- Other Underreported Stories: Analyst Integrity?
- Top Underreported IT Stories in 2007
- Szulik's Departure from Red Hat
- Red Hat Beats Q3 Expectations
- Apple Shuts Down Think Secret
- Oracle Blows Out the Quarter
- Growing Demand for SaaS
- SugarCRM 5.0 Now Available
- TechIQ: 10 Open Source Companies to Dominate 2008
- Red Hat, JBoss and Corporate Culture
December 31, 2007 | Comments: (0)
Other Underreported Stories: Analyst Integrity?
My previous posting was about InfoWorld's underreported stories for the year. As I mentioned, we had quite an interesting email thread about other stories that were suggested. Items ranged from Microsoft's end-run on open standards, Microsoft's gradual but steady entry into open source, programming language trends, convergence of Web 2.0 with Enterprise, Software as a Service (though I think that one may have been overreported!), portable devices ranging from the iPhone to the Nokia N800 and others.
But the topic that got the most discussion and the most emotion going was about the role of IT industry analyst firms. As one blogger put it:
The persistent rumor (although nobody will go on record for it in fear of retaliation)... that the major research firms exhcange better-than deserved positions for cash in the form of subscriptions to their expensive services.
There was a pretty interesting discussion with views on both sides. Some felt that the rumors have been so persistent that, well, where there's smoke there's fire. Others saying they have heard from someone who heard from someone that once they started paying their exposure improved. Others saying it's just like the rumors that magazine advertisers get better reviews, an accusation that has been levied to Ziff-Davis publications, as well as photography and stereo equipment magazines for years.
My view is that while there are some low-end firms that engage in questionable research, the big firms have a very distinct separation of sales and editorial, just as reputable magazines do. While some might wish that a good analyst report could be "bought" it just doesn't work that way. And in 20 years, I've never had an analyst or sales rep from top firms like Gartner, DataQuest, Forrester, Meta, Ovum, IDC, The451 Group, or others imply anything to the contrary. But that's just my experience.
In the end, the editors felt the story didn't really fit the bill; it went beyond the scope of an end-of-year story. For those interested, there was a similar story published at InformationWeek about a year ago. Maybe its time for an update.
What do others think? Has anyone felt that analyst firms have crossed the line by selling positive coverage? Do you think analyst firms are credible? Do you use analyst reports or services to help you make your IT decisions? Is there a story here you want to read in InfoWorld? Let me know your thoughts.
And also, Happy New Year!
Posted by Zack Urlocker on December 31, 2007 10:26 AM
December 28, 2007 | Comments: (0)
Top Underreported IT Stories in 2007
The editorial staff and bloggers at InfoWorld did some brainstorming in recent weeks to write up the top underreported stories in IT for the year. Although the topic is a bit of an oxymoron (if a tree falls in the forest, kind of thing...) I think its a good story and well worth reading. We all get busy with the normal day-to-day, week-to-week management and I think it's helpful to take a look back at some of the emerging trends.
There's also a set of slides highlighting the stories.
Some of the top items include the emergence of Java as the new COBOL, Sun's turnaround. open source's commercial strategy, renewed interest in BI, and to me the most important item, the balance of power shifting to software buyers.
There was also some interesting discussions about the stories that didn't make the cut, but I'll save that for a later posting.
Posted by Zack Urlocker on December 28, 2007 08:38 AM
December 24, 2007 | Comments: (0)
Szulik's Departure from Red Hat
It's old news now that longtime Red Hat CEO Matthew Szulik is resigning the CEO position and has appointed the former COO from Delta Airlines Jim Whitehurst. However, I've long suspected that Szulik wanted out. I remember at one of the Red Hat conferences before heading out for dinner meetings with customers he told me he was getting "too old for this sort of thing." Still his impact on the company has been profound.
Szulik has created the largest and most successful open source company and he deserves full credit for taking the company from a low-end retail box-pusher into a significant Enterprise player --well ahead of competitors. And Szulik stood up to the pressure from Microsoft, Oracle, Novell and others. He fought tough odds and helped shape and transform Red Hat along the way.
By announcing the change along with strong Q3 earnings, Szulik has timed things to ensure ongoing strong operations. The JBoss integration appears to be complete and working, despite some grumbling from those who left. And revenues and subscription growth are in good shape.
While Whitehurst has a strong operational track record, at first blush, his appointment seems a bit odd, considering that he's coming from outside the tech industry. However, earlier on Whitehurst was with Boston Consulting Group and he's generally well-regarded as a tech-savvy operations guy.
Szulik will remain as Chairman of Red Hat, but in most companies this role is largely removed from day-to-day operations. Szulik was a hard charger though and I'm sure his absence will be felt, not only at Red Hat but in the industry. He was always a class act.
