This train of thought is a little convoluted but I think it has meaning. I admit I get thinking like this on Fridays in the spring and summer when my mind wanders on Friday mornings and my three wood wanders Friday afternoons.

The thought process began as I was reading last Thursday’s “one laptop per child” PR blitz and Microsoft’s quarter 3 FY 2007 financial results.
• The crank-it-for-power, “chicklet-keyboard-for-kids” laptop runs Linux (Fedora I am guessing given Red Hat’s involvement) and some games and is intended for use all over the “third world.” Why my grand kids can’t have it I am not sure.
• Meanwhile, as I opined during February and March, Microsoft Office 2007 (and Vista) sales are exceeding financial-analyst expectations (and the cost of the Technology Guarantee program may be lower than expected). As a result, Microsoft feels it will meet its revenue and profit goals through June 2008.

(Note: I have not researched any of this myself. Since it is the virtual weekend, I am depending on the related press releases and the resultant published mass-media reports.)

Somehow the Linux open source software (OSS) angle in the one-world laptop PR segued me to thinking about my own posts during the week of April 23 concerning the now complete metamorphosis of VA Software (away from being a Red Hat competitor to its e-media niche) and Red Hat’s changing of the JBoss Group strategy (after less than a year of trying to make the inherited JBoss’ business model work). Similarly, reading about the Microsoft results led me back to thinking of how well Microsoft would do next year if it could get the EU off its back.

This caused me to ask (myself):
• I know about the millions of pirated copies of Office and Windows awash in the third world. Why isn’t there a lot of research into OpenOffice and Ubuntu piracy?
• I know the size of the OSS maintenance and support market. Why isn’t there more research into the size of the maintenance and support business that has built up around Office and Windows?

My answers:
• What piracy? All those laptop cranking kids can download OpenOffice and Ubuntu for free (Canonical will also mail you Ubuntu for free if you want to wait 6 weeks).
• As for Open Office and Ubuntu support and maintenance, the little waifs can just go to Sun or Canonical and buy a maintenance contract.

But these are third world waifs we are talking about. They can’t afford an Internet connection to get their free software? Or pay a Sun/Canonical subscription fee? In that case, they can (and I predict will) get a pirated copy of a Microsoft product like millions of others before them (the one-child laptop also runs Windows). Bill Gates is even talking about competing with the pirates in China; selling a Genuine COA’d Windows for three bucks worth of yuan.

As my thinking wandered further, it dawned on me that I have had Microsoft software for 20 years and have never needed to ask any one for any service or support. (Although I didn’t need any Microsoft support, for my day-job’s research purposes I have used the free email support that comes with my free Office Live account and it has always been prompt and actually useful almost all of the time.) Admittedly I religiously back up my work and have continued to do so even though I haven’t seen a blue screen in a decade. But that caution dates to a long retired and wise boss in my mainframe days, and I just carried it over to the PC era. And admittedly sometimes I can’t figure out how some Microsoft thing works (or why it works the way it does), but there is always another way to do what I am trying to do and I get on with my life. I believe information technology that requires no documentation will rule the world. I can’t say the same for the CompuServe, Comcast, Compaq, HP, IBM, and Verizon products and services I’ve purchase over the same time period. And as noted below, you can't say the same for the operating system recommended by the Free Software Foundation (FSF) for use instead of Windows.

So my next thought was: Why are the free softies always going on about the inherent quality of their stuff and the inferiority of Microsoft software?
• The FSF world
o would not exist if it were not for thousands of people buying expensive services and support for the free software.
o This might mean that free software breaks a lot

• The Microsoft world, on the other hand, ‘runs’ on
o millions of people buying some thing
o finding that it does a good job
o buying the next version 3-6 years later, without ever having to make an 800 call.
o This might mean, if you believe in markets, that Microsoft software does not break very often.

Of course, many of the free softies—e.g., Professor Neelie Kroes of the EU Competition Commission—do not believe in markets. If my thoughts all fit, the bigots at Wine (see the bigoted reference to the Irish concerning the potato blight that hit Europe in the 1840s) better ramp up their website as billions of kids download the Windows emulator to run their pirated copies of Word and PowerPoint (or the cranking kids will get a pirated copy of Windows too).

