Open Source Solutions

Clarise and I met with Gabriele Ruffatti and Grazia Cazzin, from Engineering Ingegneria Informatica S.p.A. to discuss, and as Clarise wrote, get demos of SpagoBI [a full BI suite], and see the new Spagic EAI tool that was announced at OSBC2007, and the Spago framework [think Spring, only different].

Once he returned to Padova, Italia, Gabriele sent out some thoughts that came to him after participating in the open source business conference. He kindly gave me permission to publish them here.

"I attended Eben Moglen’s speach at the last Open Source Business Conference in San Francisco (May, 22nd). Hearing so many times: “stand up for freedom”, I thought: "I’m here now to promote free (in Europe we say libre as well) open source projects in different domains (Spago, java framework; SpagoBI, Business Intelligence; Spagic, SOA environment; Spago4Q, a Business Intelligence domain specific solutions) at an event mainly presenting commercial open source". While there, I’ve promoted new enhancements of SpagoBI, the Business Intelligence Free Platform, comparing it to other commercial open source choices such as Pentaho and Jasper and I’ve thought again: "what is the key differentiator? Stand up for freedom…"

"A participant said to me: “Guy: you’ve made a mistake. SpagoBI is not the Business Intelligence Free Platform; it’s the Free Business Intelligence Platform”. Was he right? I think no. This is the key: SpagoBI is a free platform, commercially supported, offering a new choice despite many other commercial open source products claiming more effectiveness to the market.

"What I mean by "commercial open source product" is: a solution claiming to be open source, claiming to have a community supporting it, but offering closed add-ons for enterprise adoption with a proprietary approach to the market (i.e.: acquisition of projects and IPs to strengthen its stack, dual licensing approach, aggressive marketing). Is it effective for the BI domain? Yes, probably it is.

"What I mean by a "free platform" is: free design, free collaboration, free assembling, free adoption. Not just because the license (SpagoBI adopts the GNU LGPL license), but because the efforts are in the software development improvement direction rather than in marketing proposition, in collaborations with different projects and solutions instead of acquisitions, in integrations with many free, open source, and also closed solutions to achieve the most effective solution for the user, maintaining a totally free code base core, instead of closed specific add-ons. Is it effective for the BI domain? I really don’t know, but the market, or a very new market knows the answer.

"Stand up for freedom. Is it “against” a wide commercial adoption? Is it “against” the growth of a strong business ecosystem? I know that Europe, Asia and South America are looking for free/libre software; Public Administrations are looking for free/libre software; new domain specific applications can be built thanks to free/libre software (Spago4Q is a just a first sample of it). At OSBC in USA, I’ve heard not only of commercial open source packages, but also of “building the right solution with an assemble mindset, opposite to a buy mindset”. It’s just a new choice: with an open mind, you can build open applications offering freedom to invent, to share your own knowledge, to assemble the right solution for people’s needs."

-- Is freedom the right approach to the Business Intelligence domain? from Gabriele Ruffatti, Director - Architectures & Consulting, Engineering Ingegneria Informatica, S.p.A.

F/LOSS - Free/Libre… Open… "Free as in speech, not as in beer" is well known. But how open is open? This is a question that continues to be debated in terms of whether or not a company is truly open source, pureblood or mudblood. Andy Astor puts forth the rationale behind EnterpriseDB's strategy around openness [NB: EnterpriseDB is a company that is sometimes cited as a company using open source without being open source]. We tend to be fairly liberal in our acceptance of what is an open source company [including EnterpriseDB], while the market for enterprise open source defines itself and "commercial open source" has more denotation than connotation within that market, and I rather like the direction being taken in Commercial Open Source, cited above, by Carlo Daffara. Others may be more stringent in accepting a company as truly F/LOSS or not.

Most open source projects avoid locking their users into their product by following or providing open standards and open APIs, and/or using a plug-in, or framework, stack or platform architecture. Such an architecture allows the users and the projects community to more easily customize the product to suit their unique needs, and to more easily give-back to the project, without necessarily becoming an approved committer to the core code. We're currently following 47 open source projects that provide components that can be used to build solutions for data management and data analytics, data warehousing and business intelligence. We're following 5 open source BI suites, most of which use or can use the same core components. JasperReports for the reporting engine and Mondrian for the OLAP engine are common but not all-pervasive. All five suites provide a convenient stack of components that work together "out-of-the-box", but provide varying facility for including additional or replacement engines and components. We're still working out what is the real differentiator among all these tools, but Gabriele has a very good point: the more freedom, the more openness, the better.

Personally, I've been working in data analysis for nigh onto thirty years. The reason that we've been looking for open source solutions for decision support, data warehousing, data mining, BI, GIS, EAI, BPM, [insert latest hyped term here], etc, is that this area requires more customization and user involvement than any other enterprise application. The beginning of this millennium saw a trickle of open source projects addressing BI needs, and 2005 saw an explosion in the number of projects. Open source BI provides the flexibility that is always needed in satisfying the user needs for data analysis, visualization and sharing. The greater the freedom, the greater the openness, the greater the flexibility… the easier it is to satisfy the users. This is why data management, analytics and BI is going to be one the most successful markets for enterprise open source, bar none.

Unlike Joseph, I did not attend the OSBC San Francisco. I, however, attended the Networking Reception where I had the pleasure of meeting Grazia Cazzin and Gabriele Ruffatti of Engineering Ingegneria Informatica, S.p.A.

