- Matt and Dave Podcast on IT Conversations
- Microsoft and Zend: Faster PHP for Windows
- Microsoft makes nice with another open source company (SocialText)
- Google: Turning weaknesses into strengths (Jotspot)
- Moving from open source "project" to "company"
- Open source keeps your options open
- Analyst: Oracle won't be able to outperform Red Hat on support
- Dries Buytaert on forks vs. distributions
- MySQL: Stanford GSB case study (Part II)
- Hiring and avoiding the clown car
October 31, 2006 | Comments: (0)
Matt and Dave Podcast on IT Conversations
Matt and I were invited by Phil Windley to be a part of his Technometria podcast series along with Scott Lemon. Thanks Phil!
The group first reviews the state of customer support from technology companies. Unfortunately, some of the most well-known companies are not succeeding in making their users happy. Are bloggers having a positive impact? For example, a company may be more likely to try to make someone happy if the customer is writing in a public forum that might influence others.In addition, an assessment of alternative operating sytems leads to a discussion of the state of Novell and how they are losing kernel developers to Oracle. Is this related to Novell's becoming more of a Linux company instead of an open source company?
Other topics discussed include Microsoft's support for the SCO lawsuit, the release of Firefox 2.0, and application generators, such as Ming and Rails.
Links:
Alfresco
AC/OS (Matt's Blog)
MediaForge
the.Inevitable.Org/anism (Scott's Blog)
MuleSource Inc
Open Sources
This program is from the Technometria with Phil Windley series.
Posted by Dave Rosenberg on October 31, 2006 09:28 PM
October 31, 2006 | Comments: (0)
Microsoft and Zend: Faster PHP for Windows
Microsoft has been on an open source tear lately. Yesterday it was SocialText. Today it's Zend. (Oracle might want to take note: this is how you partner with open source companies.) (OK. No more Oracle commentary.) (Until the next post. :-)
As announced today, as reported by InformationWeek:
Microsoft has been working with Zend to get Zend's PHP open-source scripting language to run faster on its Web server. Windows Information Server has been Microsft's prime platform for its own high-performance language, Active Server Pages. But with Web sites frequently including non-Microsoft technologies, Windows Information Server was handicapped when developers were working with PHP.No more, apparently. Microsoft knew that to compete with Apache on the web server front, the all-encompassing ecosystem model wasn't going to work. Performance matters on the web - religiously devoted to Microsoft products or not, an IT guy is going to be drawn-and-quartered if the website crawls because Microsoft can't push PHP as fast as the open source alternative.
And so they're partnering.
You've got to admire Microsoft. For all the company has done to try to kill Linux and open source generally, its lack of shame is amazing...and even admirable. The company is moving on. I know people whose perspectives of Microsoft are petrified circa 1980s/1990s. That's unwise. The market is moving fast, and Microsoft is moving with it.
Now with Zend, they're moving even faster.
Posted by Matt Asay on October 31, 2006 03:31 PM
October 31, 2006 | Comments: (0)
Microsoft makes nice with another open source company (SocialText)
If you've followed this blog, you know that I feel Microsoft's Sharepoint is the future of Microsoft's enterprise lock-in. Intriguingly, an open source company, SocialText, has joined forces with Microsoft to help them achieve that goal even faster. Ross and company are even providing an easy migration from JotSpot (acquired by Google), just in case you wanted to expedite the lock-in. :-)
Ross is a smart guy and understands that open source is still too young to be linking up solely with open source companies. Microsoft, for its part, recognizes the future, and so has partnered with SugarCRM, Zend, MySQL, and is undoubtedly working toward partnerships with others.
It's a good move. I'll feel bad when I put Sharepoint out of business, but maybe then Ross can come partner with my company. :-)
Posted by Matt Asay on October 31, 2006 02:35 PM
October 31, 2006 | Comments: (0)
Google: Turning weaknesses into strengths (Jotspot)
Google has been in the market lately for features that could never make standalone companies. Jotspot is the latest find. By itself, JotSpot would have struggled to make a viable business - enterprises buy products with wikis as features (like content management systems or office productivity suites), and consumers aren't going to pay enough to fund a wiki company.
But with this acquisition, Google has really pushed its "office 2.0" agenda much further ahead. Alone, JotSpot was just an interesting curiosity. With Google's other cloud-based applications, it starts to look like an interesting application.
Thus, in Innovator's Dilemma fashion, the original innovator, Microsoft (or, more appropriately, WordPerfect/Novell) has been supplanted by the upstart, Google. Not financially. Not yet. But mindshare is on Google's side on this one.
It will be intriguing to see if Office Live will prove an effective counterattack. My bet is that it won't, over time, because Microsoft is too wedded to its Office revenue stream. Maybe the fact that Google mostly fails to sell anything except advertising, which advertising has proven weak for everything except its core search business, will work to its advantage on this one. Just as Red Hat was forced to sell service instead of bits because it couldn't do otherwise, so, too, might Google be forced to succeed in the office 2.0 world precisely because of its inability to monetize it in traditional ways.
UPDATE: Here's an interesting piece from Bill Wise on how Google uses the buzz from failed experiments to drive its core search business.
Posted by Matt Asay on October 31, 2006 01:42 PM
October 30, 2006 | Comments: (0)
Moving from open source "project" to "company"
SandHill.com is running an opinion piece I submitted "Launching an Open Source Business" which outlines some of what we went through in our funding process.
It's a decent read despite the fact that I wrote it after a severe bout of food poisoning last week. It includes a short list of things to consider during the funding process. The most important thing is that you must be the expert on your product and your market.
Posted by Dave Rosenberg on October 30, 2006 07:44 PM
October 30, 2006 | Comments: (0)
Open source keeps your options open
Mule project founder Ross Mason has an article running over on Dr. Dobbs called "Five Ways Vendors Gouge Customers on Integration."
1. Up-Front Coding, Proprietary Toolkits
2. Astronomical Licensing Costs
3. Ongoing Services/Support Costs
4. Getting Locked Into a Prescribed Approach
5. Feature Overshot
And while integration is clearly an expensive pain, the five points he elucidates are applicable to pretty much any market segment dominated by proprietary vendors.
Posted by Dave Rosenberg on October 30, 2006 07:38 PM
October 30, 2006 | Comments: (0)
Analyst: Oracle won't be able to outperform Red Hat on support
I was reading a report from Hapoalim Securities today on the Oracle move (to support Red Hat Enterprise Linux). They clearly understand how open source works (which is atypical - Oracle clearly does not):
Yep. I do think that Oracle has good Linux expertise in-house. But translating that into a compelling support offering, especially for someone else's product (which, as I've said before, is more about service than it is about bits and bytes), is going to be a long, expensive road for Oracle, and not financially advantageous for its customers.
- We reiterate our Outperform rating on RHAT because we believe ORCL is likely to fail in its bid to take over the Linux market, or even to capture a significant amount of share from Red Hat. We thus see this as an attractive buying opportunity for RHAT.
- Previous efforts to “clone” Red Hat Enterprise Linux have all withered due to the same factors we see dooming the Oracle effort: lack of independent software vendor support, insufficent economic return to justify the risk of switching suppliers, and difficulty building a support organization that can deliver high quality support.
- ORCL’s raisons d’etre for this offering – lower support prices, longer support period for old versions, and a larger support staff – are not likely to be compelling to customers, in our view. Customers can (and do) save far more money by getting a bigger discount on Oracle’s database than they could if all Linux support were free. Customers know the overall size of the vendor’s support organization doesn’t matter, only the number of trained specialists with the necessary expertise does. And the fast-moving Linux marketplace places less value on the need for support for older versions than the moribund application market where this message resonated well.