I haven’t seen the whole one-laptop-per child plan but I guess the billions of little kids will be doing their downloading via their free UN-sponsored WiFi connection. This will be provided in a program to be run by the EU’s Neelie, who will quit the Competition Commission in a few months frustrated by the European Court’s upcoming finding in favor of Microsoft. As part of that finding, Microsoft will have to give away free MSN vouchers to billions of little one-laptop kids.

This will be familiar territory for Microsoft; a few years ago it was “forced” to give millions of U.S. kids free Microsoft software as part of U.S. state and Department of Justice settlements. What a coup then and now. A generation of third-world kids, like a generation of U.S. kids, will grow up with Microsoft while those poor EU kids will be trying to figure out what these directions mean on the gNewSense site (the FSF recommended operating system):
“gNewSense is derived from Ubuntu, and thus has most of the same functionality… There are a number of differences though… Note our goal is to produce a fully free distribution, not to have as many features as possible… To create gNewSense we needed to produce a set of scripts. We called this building method Builder and these scripts allow anyone to build their own distribution. Try it here: create a GNU/Linux Distribution. This is not how you get gNewSense, for that follow the download link above, but you can do this if you want to see how we did it.”

If anyone else thinks there is some meaning in my train of thought, post a comment or send me an email.

Posted by dennisb in OSS Culture | Permalink | Comments (1) | TrackBacks (0)

I am not sure whether this is an open source software (OSS) culture or OSS business subject. And the fact that it is hard to tell says something about the OSS movement.

VA Software has completed its transition from information technology (IT) supplier to solely being the “MySpace for Hackers.” VA Software—the parent of Open Source Technology Group (OSTG)—has sold the software business related to SourceForge Enterprise Edition to Collabnet. SourceForge Enterprise Edition is a J2EE-based integrated development environment providing project documentation, development process controls, and visibility into project status for in-house IT departments. SourceForge Enterprise Edition was the remnant of the original VA business, VA Linux, a company that sold Linux-based hardware and related services from 1995 to 2001. Ironically, given where VA Software has moved to, SourceForge Enterprise Edition is not an OSS project.

This asset transaction leaves VA basically in the media business with the popular open source software (OSS) web destinations freshmeat.net, Slashdot.org, linux.com, newsforge.com, and itmanagersjournal.com [sorry--cut and paste them; they are competitors :) ]. VA also retains the site sourceforge.net (as opposed to the software product). VA will own a piece of Collabnet and Collabnet signed a 30-month agreement to advertise with OSTG, presumably on sourceforge.net.

Collabnet, on the other hand, spun out of the media business in 1999. Tim O’Reilly, of technical book fame, and Brian Behlendorf, who is now CTO, were founders. Behlendorf had previously been CTO of Organic Online, a Web design and engineering consultancy, co-founded and contributed heavily to the Apache Web Server Project, co-founded and supported the VRML (Virtual Reality Modeling Language) effort, and assisted several IETF working groups, particularly the HTTP standardization effort. Before starting Organic, Behlendorf was the first Chief Engineer at Wired Magazine and later HotWired. Behlendorf is currently a Director of the Mozilla Foundation and a retired Director and President of the Apache Software Foundation.

Collabnet offers a product of the same name competitive to SourceForge Enterprise Edition. Collabnet became a factor—perhaps the leading factor—in OSS lifecycle management with its 2003 sponsorship of the Subversion OSS version control project; Subversion underlies Collabnet’s commercial product offering. The current president and CEO of Collabnet is Bill Portelli, a long time executive at Cadence, the electronic design automation folks (that is, he is from an application area where version and quality control really mean something). Subversion is “housed” at tigris.org (run by Collabnet of course as are many other OSS sites) and is licensed under Debian guidelines (essentially GPL V2 but there is likely some arcane difference or they would have been more specific).