Grazia provided us with a demo of Spago BI. The features of SpagoBI include:

* Portal
* Report
* OLAP
* Query By Example (QbE)
* ETL
* Dashboard
* Document Management
* Metadata Management
* Versioning
* Administration
* Data Mining
* Collaboration
* Geographic referenced information analysis

SpagoBI is positioned as an integration platform that provides BI solutions to the enterprise. It has no professional version that one can purchase. All the functions and product releases provided are available as Open Source. It is as is released under the GNU LGPL license and is hosted by ObjectWeb Consortium.

It is an impressive OSS BI solution. You can try out the demo yourself and tell us what you think ...

Open Source, SOA, and the Next-generation Data Center
Panel:

• Akash Garg, hi5 Networks
• Dave Rosenberg, MuleSource
• Sanjiva Weerawarana, WSO2
• Moderator: Michael Coté, RedMonk

What have the panel found what an SOA actually is?

Hi5 is a service to their end users, and not an enterprise SOA, so speed is key.

Mulesource thinks SOA is like a Unicorn, a beautiful idea not easily captured - OK, Travis, I want to see the cartoon of the Mule with a horn.

WSO2 sees that SOA allows the entire structure is internally service oriented, and allowing end users to aggregate needed data.

Customization and flexibility; open standards are even more important to SOA than open source; XML is a key standard, especially looking at social networks as a platform upon which folk can build.

WS*

Dave thinks trying to define something so complex is a morass, and that web services got somewhat out of control. Another interesting point is how Amazon, Google etc can publish web services allowing the world to work with them.

Sanjiva: WS* became a marketing tool for too many, but there is a core of about 15 documents.

Akash: For hi5, simple JASON, http, etc is enough.

From the buy side and the sales side, does it seem that customizing those 7-15% of WS* is a prime area for commercializing open source or is battling the closed source vendors where it's at.

Dave: if the point of SOA is to have things loosely coupled and use best of breed, going to the large stack vendor makes no sense. Start with Oracle Fusion, try to take one thing out, and it all falls apart.

Akash: agrees

Sanjiva: The community makes for the open source advantage.

One argument made against this is that the closed source vendors have 100s of trained QA folks. Dave is also seeing that sometimes the speed and volume of fixes and changes can be scary to the user.

Linux and Apache proved that the community works.

hi5 is not an open source project, but a site that uses a lot of open source and have found the community, such as PostgreSQL, to be very responsive.

Questions from the Audience

Scripting vs. Java?

Sanjiva: everything is in Java and C, and they are working with Zend for PHP binding, as well as ActiveX bindings. There is clearly a lot of interest in dynamic languages, and those communities have been underserved.

Dave likes the LAMP stack, but there may be questions of scale for Ruby, etc. Why not use Java and Spring? Java provides the scale. Mule supports PHP, PERL, etc. Build an app the right way - PERIOD.

Do you make a distinction between online businesses and more traditional enterprises?

The traditional enterprises often have legacy apps to worry about. The online businesses may have scaling issues.

Sanjiva has seen enterprise customers want REST and other lightweight services.

How will SOA and Virtualization work together to help enterprises support remote teams and distributed workgroups?

Akash: server farms, Akamai...

Dave: Everything is an endpoint, the whole goal of SOA is to allow things to work together. Companies do work through the cloud. Virtualization does bring a set of other problems for SOAs and how to work through the VM.

Sanjiva brings up Amazon EC2 and a customer of theirs that is working to unify virtualization and SOA. Virtualization naturally has a component of unification that works well with SOA.

Greenfield and Legacy; Cultural and Data Center Transformations

Dave has seen where SOA was the method to move from legacy to greenfield and this is one of the problems with WS* confusion. Silo'd information is a nightmare; and regardless of how implemented, the idea of stand-alone applications is dying.

Sanjiva: SOA as a technology platform is suitable to do what ActiveX, fuzzy beans, etc have tried with reuse. The only advantage that services has over these other technologies is the higher level of granularity forces more detailed thinking of the business.

Managing Services/SOA?

Dave - Mulesource built in MuleHQ but still not that easy yet.

Sanjiva - key part of making an SOA works

Akash - abstracting one layer up so that it is well understood how what they are monitoring affects the business; in many ways, hi5 is the next generation database, open source is core, SOA was home grown, which led to the question of what Mule and WSO2 are seeing: WSO2 small and on the web to large and internal; Mule sees that customers run different ESBs and multiple Mule instances, e.g. as H&R Block showed in yesterday's keynote, they have at peak tax season, 13,000 instances.

How would it work with ESB/SOA as a Service?

Dave: they're launching this later this year, but some things will always reside behind the firewall

Sanjiva: SaaS for middleware is definitely happening. But beware that SasS is another form of lock-in unless the identical software is available as open source. Discussion and debate ensued.

Coté changed the question a bit for Akash to outsourcing the data center. Who doesn't today? Rack or leased servers, etc.

Community Development: Business Development for the 21st Century
Panel:

• Danese Cooper, Intel Corporation
• Chris DiBona, Google Inc
• Andy Dreisch, SugarCRM, Inc
• Dawn Foster, Jive Software
• Moderator: Raven Zachary, The 451 Group

Danese objects to the term "Community Manager" as one never really manages the community. Intel is a huge contributor and user of open source. And you might never know this in that Intel wisely separates the open source projects from the corporate presence.

Andy feels that SugarCRM does a pretty good job at garnering contributions; they also reach out to the community to assure that design decisions are in accord with what the community needs. The community acts as a force modifier for the development team. Andy is in charge of the community, the Sugar Exchange, and the online training for commercial customers.