- Our belief is that ORCL will at worst end up taking only about two to three market share points from RHAT, almost exclusively in the area of providing the Linux distribution under the database servers in an application, not enough to justify the 40% drop in the stock in the last month as rumors of Oracle’s entry into the Linux market have resurfaced in print.
(Btw, Oracle, if there's even a shred of credibility to your claim that you made this move "for your customers," then I'd recommend you working more closely with Red Hat to provide the support these customers apparently crave and can't get from Red Hat alone. That would be the right thing to do. Not what you did.)
Posted by Matt Asay on October 30, 2006 10:27 AM
October 30, 2006 | Comments: (0)
Dries Buytaert on forks vs. distributions
Dries Buytaert, lead developer on Drupal, has written an insightful post on the difference between forking a project and different distributions of that project. As he notes, the differences can be subtle, but ultimately come down to the intentions behind the divergence, revealed in how much the fork/distribution relies on the original project's foundation for future development.
The distinction is critical in light of Oracle's fork of Red Hat Enterprise Linux: customers that buy into Oracle's support are effectively buying into a fork, and one that Oracle is ill-equipped to support.
But the distinction plays out every day in a myriad of different projects, and occasionally turns into a full-fledged fork, which tends to benefit no one (which is why it rarely happens in the community-minded open source world. Dries writes of Drupal:
It is important that Drupal distributions collaborate, and not compete. To do so, we have to provide Drupal distributions an environment that encourages collaboration, and that allows for specialization (such as custom documentation and support) without introducing incompatibilities that drive competition.In short, distributions are fine because they add value to the core. Forks are bad because they splinter communities and thereby fragment value, attenuating it until the point that either the fork dies or the originating project dies.The good news is that we know how to do this. We've been through this already with CivicSpace (previously called "DeanSpace"), a Drupal distribution for online campaign management and grassroots activism.
They were quick to realize that the success of the CivicSpace distribution depends on the success Drupal core, and vice versa. The decided they shouldn't fork core development. Instead, CivicSpace decided to do all its development on the drupal.org infrastructure, to synchronize releases, to submit all patches upstream, to centralize bug reports, and to share documentation where possible. Collaboration, not competition.
The bad news is that can be hard work. People will find that creating a distribution is fun and easy, but that being a responsible maintainer might be a lot less fun. Who wants to track changes, write documentation, maintain modules, provide upgrade paths, manage releases and provide support for years to come?
For this reason, I agree with Dries. What's true of Drupal is true of other open source communities. He writes:
As a community we should disapprove Drupal distributions that do not intend to collaborate, that have no signs of long term commitment, or that risk locking people in.Amen, Dries.
Posted by Matt Asay on October 30, 2006 09:39 AM
October 30, 2006 | Comments: (0)
MySQL: Stanford GSB case study (Part II)
Christof Wittig, CEO of db4o, and former student at Stanford's Graduate School of Business, has written an excellent follow-up to his original MySQL case study.
Christof does a great job of tracing the challenges facing MySQL, as well as the way the company has successfully responded to many of them. He also identifies an interesting, parallel trend to how Linux spread:
The first enterprise-wide users of MySQL were Internet-enabled start-up companies which turned to free open source software stacks such as LAMP to get their business off the ground. These companies not only saw the lower cost advantages of open source, but also valued the ability to modify parts of the underlying software to differentiate themselves in their own markets. Over time, some of these companies grew into very successful and profitable large enterprises, including Google, Yahoo!, and Sabre Holding’s Travelocity.In 2002, it was absolutely the case that Linux was an "edge of the network" phenomenon. Not anymore. It's everywhere, just as I expect MySQL will be.
In more conventional companies, initially only peripheral installations of MySQL were used to run websites or departmental solutions. Fueled by positive experience, companies then slowly started to “scale out” MySQL into other departments. With MySQL’s growing brand recognition and enhanced service offerings, even more conservative users started to look into MySQL. MySQL actively promoted the intra-company dissemination of its products.
And, somewhat humorously, as MySQL has added functionality and tweaked its business model to target traditional enterprises, it's starting to get complaints, as Marten Mickos, CEO of MySQL, notes:
For the first time we have customers that are complaining. And we used to have no customers complaining. It's a question of expectations: When you have a five-star hotel and you charge virtually nothing, then nobody will complain - even if there is something to complain about. Now as we charge more, they also demand more and we pay even more attention to their requests.This is obviously a good problem to have.
Christof's case study isn't yet publicly available, but watch the URL above for when it is.
Posted by Matt Asay on October 30, 2006 09:10 AM
October 29, 2006 | Comments: (0)
Hiring and avoiding the clown car
Starting a company is challenging in many ways, and so far I would say that hiring the right people is by far the most difficult. As MuleSource is still very early stage, we have to consider the role being filled, the personality fit and how it will jibe with the rest of the team. You also have to be extremely careful not to hire someone just because you want the problem solved right away. There is one golden rule that I have learned about hiring:
You bring in one bozo and the next thing you know it's a clown college.
Sure, it sounds funny, but I can vividly recall almost every startup I've worked for bringing in that one person or role that started a downward spiral. Some people call it the "B-player". And typically, B-players are not workplace educated enough to know to hire A-players. They hire B-minus or C-players and so on until you have a team of duds. Well-managed companies move quickly to rectify the situation whereas poorly managed companies empower further incompetence by allowing the situation to remain.
Of course, you have to be realistic about hiring otherwise it will just be you doing everything. Which is somewhat of the position we've gotten ourselves into-dramatically understaffed in everything but our support team. In my case this makes sense; support is the core of my business and we treat it as gold. But other areas, such as business development and product management are more challenging. As the CEO I have no choice but to triage until I can find the right people to bring in, and we've made some key hires with more in the pipeline.
Being that commercial open source is such a new concept it's very difficult to find people who are well versed. In many cases it doesn't matter--but it's all dependent on the individual. But I continue to wonder if the people coming out of companies like Oracle and Siebel really get it. Using the example of Oracle I wonder if the rest of the company treats open source the way that the management does. Matt and I have been talking about this for over a year. And now that I am past funding and into hiring I am concerned that we are a still in the same boat. Lots of jobs, few realistic candidates.
All this leads up to an interview in Fortune with author Michael Lewis. I read Moneyball several years ago and didn't get into it-mainly cause I could care less about baseball. But Lewis has great insight into the development of leaders and managers and how individuals drive innovation.
Have you found a common link among the innovators that you have written about?In all these cases, just about, necessity breeds innovation. People are put in natural underdog situations where if they do things the way that everybody else does, they are certain to lose.
Where I have encountered greatness, there is an ability to go a different direction from everyone else while behaving with confidence and assurance as if you're just doing it the way things should be done. It's now very fashionable to be an innovator and to be a change maker. You get lots of people throwing the terms around. People think they are more unusual than they are. Whenever you see someone say, "I like to think outside the box," you know that they are so deeply in the box that they'll never get out.
Posted by Dave Rosenberg on October 29, 2006 03:10 PM
October 27, 2006 | Comments: (0)
Red Hat isn't cutting prices, and here's why
Matthew Szulik is on the record today saying that Red Hat has no plans to cut its prices. Good for Red Hat. When you're already #1 in customer value two years running, a shot over the bow shouldn't lead to widespread panic.
Oracle claims to be offering for support for Red Hat Enterprise Linux. They're not. As Red Hat notes, and as even a cursory investigation would reveal, open source software is not about the bits. It's about the support and the innovation you put into the bits. That's what Red Hat charges for.