Matthew, you will be missed.
Good blog postings by Matthew Szulik and Michael Tiemann.
Posted by Zack Urlocker on December 24, 2007 05:13 AM
December 21, 2007 | Comments: (0)
Red Hat Names Jim Whitehurst, former COO of Delta Airlines, as CEO. Szulik to Continue as Chairman. More here and here.
I missed the Red Hat earnings call as I was more interested in RIM (a good Canadian company with awesome products, and whose stock I've been holding for some time now....yes, I'm happy today).
Red Hat beat expectations on quarterly revenue of $135.4M (+28% YTY), and net income of $20.3M (+39% YTY). You can read the transcript of the earnings call here. I used it to CTRL-F for 'JBoss' to see if Red Hat would give any details on how well JBoss is doing. Red Hat provided no additional detail other than "JBoss pipeline looks good". Maybe I missed the memo, but isn't the pipeline for OSS products always "good"? You have thousands of users who are doing so without paying. The mere hope of converting them to paying customers means your pipeline is "good". But I digest.
As I commented on Matt's blog...30% is good (and I shouldn't minimize how good it is), but are we fooling ourselves thinking that 30% on a ~$500M 12 month revenue base is "proof that OSS changes everything"? Just for a little perspective [1] RIM grew revenues by 100% YTY in the quarter, on a 12 month revenue base of ~$3B. [2] Twelve month stock returns for RHAT, IBM, MSFT, ORCL.
I am more and more convinced that the OSS business model is broken. If we agree that the typical OSS vendor goes though these stages:
Stage 1: Get attention (total downloads is the key measure of success)
Stage 2: Get some revenue (tens of millions)
Stage 3: Get lots of revenue (i.e. > $250 mil)
A Support Subscription business model is the best balance of driving revenue and giving your users free stuff to build brand awareness during Stage 1 & Stage 2. But to get to Stage 3, the support business model does not scale (because people who would have paid to get your product will get your product without paying for it). OSS vendors need to sell products, not support. That is what Red Hat is trying to do (and has been doing very successfully with RHEL). The problem is, in getting to Stage 3, Red Hat (and JBoss) has trained a large set of its users that they can get great software without paying for it.
I wish I had a solution to this dilemma, other than "just use the commercial software business model from day 1".
PS: I should state: "The postings on this site are my own and don’t necessarily represent IBM’s positions, strategies or opinions."
Posted by Savio Rodrigues on December 21, 2007 06:23 AM
December 20, 2007 | Comments: (0)
It was great to read how Apple and Think Secret have settled their long-running law suit and delcared "positive results for both sides." The only downside is that the ThinkSecret site is being shut down. It's like saying the operation was a success, but the patient died. If this is a victory I'm not sure who won other than corporate censorship.
Nick Ciarelli, a long-time Apple fan-boy founded the site when he was in junior high school ten years ago and continued to run it as a senior at Harvard. Apple sued earlier in 2005 claiming violation of trade secrets and asking Nick to name names. With the settlement, Nick says no sources were revealed and he wants to move on.
Is there more to the story? Let me know your thoughts.
Posted by Zack Urlocker on December 20, 2007 09:39 AM
December 19, 2007 | Comments: (0)
I've been tracking the "Demise of the traditional software market" for some time now. One more data point from Oracle today.
Second quarter total revenues were up 28% to $5.3 billion, while quarterly net income was up 35% to $1.3 billion."In Q1 we reported new software license revenues up 35%, the strongest growth of any quarter in ten years," said Oracle President and CFO, Safra Catz.
Yes, 35% growth on a few billion dollar base (give or take a few hundred million)...not shabby. Now, before you think, "that's just acquisition related revenue growth", keep in mind that new license growth for database and middleware software grew by 28%. The applications business, which would be impacted by Oracle’s acquisitions, grew by 63%.
Oracle didn't mention anything about Unbreakable Linux during the earnings release conference call. What's more, no analysts asked about Unbreakable Linux or competition from OSS. The analysts were too busy congratulating Oracle on the quarter.
PS: I should state: "The postings on this site are my own and don’t necessarily represent IBM’s positions, strategies or opinions."
Posted by Savio Rodrigues on December 19, 2007 08:51 PM
December 19, 2007 | Comments: (0)
Maybe this is obvious to everyone already, but a recent study of Microsoft channel partners by IDC shows that Software as a Service is seen as a big opportunity by traditional Software ISV channel partners.
- 76% view that SaaS will dramatically impact the partner landscape
- 70% of solution providers view it as an opportunity
IDC expects more companies to promote SaaS next year, including SAP, Microsoft, Cisco and IBM. While SaaS is small today (2% of all software) it is expected to grow 25% annually and become a $14.5 billion industry by 2011. Sarah Frier at Goldman Sachs is also bullish on SaaS saying its a $100 billion opportunity that will be a gradual evolution.