Dawn at Jive Software, which is mostly proprietary, has a slightly different dynamic, and mostly developer relations. On the one hand they have the completely open source Ignite, which is a fully community driven project. On the other hand, the proprietary Jive software, does provide the source code to customers for modifications, and supporting those developers if is very important.

Chris, at Google, needs in some ways, to address all humanity on the web.

Once upon, Raven was a web master, a mostly defunct title now. Where is the concept of "community manager" going, and where should it report in the organization? NOT MARKETING. SugarCRM split it off into a totally different organization, outside the normal corporate structure.

Many feels that there needs to be a strong community manager; is this true. Chris feels this is really dependent upon the type of project. Danese looks towards Apache, which is a federation of developers; projects may need a strong community leader at the start, such as Linus for Linux, but there may be less of a need once the project matures and the community grows. Dawn goes along with this in general. Andy likes the idea of a facilitator better.

One idea is going beyond development communities and getting the community involved in all aspects of the business. Dawn had brought this up at 451 Group's Monday Night event, and she feels that this is very important to the growth of a proejct; many communities have huge nationalization groups. Danese looked at CVS, subversion, and the newer tools like GIT which are not hierarchical.

Audience Questions

How to get customers involved in the community?

Chris: make it easy and don't force developers to upgrade in response to vendors' marketing.

Danese: human nature, maybe, to join

Dawn: expects the enterprise customers to join the community, due to its nature

What is the complete set of features offered to the full [not developer only] community?

Dawn sees it evolving from a code management tool and mailing list to forums, blogs, and wikis.

Andy looks at the Exchange where community members can post and find projects.

Danese cites Launchpad.

Chris brings up the next question as to who speaks for the customers in blogs, forums, IRC, IM, etc.

Attracting and growing?

Danese went through a variety of examples from running contests to having proselytizers running around the globe. Figure out why you want the community and where you want the project to go.

Chris and Danese agrees that the community manager or diva can kill themselves travelling.

Dawn brought up Facebook; the ways to pull in people is as diverse as the communities.

Other than hiring developers, do any of the panel hire consultants from the community?

Chris: always, or funding community projects.

Andy: bounty program.

Danese: remember that typical consulting contracts won't fly. In the early days, FSF had an attribution clause, but other companies may need to consider other legal vehicles.

Dawn has experience in getting creative with supporting individual community members other than monetarily. This is not summer of code, btw.

It's easy in the global political sphere to see if a country is democratic, but not if it has effective checks and balances to make democracy work. Open Source is somewhat the same.

Chris doesn't have a problem with communities forking project; in some ways it's a success metric. Open Source is self-selective.

Danese feels that the questions stems from companies entering open source strategies, but have difficulty loosing control.

Dawn agrees that companies that want to control and strongly influence their community are not strong candidates for open source-ing their product.

Any stories of where one tried to highlight controversy within the community to spur development; how to assure that there is enough conflict happening?

Bugs, contests, questions on the forums.

Some feel that the open source strategy is about making development cheaper and reduces sales and marketing need.

Danese: open source is about the community and doesn't really reduce these costs.

Dawn agrees that community may not offset marketing costs, you get word of mouth, but if your product sucks, it won't be good.

Andy feels that it takes time, effort and cost to spin-up and maintain these communities; but the benefits are tremendous; training content really helps to build the community.

Chris states that community is a cost of doing business in IT, open source or not.

The role of community facilitator is a unique blend of marketing and technical expertise.

Raven pointed out the some companies hired a facilitator, but didn't give them any backing, funding, authority, anything. Danese told the OpenOffice.org story of starting with a full marketing staff of 6, and moving towards allowing anyone who wanted to advocate OpenOffice.org even downloading a business card.

Andy: the community for SugarCRM has a lot of business users, but the development community is a key focus.

Dawn sees groups forming within developer communities, such as GUI.

Danese points out that Ubuntu maintains separate user and developer communities.

SugarCRM started out on Sourceforge and in parallel built a commercial side.

Dawn thinks that the key people of a product must be a part of the company.

Danese brings up Ubuntu and how Canonical and Debian interact.

There is a new mean of product and project management coming about from open source. Product managers are very much the one group that could be named as being upset about open source.

The Art of Picking Your Poison - Open Source and the Choice of an Application Architecture, Eugene Ciurana, Leapfrog Enterprises

Eugene has a rich background in IT infrastructure, and has been implementing open source and contributed code to various projects since 1997, as well as being an advocate for OpenOffice.org, Mule ESB, Apache and other open source projects.

Eugene described the job of an enterprise architect, starting with... a joke, and ending with the ability to describe your architecture in both technical terms and business terms. The best enterprise environments are a mix of closed and open source solutions - choosing the right mix is the job of the architect.

Beware of vendors being Marketectures.

Architecture is vendor and technology agnostic.

Today, open source is often the cutting edge technology in any given segment. This can be a barrier for some risk-adverse organizations to adopting an open source solution. Education, and non-vendor education in particular, is the path to overcoming this barrier.

Eugene recommends spending at least 30 minutes a day following the trends in your industry, and not just through traditional means such as trade journals, but check out digg, reddit, slashdot, etc.

Evolving an architecture from a typical point-only integration is very difficult, and Eugene went through a case study from early 2000, to prove the point. Of course, one problem with many architectures, that I've seen, is that they often lead to silos of information, duplication of information, and poor data quality. With point-to-point, each problem becomes a new, involved project, each with their own interoperability problems, and with higher expense. [Eugene's case study reminded me of a situation we hit, where a customer had rolled out a brand new order management system - in Pick/Universe - 10 years after Dr. Pick had died.]

