business processes.

In particular ERP automates the reconciliation of money, products, people, and time to provide a clear record of what is and has happened in your business, with the full audit capabilities required by regulations such as Sarbanes-Oxley. Without this foundation added systems like CRM, SCM, and BI are only marginally effective.

Today, I find that business leaders have a better understanding of best practices for successful ERP implementation. They realize that ERP means changing how the organization does business. They usually recognize that top management needs to be committed and that it will require participation by all affected functions. They often realize that it is best to pick a system that fits the business, and as much as possible to avoid customizing software code.

That ERP is about the business, and not about the technology is a point that cannot be emphasized enough. Too often ERP decisions are driven by technology based factors (cloud vs on-premises, .Net vs PHP) while the core business requirements remain unexplored.

According to our 2011 survey, 38% of ERP projects exceed their budgets for total cost of ownership. Furthermore, as I indicated in my keynote, the risks of ERP go beyond cost overruns: ERP is particularly subject to functionality risks (the project was within budget, but the system doesn’t satisfy key requirements), adoption risks (the project was within budget, but the organization is not fully using it), and benefit risks (the project was within budget, but the expected benefits are not realized).

That is when technology is placed before business need, the ERP disasters we all fear come to pass. Learn the lessons Frank shared, focus on your business needs, and benefit from your ERP.

Posted in Uncategorized | Tagged , | Comments Off on ERP Lessons Learned

Head in the Clouds

ERP analyst Derek Singleton posted a paean to the SaaS variant of cloud computing. With his head firmly in the clouds looking for rainbows covering five points, Derek does trip over one salient feature of modern computing—the user experience should be the focus of vendor efforts (everyone has feature / function these days). For Derek, here is some ground level perspective on his key points.

First, he notes that cloud companies “have momentum and attract great talent.”

SaaS companies have an intangible that’s working to their benefit – they’re recruiting exceptionally bright, young talent.

The same could be said of the “Dot-Coms” of a decade ago, images of a bubble economy are hard to avoid here.

His second point is the wonder of multi-tenant architecture and ”smoother scalability.” To the usual claim of overnight upgrades across an entire customer base, Derek adds a new one:

A further advantage of multi-tenancy is that company-specific customizations are left in tact when the system is updated. This is because customizations are made in metadata. For the non-techies out there, metadata is data that defines the settings and customizations for each customer, but is maintained separately from the core application code.

Multi-tenant architecture is a mixed bag for ERP installations – one swoop upgrades can cause problems for any firm covered by Sarbanes-Oxley (whether directly or through bank loan terms). The meta-data approach is indeed an appropriate development technique, but preceded multi-tenant architecture and SaaS by at least a decade. Traditional ERP vendors have employed meta-data approaches for quite a while. For example HarrisData restructured its applications to a meta-data base starting in 2000, while HarrisData’s RTI Software division began using meta-data in 1992.

The third point that the “cloud is changing enterprise software consumption” is a recycling of a three (four?) decade old argument over how to license software for enterprise use. Many pricing models exist, from the enterprise wide license, to the cpu license, to various user based license schemes (including SaaS licensing). For each pricing model there is an equivalent array of payment options from upfront cash to lease terms. The arguments and distinctions between the pricing / payment options amount to deciding whose ox is being gored at any point in time.

The fourth point on the importance of user experience is a very strong point, “great user experience equals happy customers.”

Historically, enterprise software hasn’t been overwhelmingly friendly when it comes to the user interface (UI) or user experience (UX). It’s why a standing army of consultants and professional services firms exist to help buyers customize their systems and learn how to use the software.

Too few traditional vendors take this point seriously, focusing instead on capturing the monopoly rents derived from providing the standing army of consultants and professional service people. However as with meta-data above, a focus on user experience is independent of cloud and SaaS deployment of applications. HarrisData emphasized the user experience beginning in the mid 90s, and has held training and services costs to less than 5% of revenues annually since 1996. Reducing training and services required to deploy our applications is still number one on HarrisData’s to do list.