With VA assuming an equity position in Collabnet, it joins Benchmark Capital, O’Reilly New Ventures, and Norwest Ventures as lead investors in Collabnet. If you follow all this interlocking connection between advertising, content, and OSS technology, you realize how completely commercialism has taken over OSS from the community.

Posted by dennisb in OSS Culture | Permalink | Comments (0) | TrackBacks (0)

Red Hat is explaining its JBoss middleware strategy at a press conference this morning. I am live blogging it, but a recording or transcript may be available later on the Red Hat site if you don't want to take my word for it.

Red Hat calls the conference subject the "middleware migration" program. Red Hat executives say the company is now a full end to end stack provider. They want JBoss to do for legacy middleware deployments what RHEL did for legacy Unix sites (recover investment, take back value). Red Hat's upgraded approach, they say, provides much better value than legacy client/server middleware.

This is what I heard explicitly or implicitly:

-- An update to the JBoss strategy is needed because a year after the acqusition, JBoss has not been the revenue boon Red Hat expected.
---------- Studying Red Hat's financials, which is what I do in my day job, it's clear that the press reports about JBoss revenue stream prior to the acquisition (that is, for the period 2000-2005) were grossly overstated. I am embarrassed to say that even my own estimates of their size when I was making such estimates for IDC between 2003 and 2006 were a little on the high side. And I was always being criticized by JBoss management for ranking the open source software (OSS) middleware "wunderkind" too low on my share tables.
---------- At its most recent quarterly conference call, Red Hat's management revealed that its JBoss division did not meet the $50 million bookings target for CY 2006 that would have triggered bonuses for all the ex-JBoss guys.

-- The good news for JBoss users is that users found (both pre- and post-acquisition) that the core application server product was so well built that it did not need a lot of subscription maintenance support (and that of course is how JBoss and now Red Hat makes its money in an OSS business model). This fact showed up in user survey work I did between 2003 and 2006 and that was the basis of my estimate of JBoss.
----------- Therefore, Red Hat is making changes to the jboss.org, bifurcating the community from the distribution just as Fedora is bifurcated from RHEL. This will allow the community to do more with Apache and others. Bob McWhirter will be to Jboss.org what Max Spevack is to Fedora-project.org (see my February 3 post)
----------- On the product side, a separate organization will package integrated JBoss Platforms, one for applications and another, later in the year, for SOA and BPM (so this organization is more like the Red Hat group that gives you RHEL and separate from jboss.org)

-- Red Hat should be congratulated for moving to a Plan B very quickly. Based on comments at their recent quarterly financial-analyst conference, I believe
------------ Red Hat is kicking off a program of implementation and training services (rather than subscription maintenance services) to support the entire JBoss stack (not just the rock solid J2EE server).
------------ Red Hat also announced middleware development tools to let people work with the JBoss stack more easily.(More broadly, in terms of likely future acquisitions, Red Hat cautioned at its financial analysts meeting “don’t think as much up the stack but across the enterprise.” That would mean they will look at answering all kinds of professional development needs in addition to the deployment and infrastructure software they currently offer. They are probably not looking to acquire applications suppliers. So they acquired Metamatrix data management software supplier, which had no OSS connection.) It is the first example I can think of a proprietary product being kidnapped and moved into OSS. Metamatrix will be initially available as a subscription and eventually moved to OSS (depends on license restrictions, etc.)
------------- Red Hat will build any new services effort around development and deployment partners of all types. For example, Red Hat also announced an arrangement where Vitria will ship its next-generation SOA software with and on the Red Hat JBoss stack.

My open question going in was this: who will Red Hat criticize at the press conference My experience with Red Hat Enterprise Linux (RHEL) events is that Red Hat management begins by criticizing Windows. Since Microsoft "gives away" most of its middleware with its operating software (BizTalk is an exception), it was hard for Red Hat to argue that the Microsoft middleware software does not have the advantages of OSS.

I jest. At the point where they usually criticize Redmond, the executives basically contrasted the JBoss stack with legacy middleware (implicitly Websphere, Weblogic) the way Covalent did with Gluecode a few weeks ago.