This brings us to SOA and resource oriented computing [ROC], wherein services provide not only data but computational capability. One can leave their enterprise applications in place, and supplement with open source software within a SOA using ROC.

When looking at open source, assure that there is a strong, active community.

The first question when evaluating any technology, any product, is "does it solve your problem?" and if not, it doesn't matter if it's open source or not.

Eugene feels that open source is very good for infrastructure, because it's not domain specific. As an adjunct to this, Eugene feels that open source is not, and will not for at least five years, be good for domain specific applications. With this, I disagree.

Risks and Rewards: How Enterprises Are Adopting and Managing Open Source

Panel:

• Bill Whurley, BMC Software Inc
• Tim Golden, Bank of America
• Jon S Stumpf, AIG Technologies
• Moderator: Stormy Peters, OpenLogic

Primarily the panel discussed the policy decisions that must be made in adopting a disruptive technology, and how to assure that the chosen approach scales. The important thing to realize here is that these CIOs consider Open Source to be disruptive. In talking to sales folk for open source companies, they like to avoid the term "open source", in some cases, and focus on features and technology. But the providers must realize that the buyers, decision makers, do consider open source disruptive. To me, it's the middle choice between the traditional build vs. buy decision - but that's another post.

Legally, the lawyers simply need to understand licensing, and how that license affects or effects the use of the product in relation to your needs. There can be things buried in licenses, such as, real example "Say a prayer for my beautiful girlfriend", that the buyer may not wish to endorse. The legal counsel is not accustomed to the fact that open source licenses are non-negotiable, and the buyer becomes a second party to the license. The lawyers may need to be educated on notions of copyleft, and other provisions of open source licenses.

Any approval board must be so constructed that the board is not a bottleneck.

1. minimum number of members: purchasing, legal, chief architects of affected areas, etc
2. proceduralize patterns of behavior
3. make membership fluid
4. don't have an open source review board, make the decision to consider open source products and flow those products into the standard procedure
5. exception process to the standard procedures may be required for open source

Open source can sneak in very easily [so did WiFi, eh?] but to keep users compliant to internal standards and architecture, IT can't impede the users trying to do their jobs. This is a people management problem that has nothing to do with open source.

Tim makes the point that open sourc software is the same as closed source software, except in one's relation to the software.

A question from the office concerning indemnification... turns out that it's not seen as much value to the CIOs, though it is an issue, but it's also an issue with closed source.

Managing open source software, again, much the same as managing any software: security impact? benefit of the upgrade? This applies to applications. Tracking and keeping abreast of framework components must be treated differently.

There is a quality assurance concept known as two-way traceability which I think is very much underused in software development and configuration control.

What happens when a community goes dormant? A point was brought up that when working with a small company, purchases often include software escrows. One value of open source, is that the code is already there. So, one of the first responses is to contact the license holder, and see why the project may have gone dormant; then determine the remediation strategy. It can range from hiring the IP holder, to reviving the community.

Security teams views of open source? They use it, no opposition. The security department assesses the risk associated with anything being brought inside. Open source should be thanked for inciting more transparency from closed source companies to help with security problems.

Given that open source allows for change and modification, are there any operational Achilles' heel and how to protect against it? Unit testing.

Tim: the biggest surprise that Tim had was the amount of heat that is generated by advocating open source; the community phenomenon is very complex and one must come far outside of one's comfort zone to understand it.

Jon: dealing with the misconceptions about open source has been difficult, though the amount of documentation and knowledge about open source is helping to dispel these; Jon recommends pushing it through the standard processes and realize the licensing differences.

whurley: recognize that you're already using open source; change the way you do business and learn from the philosophy surround open source, community development, and sharing knowledge, as well as adopting the software.

The Bazaar Cathedral: A Look at Open Source at E*TRADE FINANCIAL - Past, Present, and Future, Lee Thompson, E*TRADE FINANCIAL

Today is the decade-versary of the publishing of The Cathedral and The Bazaar.

E*Trade announced their open source strategy in 2002. While it raised some eyebrows, Linux leads to dramatic improvement. Website speed has a Keynote transaction at around 3 seconds now. Days with extreme trading volumes have been handled effortlessly by using Linux and even made Investor's Business Daily when Dow lost 400 points in late February, 2007.

International customer evolution leverages the lessons learned in the USA, and bringing those services back.

Dang - low battery and no power outlet in site. BFN...

Open Source: Why Freedom Makes a Better Business Model, Marten Mickos, MySQL

Success is going from failure to failure without loss of enthusiasm
-- Winston Churchill

By doing things wrong, we will learn to do things better.

Freedom of speech vs. Free Beer: Open Source doesn't give much free beer away, but they give the customer the flexibility to use the product freely.

Liberating products goes beyond software: eBay freed Trade, Second Life freed socializing, IKEA freed furnishings and the USA freed entrepreneurism. Software freedom is so powerful as ~100,000 mostly white males now between 40 and 60 years of age created the current information society and there are now ~30,000,000 developers with much broader demographics on the Internet today.

Outsized software profits come from

1. Innovation
2. Netowrk Effect
3. Scale
4. Lock-in

The first three apply well to F/LOSS but the fourth is not compatible with open source licensing.

Filtering results for software companies, closed and open, through the philosophy of open source business models with lower sales and marketing costs, and it may be that open source companies can scale in size with profitability the same as closed source, but get to higher levels of profitability at faster rate, at a smaller size.

Open Source is not a business model, but a smarter way to produce and distribute the goods.

