September 2007


Red Hat reported great numbers today. The stock was down in after hours trading, and I suspect it will be down tomorrow. The stock movement will likely be based on comments about the JBoss division by Red Hat’s CEO. Forbes reports:

“Sales came in slightly higher than expected, with Red Hat reporting revenues of $127.3 compared to revenues of $99.7 million during the year-ago period. Analysts had expected sales of $125.7 million.

Over the past year, Red Hat’s shares have fallen more than 25% as the company has toiled to transform from a Linux vendor into a peddler of a broader array of software.

One big reason: Sales of the so-called application server software Red Hat acquired with its purchase of JBoss last year have disappointed. “The rate of JBoss bookings and revenue growth has not met my expectations,” said Szulik in a conference call with investors. “We know we can do much better.” “

Red Hat Linux is definitely firing on all cylinders. JBoss, in the words of Red Hat’s CEO “could do much better”.

Further comments via CNet:

“However, not all believe the JBoss deal is going swimmingly. Credit Suisse analyst Jason Maynard downgraded Red Hat from “outperform” to “neutral” Monday. “We believe our thesis of improved field execution and meaningful JBoss acceleration won’t materialize and deliver the anticipated upside to our forecast,” he said. “Our checks indicate that the organization continues to be in a state of flux as the company struggles through its transition to a multi-product company.””

Here’s something I found funny via Reuters:

” Chowdhry of Global Equities Research said the prospects for JBoss were dim because its software was designed to work with the Java programming language. He said Java was losing market share as businesses embrace Ruby, a programming language that is easier to use and works with an open-source rival to JBoss known as Ruby on Rails.”

LOL. RoR is going to kill JBoss…Seriously?!?! If RoR is such a threat, why are other Java app server vendors “doing okay“? Is there something more here? (Hint: yes – I’ll get into it tomorrow).

Via Stephen Shankland at CNet – Very cool news about an effort to run Linux more efficiently on servers, laptops & mobile devices. The effort was kicked off by Intel, but is open to all.

According to Stephen’s report:

“On a current laptop, running Fedora 7 from Red Hat uses about 21 watts. “If you apply six little changes we propose, that same laptop takes 15.5 watts,” Hohndel said. “You have just added a more than an hour to your battery run time.””

“Taking Intel’s advice and fixes can trim about 10 watts of power consumption off a modern dual-processor server, said Dirk Hohndel, chief technologist of Intel open-source technology center. That’s not a gargantuan amount–until you consider that if done correctly it’s free power savings, that each watt of server energy saved cuts another 1.3 watts from air conditioning (according to Intel figures), and of course that 10 watts per server is a lot when multiplied by the thousands of servers that populate larger data centers.”

There is a lot of pressure to consolidate commodity hardware onto, bigger more efficient systems. The competition between commodity systems & larger systems may well result in Linux becoming “the most energy efficient” operating system on servers. It appears that Intel is focusing on this goal from a commodity server standpoint. How long until the larger systems vendors decide to do the same with Linux? (Note: they may be doing so already – so please comment and update us all!)

Last week I spoke with Todd Hay, VP of Marketing at ActiveGrid.

“Too often, we fight ourselves in the AJAX community. For customers, there is too much confusion between AJAX libraries such as Dojo, JQuery, Prototype or the tens and hundreds more. Customers get overwhelmed and look towards Adobe or Microsoft. Other times, customers want to use an open source library and there is no support option available.”

Considering that we’re still early in the product lifecycle for Web 2.0 runtimes/libraries/technologies, the average customer doesn’t want to make a bet on a product (or company) that may be gone in a few years. In these cases, choosing Adobe or Microsoft feels safer. I can understand when a customer selects Adobe AIR or Microsoft Silverlight because of some capability or feature not sufficiently available in an OSS AJAX framework/library/product. But selecting AIR or Silverlight to reduce the confusion between OSS AJAX frameworks/libraries/products signals a problem for AJAX communities and vendors. There appears to be a need for some overarching standards for AJAX-based frameworks. These standards would enable customers to choose between various implementations with confidence. The OpenAjax Alliance is working towards “Standardizing Ajax Development” (as their logo says – note that ActiveGrid, Adobe & Microsoft are all members of the Alliance). But there is definitely more that could be done, and done faster.