But a real concern might be how a Red Hat JBoss partner program dovetails with the RHEL partner program announced in March. I will follow up by "talking to..." a few Red Hat partners over the next few weeks and see what they think.

As an aside, at the recent quarterly conference call, Red Hat signaled likely acceptance of the GNU General Public License Version 3 (GPL V3) based on its first look at the third draft (as I described at the time, there will be one more final-call draft after a minimum 60-day comment period on the third draft). Red Hat thinks GPL V3 “is better on the patent stuff.” The company also said that “like any OSS project, it gets better with each iteration.” I take that to mean that it still isn’t where Red Hat would like it to be but they would not comment further, saying the company prefers to work through the relevant Free Software Foundation license committees. Red Hat uses a hybrid legal document; different products are licensed under different open source software licenses, including one use of GPL V2 and one use of the Lesser GPL (which goes away with GPL V3).

Posted by dennisb in OSS Business Issue | Permalink | Comments (0) | TrackBacks (0)

Today, I’d like to point you to the TIBCO General Interface (GI) rich internet application (RIA) toolkit. The pointer is based on a session I had with TIBCO recently that was primarily about their upcoming BPM Business Studio and recently announced ActiveMatrix foundation software. But as always I followed the open source software (OSS) angle while I was talking to TIBCO.

TIBCO made its Ajax-based RIA development environment open source in the fall of 2006 with release 3.2. There was a little buzz about the move at the time but basically TIBCO is not really into the OSS culture yet so a reminder is worth while. With the recent 3.3 release, the code is out of a beta stage and is now what I would call truly open source [that is, GI has been available under Berkeley Systems Distribution (BSD) terms and conditions for six months but that did not mean much until recently because you could not actually get your hands on the source code]. The latest release primarily extends capabilities to Firefox 2.0 and Explorer 7.0. These release notes include all the other key features rolled out in versions 3.1 and 3.2.

Increasingly, these “Talking with…” sessions, where I recount some conversations with OSS suppliers, will be headlined with a slash mark as in TIBCO/General Interface above and Iona/LogicBlaze on April 17. I illustrate in my OSS scorecard the close relationship between almost all successful and useful OSS and some critical corporate backer. More and more in 2007 these connections are becoming legal (acquisitions or whatever) as well as just a matter of critical corporate backing. That trend can probably be traced to IBM’s 2005 acquisition of Gluecode (the leading proponent of the Apache Geronimo application server) two years ago next month but it looks like the trend will hit high speed from here on out.

Don’t panic when you see your favorite OSS come under some “impure” corporate umbrella; it’s necessary to the overall success of the movement. Most of your “impure” corporate bosses simply need to be sure that they can go to an IBM, an Iona, or a TIBCO for support before they bet critical business processes automation on OSS.

(By the way, TIBCO Business Studio is freeware although not OSS. It is also available from TIBCO’s web site.)

Posted by dennisb in OSS Development | Permalink | Comments (0) | TrackBacks (0)

Following up on the April 10 announcement that Iona acquired LogicBlaze, I “met” this week with James Strachan, the Chief Architect and co-founder of LogicBlaze, and a mover and shaker of the Apache Software Foundation (ASF). Like Jim Jagielski of ASF and Covalent whom you met last week, Strachan came to open source software (OSS) from the IT end-user side (is there a trend here?). Strachan began his IT career out of the Univ. of Manchester in the hothouse of middleware, the financial services industry. Early employers included JPMorgan Chase, Nomura Research Institute, and Dresdner Bank. But, not a total neophyte to the supply side, James also developed software for NEON (later acquired by Sybase), and founded SpiritSoft, a UK company that provided enterprise Java middleware between 1997-2005 and that was acquired by SeeBeyond (later acquired by Sun). The four appearances of the words “acquired by” in this first paragraph tells you all you need to know about the middleware business (I can put you in touch with the founders of HyperDesk if you send me an email).