Success in open source requires that you serve two disparate groups.

1. Those who spend time to save money
2. Those who spend money to save time

People in group one may recommend you to those in group two, but are unlikely to convert; still there is benefit to both groups. [To me, this is the key to open source - recognizing, serving and benefiting from both groups. This is why community is so important.]

Marten found a baker's dozen of open source business models in active use today. I'll link to the online version when available. All of them are hybrid models that provide free software, while charging for something else, from services to hardware.

Hacking the Newspaper: How an Open-Source Nerd from Kansas is Revitalizing Journalism, Rob Curley, Washingtonpost.Newsweek Interactive.

Yesterday, Rob gave this talk in four hours at Berkeley - today, he's drinking Red Bull, and he'll be giving it in 30 minutes or he'll be pissing himself. There you go. OH, and he's using a Mac. No problem then.

Open Source is rarely used in traditional journalism. But, now, meet Django.

Wahingtonpost.Newsweek Interactive is all about openness in everything.

The Jayhawk Basketball ticket controversy... How not to go from a system where the only way to get into a game was to have your grandparents die, leave the tickets to your parents, who die and leave the tickets to you, to a system where you can get tickets based on how much you donate to the University. Internology - get a student with a digital camera to sit in every seat in the stadium with a digital camera, and post the results as drill-down from a seating chart of the stadium. That's great, but don't post the internal memos about the decision making on the same page.

Starting in 2001, they began using SMS text messaging from the web site to get information out about things like T-ball games for the kids. It evolved to send things like tornado warnings, but really, the most important thing is T-ball.

Stats to the max, with bios unlike most you will see - like arrest records. And FutureHawks, where kids could send in and sign letters of intent to play for the Jayhawks, and then report on how they played in their last game. All with supplemental data gathered in the most accurate manner possible - calling the players' Moms.

Forecast - comes from wind blowing by local landmarks easily recognized by students and alumni - e.g. the doughnut shop.

Live chats in Django.

Getting virtual - or a good excuse to expense an Xbox. Publish a full game story, with video, from a simulated game with the same teams playing that day. For the first three games, they came closer to predicting the outcome than Las Vegas.

They took these ideas to Florida's retirement community - can these ideas and technologies meet the needs of a diametrically opposed demographic. Yes.

The restaurant guide in a database that answers questions like Vegetarian Friendly and is there a dock to park the boat, but most importantly, a time-sensitive database to discover when the restaurant starts and stops cooking. When they saw that all the old folk in Naples, FL had iPods, they moved the restaurant guide to that white piece of plastic. And so the team could find pancakes at 2:00 a.m. [when the bars close, isn't it?] they put it on the mobile phone. And since you needed pancakes at 2:00 a.m. they started GODcasting - services on the go.

Contents how you want it: iPod, mobile, PSP. In the newspaper industry, if you can get your content on a device, you can expense. Expect many more devices to be supported soon.

OnBeing gets the real scope from the horse's mouth - the person on the street and the videographer and the audience's comments. The interviews take an hour, and are edited down to 90 seconds.

Questions

Paying for support? One programmer on staff at their skunkworks, who write Django and supports it.

Why was Django created? They needed the "cliff notes" of programming, so that they could have an idea at lunch and go live with it by supper.

The real story for me from this high energy presentation was the power of innovation in custom software development. 'Tis great that Django is available for others to grow and build upon, and I would like to hear from them.

Matt's pondering the question "How to get the second day people to follow the schedule?" The rooms almost empty - everyone is off breaking their nightly fast.

Hmm, Blackberries interfere with the audio system in the room, Treos, etc are fine. So, Blackberries should be turned off. Finally, after years of missteps, score one for Palm. My Lifedrive is happy.

The first Keynote of the day, the fourth of the conference, will give us some of the cutting edge prognostications as to where Open Source may be in the next two to five years.

Yesterday, I didn't think to mention what a great location the Palace Hotel is. First, it is very historic, having been established in 1875. Second, it is a beautiful venue. Third, it is very conveniently located, at the corner of Market Street [the main drag in San Francisco] at New Montgomery, which is between 2nd and 3rd streets, making it right in the heart of the Financial District. Around the corner, at 2nd and Mission, is Mondo Caffé, which I've enjoyed since I worked in an office above circa 1995. The Thirsty Bear is another block down New Montgomery past Mission, on Howard. Great coffee, freshly brewed beer and marvelous tapas, surrounded by everything you could need - like when I ran across Market to the Radio Shack to buy new batteries for my podcasting lavaliere microphones. What could be more convenient? BART to the Palace - take the Montgomery BART stop, exit via the Market | Post stairs, which lead to the turnstiles to exit BART, keep to the left and leave through the New Montgomery Street stairs, which bring you to Market Street in front of the Palace Hotel. That's convenience.

I hear the rock music now - it's time to go in for Matt's day 2 Welcome.

Copyleft Business Models: Why it’s Good Not to Be Your Competitor’s Free Lunch, Eben Moglen, Columbia University Law School.

Anyone who has heard Eben knows that it's useless to try and capture the nuance and delightfulness of his speech.

The story he's telling of the history of the computer industry and it's evolution out of and back to open source is wondrous.

One point that Eben makes, which touches on an argument a friend and I have been having via email, is how the United States in the 18th, 19th and early 20th centuries had the problem of attracting people to this land of opportunity, and the impact on Intellectual Property law of today. Quite different from the Immigration laws of the past 50 years, or the debate that's going on in our Congress today.