It's what Oracle can't replicate with a keynote and a few webpages. Oracle could support Linux - arguably very well. But Oracle didn't choose to support Linux, a technology. It announced that it is offering support for someone else's product, but in the process confused the old world definition of "product" (bits which we can charge big upfront license fees for, as Oracle does) for the new world's definition (support, RHN, etc., for which Red Hat earns its keep over time through a subscription).
That is, Oracle announced support for...Red Hat's support, thinking that it was announcing support for Red Hat's bits and bytes.
It's a bit like me saying I'm now offering support for IBM Global Services. Sure, I could try. I could "fork" those services. But would you buy them? Neither would I.
In like manner, Oracle can offer huge discounts on Red Hat Enterprise LInux - the bits and bytes. But it's hard to discount beyond what the discounts Red Hat already offers: 100%. How much lower than free can Oracle go?
No, Oracle tried to confuse the issue, or was simply confused. It can't support Red Hat Enterprise Linux, because really Red Hat's product is the service it puts into the kernel, both pre- and post-sale. Oracle can declare that it will answer the phone faster, but who cares? it won't (and can't) have the answers when you call? It can say that it's supporting Red Hat's...support, but it isn't.
Oracle is offering nothing with this move. But at least it's offering 50% off.
Posted by Matt Asay on October 27, 2006 04:55 PM
October 27, 2006 | Comments: (0)
News: Polycom linking up with Digium
Cool news on the telephony front: Polycom and Digium are partnering to provide telephony solutions for the SMB market. From the press release:
Polycom's award-winning SIP desktop and conference phones will be combined with Digium's Asterisk Business Edition, the professional-grade version of Asterisk, the industry's first open source PBX. Through the combined Digium and Polycom offering, SMB customers will have access to advanced telephony solutions which will be more affordable compared to proprietary systems. Additionally, the new offering will give customers the control, rapid feature development and deployment, and rich feature base that the Asterisk open source community and its partners provide. Per the agreement, Polycom will also be the preferred VoIP phone provider for Digium solutions.Great work, John, Mark, and team. Nothing like knocking down a massive partnership in your first few months of venture funding. :-)
Posted by Matt Asay on October 27, 2006 04:30 PM
October 27, 2006 | Comments: (0)
Paul Graham: 18 mistakes that kill startups
I've had this Paul Graham post bookmarked for a week or so, and only just got around to reading it. Pithy, and dead on. Maybe even his location point. ;-)
I particularly liked this point, because it applies so well to open source, where no one has figured out The Right Model (in most startups it's hard enough to figure out the right product):
In some fields the way to succeed is to have a vision of what you want to achieve, and to hold true to it no matter what setbacks you encounter. Starting startups is not one of them. The stick-to-your-vision approach works for something like winning an Olympic gold medal, where the problem is well-defined. Startups are more like science, where you need to follow the trail wherever it leads.Great insight. Here's the list, but you really need to read the commentary to learn from it:So don't get too attached to your original plan, because it's probably wrong. Most successful startups end up doing something different than they originally intended—often so different that it doesn't even seem like the same company. You have to be prepared to see the better idea when it arrives. And the hardest part of that is often discarding your old idea.
1. Single Founder
2. Bad Location
3. Marginal Niche
4. Derivative Idea
5. Obstinacy
6. Hiring Bad Programmers
7. Choosing the Wrong Platform
8. Slowness in Launching
9. Launching Too Early
10. Having No Specific User in Mind
11. Raising Too Little Money
12. Spending Too Much
13. Raising Too Much Money
14. Poor Investor Management
15. Sacrificing Users to (Supposed) Profit
16. Not Wanting to Get Your Hands Dirty
17. Fights Between Founders
18. A Half-Hearted Effort
Posted by Matt Asay on October 27, 2006 04:07 PM
October 27, 2006 | Comments: (0)
Alan Cox: Open source is not divinely ordained to be more secure
In the interest of putting things in soundbite-sized nuggets of "wisdom," people sometimes say dumb things like "proprietary software is not secure" or "open source is secure." As Alan Cox pointed out at LinuxWorld London this week, however, this is just not true, and doesn't really help anyone's cause:
Things appear in the media like open source software is more secure, more reliable and there are less bugs.While his statements may be anathema to the open source delusional (who are embroiled in the political open source vs. closed source fight, as Dana notes), they just make sense.Those are very dangerous statements. That analysis just looks at well-known projects. If you take 150 projects from SourceForge [a repository for open source code], you do not get the same marks as you would with the Linux kernel. The debate of Microsoft saying 'Look how secure we are' versus Linux saying, 'We're more secure' is not looking at the important points.
High quality only applies to some projects — those with good code review and those with good authors.
The fact that I write something and release it under an open source license says nothing about its inherent quality. The fact that several billion people on the planet can opt to look at it doesn't make it more secure, either. And even if 99% of the world's population did stare long and hard at that code, it wouldn't make any difference, because 99% of the world are just like me: code illiterate.
Open source software, in its optimal form - great authors and a great community around it - can't be beaten for security. But open source, by itself, is just a licensing and development methodology. It's not inherently good or bad, secure or not secure.
Dave and I made this point during our Oracle "Open" World presentation this week, and the IT-centric crowd didn't bat an eye. They understood that quality of code can't be retrofitted onto software through process - it has to start with good authorship.
I just wish the political nuts on both sides of the debate got it, too, so that we could go back to talking about features, price, and support. You know, the things real buyers care about.
Posted by Matt Asay on October 27, 2006 04:04 PM
October 27, 2006 | Comments: (0)
Attribution: Full-blooded open source
Ah, back to the delightsome days of accusations of mudblood open source. The newest whipping boy? Attribution clauses.
Who uses attribution? Several of the most promising open source companies...as well as the European Union (EUPL). They seem to like open source, as does the Creative Commons (which also uses attribution as a core licensing element). Oh, and so does Mozilla, which is in a tussle with Debian about improper use of its trademarks. (See, you have to police trademarks if you want to keep them and, no, trademarks are not from Satan.)
So, attribution is actually fairly common in open source. You might not know this, but the OSI approved attribution licenses way back when (e.g., Attribution Assurance License, as well as the original BSD license). And even recently (CDDL). But they're not in vogue now with "the community," I guess.
Here's why they should be:
- If someone benefits from open source, they should contribute back. In Marten Mickos' language, users can contribute cash or code (bug reporting, bug fixes, functionality improvements, etc.). But what if the license doesn't nudge them to do so? Then they won't (at least, 99% of the time).
(This is why I've complained about Google and Yahoo!, two of the industry's biggest beneficiaries of open source. They and nearly every other Web 2.0 company use and modify open source software and provide commensurately little value back because they don't have to - the ASP loophole in the GPL allows them to benefit without giving back their modifications, or cash, or...
Attribution. Attribution is a way of claiming some value back in return for modification and use of one's software, even if the user opts not to contribute code or cash. It's not onerous. It's fair. You don't want to give me money for my software, or modifications you make to my software? Fine. At least reflect where you got it.
- Attribution does not burden distribution. If you think it does, I invite you to talk to the scads of hosting companies and others that are using Alfresco's Community product, for example. (Try Intelliant, for one.) We find out about them all the time - they abide by our MPL license and have proper attribution. It's a win-win.
Even if it were burdensome, since when does an open source project need to concern itself with facilitating generous financial returns on its software? Red Hat probably would have loved to have had a non-GPL version of Linux to sell when it started. Instead, it adapted the traditional software model and is the better for it.