Ok, I admit that analyst math doesn't always come true. After all, every market they study is a billion dollar market, but some end up being zero billion. But at least for the next few years, I think there's great opportunity for IT organizations to leverage software as a service in order to introduce applications more quickly and at a lower cost.
And what's powering SaaS? You guessed it: open source.
While the IDC study doesn't specifically comment on that, we see more and more of our ISV customers evolving their business towards a model that includes traditional on-premise software as well as on-demand options.
Posted by Zack Urlocker on December 19, 2007 02:49 PM
December 18, 2007 | Comments: (0)
SugarCRM has announced the GA version of Sugar 5.0. This includes new features such as improved dashboards with charts and graphs, an AJAX email Client, and a new Module Builder that allows easy customization. It's a major new release and if you haven't looked at SugarCRM in a while, it's worth checking out. There's also an online flash demo that walks you through the new capabilities in about 15 minutes.
SugarCRM has also announced an on-demand architecture that supports each customer having their own separate instance for greater security. Users can chose between having Sugar hosted on-premise or on-demand, which is an increasing trend among software ISVs. You can download the Sugar Community Edition or an evaluation version of Sugar Professional or Enterprise editions from their web site.
Also note that SugarCRM has their developer conference coming up February 6-8 in San Jose. It's a good opportunity for those who want to learn more about SugarCRM or are already using it and want to get best practices from other experts.
Posted by Zack Urlocker on December 18, 2007 06:47 AM
December 17, 2007 | Comments: (0)
TechIQ: 10 Open Source Companies to Dominate 2008
TechIQ, which focuses on the VAR IT channel business, has identified the top 10 open source companies they expect to succeed in 2008. It's a good article and although it includes some obvious choices, it also includes a few well-thought out surprises. So it's not your typical end-of-year round up. This one has some creative thinking behind it. I think there's a lot of potential for VARs to get behind the open source movement and help customers with implementations as well as integration of open source offerings.
(Disclaimer: I admit that I blushed to see MySQL in the list. But I would have posted this article regardless.)
Posted by Zack Urlocker on December 17, 2007 02:43 PM
December 17, 2007 | Comments: (0)
Red Hat, JBoss and Corporate Culture
By now, some of you have read that several analysts downgraded Red Hat due to Red Hat's execution of the JBoss acquisition. I chose not to blog the story because, well, it's about the performance of a competitor (JBoss) to IBM WebSphere Application Server.
Matt blogged the downgrade and his belief that JBoss is doing fine. Then things got interesting with comments CNet readers. I've clipped a few because they highlighted key points that enterprise vendors should consider before an OSS vendor acquisition. As Roy points out, an OSS acquisition is truly about the people, not the technology, being acquired. In Red Hat's defense, it was unlikely that JBossians would find the culture at any larger vendor, including RH, to be close to the freedom of the JBoss culture.
Roy:
"They're looking at the amount of money RHT spent on the JBoss acquisition and the poor revenue return on it. That sorta tends to happen when you destroy the sales and marketing machine JBoss had built."
Matt:
"I'm not looking for any blame. I've yet to see any reason to believe that Red Hat isn't getting JBoss moving. I know there are plenty of JBossers who have left the company who are happy to dig at Red Hat not loving them enough, but I'm not overly bothered by their outside perspective on what's going on inside the company."
Roy:
"Remember... what was RHT buying? They weren't buying code (OSS). So they were buying... yep... people! As luck would have it, once the "special" people started leaving, the investment's downward spiral began. Today it seems the surprise is on RHT."Then, a rather long comment from a (likely ex or current) JBossian:
"RH is more concerned with creating policies and initiatives that ran contrary to the spirit which Fleury and his team created - one of openness, caring for their work, and the motto have fun which was present every day till the day RH took over. RH was more concerned with making sure no one on @core used profanity, or communicated anything with anyone about any of the projects we were working on. It became the Borg Collective, and that hardly inspires creative and talented people to keep on doing what they do best.....
Did JBoss'ers think they were special, sure we did, and we were. We were the trailblazers, doing something innovative, and different. We fought the likes of IBM, BEA who tried to drown and kill us at every turn, and we survived, and even managed to come out on top in many ways. So yes, we are a special bunch, and proud of it. It's to bad RH management tried to take that feeling away. They are paying the consequences for it now, and there will be more to come."
Let's hope that comment didn't come from Marc speaking in the third person ;-)
Posted by Savio Rodrigues on December 17, 2007 05:25 AM
December 14, 2007 | Comments: (0)
Here are three points from Tim Bray's recent post. They very relevant to OSS. (I added the numbers to delineate his points):
[1] Getting started should be free: Also, it shouldn’t take more than a few days.[2] Some popular tools will be Open Source: Which raises a question: would you rather provide the popular OSS tools, or compete with them?