The source of the strength, and the fundamental model, of the USA century, was that after World War II, the USA built the Interstate Highway system, invested in public medical research facilities, and created a secondary, tertiary and quaternary education system second to none.

How Not to Be Your Competitor’s Free Lunch

1. Don't tolerate software or service monopoly.
2. Stand up for Freedom. Freedom here is not some empty generality; it has content, meaning: share and share alike; what's ours is yours too, as long as you play by the rules. e pluribus unum or by other words GNU/Linux

CIO Hotseat, Niel Armstrong, Activision, Eugene Ciurana, Leapfrog, S. Christopher Gladwin, Cleversafe, Ilan Kinreich, RadView Software, Inc. Oliver Marks, Sony Playstation, Dave Rosenberg, MuleSource, and Moderator: Steve Fox, InfoWorld

The Open Source Companies make a pitch, and the CIOs decide if they would buy. From the fine Tiffany bag, er, or not, the decision is made, who goes first...

CleverSafe

Brings open source IDA technology: inherently secure, inherently private, dispersed storage for all the world's data.

The benefits of dispersed storage:

• Secuity/Privacy
• Reliability
• Scaling
• Cost Effective
• Infrastructure for all sorts of Mobile to Enterprise solutions

Five minutes and done with the pitch. Now the grilling begins.

MuleSource

Dave showed the normal ESB architectural diagram, and told the CIOs that if they hadn't seen before, they're not reading enough trade magazines...

235 reasons from Microsoft to retain FIVE law firms.

Done in four minutes, and Eugene, a two-time Mule production user, wants to know what will give him a warm fuzzy. MuleSource has so much money, Dave's going out to buy a boat tonight. [I want to know what kind of boat and when's the party? Update: MuleSource Closes 12.5MM$US in funding.] And they use Palomino and other tools to assure that all their IP is rightfully theirs. There is a clause in their contracts for subscribes that MuleSource will bear all the legal costs if there should be a problem.

Mule works with all other ESBs; if there's a way to get data into and out of the application, it can ride the Mule.

MuleSource advocates multiple ESBs - isn't this adding complexity not reducing it? Putting a box in front of some wonky application and let it talk to Mule and Mule talk to everything is much less complex than trying to work with funky data.

RadView WebLOAD

Identify performance bottlenecks, pinpoint the root cause and monitor the production environment. JasperSoft is their reporting/analysis tool.

Five minutes and the CIOs want to know more - like what the product really does.

RadView was closed and is taking WebLOAD open - how are they building their community? webload.org has a lot of information, and working with partners and customers

Wrap-UP

WebLOAD looks great for downloading, but not being mission critical, there may not be a reason to pay for it; the presentation wasn't convincing.

MuleSource makes a lot of sense, being used - kick the tires; marketing message is confusing to some.

CleverSafe, from a regulatory stance, and whether or not one can recreate their data no matter what, is scary; needs a better business case, especially regarding governance.

Online Strategy for Open Source Business, Mark Burton, MySQL, Mark de Visser, Zend Technologies, Inc. Michael Evans, Red Hat, Inc. John Roberts, SugarCRM, Inc. and Moderator: Steve Fox, InfoWorld

Steve Fox provided the introduction, discussing the synergy between Web2.0 and Open Source, and then the panel introduced themselves.

The first question is how the open source communities use online strategies.

1. Get the word out from downloads to community forums.
2. The online world is the starting point for anyone interested in an open source project.
3. Starting with a $25/month shared account and Sourceforge, and recognizing that traditional sales model wouldn't work, SugarCRM provides a strong self-service infrastructure, including the exchange for selling extensions and their forge - three years later
4. Provide a global reach
5. Making the technology easily available: downloads, documentation, conversations
6. marketing
7. Red Hat has learned that you can't half-ass, part-time your online presence; you must have dedicated, knowledgable people

Who downloads the software, and what do you know about them?

1. Fedora is free and there is very little tracking, especially as Fedora is available from a variety of mirror sites; RHEL line is not a free download; JBoss brought a lot of tracking sophistication
2. SugarCRM was impressed with how strong the MySQL and Red Hat brands are, without advertising; information from downloaders is voluntary; proving the value of the software is the main strategy; SugarCEM has been translated, by volunteers into 70 languages, and has been downloaded into over 100 countries
3. MySQL has massive distributions through mirror sites, but once they come to MySQL, they start to gather information about uses and needs; about 80% of paying customers have already used the product
4. Zend doesn't know who download until they come to Zend for services or more information

What customer data do you have, and how do you use it?

1. To be open source, you must be very open, totally transparent, the code, development plans, and bug databases are all open to everyone
2. MySQL tries to understand use trends and some demographics; e.g ~three-quarters of their customers are SMB
3. The folk who reveal that they are a competitor when they download an open source product are the best
4. By capturing demographics, e.g. developer, DBA, etc, Zend can target content to particular users
5. SugarCRM tries to look more at the ideas that users bring to the forums, as well as generic data about underlying databases and platforms, and uses; they also use newsletters and developer groups; there is a mechanism through the "heartbeat" information from the code, as to what is being used: version of the underlying infrastructure, how is MySQL being stressed, etc
6. Web tracking tools are used at Red Hat provide some interesting data, but the customer data through sales is the most interesting; a subscription model causes an interesting problem as customers want to co-terminate subscriptions that were bought at different times from different channels
7. Open Source companies must always be thinking about the customer, and how to make subscription renewal attractive and easy

How do you convert online downloads to sales?