To this point, Todd explains:

“We recently acquired TurboAJAX Group, a company that had 2 of the top contributors to the Dojo project. We are committed to the Dojo community and will be offering subscription support offerings for Dojo and for TurboAJAX products, which are built on Dojo. Next, we’re going to release TurboStudio under the GPL to minimize barriers to adoption. We’re also going to work with other members of the AJAX community on ensuring a greater degree of collaboration.”

Releasing TurboStudio under the GPL (coming soon) is great news . However, I think that the more important goal is getting some degree of collaboration and standardization across the AJAX world. Choice is good, but without standards, choice leads to confusion, which in turn leads to many customers making the ‘safe choice’.

InfoWorld reports:

“IBM is to offer the world a free word processor, spreadsheet, and presentation program in yet another bid to upset the dominance of Microsoft’s Office suite.

IBM says it will contribute 35 programmers to the Symphony-cum-OpenOffice development effort….”

The Lotus Symphony FAQ states:

“Lotus Symphony is based on the Open Document Format (ODF) standard-which means you’re not locked into proprietary file formats, software licensing agreements and upgrades”

Lotus Symphony products are standalone versions of products that are being bundled in Lotus Notes 8, and are built from OpenOffice technology.

The download process requires that users enter an IBM developerWorks or PartnerWorld ID. I know this sucks, and we deal with this issue when users want to download WAS Community Edition. However, we’re required to adhere to export laws, exclude downloads to embargoed countries (i.e. North Korea), and IBM’s legal team is a little more cautious when distributing OSS-based products for some reason ;-).

I just read that Carl Icahn owns a 8.5% stake, worth $426.5M, in BEA. He wants to see BEA sold in order to maximize shareholder value. While I can understand this point of view, I’d like to offer another option. Transforming OSS from a threat to an opportunity could be one step towards a BEA recovery. The recovery would surely help maximize shareholder value.

How would BEA get there from here? First, evolve beyond the “Blended Strategy” to OSS, which essentially entails supporting 3rd party OSS products such as Tomcat, when used with BEA products. Next, get behind Apache Geronimo and deliver a BEA distribution of the JEE5 certified open source app server. To be fair, BEA groks the value of OSS, and has supported various OSS projects. However, BEA’s OSS efforts have not provided them with a way to combat JBoss in BEA accounts. Getting behind Apache Geronimo would address this strategic gap.

Before I go on, let me fully disclose that I am an IBMer, working in the WebSphere Application Server product management team. Our products compete directly against WebLogic Server (WLS). That said, the postings on this site are my own and don’t necessarily represent IBM’s positions, strategies or opinions.

Over the past 3 years, BEA has faced stiff competition from leading enterprise vendors such as IBM. But BEA has also been the target of JBoss. Unlike competing with enterprise software vendors, competing vs. OSS with new product features, improved performance, additional marketing, sales strategies, isn’t always the right answer. For example, you can’t beat JBoss by adding a new higher performance feature to WLS.

Essentially, in certain projects,  BEA customers have said: “I’m looking for a Toyota Yaris (small, cheap, not luxurious)” and BEA has countered: “We can sell you a Toyota Camry (bigger, more expensive, more luxurious – i.e. WLS), but we will also provide regularly scheduled maintenance if you want to drive a Vespa Scooter (fun, cool, for short non-highway distances – i.e. Tomcat) along with the Camry”. JBoss has won in BEA accounts because they offered a “Toyota Yaris”. I am not disparaging any products, I am merely pointing out that different projects require different infrastructure, similar to Zack’s positioning of MySQL vs. Oracle.