The final point is that since cloud and SaaS is on the web, and young people “get the web,” the cloud is youthful and innovative and magically better than sliced bread. Apparently young people want

web demos, trial versions of the system and user ratings of the product.

This is hard to argue with, as young (and old) people have wanted these thing in the 80s and 90s as well. Perhaps all people would be better off focusing on the business value of various ERP options rather than the hype and fluff of analysts.

Posted in Cloud computing | Tagged , | Comments Off on Head in the Clouds

PHP on IBM i – Adding new value to old applications fast

The IBM Systems Magazine PowerUp blog is talking about PHP on IBM i, a thread started by Laura Ubelhor with her post “There’s No Doubt PHP is an Awesome Fit on IBM i“.

For the average IBM i shop, PHP is a great way to extend existing applications to the web. Laura’s post highlights success in dramatically improving efficiency of AP – using a web application to deliver the details of the remittance advice when no paper check is used. Self-service applications delivered through the web offer significant savings in reduced paperwork, eliminated mail charges, and fewer inbound phone calls. PHP allows IBM i developers to quickly deliver self-service applications to new groups of users – such as vendors, customers, and suppliers.

One HarrisData manufacturing customer used vendor self-service applications over the web to share raw material requirements plans (from MRP) directly with suppliers. The results? Suppliers were able to better manage their own manufacturing capacity, improving efficiency and reducing costs at the supplier. Further, because the suppliers could see the forecast and any changes, they were able to substantially improve on-time delivery, giving manufacturing management the confidence to optimize processes and lower costs.

Success stories like these are really based on delivering web-based applications that provide external users significant value, and only require the external users to have a web browser available. Well-designed self-service applications are available anywhere, anytime, with no training. IBM i shops can quickly and successfully extend applications to the web using PHP. The skills are easy to learn, and focused applications can be developed and deployed quickly – with the entire application stack (including the web server) on your IBM i.

At HarrisData, we made the choice years ago to move our user interface to a web paradigm, to take advantage of the how quickly and easily users took to browser-based solutions. We’ve developed comprehensive ERP, CRM, and HRIS applications using PHP to drive browser-based interfaces to our traditional RPG applications, allowing us to deliver high-impact solutions without having to completely re-write the RPG business logic that has been tested and working for our customers for decades.

Posted in PHP | Comments Off on PHP on IBM i – Adding new value to old applications fast

Is Enterprise Integration the Next Hurdle for the Cloud?

Stephanie Neil poses an interesting question on whether SaaS and Cloud solutions create as much work in enterprise integration as they solve in reduced hardware and storage management.

Enter software as a service (SaaS) applications, which might seem to be an IT manager’s dream: no server and storage systems to buy and maintain. But their emergence presents a whole new integration problem between on-premise legacy apps and those that live in the cloud.

According to the recent InformationWeek Analytics 2011 Enterprise Applications Survey, 43% of SaaS users are very happy with the ability to deploy the applications quickly, but are much less satisfied with the complexity of integrating hosted apps with on-premise systems and data sources.

To the extent that enterprise vendors (either cloud or on-premises) utilize Service Oriented Architectures and document required data structures this may seem like a small problem. However, all it takes is one key application which is not SOA and the enterprise integration problem rears its ugly head. At that point IT skills developed by integrating current applications are useful. Unfortunately in many SaaS and cloud implementations, the IT staff does not have access to the application source code necessary for successful integration efforts.

An appropriate vendor offering in today’s environment should take advantage of what the cloud offers (no server and storage systems to buy and maintain) while providing source code access IT still requires for successful implementation. HarrisData offers ERP through a Platform as a Service structure to directly address this problem. In addition to managed hardware and source code access, PaaS allows the customer to control application upgrade timing – an important consideration in the era of Sarbanes-Oxley.

Posted in Cloud computing | Tagged , | Comments Off on Is Enterprise Integration the Next Hurdle for the Cloud?