The “acquired by” mantra also probably has a lot to do with James Strachan’s philosophy of OSS. Realizing from both his supply- and user-side work that middleware was rapidly becoming a commodity, he was one of the founders of the Apache Geronimo J2EE application server project, Groovy, dom4j, jaxen, and Jelly as well as the Apache ActiveMQ project, with which LogicBlaze is so closely aligned. He is also an active committer on a variety of leading OSS projects including Spring, Maven and Jakarta Commons, and has taken the specifications lead on JSR 241 (while “sitting in” on JSR 52, 107, 127, 152, 174, 208, and 223 meetings).

From his work in the “the City” (UK for “on Wall Street”) and with the software suppliers, he realized that in the long term middleware would go open source. In fact, someone would have had to invent the OSS business model for middleware if it had not already been invented for commodity software such as utilities and operating systems further down the stack. James feels that there might be a long-term need for proprietary high-performance middleware (e.g., for spooks and stocks) but that otherwise it is hard to justify commercial software business models for middleware today, especially for midsized enterprises. That works two ways:
-- Midsize companies can’t spend a lot of money to do the simple things that application servers and message brokers do. These companies, he says, cannot afford to pay large up-front license fees by the server or platform but will spend on consulting, training, etc. to support OSS middleware.
-- Similarly, smaller software vendors with good integration middleware technology can’t survive outside the OSS model because of the cost of sales finding prospects and the length of negotiations in big WebLogic or WebSphere type deals. But with an OSS model, it is much easier to get customers (in fact, he says, LogicBlaze can’t keep up with requests for service).

That’s why the Iona deal was a good fit with LogicBlaze. Iona brings the resources LogicBlaze needs to handle all those requests for service. And James believes Iona “buys into” the OSS model. Unlike large systems middleware suppliers such as BEA and IBM, whom he says see OSS as an “on-ramp” to their proprietary middleware products. Iona sees its “commercial pieces” fitting well into an OSS “foundation” such as ActiveMQ. Even before the acquisition, Iona used ActiveMQ code in its Celtix enterprise service bus.

Strachan recalled that the ActiveMQ started 4 years ago as side project to the Apache Geronimo effort. He joined from his work with ASF although he notes that the JMS container that became ActiveMQ started at Codehaus. That type of cooperation within the OSS culture is fascinating to watch (as is the opposite trend; e.g., how the other guys that hatched Geronimo left the JBoss community and went through very un-OSS-like legalities).

Rev 5 of ActiveMQ is a couple of months away. James believes it is as good as most commercial message brokers since the development community has cherrypicked good idea from the last 20 years. He says the group is also very proud of innovate features that even commercial message brokers do not have such as:
-- Message groups
for web style “sticky” load balancing
for correlating different threads in database caching
-- A new form of failover support
As with Covalent/Apache Tomcat, Iona/Apache ActiveMQ is a great example of how the business side and community side of OSS come together.

Posted by dennisb in OSS Development | Permalink | Comments (0) | TrackBacks (0)

I “met” this week with Jim Jagielski, a great example of how both the business side and community side of open source software (OSS) comes together, as described in my post about the OSS Scorecard. Jim is both the CTO of Covalent and a director of (and I think one of the founders of) the Apache Software Foundation (ASF). The proximate cause of hooking up with Jim was the announcement on March 29 that Covalent and IBM had released JBoss migration code to ASF.

As noted in my research at various places, ASF is one of the most community-oriented of the OSS organizations. Community over code is almost a motto, according to Jagielski, who says “Good collaboration causes good code.” But almost everyone in the community still needs a day job. He notes that almost all ASF “managers” are also involved in commercial enterprises. In fact, it looks like to become an ASF manager, ASF needs to trust you to keep the distinction clear. Jagielski says vis a vis Covalent, maintaining the wall is easy because Covalent does not have a conflicting “product;” its business is to service OSS code, primarily Tomcat, the Apache web server, and increasingly Geronimo—the Apache application server. He also points out that with its business-friendly license, Apache encourages more people to use ASF code, again a good fit with Covalent.

So with his Covalent hat on, Jagielski talked about the recent announcement. I was interested in the IBM Gluecode (known as WebSphere Application Server Community Edition) relationship with Covalent. Jim said he had mentored the Gluecode community before the 2005 IBM acquisition. I think Covalent would probably have supported Geronimo the way it supports the HTTP server even if IBM hadn’t adopted (my word, not his) the project. But the IBM involvement with Geronimo adds a lot of intellectual power and credibility to the project.