Here's what the Open Source Definition actually requires:
The license shall not restrict any party from selling or giving away the software as a component of an aggregate software distribution containing programs from several different sources. The license shall not require a royalty or other fee for such sale.
Do you see anything in there that would restrict attribution? Me, neither. Read the other nine elements of the OSD - they don't impinge on attribution, either. - Attribution fits a model of Droit Morale. Maybe not so important in the US, but very important in Europe, especially France and Germany. If I make a nice painting in France, for example, I have a moral right to attribution, among other things.
Again, this sort of right hits home with applications, which have a UI, and less so with infrastructure software like a database. But it's important for many developers that the UI they create be attributed to them.
- Attribution protects customers. Developers want to distribute their software widely, but they also want to ensure they get credit (or, in some licenses, if you change the software they want to make sure you don't blame them for the problems that might result). It's also important to many end-users that they're getting the genuine brand (trademark) they trust. Just as Red Hat from a source other than Red Hat is something much less, so, too, is SugarCRM from another source less than SugarCRM. Attribution is a way of protecting the customer experience without burdening distribution.
These are just a few reasons to welcome, not discourage, attribution. If you disagree, I invite you to comment.
Btw, the OSI is actively considering attribution clauses (I'm on the board), so stay tuned to see how that debate plays out.
Posted by Matt Asay on October 27, 2006 11:04 AM
October 26, 2006 | Comments: (0)
Is your MacBook dozing off during email? Get your firmware upgrade here.
The SMC Update improves the MacBook's internal monitoring system and addresses issues with unexpected shutdowns. This update is recommended for all MacBook systems, including those that received warranty repair.
Man, I wish I had something to say about all that Oracle vs. Red Hat noise.
Posted by Dave Rosenberg on October 26, 2006 08:46 PM
October 26, 2006 | Comments: (0)
European Commission testing quality of open source software
The European Commission has raised over $3M to fund a project that will test the quality of open source software.
Among its goals, the SQO-OSS will benchmark the quality of source code to help to prove its suitability for deployment in businesses. It will also publish a league table rating open source applications according to their perceived quality.Quality is always in the eye of the beholder (or, rather, in the eye of the IT buyer that is implementing the software), but this study should be a useful addition to the conversation."An industry matures when its products become standardised," said Diomidis Spinellis, project lead, and professor at the University of Athens in Greece. "Through the objective evaluation of open source projects, SQO-OSS will provide many smaller and less known projects with the visibility and respectability they deserve," Spinellis added.
Posted by Matt Asay on October 26, 2006 02:03 PM
October 26, 2006 | Comments: (0)
Breaking news: MySQL to offer support for Oracle databases???
In a dramatic twist on Oracle's news yesterday that it will be offering support for Red Hat's Enterprise Linux, word on the street is that MySQL is considering offering support for Oracle's venerable database.
The hitch, however, is pricing. Early estimates suggest that MySQL would need to charge 1000% the cost of Oracle's typical pricing to make a profit on the support, given the persistent bugginess of proprietary databases compared to MySQL.
/tongue in cheek :-)
Posted by Matt Asay on October 26, 2006 11:43 AM
October 26, 2006 | Comments: (0)
Hidden costs in Windows upgrades
PCWorld has an analysis of the costs associated with upgrading to Vista. Hard to figure out the total cost when you consider the time sink and the fact that they suggest install cleanly.
How much you'll have to pay for an upgrade ultimately will depend on where you buy your PC. As reported today by The Wall Street Journal Online, major OEM's will offer different pricing structures. For example, Hewlett-Packard will allow you to buy certain models and upgrade for free. Depending on the HP retailer, however, you may have to pay shipping and handling fees.And the Journal reports that Dell will also allow you to upgrade by paying shipping and handling fees for systems using Windows XP Professional or Media Center editions. However, the Texas-based company plans to charge you $45 for an upgrade from Windows XP Home Edition to Windows Vista Home Basic.
Posted by Dave Rosenberg on October 26, 2006 10:59 AM
October 26, 2006 | Comments: (0)
Walmart.com respins with open source (OpenLaszlo)
Walmart.com is relaunching, and has open source at its core. MySQL? Well, probably (that's a given these days), but this news is even more significant, because it's OpenLaszlo from Laszlo Systems. OpenLaszlo is a great way to add rich media functionality to your website. WalMart agrees.
The real question, given WalMart's reputation for driving down vendor pricing, is how much lower than free did it require? :-)
Posted by Matt Asay on October 26, 2006 09:50 AM
October 26, 2006 | Comments: (0)
Open source Java for Christmas
Jonathan is now on the record as saying that we'll have open source Java within the next 30-60 days. As reported by InfoWorld:
Demonstrating a perhaps more aggressive path than anticipated, Sun Microsystems is set to announce the open-sourcing of the core Java platform within 30 to 60 days, Sun President and CEO Jonathan Schwartz said at the Oracle OpenWorld conference on Wednesday morning.The core platform encompasses the Standard Edition of Java, and it will be offered via an open source format under an OSI (Open Source Initiative)-approved license, likely the same one used for Sun's open source Solaris OS. Sun officials, including Rich Green, Sun executive vice president for software, have talked about Java being offered via open source in stages later this year and into 2007. Parts of it, such as the Java Enterprise Edition, already are available via open source, with the GlassFish application server constituting the open source enterprise variant.
Posted by Matt Asay on October 26, 2006 09:46 AM
October 26, 2006 | Comments: (0)
Crazy thought: Red Hat could provide Level 3 support for Oracle's Red Hat Linux
I haven't thought this through yet (obviously), but thought the idea was amusing, if not useful. My bet is that companies are going to want to keep going to Red Hat for support given Red Hat's leadership on the Linux kernel. But some may be tempted by the alleged 50% discount Oracle is offering.
But let's say the discount is real. Here's a crazy thought: why couldn't Red Hat charge an additional 25% to offer Level 3 support for Oracle's Red Hat Linux?
In this way, Oracle takes on the heavy lifting on support (Levels 1 and 2), Red Hat offers peace-of-mind for the thorniest issues (Level 3), and both get paid for their value.
This, in short, gives customers what they want (Red Hat backing up Linux) and allows customers the discount they supposedly have been clamoring for. In short, it gives them
Red Hat Enterprise Linux, in other words, but at a 25% discount. That would be a fun stick-in-Larry's-eye for his troubles.
(See Red Hat's response to Oracle's move.)
Posted by Matt Asay on October 26, 2006 09:01 AM
October 26, 2006 | Comments: (0)
Mark Shuttleworth on Oracle's Linux: "Free software in a proprietary wrapper"
The451 is carrying a great analysis of the Oracle Linux move by Mark Shuttleworth, founder of the leading community-based Linux distribution, Ubuntu. Here's a gem from Mark:
I think we can expect Oracle to get frustrated with supporting someone else's codebase. If it takes off from a business perspective, then fine, but I really doubt that large numbers of people will switch from Red Hat to Oracle as a provider of support for Red Hat. Red Hat, certainly, is not going to make it easy for customers to live on both sides of the fence, and most companies will want SOME access to Red Hat. So if Oracle wants to make money from "owning a linux stack" then I don't think this is a winner. On the other hand, if they just want to keep a lid on Red Hat, without rocking the boat too much, then this works fairly well.Also see The451's pre-Oracle announcement discussion with Mark - Mark told The451 that an Oracle/Ubuntu link-up is a matter of when, not if.Fundamentally, though, this is still free software in a proprietary wrapper. The pricing may be different, but it's still old-school thinking. I don't think anybody who will consider jumping to Ubuntu from Red Hat will pause very long on the Oracle option.