1. Poorly... Zend, if we could double the conversion rate... make it easy and compelling; add value so that the customer wants to further engage; make your message clear
2. SugarCRM went from counting downloads and modeling conversions to realizing that providing great software and a great service, regardless of open source or not, regardless of the license, is how coversion happens
3. Make it very easy to complete the transactions, from phone calls inbound to inside sales, to click-through transactions
4. Red Hat makes things easy at all levels, and provides options to customers to allows them to self-select up; at times there are too many leads for any size sales force or automated system to follow
5. MySQL doesn't even like to talk about conversion; one thing is that the differentiation between the community and enterprise versions must be very clear
6. If an open source company runs out of leads before they meet their sales number, they should go back to their product and figure out what is wrong with it or how it can improve
7. The product must be easy to download and easy to install and easy to use, and the blogs, forums, online interaction, and free training really help
8. A top priority is self-service from subscription renewals to security updates
9. Providing help through forums and documentation, targeting the community user; there is a significant community that won't ever need more than what is freely available
10. The opportunities for enterprise apps may be winding down; the time when three people could work together for a summer and grow an enterprise application may be gone, though timing wise, 'tis an excellent time to start an open source company as the models have been proven
11. A question was asked how SugarCRM serves the business user
    1. IT is still important
    2. They know a good CRM app from a bad one

How does one develop the services?

1. Under GPL, once can't sell a license, so Red Hat has always been service oriented, which have developed into a subscription model; the Red Hat network is at the core
2. The Red Hat Exchange is all about making it simpler: support, licensing, reviews - a fully online global marketplace - only two weeks in, the reviews are 80-90% positive
3. MySQL has done something similar to the Red Hat network, by adding a lot of service value to the MySQL Enterprise Edition that makes it easy to install security patches, updates, and tuning
4. Zend follows suit with the Zend Network, and the Zend Network makes management of the application stack built on Zend PHP even easier
5. SugarCRM does ondemand and onsite subscriptions; the subscription model forces a providing company to stay sharp and keep adding value

Open Source in the Battleground for Mobile Dominance, Benoit Schillings, Trolltech, Inc

In many ways, I'm glad that my first choice for the second session, see my schedule, was so packed, I couldn't even get into the room after talking with Nick and Ian of JasperSoft during the break. As anyone who reads our TeleInterActive Lifestyle blog knows, I'm very interested in mobile and wireless solutions, so listening about the open green phone for mobile development is great.

Benoit's talk brought forth the ways in which open source can overcome some of the barrier to mobile development. One of the major frustrations for enterprises in extending their data analytics and collaboration platforms to their increasingly mobile workforce is the lock that carriers have on what can go onto a device and what can go over the air waves. Some of this has been mitigated with smart phones using PalmOS or Windows Mobile or Symbian on GSM networks. Unfortunately, there have not been much uptake in the enterprise space.

Another very interesting discussion was around language: C/C++, Java and its versions, Python and even Ruby.

There has been talk of convergence for a decade, without much real delivery. Open Source could make this happen. Standards is very important in this area as well.

The Convergence of OSS and SaaS: The Business Model of the Future, William A Soward, Adaptive Planning

Discussion of Open Source converging with SaaS

• Use before vs. Try before you buy
• convert downloader or trial into paying customer
• plus subscription, consulting, training
• lower sales with slightly lower R & D vs. Lower R & D and maintenance
• most SaaS is North America

Adaptive Planning sees customers start on demand & bring it in house later

New SaaS about customization & feedback; rapid development cycle

SaaS & OSS culturally very similar

William went into financial and business performance metrics; for me, the most interesting point & the least proven is awareness of the lifetime value of a customer

A high percentage, perhaps one-third to one-half, of their downloads from sourceforge are business users who need help getting into production

Emerging Distribution Options

1. salesforge.com appexchange
2. sourceforge marketplace
3. Intel Suite 2
4. Red hat
5. Spikesource
6. Goggle?
7. I would add Amazon EC2 and S3 and future possibilities for a marketplace from them

Open Source: Leveraging the Capabilities to Drive Consumer Product Innovation by Marc West, SVP & CIO of H&R Block, Inc. Marc states that he'll not have the passion of Matt and Matthew, but his customers are open source's customers, and he brings the innovation of open source to their customers.

As Marc was told as a young programmer... it's about the customers, shareholders, stakeholders, and your fellow employees.

More than 10 years ago, if you wanted information you went to a training class, now you go to your peers via the Internet.

Niche is good, but scale is critical.

Marc states that we are now in our third generation of driving core technology changes and have not yet fully delivered on the value that "technology" has long pormised to bring.

Marc has a very interesting slide on how the F500 view Open Source...

Brigh Shiny Objects: "newest is not always best, neither is most mature. It's the mix of the best that wins. Without standards you will fail, because you will create a middle layer that you cannot unwind. H&R Block has used open source to make fundamental CHANGES in how they serve their customer. Marc then gave some context on H&R Block. Large market, competitive and highly regulated. They open 10,000 of their 13,000 offices and hire about 100,000 additional seasonal employees for TAX SEASON, with training and retraining to keep up with regulatory changes that happen every year, often at the last minute, which requires rebuilding their systems to match those changes... while serving customers. They make about half of their annual revenue in 12 days, near the beginning of February.

H&R Block is innovating for consumers, using mostly open source or [in-house] open sourced solutions.