I tweaked him about his quote in the PR release that was a subtle dig at Red Hat’s JBoss. He is implicitly saying that JBoss is a “traditional J2EE application server.” I don’t think the BEA and Oracle OC4J guys would think that way. But Jim’s point was that JBoss is constructed simply as a monolithic J2EE instantiation whereas Geronimo is a framework for entity connection using what the community calls the G-bean (what EJBs are to vanilla Java). He likes the way it is easy to strip pieces out and switch pieces into Geronimo; it can be stripped down to tomcat, jedi and the transaction manager if that’s all the ISV or IT department needs.

As for the March 29 announcement, IBM and Covalent see it as another way to to promote the viability of Geronimo (which both want to do to further their own co-opetitive services businesses). Jagielski feels that many companies feel that moving from one application server to another is a major pain point. With the newly released JBoss migration tool that Covalent has open sourced, Jim says it is a straightforward transition and migration. It appears Covalent and IBM chose to start with JBoss because JBoss users are already comfortable with OSS concepts. But it also appears that a similar migration tool could work for those other “traditional J2EE application servers.” Stay tuned, WebLogic and Fusion users.

Posted by dennisb in OSS Development | Permalink | Comments (0) | TrackBacks (0)

One thing I stress in my day-job research as well as in the theme of my original post here on ebizq is "open choice" as opposed to open source. All kinds of quantitative and qualitative research that I have been involved with indicates that enterprises are really looking for
-- the flexibility and ,development power of open source software (OSS) as well as
-- the ability to keep certain long-term IT assets on their mainframe (be it a real mainframe, AS/400 or other older "server") and
-- keep supporting their user base (be it actual employee, customer or partner) via the pervasive power of Windows applications such as Office, Project and so forth

I came to this very interestinng exhibit and conference in NYC, the Search Engine Strategies meeting (there is one coming to a city near you) to support my day job but took a few minutes to cruise the show floor looking for OSS-based products. As I have indicated in my OSS scorecard, such applications are few and far between. (Google is here at the conference and of course they are a big exception to that gross generalization.)

But I did come across an interesting company that offers open choice. Called Hot Banana, they offer a web marketing application that runs on both Windows and Linux. It is based on what I had long ago dubbed the open choise middleware, Macromedia Coldfusion (now part of Adobe). It's always good to see theory backed up with real life examples.

Posted by dennisb in OSS Development | Permalink | Comments (0) | TrackBacks (0)

IONA, one of the original CORBA guys, which has been re-inventing itself for the last few years based on both a proprietary ESB called Artix and with and an open source ESB project called Celtix, has acquired LogicBlaze, the corporate sponsor behind the ApacheMQ MOM. Iona says Artix and Celtix both have pieces of ActiveMQ in them. Conveniently, Celtix was moved from the ObjectWeb OSS community to incubator status under the Apache Software Foundation two months ago.

Financial analysts on this morning's conference cited my estimate of $1M for LogicBlaze's 2005 revenue (mostly subscription revenue) and Iona did not dispute that. In fact, I sense I might have overestimated LogicBlaze's 2005 results given that 2005 was not a full shipping year for LogicBlaze. $1M was probably more representative of LogicBlaze's 2006 results. Iona emphasized that this is about getting into the OSS business model and culture and less about 'buying market share." This is similar conceptually to IBM's acquistion of Gluecode in May 2005.

LogicBlaze co-founders Hiram Chirino, Rob Davies and James Strachan will join IONA's "Open Source business." The three are "key contributors" to Apache ActiveMQ and Apache ServiceMix, which Iona says many IT organizations and ISV's are incorporating in enterprise SOA environments. ISVs in this category include eCube and MySQL. Iona had been a partner because it's idea is to get away from software stacks and back to the original CORBA-heritage distributed view of infrastructure software functionality.