Posted by Matt Asay on October 26, 2006 07:27 AM
October 25, 2006 | Comments: (0)
Whose problem is Oracle trying to solve?
I'm actually trying to get some sleep, but I can't let this go. The more I think about Oracle's move, and the more emails/comments flood in, the more I want, with Dave Dargo, to call "foul."
Here are a few reasons:
- Red Hat has for the last two years been rated #1 in customer value. By customers. Regardless of where Oracle stands on that list (#39 of 41), being #1 is a Big Deal. In fact, I'd probably interpret #1 to mean that customers like doing business with Red Hat, feel like they're getting more than their money's worth, and generally don't plan on leaving Red Hat any time soon. Perhaps this is part of Oracle's reason for the move...?
- The pricing Oracle gave is completely bogus, as Don Marti's comment to my original post indicates, as does common sense. Red Hat has been selling to the enterprise for over seven years. VPA and discounts apply. Oracle uses its discounted pricing against Red Hat's list pricing. It's a clever but deceptive tactic.
- Bug fixes. Oracle said that Red Hat forces customers to upgrade to future releases to get bug fixes (effectively killing their ability to stop paying for Red Hat Network/support). This is 100% wrong, as Oracle knows (and as every Red Hat customer knows).
- Certifcations. What will happen to Red Hat Enterprise Linux implementations that Oracle patches? Fine, Oracle says that it will support its own patches, but this says nothing about the other applications that these applications may well break. Interestingly, this is a problem that Charles Phillips identified last year:
[Customers] are bogged down by the dependencies between applications. Add a patch in Linux and five other things break.
In other words, once a new patch is applied, all bets are off on whether all those non-Oracle applications (or even the Oracle applications) will still work. Are you going to trust Oracle to make sure your non-Oracle applications work? Like your SAP, IBM, etc. implementation? Red Hat serves as a useful third-party platform provider that Oracle cannot hope to match, simply because it has too much at stake (Middleware, database, applications, etc.). - What happens to support for Tomcat/Hibernate/Struts/Seam/etc. running on RHEL supporting Oracle?
- Could it be that Oracle's real goal is to contain open source penetration? Few would lose more (or could gain more, if they'd just embrace it) if open source continues its move up the stack. The Oracle franchise is increasingly threatened by open source, which is isolating Oracle's relational database as a dated container, while web services, contemporary extraction tools, virtualization, and the cache file system hamper Oracle's growth and economics.
- And what happens in all this to the value of Open Solaris? HP UX? AIX? Oracle just shot them.
All of which begs the question: What (or whose) problem is Oracle trying to solve?
Posted by Matt Asay on October 25, 2006 09:33 PM
October 25, 2006 | Comments: (0)
Analyzing the Oracle Linux move...from the guy who developed Oracle's Linux strategy (Dave Dargo)
I should have looked to Dave Dargo's blog first when the Oracle news hit. Dave, for those who don't remember, was a longtime Oracle employee and started and ran Oracle's open source program office. He knows quite a bit about Oracle and Linux.
And he's calling "foul" on Oracle's announcement today, with a great deal of insight into why.
He persuasively argues that Oracle's discounted support will actually be dramatically more expensive (like a gazillion percent more :-), the service won't be as good (for reasons similar to those that I argued), and that Oracle has no history of providing superior value at a lower cost, while Red Hat has that in spades:
But what about the other part of that quote, that support has to be better. There's a survey from CIOInsight [PDF] that shows Red Hat is the number one vendor for value as rated by CIOs in 2004 and 2005. Where does Oracle fit on that chart? Glad you asked, they ranked 39 out of 41. [Emphasis mine.]Good points, Dave. Head over to his blog to get more. It's fascinating analysis.The other thing I'm most curious about is the concept of Oracle's Unbreakable Linux Network (ULN). The claim is that it takes less than a minute to switch from Red Hat's Network (RHN) to ULN. It's going to take more than a minute, and a fair amount of cost, to get through the legal agreements and process of switching over. But even with that aside, I'm mostly curious as to why Oracle's first real support network is for someone else's product. Where's the Oracle Database Network and Applications Network and PeopleSoft Network and Siebel Network? Where are the support infrastructure networks for Oracle's own products to automatically distribute fixes, patches and alerts? It's amazing that they can provide all that for a mere $399 for a competitor's products, but not for their own $200,000 product.
At the end of the day they still haven't answered the basic question of how eliminating choice benefits the customer, and that's bull*&#%.
Posted by Matt Asay on October 25, 2006 09:24 PM
October 25, 2006 | Comments: (0)
With friends like these....(Oracle goes after Red Hat)
I'm here at Oracle "Open" World, and am a bit shocked by what I'm seeing. Oracle, longtime partner to Red Hat, is rolling out the next phase of its Unbreakable Linux program, designed to kill Red Hat and Novell (whatever Larry Ellison might say to the contrary). With partners like Oracle, who needs competitors?
Now, Oracle will say that it's offering a level of support unmatched by Red Hat, and it will also say that this program is not designed to kill Red Hat. Maybe. I talked with some Oracle employees, and they were convincing that some Oracle customers have complained about Red Hat's service/support.
But even if true (and I've heard as much grousing about Oracle's support as I have Red Hat's, or anyone else's for that matter - the fact is, no one likes support if they have to use it), there were better ways to deliver superior support without undermining one's partner. Like offering it as a value-added feature to the partner's standard support. But I digress....
Maybe this is why Oracle has been aggressively recruiting kernel developers out of Novell. Novell has lost three in the last several months, making it hard for Novell to claim any leadership against Red Hat, which is a hard-core innovator on the kernel. Oracle understands that to support a community-based product, it has to be part of that community.
This, incidentally, is still the best reason for Red Hat customers to stay with Red Hat for support: Red Hat is doing more to innovate and develop the kernel than anyone else, including Oracle. Source of code matters more than source code in Linux, and Red Hat is the predominant source.
Oracle customers need to ask themselves if they want a forked version of Red Hat Linux. Like Red Hat or not, they have served a useful function of consolidating interest in Linux and thereby fostering commercial adoption of Linux. By providing non-standard kernel patches, Oracle is forking the kernel and setting themselves up to support the fork forever.
Edward Screven, Oracle's CTO, denies this, saying that they're only providing bug fixes, not new features:
We think it's important not to fragment the market. We will maintain compatibility with Red Hat Linux. Every time Red Hat distributes a new version we will resynchronize with their code. All we add are bug fixes, which are immediately available to Red Hat and the rest of the community. We have years of Linux engineering experience. Several Oracle employees are Linux mainline maintainers.That's welcome news, but hard to implement in practice. I heard from several Oracle employees today that one of the reasons for the move was that customers complain that Red Hat doesn't fix bugs fast enough and doesn't add requested functionality fast enough.
Note to such customers: neither will Oracle.
I find it difficult to impossible to believe that Red Hat, with more control and insight into the Linux kernel than any other company, can be beaten on providing timely bug fixes. Regardless, the larger question is on feature requests if, in fact, this is something that Oracle plans to do (which it is not planning, according to public reports).
If there is an intent to move these patches into the kernel, Oracle won't be able to force its will on the kernel any more than Red Hat is able to. Oracle's alleged big customers that are demanding this level of support from Red Hat are going to find Oracle equally impotent to be able to absolutely force its will onto the kernel, leaving the customers...forked. I suspect that no one wants this.