• Organizit: web product using Lazlo but store your data online or on your desktop, 10 weeks to design, build, launch with over 2 million copies in distribution through download or CD.
• Tango: helping understand what you're doing when filling out your taxes, not just filling out boxes. An emotionally designed product, using Hollywood producers and consumers. Without open source, this product would never had happened. Five months to build.
• Zimbra: email, think about rolling out and rolling back all those email accounts for those 100,000 seasonal employees
• Vpro: virtual [tax] professional is a new platform to connect the consumers with tax professionals real time; eight months using multiple open souce components integrated into a simple platform that provides the right expert; a key part to "Taxes your way" and accounting your way.

Marc went into all the things that H&R Block learned, some of which are somewhat obvious, like open source still needing support, and some are less obvious, like the value of SCRUM, getting standards right, and that open source is a viable option for them. Within five years, the key Open Source components will be in the same place as UNIX was in the late 1990's; SasS will get it right and applications/computing in bite-sized chunks, on-demand, will be usable and trusted; the shift towards I creation over writing interesting code - different pipelines to manage; innovation pipeline continues to get compressed through features/competition and ubiquitous access; global teams are required to make it all work - talent is critical. Another interesting point that Marc brought up, is that there is more computing power for less dollars - do we know how to use it?

Open Source is not about saving money, it's about performance and flexibility. So says Marc, so say we all.

Matt has signed off, and the music is back as we await the first Keynote Presentation. With the music, an interesting slide show is going on, with pictures from the late 1800's and the 1900's and statements from folk like the chief of the U.S. Patent Office and Thomas Edison, all mistakenly saying that innovation was dead, but then showing things like the Wright Brothers, and Amelia Earhart and a very young Bill Gates.

And now "The Evolution of Open Source in the Enterprise by Matthew J. Szulik, Chairman, CEO & President of Red Hat.

There are more opportunities and more choices for the enterprise than at any other time, in part because open source is now in the second and third generation of providers. [As an aside, the room is really filling up - I'm crunched between two others, one typing, one not, and I can hardly keep Vate on my lap. Er, if you don't know, Vate is my MacBookPro. ] The industry has changed, and it's a wonderful thing that some folk in the industry today doesn't know what VSAM is. The opportunities being presented to IT customers today cover new abilities in BI to virtualized appliances from developers around the world. The opportunities for the industry is to unlock trillions of dollars in critical data still riding in legacy systems like Pick. [This I know from experience when we did a dW for a large, F100, organization whose ordering data was still in Pick/Universe.]

For the customers with whom Matthew speaks, it's not the bits, it's how fast open source can scale and apply project management skills to solve their needs today and in the future. All interesting systems change. But where do these skills come from to meet these opportunities?

Lock-in vs. interoperability, proprietary vs. open source... There are no open source vendors that have a problem with interoperability. The transparency of the source code leads to open standards, federated data models, better BI, real-time, effective decision making information.

Governments around the Globe are putting money into thinking about and implementing open standards, such as ODF. Customers are thinking about new data models, federating data, and how this will happen in 2010, 2020 with thousands of virtualized clients, virtualized servers and virtualized appliances. Companies like Greenplum are looking at this now.

Patents... Focus on what is original, what is truly an invention. Patents can stifle innovation. The EU is addressing this in a multi-year legal action. Another point brought up by Matthew is that the relation between the customer and the provider of IP has changed, the customer is much more involved. Matthew brought up a project Red Hat is doing right now with J.P Morgan. Another example is what's occurring in DRM free digital music. One of the most compelling arguments for open source is that the customer is an active part of the process. Education about patents, trade secrets, copyrights, and differing patent law from country to country. Open development is creating an open, federated portfolio of intellectual property. Open source is building on the last 30 years of the IT industry without barriers to sharing knowledge. An hardware example, as brought up by 60 Minutes last week, is the One Laptop per Child project - all open, all about sharing. The OSBC assemblage has an opportunity to meet this challenge through dialog and discussion. Matthew also brought up generational issues; the younger generation embraces sharing much more than those over 40. [As a 51 year old, once hippie, I dispute this idea by the way. Though as hippies became YUPpies, hmm].

The discussions don't center on VC funds and the exit strategy, but on innovation strategies around technology. Matthew knows he has been chasing windmills for over 10 years, and he is amazed at faces in the audience today, of people who wouldn't return his phone calls 7 years ago.

Projects coming out of Rice challenging the traditional text books and their cost... Indiana University Sky project... Throwing away 11 years of Breast Cancer research because of system incompatibility...

How will we create a vibrant culture around the value system that is open source? Don't be held hostage by the past. Really think about this exciting industry, this value system, that can unlock human potential, solve new problems, in all areas.

The rock music has stopped and Matt Asay's Welcome is happening as I type. This is the seventh show and the fourth year for the OSBC. There is a revolution going on in Open Source solutions, with over 2 billion US$ in V.C. funding and 65% of enterprises stating that open source has spurred innovation in their industry.

Open source's only real friend...
... is the customer
A friend of Matt said this

A very interesting point is that proprietary vendors create ecosystems around their products that do create value for their customers, but are also designed to keep competitors out of the ecosystem and out of the market. Organizations, such as the Open Source Alliance, are creating ecosystems for Open Source projects and companies, that create value for customers and communities across open source projects, but invite all open source providers.

I've arrived and registered at the OSBC2007. There are several conferences happening at the Palace Hotel in San Francisco today, but OSBC seems to be the biggest, at least with the longest Agenda and most rooms.

I've found the WiFi and have checked out the equipment. I'm just watching the passerby and looking for folk that I know.

I have heard back from Lidia Fiorini of Engineering Ingegneria Informatica, S.p.A. who told me to speak with Gabriele Ruffatti and Grazia Cazzin about SpagoBI.

And now I'm off to find coffee and get settled in from Matt Asay's Welcome.