Now, monetizing OSS is hard. But on the other hand, monetizing software is hard, always has been. And doing it while you’re competing with Open Source is especially hard.
[3] To make money, give things away: Or as this pony-tailed executive I know says: "Monetization at the point of value."
There are two kinds of people in the world: those who might do business with you, and those who never will. For the second group, you lose nothing by letting them grab your stuff for free.
The first group is more interesting. On one hand you have the risk that they’ll go ahead and use your product or service without ever paying a penny for licensing or support or indemnification or whatever. On the other, there’s the risk that if you make them pay up front, they’ll try something else that is free-to-try and it’ll be good enough.
Very insightful. Give customers something free that is easy to get started with. Wait for the customers who would likely do business with you to follow through with that intent. And hey, if these customers buy commercial licenses, that's cool. Don't sweat the customers who will never pay for 'value'.
PS: I should state: "The postings on this site are my own and don’t necessarily represent IBM’s positions, strategies or opinions."
Posted by Savio Rodrigues on December 14, 2007 12:20 PM
December 14, 2007 | Comments: (0)
The Internet May Be a Sucky Customer to Base Decisions
Back in the day, Sun was riding high as a result of its success with Dot Coms. It was all going to be different; brick and mortar customers were laggards that "didn't get it", the future was in the hands of Dot Coms. Sun bet on the dot coms and well, you know the rest.
It's important to revisit that before validating strategies based on "Internet companies" of today.
Jonathan reports the following and Matt uses it to validate his beliefs that OSS is the only way forward and that the support-based business model rocks:
"We held an event a few weeks ago to which we invited a broad spectrum of such customers, one room filled with CTO's from some of the world's largest on-line companies (whose brands nearly everyone would recognize - the internet as a social phenomena is in full swing);"
"Not a single company in the CTO room paid for software."
"In contrast, not a single company in the CIO room allowed free software without a commercial support contract. Not one. Validating the notion that for more mature/diverse companies, the cost of downtime dwarfs the cost of a support contract."
It's surprising that Sun/Matt believes that the support business model is vindicated using this information from "Internet CTOs". I say this considering Red Hat, JBoss and MySQL, the 3 best known OSS brands/companies have all shifted away from a support business model. These vendors are selling software, they may call it something else, but in the words of some British guy, "would a rose by any other name be less tasty?"
PS: I should state: "The postings on this site are my own and don’t necessarily represent IBM’s positions, strategies or opinions."
Posted by Savio Rodrigues on December 14, 2007 12:04 PM
December 14, 2007 | Comments: (0)
Sam Ruby has a knack of making simple statements that convey so much. Here's an example:
"Standards that Matter are Standards that Ship"
So true. Sam talks about the <video> element in the HTML5 spec and how certain vendors appear to be addressing this spec requirement. This statement caught my attention:
"Fundamentally, Microsoft’s strategy is sound. Ignore standards that you find inconvenient, and focus on producing and enabling the production of content people want."
This is where OSS can play a role. The appropriate OSS community can implement standards (and specifications) that other vendors find inconvenient. This isn't new news to any of you I'm sure. But are there differences when we consider an OSS vendor vs. and OSS community?
An OSS vendor can choose to implement the "inconvenient standard/spec" and use it as a competitive differentiator versus the incumbent vendor. But are there "inconvenient standards/specs" for OSS vendors themselves? If there are, or will be in the future, do we expect OSS vendors to act differently than Microsoft is?
PS: I should state: "The postings on this site are my own and don’t necessarily represent IBM’s positions, strategies or opinions."
Posted by Savio Rodrigues on December 14, 2007 11:12 AM
December 14, 2007 | Comments: (0)
United Million Miles: Reward or Punishment
United Airlines, my carrier of choice, sent me an email this week indicating that they are now going to offer a special reward for flyers who achieve a million miles lifetime flying on United. Having flown over 900,000 miles with United so far, I am close to achieving this milestone in the next year. But if the rewards encourage more air travel, I'm not sure it's such a good thing. The last thing any frequent flyer wants is more air travel!
Posted by Zack Urlocker on December 14, 2007 07:17 AM
December 13, 2007 | Comments: (0)
Easier IT Recruiting with Jobvite
Q4 is usually a time when you're doing your planning for the next fiscal year and looking at staffing up in Q1. The only problem is it always takes longer to hire than you would expect.
Jobvite is an interesting potential solution. It's a software as a service offering that helps companies better focus and manage their recruiting efforts by getting the whole company oriented towards talent acquisition.