To address the JBoss threat, BEA could take a page out of the IBM playbook. We compete with JBoss with WAS Community Edition, built from the open source Apache Geronimo project. Additionally, since we don’t believe in a one size fits all approach, we offer customers choice at the project level, and are able to protect their investments using the breadth of our WebSphere Application Server family of offerings. This has been a successful strategy for winning in the face of OSS competition.

Offering BEA’s own Geronimo distribution would enable BEA to “offer a Toyota Yaris” when the developer or customer seeks a “Toyota Yaris”. Additionally, BEA’s distribution of Geronimo could be used to seed their developer base with a *BEA product* when the developer seeks to use OSS.

BEA executives will surely be worried that the BEA Geronimo distribution will cannibalize WLS revenue. BEA shouldn’t let risk cloud opportunity. Many customers initially choose JBoss for projects that don’t have funding or don’t need the capabilities delivered by higher-value enterprise application servers. By providing an offering that meets the price, simplicity and performance needs that customers evaluating JBoss are signaling, BEA could drive additional revenue and attract/keep developers.

Why do I want to help a competitor? A stronger BEA helps to keep WebSphere on our toes. A stronger Apache Geronimo product and BEA winning more against JBoss will also force JBoss to raise their game. In the end, customers benefit, and that’s why we’re all here, right?

So BEA, move beyond the Blended Strategy. Get behind Apache Geronimo. You can thank me later ;-)

PS: I should state: “The postings on this site are my own and don’t necessarily represent IBM’s positions, strategies or opinions.”

The Motley Fool has a pretty interesting article on IT analyst firms such as Gartner & Forrester.

I’ve worked with these firms for nearly a decade and do believe that they provide a valuable service to their customers and the IT market in general.

Some metrics for Gartner:

“Each year, Gartner’s 650 researchers attend 18,000 vendor briefings, along with answering 240,000 client inquiries.”

Obviously, not every client inquiry is of equal weight, but these numbers work out to 1.5 client inquiries per researcher per weekday.

Let’s set aside the client inquiry figure, which, even if each inquiry was from a different customer we’re only covering a very small subset of IT buyers. Even if a company is not a Gartner, Forrester, IDC, RedMonk, 451 Group or Entiva (etc.) customer, the company will be influenced by what these analysts say in the public media.

Analysts help customers make purchase decisions. Analysts also educate the market on new technologies. Even in open source land, the IT manager and CIO of a developer using your OSS product is much more likely to take something serious if they hear it from Gartner, Forrester or IDC first. Working with the large analysts will help your OSS business if you want *paying* customers in the enterprise. Boutique analysts like RedMonk, 451 Group or Entiva are going to be crucial in developing/refining your OSS business strategy. However, because of their size, these boutique analysts don’t have the reach of a Gartner or Forrester in terms of customer purchase decisions. You’ll need to work with both classes of analysts, because each bring a lot of (differentiated) value to the table.

BTW, the article also makes the case against using analysts:

“The current model of analyst-intermediated, opinion-based technology buying and selling produces poor financial returns. Across the world, corporate managers spend more than $1 trillion a year on technology. To help managers invest, and technology marketers persuade, $2 billion a year is spent on technology analysts. Yet 70% of projects fail to deliver a financial return, according to the Standish Group and other commentators.”

(Funny enough, Standish Group is also an analyst firm). Should analysts bear the weight of these 70% of projects that fail? Would fewer projects fail if analysts were not involved in the equation? What is the ratio of project failures for similar companies that use analysts vs. companies that don’t?

Questions, questions, question…I better call an analyst ;-)

PS: I should state: “The postings on this site are my own and don’t necessarily represent IBM’s positions, strategies or opinions.”

From the InfoWorld story about Sun OEMing Windows Server with the Sun Fire server line for customers that want to use Windows.

“The Sun Fire servers already support Windows Server, along with Red Hat Linux, Novell’s Suse Enterprise Linux, VMware’s ESX virtualization system, and Sun’s own Solaris operating system, according to Lisa Sieker, vice president of marketing at Sun Systems.