LogicBlaze had "incubated" out of Simula Labs founded by Winston Damarillo, also the founder of Gluecode. Simula was created to commoditze, distribute, and support OSS projects.

LogicBlaze hopes to be better able to deliver enterprise-level support to companies deploying its OSS technologies because of Iona's long experiences supporting mission critical applications, especially in the telecommunications industry and financial services. LogicBlaze's 25 customers are in a similar mix of industries.

The companies also positioned the announcement as another example of how the OSS community side of things is merging with the business side of the IT world. I have written extensively about this phenomenon at my site, itinvestmentresearch.com and recommend particularly the free microanalysis I have done on the World Domination 201 OSS treatise.

Iona also becomes a more direct competitor with Cape Clear, founded by Annrai O'Toole, who was also an Iona founder. ApacheMQ and Celtix also compete directly with all the major IT suppliers offering ESBs (BEA Systems, IBM, Oracle, SAP, SoftwareAG/webMethods)

Posted by dennisb in OSS Business Issue | Permalink | Comments (1) | TrackBacks (1)

Per a post I put up back in February, I said that one of the biggest issues in blogging about open source software (OSS) is keeping track of the products, communities, and commercial backers. There's pretty much a one to one to one relationship among the three but that relationship is not always obvious.

Even for the simplest connections. For example, it's probably intuitive that the Mozilla portal is built by the Mozilla Foundation which is backed by the Mozilla Corporation. But it is probably not obvious that whole string really unravels all the way back through AOL to the original Netscape browser. As for something more convoluted but not unusual: the Sleepycat OSS database project is really a derivative of the Berkeley database that forked out of UNIX at System V time but the community is now backed by Oracle, which purchased Sleepycat in 2006.

Some OSS has a community but no commercial backing (such projects don't usually last too long). Others have commerical backing but no community. Sometimes, that is intentional and the product is really freeware rather than actual OSS. But in other cases, it is is because the project is just getting off the ground.

So, back in Feburary, I put together a scorecard to keep the players straight and just in time for baseball season I have updated that scorecard. It is available as a free download at my web site itinvestmentresearch.com (link from button to the right where it says "My Other Work"). I have notes so far on more than a dozen OSS applications (including SplendidCRM, about whom I posted this week and Compiere about whom I posted a few weeks ago), a half dozen OSS databases, a dozen pieces of OSS integration middleware, a half dozen OSS develolpment tools and toolsets, and nine pieces of OSS infrastructure. Other projects tied together for you if it will help include (but are not limited to) Jaspersoft, openoffice, Zimbra, jackrabbit, Geronimo, Celtix, Eclipse, Debian, Fedora and Xen.

Let me know if there are other OSS projects you would like to see added--yours or something you want me to research.

Posted by dennisb in OSS Culture | Permalink | Comments (0) | TrackBacks (0)

As part of a continuing series of “meeting” with people in OSS development community, I spoke with Paul Rony of SplendidCRM, RTP, NC, on April 3. Paul is a true open source software (OSS) entrepreneur with a twist. He embraces the “free as in speech, not beer” OSS code of conduct and, like all good ‘hackers,’ seems to be totally enjoying his effort to build a better customer relationship management (CRM) mousetrap just for the fun of building it. But he doesn’t enjoy working with PHP, really feels Java failed to deliver on its write once/run anywhere promise, and only begrudgingly works with Linux. Paul is a rare breed in the OSS community: he is a Microsoft bigot.

SplendidCRM Software provides a Microsoft-centric SugarCRM knock-off. It is derived from Sugar’s images and icons, some of its HTML code and Javascript, and the SugarCRM stylesheet. The knock-off claim is specific and intentional. (NOTE: I have not used either product so I cannot verify the second half of that claim. I can confirm its Windows centricity. If some reader who uses SugarCRM takes a look and finds the “SugarCRM knockoff” part of the SplendidCRM claim inaccurate, please send in a correction.) SplendidCRM adheres to the SugarCRM license and OSS concept and provides an OSS edition which can be downloaded from the SplendidCRM site at no charge. Only Windows Server 2003, Windows 2000 Server or Windows XP, Internet Information Server (IIS), a 1GHz processor with 256 Mb RAM and 300 Mb of hard disk space are required. A professional edition with a strong SQL tie-in, supporting SQL Server, Oracle 9i or 10G, IBM DB/2, and/or MySQL is available for a fee.