While I haven't used Jobvite, it is an interesting offering and it has been used by quite a few Silicon Valley firms, ranging from TellMe to Tivo. Jobvite continues to develop momentum and has recently announced Series A venture funding of $7.2 million from CMEA ventures.
I think its a good sign of how new applications are being tackled successfully as services. This is an increasing trend and should be a part of IT planning for 2008.
Posted by Zack Urlocker on December 13, 2007 03:43 PM
December 12, 2007 | Comments: (0)
How do you disrupt a billion dollar hardware industry? Open source it. That's just what Sun has been doing, rolling out their latest OpenSPARC T2 processor with an open source design available under the GPL.
You can download the design, specs, documentation and related tools for their T2 processor, which has eight cores and a 64 thread processor. The T2 processor is essentially a server on a chip with multi-core CPU as well as networking, security and I/O built in.
Sun also announced that the processor design is being used in several major US engineering schools including University of Texas at Austin, University of Illinois, Urbana-Champaign, Carnegie Mellon University, University of Michigan, Ann Arbor (Go blue!) and local faves UC Santa Cruz (Go banana slugs!)
While Sun has long been a proponent and supporter of open source, I don't think they always get credit for the level of commitment that they have. And it's not about open source altruism. The more open source is adopted, the more Sun benefits with it's platform strategy and expanded demand for its products. More information is available at www.opensparc.net.
Posted by Zack Urlocker on December 12, 2007 10:40 AM
December 12, 2007 | Comments: (0)
:-) Hey, that's what Marc's post says....Marc is becoming an advisor for RIA vendor Appcelerator.
Fleury writes:
"And finally, what sets them apart, at least for me, is that they are friends and a group of my favorite peoples. So let's go, I will be helping with as much visibility as I can bring them, mainly PR and this blog. They will need visibility. I am helping with messaging a bit, I am helping with the business model a bit but really all I do is listen to them and what they do and then repeat to them what they told me. This advisory thing is easy when the people in front of you know what they are doing."
From the Appcelerator website, their business model includes professional services, education/training, technical support and an option for commercial licensing as an alternative to the GNU-based OSS code.
Marc says:
"So what sets these guys apart? 1- First of all, 100% Open Source. No one else in the space is 100% OSS. ... "I do find it interesting that Marc points to the 100% OSS angle as setting Appcelerator appart. Considering the following comment from Marc, that I found this on Fleury's RSS feed around Nov. 26th (via Google Reader):
"I don't want to come across as the ultimate realist but as far as scalable business models go, we still haven't found anything better than a proprietary distribution of open source software."Marc, I'm with you 100%. There is no shame in making $$ through gated access to some version of your code. If Appcelerator is successful, you'll be advising them to shift from a Support Subscription business model (where 100% OSS matters) to a Software Subscription business model (where you have some proprietary products).
So why the big focus on "100% OSS"? Well, I know why..you need to be successful with a Support Subscription business model before you can shift to a Software Subscription business model. I just wish the OSS discussion wasn't always an either/or discussion.
Anywho, best of luck to Marc & his friends at Appcelerator.
PS: I should state: "The postings on this site are my own and don’t necessarily represent IBM’s positions, strategies or opinions."
Posted by Savio Rodrigues on December 12, 2007 08:24 AM
December 12, 2007 | Comments: (0)
Sun's $1M grant to Sun's FOSS Communities
Simon (Chief OSS Officer at Sun) talks about Sun's $1 million dollar grant to Sun controlled FOSS communities.
Net/net: Sun will take $1M and divide it amongst six Sun owned/controlled FOSS communities. The communities will propose how to use the cash to award contributors. Sun will approve these proposals with modifications as required. The program seeks to reward innovation.
Simon states:
"Sun will be announcing a multi-year award program in support of fostering innovation and advancing open source within our open source communities.....
This year's participants include OpenSolaris, GlassFish, OpenJDK, OpenSPARC, NetBeans, and OpenOffice.org."
Yes, the 6 OSS communities are Sun owned/controlled projects. This shouldn't be a surprise. Sun is a business, and it's their legal duty to improve returns for shareholders.
I'm not sure why a vendor would have to pay OSS developers to contribute to a project if (A) doing so is not already part of their day job, or (B) the developers isn't getting more benefit out of the project than they are inputting. I'm all for hiring the top contributors to a given project if they aren't already paid to work on the project.
This seems awfully like an attempt to use cash to drive interest around projects that may need more interest. Does anyone remember Red Hat/JBoss, MySQL, or other leading OSS vendors doing something similar? I can’t, but I don’t know/remember everything.
The only obvious point, as highlighted by the announcement in Bangalore, is that this cash could help attract developers from emerging countries into Sun controlled OSS projects. But the following statement from Simon suggests that the awards will go to current community members.