Asked if Sun’s latest alliance shows its intent to compete more directly with Linux and VMware, Sieker said, “We are responding to strong customer demand for this…. It’s a complex marketplace out there.””

So very true, it is a complex marketplace out there. Simplifying the marketplace to “…oss is the right answer in all occasions” really minimizes the prevalence of heterogeneous environments and needs that customers have.

PS: I should state: “The postings on this site are my own and don’t necessarily represent IBM’s positions, strategies or opinions.”

Shaun Connolly at JBoss has a post about “What’s in a Subscription“. Shaun states:

“Put simply, a Subscription is comprised of:
1. Software bits
2. Patches and updates to the bits
3. Support in the use of the bits
4. Legal assurance”

The fact that OSS vendors have to use legal assurance/indemnity as a method for driving subscription purchases is due to the general FUD around OSS licensing, patents and copyrights. Things like copyrights & patents are vendor issues and should not be made customer issues.

Do any of us care that the Blackberry of iPod we use may have patent infringing technology inside? Or that these devices may include some copyrighted material (a la the software) inside? No, we expect that the vendors who sell us these products have taken care of that for us. Staying on the right side of the IP law line is another cost of doing business for RIM & Apple in this example. As a customer I may pay for this cost in the total cost of the product, but IP assurance is not used as a customer benefit, just as “we didn’t pollute more than government guidelines allowed when building this product” is not used as a customer benefit. Again, isn’t IP assurance a vendor cost and issue? Why have we made this a concern for customers?

What am I missing?
PS: I should state: “The postings on this site are my own and don’t necessarily represent IBM’s positions, strategies or opinions.”

I just read the JS post about the continuing rise of OpenOffice.org.

Why do vendors like Sun, IBM (officially now), Google, etc. support OO.o? Well, providing customer choice and openness are two of the key reasons quoted. These are definitely great reasons that we all want to see OO.o succeed.

Most vendors backing OO.o compete with Microsoft in markets other than Office Productivity software. So, what about the competitive benefits for vendors supporting OO.o? Let’s face it, $1 diverted from MS Office could become $0.70 diverted from the budgets of Microsoft’s other business units.

Here’s a look at Microsoft’s 2006 FY financials by reporting division:

- MS Office is inside of “Information Worker”
– MS Windows Client operating systems are inside of “Client”
– MS Windows Server operating systems are inside of “Server & Tools”
– MS SQL Server is inside of “Server & Tools”
– MS Ad revenue is inside of “MSN”

When you look at the operating margins of Microsoft’s business units, it’s quite easy to see why attacking MS Windows profits with Linux and MS Office profits with OO.o is the strategy of choice for vendors that compete with MS in other markets.

PS: I should state: “The postings on this site are my own and don’t necessarily represent IBM’s positions, strategies or opinions.”

Matt has a great post on the sources of deals & leads for Alfresco, which could be useful to a broader set OSS vendors.

What is surprising, but not unintuitive, is the low percentage of Hosted Trials that turn into deals. Namely, Hosted Trials represent 30% of leads, but about 5% of deals.

The conversion rate between Enterprise Trials leads (15% of all leads) and deals (10% of all deals) is much higher.

If you’ve tried the SugarCRM Hosted Trial “just to see what all the fuss is about”, then this data shouldn’t be a surprise. If I had to download, install and configure SugarCRM just to form a few general opinions on the product, I would have found something else to do with my time. Only (more) serious users would go through the hassle of downloading, installing and configuring the product.

So, don’t use this data to prioritize a Hosted Trial (SaaS) lower on your list of to-dos. Even if you don’t get deals via Hosted Trials, you’ll expose your product to a lot more people (especially of the kind that write checks vs. those that write code) than requiring that they install the product to try it out.

Next Page »

Follow

Get every new post delivered to your Inbox.