Despite his overall feelings for the LAMP stack and derivatives, Paul does admire the MySQL business model and is trying to imitate it. On March 16, SplendidCRM made its platform available to enterprise software system integrators under a choice of Free-Runtime and Free-Hosting Licenses. Note that Paul thinks of his product as an integration/business process platform more than an application. He feels all business process management is rightfully coalescing around the front office application. SplendidCRM’s license to system integrators includes two distribution methods: a runtime license, enabling integrators to sell the application via CD or download; and a hosted, on-demand license letting the integrator use the Software-as-a-Service (SaaS) model.

Although I agree with the integration platform approach, the lack of MRP and similar manufacturing business process support means SplendidCRM as a platform is more aimed at services industries and organizations currently. However, the fact that it was built on Microsoft’s infrastructure with Microsoft tools means companies of any industry that are also Microsoft bigots and can get right in and mix it up with the product very quickly, adding their own industry-centric add-ons and tie-ins. SplendidCRM Open-Source and SplendidCRM Professional have identical features, but the Professional version includes the SQL source code which is typically needed when integrating. Also included in the Professional version is the source code for the SplendidCRM Plug-in for Outlook 2003. However, the Outlook 2003 plug-in is available for a fee for open source edition users. The code can run on SUSE Linux through its Mono layer, but as Paul implied more than once, “Why would you want to?”

The aversion to the LAMP stack and fondness for the .NET stack aside, SplendidCRM has a couple of major problems going forward. It needs a community to exercise the platform and build it out. Although as developers, the SplendidCRM founders think in terms of what MySQL is doing—as noted above—they may want to model themselves more after what Compiere is doing, as discussed in this space last month. But SplendidCRM has a bigger challenge than Compiere. Compiere is taking on SAP and the like but at least it is using OSS tools rather than NetWeaver and the equivalents. SplendidCRM, although it seems to position itself against SugarCRM, is really taking on that little company in Redmond in the CRM application arena using that little company’s crown jewels.

I’m going to watch this OSS process with interest.

Posted by dennisb in OSS Development | Permalink | Comments (0) | TrackBacks (0)

Everybody loves lists psychologists tell us... if only to argue with them. PC World has released a list of what it calls The 50 Best Tech Products of All Time and I want to enter my dissent. This is left coast/SV navel gazing of the worse kind.

First, although the headline of the PC World press release says best "of all time," the first sentence indicates that the editors have only gone back and reviewed the mere few years of what they call the PC era (it is PC World after all). Since there was nothing unique about the PC era as opposed to all of the information technology that came before it other than the invevitable and ongoing downsizing of footprint, it is not surprising that I can quickly identify products that predate the PC era: Wordperfect, Compuserve, Tetris (I think I remember it floating around on mini networks), and Red Hat Linux. Smarter guys than I can probably trace the laser printer, the Hayes Modem, and a few other entires back to their proper pre-PC-era roots.

I think the editors mix up the PC with workstations where convenient. The software technology fork between the mini and the workstation, Multics descendants and Unix descendants, hierarchical timesharing and peer to peer networking, support for dumb and smart clients, and so forth is meaningful. The ability to squeeze circuits on to smaller and smaller bits of silicon is just physics.

Which is why I debate Red Hat Linux's placement on a list of great PC era inventions. Linux is Unix. Granted it was written to run on the 386 but that is not the same as saying it part of the PC era. Unix was the workstation operating system developed in the late 60's and throughout the 70s as the un-Multics. Linux is just the natural descendant of that process. In particular, Red Hat Linux is marketed as a server operating system (and was from the beginning I believe), rather than as personal operating software such as DOS or CP/M.

Agree or disagree? Great, let me know via the comments button below.

Posted by dennisb in OSS Culture | Permalink | Comments (0) | TrackBacks (0)