Anyway, time to let Oracle tell its own story. Here's why they're doing it:
Larry Ellison cited a range of problems with current Linux support:
Oracle thinks it can do better on service levels:
And price (though this has to be a complete canard because I've yet to find a big enterprise that thinks Linux - Red Hat or otherwise - is pricey compared to the alternatives (remember Unix?):
And, lest you missed the point that Oracle's support is cheaper than Red Hat's, Ellison pounded home on the theme:
So, where does this leave us? With Oracle claiming to provide better support on a product that it has less control over than Red Hat does. I'm not seeing the logic there, but I do understand that Oracle's scale allows it to undercut Red Hat on pricing. Even without that scale, the fact that it isn't investing the same resources into Linux that Red Hat is leaves it able to charge less for the support.
It will be interesting to see how Red Hat responds. Given Red Hat's strength in creating and supporting the Linux kernel, as well as its newfound JBoss arsenal, I don't think it's time to head for the exits. Red Hat is still the premium provider of Linux. This complicates things, but doesn't significantly change them.
(Btw, any one else think that Oracle's failed acquisition of JBoss is a big part of this announcement? It would be sad if personal pique could be the motivation for such a big announcement, but I wouldn't count it out.)
Posted by Matt Asay on October 25, 2006 01:54 PM
October 25, 2006 | Comments: (0)
Just when you think the world is getting more savvy on open source...
I was reviewing a contract the other day with a major Fortune 500 institution. This is a forward-looking institution with a massive IT budget and, no doubt, Linux and other open source running rampant throughout its organization.
I was therefore amused to see this clause in the contract they provided to us:
Alfresco hereby warrants that the Software contains no open source code and that it will not provide any open source software as part of its Services or Software.Um, we have a problem here....
I explained to them that all of our software is open source, and why they should be happy about this fact. They now are, but it's always surprising to find persistent misconceptions about open source in the savviest of companies.
It reminds me of a presentation I delivered years ago to the IT workers of the State of Texas. My presentation followed that of the State's general counsel (for IT, anyway), who spent her 45 minutes warning the IT staff to not use open source and to wait until the State had a formal policy in place for adoption of open source.
As soon as she left the room, I asked for a raise of hands for "All those currently using open source in your work." 80% of the hands went up (and I suspect the other 20% were lying :-).
Open source is everywhere. Might as well get used to it.
Posted by Matt Asay on October 25, 2006 10:54 AM
October 24, 2006 | Comments: (0)
New Mac Core 2 Duo: Will I or won't I?
My MacBook saga is still not quite over. I did hear from both the Apple PR exec who deals with the Apple Stores and the store manager from Palo Alto (I will call you both back this week, I have been too busy the last few days--just imagine how busy I would have been if I had given you my laptop :>)
Anyway, I am thinking about going for one of the new Core 2 Duo. The MacBook I have remains an expensive toy. Usable but unlikely to last. The spacebar is squeakier than ever and one side of the mousepad bar is an unknown as it seems to go in and out of the top case. My kingdom for a Thinkpad running OS X.
My big fear is that if I go for the 15" Pro (which I think is too big) that they will come out with a new one in a few months that will be a size more to my liking. Isn't life tough?
Posted by Dave Rosenberg on October 24, 2006 10:24 PM
October 24, 2006 | Comments: (0)
Positive thinking may not always be a plus
Finally, justification for my moribund attitude. A recent study says that those that harbor ambivalent feelings -- the simultaneous anxiety and excitement of starting a new job, say -- may result in enhanced creativity. That's right, it's us glass half-empty guys that make things happen.
What's the link? MIT neuroscience professor John Gabrieli explains it this way: Problem solving requires access to all our mental resources, and an ambivalent state broadens thinking, perhaps allowing us "to cast a wider net" over the information in our brains.
Link: The Power Of Ambivalent Thinking
Posted by Dave Rosenberg on October 24, 2006 10:19 PM
October 24, 2006 | Comments: (0)
Jason Maynard's report from Oracle Open World (Day 1)
Jason Maynard, CSFB analyst extraordinaire, is at Oracle Open World and is finding some common themes surfacing:
Good intel, Jason. I continue to believe of the large enterprise ecosystems, Oracle has the best chance of being the hub of a large open source ecosystem.
- "Scale matters." The company has enough R&D to support ongoing development efforts for all their applications and technologies.
- "Better information gives better results." The volume of information is increasing at a phenomenal rate, creating the need for technology solutions that can harness the information explosion. Oracle’s technology solutions address this issue, as seen in its roadmap for Oracle 10G and 11G.
- "Ecosystem & Partner Relationships." Oracle is working hard with its partners to provide interoperability. Relationships with rivals like IBM and Microsoft is still intense but these vendors are finding pragmatic common ground given customer needs.
But I still think that Oracle needs to choose carefully how it deploys Linux as part of that ecosystem strategy.
Posted by Matt Asay on October 24, 2006 01:57 PM
October 24, 2006 | Comments: (0)
One man's freedom is another man's... (GPL)
I was on a call the other day, and we were discussing whether it was appropriate to look into the intent underlying an open source license, or whether one should accept it "as is," on its face. I'm in the latter camp, believing the Open Source Definition applies to the words of a license, and not to any inferred intent behind those words.
The point was made, however, that it's impossible to read the GPL without seeing Richard Stallman's intentions/philosophy all over the license. Intentions certainly matter to Richard. However, as can be seen with GPLv3, it's very difficult to bound one's intentions once efforts kick in to impose a political view on the world through software.
It is this very grasping nature of Richard's, however, that makes the GPL the software capitalist's best friend. Richard never intended this. His purpose is freedom. But the problem with trying to control outcomes is that one can't help but control the means to those ends, as well. (I wrote about Machiavelli during my Masters program along this very same vein - to control the ends we must control the means to those ends. There is no other way to do it.)
So, Richard, by trying to guarantee freedom has instead created a license that mostly constrains it. Though people debate this point, the BSD actually offers more freedom. Last time I checked, freedom meant:
By this metric, then, the GPL is hardly a free software license. Forcing people to share's one's version of freedom is not...free. In fact, I'd argue that it's the closest thing to traditional copyright that the open source world has ever devised.
- the condition of being free; the power to act or speak or think without externally imposed restraints
- exemption: immunity from an obligation or duty
Don't get me wrong: I'm a capitalist. Therefore, I love the GPL. The GPL offers vendors a way to freely (as in cost and "freedom") distribute their software without any fear that a competitor will use it. Why? Because to do so would be like dropping a neutron bomb on their own software. It's just not going to happen. (I wrote a long paper on this back when I was still Larry Lessig's student at Stanford Law School - you might enjoy it. He gave me a good grade. :-)
In the ongoing debates about whether attribution licenses are open source, whether GPLv3 is open source, etc., please keep in mind that the supposed paragon of software freedom is also the license that most tightly imposes a distinct lack of freedom on downstream users. If you're a capitalist like me, you probably like this fact. But if you're a software developer...?
Posted by Matt Asay on October 24, 2006 07:26 AM
October 23, 2006 | Comments: (0)
Come join us for: The Truth and Myths About Open Source at Oracle Open World
Matt and I have the privilege of doing a session at Oracle Open World on the world of open source. Matt actually did all the slides. I am merely a leech on his genius. That's why he receives 15% of all my future earnings.
Stop by and say hi. We'll be the ones in penguin costumes.