"We are wanting to recognise and reward innovation, which we fully expect to come mainly from existing community members including the many already employed to work on software (though not Sun employees since Sun is the sponsor of the awards). It's about sharing the wealth."
Maybe I'm over thinking things and it's "just about sharing the wealth". Stranger things have happened :-)
PS: I should state: "The postings on this site are my own and don’t necessarily represent IBM’s positions, strategies or opinions."
Posted by Savio Rodrigues on December 12, 2007 07:31 AM
December 11, 2007 | Comments: (0)
Having worked on a broad range of IDEs for many years when I was at Borland, I'm always interested in seeing the latest developments in this area. To a large extent, I think IDEs and programmer's editors are largely a matter of personal taste. One man's emacs is another's poison, so to speak. Passions run nearly as high in picking development tools as in programming languages. Which makes sense. The tools and language are where programmers live and influence what they can do; or at least the ease with which it can be done.
I was pleased to see Red Hat has announced an Eclipse-based IDE for JBoss, building on some great work done at Exadel. Exadel has always been a bit ahead of their time with powerful visual tools for the latest Java frameworks, so it's great to see their tools get wider exposure.
Red Hat's JBoss Developer Studio provides a lifecyle development tool with support for Java Enterprise Edition, Hibernate, Struts and Spring all within a single environment. Frameworks like Struts are notoriously complex to use and maintain, and so having a visual representation is key to making it work. You can visual develop the user interface of your application and the underlying interaction model. And when you want to get to the code, it's all there.
JBoss Developer Studio is available for both Linux and Windows for $99 as a subscription with updates. If you're looking for a way to streamline J2EE development and the associated frameworks, this is worth looking into.
Posted by Zack Urlocker on December 11, 2007 11:13 AM
December 07, 2007 | Comments: (0)
Internal vs. External Testing?
When I first arrived at MySQL more than 4 years ago, we didn't have much in the way of internal Quality Assurance. There were a couple of test suites that had been created and many engineers wrote unit tests. We supported a wide range of platforms, but formal testing was pretty spotty. As bugs were identified and fixed, the test suites were extended to help identify regressions. It was by no means a bullet-proof process; mostly the team depended on external testers to report issues. Sometimes a build would be created that would have some embarassing bug and we'd scramble to fix it.
Since that time, we have introduced many more internal QA processes. The test suites are now 10x larger than they were just a few years ago. Our build system and use of automated tools for static and dynamic code analysis has expanded. And we also eat our own dogfood, running beta and RC software more routinely on internal applications. For example, we've been running MySQL 5.1, now a solid release candidate, for more than six months on our web site and internal applications.
While it's taken a few years, I think we have arrived at a good balance between internal and external testing. There's no way we could ever hire enough QA people to make up for the tens of thousands who download MySQL every day. They report unique corner cases or weird combinations that internal testing would likely never find. And our internal testing has now made it so that quality is now more consistent and we have fewer regression errors.
So I guess it's not a question of Internal vs. External testing, but more a question of finding the right balance. Interesting that there's also a trend among closed source companies, like Microsoft, to adopt some of these techniques that used to be only found in the open source world. Microsoft has expanded their external testing with the Community Technology Preview (CTP) programs. Good to see the learning on both sides...
Posted by Zack Urlocker on December 7, 2007 06:36 AM
December 05, 2007 | Comments: (0)
As a follow-up to my earlier blog, I happened to get 15 minutes hands-on use of the Amazon Kindle, their new portable eBook reader. As it happens, Amazon's Lab 126 development team is located in the same building as MySQL's US headquarters, and so when I happened to spot a team of folks with a Kindle in hand at a local restaurant for lunch, I dove in and asked if I could try it out.
First point to note: the display is fantastic. To me, it's as easy to read as a book. While you can argue that the text is dark grey rather than black, it has excellent contrast. The underlying e-Ink technology has improved since its first introduction in the Sony Reader, but there is still a split-second flash of black when you turn the page which detracts from the smoothness and shatters the illusion of it being a real book. (Imagine if your ipod hiccuped every minute while playing music.)
Also, the device is not nearly so ugly in person as it might appear in photos. In fact I'd say it looks kind of cool, especially given the quality of the display. The Kindle may not be as elegant as an iPhone, but you're not going to feel like you're lugging a 1980's Kaypro around. The cover is a bit cumbersome and to me adds some unnecessary bulk, but that may just be a matter of personal taste. Overall, the Kindle is lightweight and easy to hold and operate in one hand. Performance appears to be quite good and downloading sample chapters or entire books is a snap. The user interface is not quite as elegant as an iPhone, but in about 5 minutes you know everything you need to know to use it.
My limited hands-on evaluation does not change the fact that at $400 the Kindle is still very expensive. If for that price it included some kind of Netflix-style book subscription, then I would probably go for it. But at $400 plus $10 a pop on books, the economics don't work for me. I don't have the need to carry more than 2-3 books at a time when I travel, and I'm willing to trade off convenience for price at this point. But if you're schlepping ten pounds of reference books around, then you may view it differently.
And as my lunch companion pointed out: if you've already got a cell phone, a Blackberry, an iPod and a laptop, then adding one more device seems like overkill. But if the Kindle could be a general purpose browser and email device, that would be a different story.
Maybe that's Kindle 2.0?
Posted by Zack Urlocker on December 5, 2007 01:17 PM
December 04, 2007 | Comments: (0)
One of the virtues often touted of open source is the idea of "release early & release often." Eric Raymond documented the Linux kernel team's use of this approach in detail in his book "The Cathedral and the Bazaar."
While "release early & release often" is not specifically tied to open source, it is a good practice to keep software in a ready to release state. Microsoft has long used the notion of nightly builds in order to make sure that new check-ins don't break existing code. (Or that if they do, whoever broke the code is on the hook to fix it.)
At MySQL we have a build team that helps keep our various software products in good shape, but there are often complex code merges across multiple teams required.
We're recently introduced the idea of "feature preview versions" that enables teams to have a bit more flexibility in releasing software more frequently on the MySQL Forge. The idea is that a team can release something which is a feature version, for example, the recently released MySQL 6.04 preview release of the Falcon storage engine. In this case, the team released this preview version ahead of the normal Beta release and ahead of code merges with other teams. There's a number of internal improvements in this release of Falcon that improve performance, especially on multi-core, large memory systems.
The feature preview version is not intended as a production release, but it does enable us to continue to release more regularly and ensure that those who want the earliest access to new software can get it, even though it may be a bit rough around the edges. Those who are more conservative can wait until all of the code merges are completed, fully tested across multiple platforms and then officially released as part of the Falcon beta.
Posted by Zack Urlocker on December 4, 2007 02:23 PM
December 04, 2007 | Comments: (0)
ESTD: Converting Your OSS User Base
Easier said than done...converting your OSS user base. Read on, but that's the summary of this post.
Joel Jackson of Red Hat said:
"Do you think the genius of OSS is how we sell it? Seriously. How did you get a job blogging on open source my friend.We already know that to scale in OSS you start off by doing two things: (1) writing great code and (2) getting market share (by selling what you call "Support" subscriptions"). Then, right about the time your OSS company is in Gartner's upper right "magic quadrant" with the other proprietary companies, you establish a revenue model that try's to convert much more of your user base (by selling what you call "Software" subscriptions).
Is this what you were trying to educate us on all along?
Come on and say it with me: The real genius in OSS isn’t the way we sell software. It’s the way we develop software and crush the development models of proprietary companies in the process. OSS transforms the way software is developed, distributed and supported. "
Joel, I truly appreciate your comments. Look, I am an OSS fan. But, I don't subscribe to the "it'll all work out in the end" line of thinking that seems to be the norm amongst OSS proponents. I *think* that the balanced approach is why I got a 'job' blogging about OSS ;-)
I can't reconcile that OSS is game changing, yet is only estimated to represent < 2% of customer spending in 2011. The fact that 2011 will be 20 years into the life of the commercial OSS market, if you base t=0 is when Red Hat was founded. I keep dropping those numbers to bring some reality to the discussion. Heck, let's say the <2% is really going to be 4%, or 10% in 2011...is that what we should define as "game changing"? In my opinion no. Is revenue the best measure of OSS success? Maybe not, but for those of us who get paid to work, it's probably the best measure of success out there.
Joel, I mean no disrespect to OSS proponents that have predicted the end of the commercial software business model as we know it. I like and respect many of them. But I do feel that they trivialize the tough battle ahead for OSS.
Yes, OSS is great if your company is able to (1) write great code and (2) gain share via a Support Subscription business model. But it's a completely different ball game when you (3) try to convert your user base by selling Software Subscriptions. When you start doing (3) you are competing vs. the commercial software vendors who have been doing this for decades. OSS plays on its home turf with (2), and (1)...but there are also great commercial products out there. The real challenge is for OSS vendors to attain success with (3), namely, converting users into paying customers. Oh, and as you start selling products to do this (under the guise of a service or not), you are playing the commercial software game. It's a tough game....tougher than OSS proponents make it out to be.
PS: I should state: "The postings on this site are my own and don’t necessarily represent IBM’s positions, strategies or opinions."
Posted by Savio Rodrigues on December 4, 2007 12:51 PM
December 03, 2007 | Comments: (0)
Support Subscription vs. Software Subscription
On my previous post about the scalability of the "support subscription" business model, Joel Jackson said:
"I’m confused Savio...what don’t you understand about a software subscription annuity model? $200-500mill today in base revenue could easily be (along with a modest growth rate of new customers, which Red Hat doesn’t seem to have a problem getting) over $1bill in revenue in the near future"
Joel, thanks for bringing up this point. I'm talking about the Support Subscription business model, not the Software Subscription business model. I may not have been clear on the difference in my previous posts.