Wednesday October 25, 2:30pm: Moscone South 300
The Truth and Myths About Open Source
Posted by Dave Rosenberg on October 23, 2006 03:18 PM
October 23, 2006 | Comments: (0)
IT does matter, but open source IT may matter most
InformationWeek has a great story and survey detailing where and how the industry's top enterprises spend their IT dollars. The results are illuminating, and portend great things for open source.
Interestingly, IT dollars continue to rise, even as overall company revenues fell from 2005 to 2006:
As for where those dollars are going, despite all the hype that licenses don't matter (because the real costs are in implementation), the numbers don't bear that hype out. The numbers say, "Enterprises are wasting money on licenses."
The license cost savings open source delivers (as compared, say, to Microsoft), however, would be negated if open source implementations cost more (in fact, they cost less or at least give the CIO choice as to how she spends those dollars) or if open source required more IT personnel (clearly the biggest IT cost). But experience proves otherwise, as The Robert Frances Group has written.
Besides, as The Christian Science Monitor and others have found, open source makes those IT personnel costs stretch farther, deliver more innovation, etc. Open source makes IT dollars work for the enterprise, and not for the vendor.
So, Nick, maybe it's not that IT Doesn't Matter, but rather that proprietary IT doesn't matter. Or won't, as more companies start feeling the benefits of open source. Money will still be spent on IT, but the smart money is on open source IT.
Posted by Matt Asay on October 23, 2006 12:19 PM
October 23, 2006 | Comments: (0)
Open source's fundamental right to fork...largely unexercised
Brian Behlendorf dubs it the cardinal rule of open source. I think he's right. Apply Ockham's Razor to the Open Source Definition and "right to fork" results.
Yet it rarely happens.
Joomla/Mambo. Adempiere/Compiere. BSD and all its children. These are the exceptions to the rule.
Most open source projects never fork. I suspect the primary reason has to do with the open source ethos (Cooperation first, fork as a last resort) and human logos/sloth (Easier to help propel a rolling stone than to exert force to budge an inert one). And to the extent that projects do fork, it's interesting to see how often the branch wins out over the trunk. Almost never (by my admittedly unscientific estimate).
Still, even if no one ever forks, it's still a good thing. It's a good thing because the right to do something often serves as a useful surrogate for actually doing it.
Maybe we should therefore stop acting like Chicken Little over GPLv3 and the fragmentation that may result. You can fork, but you can't stop others from incorporating your fork back into the main tree. Maybe a thousand Linux forks will simply make a better Linux.
Posted by Matt Asay on October 23, 2006 10:02 AM
October 20, 2006 | Comments: (0)
MySQL's Marten Mickos is probably my favorite CEO. He is a great leader for his company and the OSS movement. I can only hope to be half as good a CEO as he is. **Please note I am not being sarcastic.**
Check out this Q&A with MM--lots of insight into MySQL and the overall OSS marketplace.
Mårten: This is a great question. First, I think closed source vendors are proving the hypothesis incorrect, because they are the ones who have un-friendly products although they have a licence revenue stream. And open source products, which typically lack a licence fee, are the ones with the best user friendliness. Why is that? I think the reason is that popularity is worth more than the marginally improved fees you could get for a user-unfriendly product. Sure, we lose some purchase orders because customers simply don't need our support. But at the same time we gain millions of new users, and they in turn help us develop the product and drive our marketing.I think one of the reasons MySQL is so popular is that we continue to take user-friendliness seriously. We may joke about leaving some bugs in the product in order to have more support fees, but in reality we take the bugs very seriously. And when we at times have more bugs than we would like to have, the whole organisation is in pain until we get the upper hand and kill more bugs than we produce. Version 5.0 was such an example - when it first came out as GA it had more bugs than earlier versions. It didn't feel good to see the list of incoming bugs, so we rearranged tasks and expanded our QA test coverage in order to fix the situation. I hope you who have used the latest versions of MySQL 5.0.26 have seen the significant improvements.
Posted by Dave Rosenberg on October 20, 2006 10:28 AM
October 19, 2006 | Comments: (0)
The gardener's art of 'grafting' and open source hiring
The past few weeks have been a lesson in olive tree gardening. Not literally, of course. Instead, I'm referring to the analogy between how one grafts olive trees and how one grafts employees into a company. The tie between the two hit me in my morning reading today....
When we think of gardening, we normally think of planting a seed, weeding around it, fertilizing the ground, and waiting for something to grow. In the case of olive trees, however, the process really kicks in once the "something" (the olive tree) has already grown. If the olive tree is providing bad fruit (olives), or a low volume of good fruit, you can graft in branches from a different olive tree - one that produces great olives in abundance - and it can actually change the tree such that it starts to produce good olives, and lots of them.
The important thing in the process is to keep the balance in the tree. You can't over-graft a tree - its roots will only support so much. Therefore, it's important to gradually graft in good branches.
This applies to how one hires for open source companies. Yes, it applies to how any company hires, but it's particularly true in the case of open source, because open source is a dramatically different way of doing business. It's normal to look a person's resume, see exceptional experience in a candidate, and assume it will automatically translate into success at one's company, open source or no.
This is not wise, however, as hiring people is about their cultural and competency fit with a company. Much as we'd like to look at job candidates as simple math equations involving dollars and cents, they are always more than this. Hence, what Jack and Suzy Welch write in BusinessWeek about M&A holds true for adding people to a company, generally:
...[R]ecognize that the cultural fit of two companies is as important as strategic fit--if not more so. Oh, how exciting it seems when a merger or acquisition makes perfect sense in terms of products, technologies, and numbers. But what a disaster it can and will be if the two companies operate with distinctly different values. The fact is some cultures don't combine--they combust.The first rule of good open source hiring, then, has almost nothing to do with open source. You need to find people that fit the values of your company.
For an open source company, people need to somewhat ego-free or, rather, teachable, because there can't be barriers between one's community and the company. You can't assume you know what needs to be done in an open source company until you've slogged in the trenches for a spell.
I really like working with John Powell (former COO of Business Objects) and John Newton (co-founder of Documentum) because they fit this mold. Sure, they have egos. But not the kind that make them think they know all the answers. Instead, their egos make them self-confident enough to accept that they don't know everything, and that they have a lot to learn from their community of developers, users, and customers. (Some may remember that I once questioned whether these two were the right ones to start an open source company. Because of this trait, however, they've done an excellent job of learning open source on the job.)
The second big requirement is that open source companies need to be careful how they grow. Because of a shortage of supply of experienced open source business people, it's critical that an open source company carefully, gradually "graft" in new employees. Graft them in too quickly and the new branches will overcome the old roots.
This reminds me of the Larry Augustin gave me a few months back. Larry's point was that open source companies may need to grow more organically - at least, initially - than proprietary companies do. You don't have the luxury of a big upfront license fee (recognizable revenue) to build out with, so you have to patiently tend the "olive tree" until revenues kick in. So, you keep field sales to a minimum and take a lower-cost, longer ramp up to sales. This is more efficient and more in keeping with the strengths of the open source model, which is as much about business model as it is about development model.
So, again, graft carefully. An open source company is a delicate balance of community and commercialization. Too much of the latter, too soon, and you end up looking, acting, and dying like a proprietary software company. It's simply not sustainable at the lower ASPs that most open source companies have.
Third, open source companies require innovative, creative thinkers. No one has figured out The Right Model for open source. The proprietary software world is largely in stasis in terms of business model - we know how to build those businesses (though that doesn't mean it's easy to do so). In open source, however, we're in a fecund time of exploration and experimentation. Yes, there are lessons to be learned from the Siebels and FileNets, but it's critical to find people that see a future beyond these old models, because open source companies depend on bending old rules and installing new paradigms.