JBoss provides us with a good example of the two in practice:
Prior to the acquisition, JBoss employed a Support Subscription business model. Anyone could freely get and use JBoss products without paying JBoss a cent. Customers that wanted to call in with technical support questions and receive defect fixes faster, etc. had to pay for these rights. The customer paid for support around freely available products. The customer was clearly paying for a service (i.e. support, not to be confused with professional services a la IBM Global Services or Accenture).
Several months after the acquisition JBoss adopted the "Fedora-model", or, in layman’s terms, moved to the Software Subscription business model. Customers can only get certain versions of JBoss products by paying for "a JBoss subscription". This subscription includes technical support and monitoring services (which were part of the Support Subscription business model that JBoss previously employed). The additional access to non-freely available products is what sets this business model apart from the Support Subscription model.
Software Subscriptions has been a component of the commercial software business model for longer than I've been alive. FWIW, over 50% of Oracle's revenues are driven by Software Subscriptions. Software Subscriptions in the commercial software business model give customers access to non-freely available products. The customer has the right to receive technical support & maintenance around these products under the Software Subscription business model.
If you just said, "wait a minute, are you saying the Fedora model / JBoss Subscription is just like the Software Subscription business model that commercial software vendors have been using for decades?" Bingo.
The Software Subscription business model scales because customers are buying products. When a customer stops paying, they no longer have access to the product. Contrast this with the Support Subscription business model where customers can elect to use the product without support, or start with support and move to unsupported if they didn't make many support inquiries in the previous year.
There is absolutely nothing wrong with OSS vendors employing the Software Subscription business model. It's the best business model (from a revenue standpoint) that we've seen in the software industry. Just don't tell me that OSS "changes all the rules and has a cherry on top".
PS: I should state: "The postings on this site are my own and don’t necessarily represent IBM’s positions, strategies or opinions."
Posted by Savio Rodrigues on December 3, 2007 06:18 AM
December 02, 2007 | Comments: (0)
Fit for Purpose - OSS & Commercial Products
Okay, so not everyone agrees about my claim that the support-based OSS business model does not scale. :-)
Matt said:
"The open source doubters continue to underestimate the power of open source as a business model tool. And Savio, though normally I wouldn't characterize you as anti-open source, you seem to want, like many at IBM, to keep open source nicely contained in a package that is easy to leverage but doesn't impact your license business."
I am not anti-OSS by any means. But, I do question where it's all heading. I've long believed that the future is a hybrid model. Twenty years from now, I believe that leading software vendors will offer a mix of OSS & commercial products. I believe this because I've seen this model work already. When I do the math, it's very difficult to envision commercial software vendors throwing away the business model that is driving a $200 billion software industry today. I am not trying to be condescending in any way, I'm just looking at the data.
I don't question that substitution is going on, especially from startups. I am 100% sure that substitution is going on for certain uses. There are some projects/uses that OSS is the better answer. And there are other projects/uses that commercial products are the better answer. Until recently, customers didn't have much choice and used commercial products for both types of projects/uses. OSS changes that decision, and tips the scales in favor of the customer. All software vendors recognize this and some are acting on it.
The future isn't OSS - sorry to rain on the parade. The future is recognizing that customers have varying project needs and neither OSS, nor commercial products are the best solutions for all these project needs.
PS: I should state: "The postings on this site are my own and don’t necessarily represent IBM’s positions, strategies or opinions."
Posted by Savio Rodrigues on December 2, 2007 08:49 PM
- Get Started
- Port 25 Blogs
- OSS News
- Join a Project
{Open Source} Heroes Happen Here
Start today and order your own Hero Hack Pack – which includes Getting Started with Open Source, Windows Server 2008 and Visual Studio 2008 Trial. Each pack is a chance to win a free pass to OSCON 2008.
TOP STORIES
Microsoft's post-Yahoo optionsNet neutrality bill introduced
MS adds $3 million to Big Easy
AMD's Java improvement efforts
Leopard at 6 months
Intel still investing in WiMax
Yahoo tests aggregated search
Developers vs designers
Sun defends JavaFX Script
Botnet spams 60B a day
ADDITIONAL RESOURCES
- Virtualization: A Step by Step Approach to Success
- Dialing up Agility with Business Transformation
- 5 Things You Need to Know About Storage Virtualization
- Virtual Test Lab Automation: Manage development infrastructure
- Improve Resource Utilization and Lower Operating Costs
- Protect Your Data with SSL