I have loved working in the open source business community. It's a huge challenge, but I'm constantly impressed by the new open source employees who "cross the chasm" and become productive open source business citizens. Not everyone makes the leap. Most will wait for the late adopter phase when their personal returns for making the transition will be minimal (but so, too, will the risk be). For those willing to embrace the grafting process, however, there's no better time or place to work in technology.
Posted by Matt Asay on October 19, 2006 08:10 PM
October 18, 2006 | Comments: (0)
IBM is proof that (proprietary) enterprise software is not dead (yet)
Jason Maynard's CSFB report this morning has a tidy analysis of IBM's Q3 numbers. They're a clear indication of why proprietary software will remain with us for some time.
So, the good news for enterprise software vendors is that you still have several years of historically anomalous returns. But I can't help but think of Andrew Marvell's classic poem, "To His Coy Mistress," wherein the narrator lauds his lady for her virtuous resolve, but suggests that it's time to party now because the end is near:
- IBM's Q3 results should be overall positive for the greater software industry. The Q3 report showed that demand in key segments continues to be healthy, and barring any major market share gains, we believe independent vendors should be able to generate similar growth results. In addition, while services bookings were again behind plan, management indicated that the pipeline was strong and was optimistic about the future. This is a stark contrast to last quarter when management was worried about a potential slowdown in IT spending. Furthermore, we view IBM’s optimism as another sign that the stage is being set for the year-end budget flush scenario to play out.
- Within the company’s software business, total revenue came in at $4.4 billion, up 9% year over year, and was ahead of Credit Suisse’s software estimate of $4.2 billion....We found IBM’s attention and focus during the earnings call on the software business as an interesting signal that more acquisitions are likely down the road.
- Consistent with prior quarters, IBM’s WebSphere products continued to experience strong growth. In total, WebSphere grew 30% y/y (up from 17% in Q2), driven by continued adoption/demand for SOA deployments. The leading independent vendors in this space, BEA and Tibco have also experienced strong demand for their products, however total revenue growth in the last quarter (19% for BEA, 14% for Tibco) was slower than WebSphere. While it is clear that end demand continues to be strong and the WebSphere brand is somewhat of a catch-all, the growth is still impressive.
- IBM’s information management products grew 12%, and the company disclosed that database peripherals grew 50% in the quarter. IBM again declined to disclose the actual performance of the database, and we continue to interpret this omission as a sign that Oracle is outgrowing IBM’s distributed database.
And from a separate Maynard report:
- Going into the seasonally strong fourth quarter, we continue to believe the stage is being set for the year-end budget flush scenario to play out. We believe that most vendors carried strong pipelines into the fourth quarter and we feel that in general, businesses have saved a good portion of their budget for the end of the year. Looking ahead into 2007, we continue to believe that next year should be a relatively good year for software budgets. We continue to hear feedback from CIO’s that On Demand, offshoring, and virtualization are enabling organizations to lower the cost of IT operations, thereby freeing up capital for new projects. From a macro perspective, slowing productivity growth could be another factor that will cause businesses to increase investing in value enhancing technology.
But at my back I always hearNor sell software. Eat, drink, and be merry now. For tomorrow you'll be in open source land.
Time's wingèd chariot hurrying near;
And yonder all before us lie
Deserts of vast eternity.
Thy beauty shall no more be found,
Nor, in thy marble vault, shall sound
My echoing song: then worms shall try
That long preserved virginity,
And your quaint honour turn to dust,
And into ashes all my lust:
The grave 's a fine and private place,
But none, I think, do there embrace.
Posted by Matt Asay on October 18, 2006 11:22 AM
October 18, 2006 | Comments: (0)
I was going to say 'Why I blog," but Nick called to my attention what an ugly word that is. So I'll say "write," instead. (Besides, it also applies to email. I write voluminous emails, much to the recipients' chagrin.)
I've had a few people ask lately how I find the time to write (blog) so much. And a few people clearly thought that writing equates to "time wasting." If I had a demanding job, surely my blog traffic would go down.
On that account, let's just say that I manage a business that doubles revenue every quarter. Enough said.
As to why I write, that's also easy. Because it helps me to think. What may seem muddled here often crystallizes right after I hit "Publish." Or it will when I get comments back from people telling me where my reasoning breaks down. My blog traffic tends to go up the better things are going with work, because that's when my mind is running at full speed.
In short, some people like meetings to help them think. Me? I write, and read others' writings.
As for when I write, it's usually when I'm on a conference call. That time is productive for someone on the call, I'm sure. But rarely for me. So next time you call me, pardon the tapping of the keyboard in the background. I am listening. Just often not to you. :-)
Posted by Matt Asay on October 18, 2006 11:01 AM
October 18, 2006 | Comments: (0)
I hadn't noticed this when it happened, but ComPiere was recently forked, as Frank Scavo notes in a detailed, excellent analysis. The fork came from a group of freelance ComPiere consultants that were apparently worried about NEA-funded Compiere's intentions, both present and future, as Frank notes:
Understandable from their side, but also from Compiere's. At the end of the day, Compiere has to own the roadmap, code, etc. They're the ones that are going to be hammered by customers if something goes wrong. It's a tradeoff one makes when going commercial open source. (Btw, Janke demolishes the notion that ComPiere, Inc. will be crippling the community project to enhance the viability of his company.
- The speed at which ComPiere Inc. processes fixes and enhancements submitted by contributors and the refusal, in some cases, to even accept them.
- The refusal of ComPiere Inc. to provide Compiere version migration tools except in the sale of a support agreement.
- Rumors that ComPiere Inc. is planning to limit the functionality of its open source offering, to better position some future proprietary offering.
You can view the forked project here. It's called Adempiere. Lame name, but then, it's not supposed to be a marketable business. :-)
Posted by Matt Asay on October 18, 2006 08:47 AM
October 18, 2006 | Comments: (0)
Will Zimbra crack the $1 billion barrier?
TechCrunch is reporting that Zimbra is up to 4 Million mailboxes and with some simple math we can calculate the following:
4 million@$25/per mailbox = $100m in revenue.
Now consider that MS Exchange has 150million seats.
150m@$25/per mailbox = $3,750,000,000
While it's unlikely that Zimbra will take the whole 150m mailboxes there is a clear trend and market potential. There was much discussion a few months ago about open source cracking the billion dolar mark. Clearly it is possible.
Posted by Dave Rosenberg on October 18, 2006 06:22 AM
October 18, 2006 | Comments: (0)
The battle of the (closed) ecosystems (ZDNet)
Josh Greenbaum at ZDNet suggests that the war of the ecosystems is on, and that Oracle is looking weak. I don't agree, but think Josh's commentary is interesting:
The partner front is the new flash point in the battle of the ecosystem. When it comes to IBM vs. Microsoft vs. SAP vs. Oracle, it turns out that a large coterie of happy, incented, and otherwise engaged partners that are out legitmizing an ecosystem vendor's ecosystem -? and selling the products and services they develop that fill in that ecosystem's white spaces - has become a major competitive weapon. And SAP is absolutely committed to making the most of this weapon in their four-way fight against their partner/competitors IBM, Oracle, and Microsoft.True enough, though I think each of the big ecosystem vendors is missing out on the most important partner for the future: open source.
Oracle is in a great position to become the center of that universe, but needs to act now. Instead, Josh suggests that Oracle has been MIA:
What about Oracle? In the run-up to their big Open World shindig next week - where we can expect to hear a lot of crowing about the solidification of their market position and their growing revenues from enterprise apps -