© 2010 Forrester Research, Inc. Reproduction Prohibited 1
© 2010 Forrester Research, Inc. Reproduction Prohibited1
© 2010 Forrester Research, Inc. Reproduction Prohibited2 © 2009 Forrester Research, Inc. Reproduction Prohibited
OSS Adoption Patterns In Enterprise IT
Jeffrey Hammond, Principal Analyst
August 11, 2010
When it comes to Enterprise IT adoption, Open Source Has “Crossed the Chasm”
© 2010 Forrester Research, Inc. Reproduction Prohibited4
A word about the surveys used in this deck…
1. Forrester‟s Enterprise And SMB Software Survey, North America
And Europe, Q4 2008/9(2227/2165 total – 1114/940 for
Development)
Primarily Directors, VP App Dev, VP I&O, CIO
2. Forrester – Dr. Dobbs Developer Technographics Q3 09 (1298
total)
Mostly developers – slightly skewed toward systems and .NET
3. 2009/2010 Eclipse Community Survey ( 1498 total)
Mostly developers – skewed toward Java
© 2010 Forrester Research, Inc. Reproduction Prohibited5
Market Trends And OSS Evolution
Examples And Best Practices
Agenda
© 2010 Forrester Research, Inc. Reproduction Prohibited6 © 2009 Forrester Research, Inc. Reproduction Prohibited
0% 10%20%30%40%50%60%70%80%90%100%
Reduce IT costs
Improve integration between apps
Use IT to increase innovation
Support regulatory reqs.
Improve communication of IT value to business
Increase ability to meet unmet demands for IT services
Address IT staffing and skills challenges
Expand use of open source software
Reducing number of (major) software vendors we work with
Expand use of SaaS
Move some/more apps to off premise providers
1-Not at all important 2 3 4 5-Very important Don't know
In 2009 integration and innovation were top of mind
Source: Enterprise And SMB Software Survey, North America And Europe, Q4 2008
Base: 2227 software decision-makers at North American and European enterprises and SMBs
???
“Thinking of your firm‟s current planning cycle, how
important are each of the following goals?”
© 2010 Forrester Research, Inc. Reproduction Prohibited7 © 2009 Forrester Research, Inc. Reproduction Prohibited
0% 50% 100%
OSS
Business Process Management(BPM)
Mobile tools/middleware
Advanced analytics
Data Services/Information as a Service (IaaS)
Information Lifecycle Mgmt (ILM)
Application Lifecycle Management(ALM)
PaaS/Cloud
Rules
Complex Event Processing(CEP)
Enterprise Service Bus (ESB)
Expand/ Upgrade existing implmentationImplementing/ Implemented
Piloting
Interested/ Considering
Decreasing
Removing
Not Interested/ Don't know
Adopting OSS was an important technology goal
Source: Enterprise And SMB Software Survey, North America And Europe, Q4 2008
Base: 2227 software decision-makers at North American and European enterprises and SMBs
“What are your firm‟s plans to implement or expand its use of
the following software technologies in the next 12 months??
© 2010 Forrester Research, Inc. Reproduction Prohibited8 © 2009 Forrester Research, Inc. Reproduction Prohibited
0% 50% 100%
Reduce costs
Improve business process execution speed
Support company growth
Improve integration between apps
Support regulatory reqs.
Be more nimble as the firm change business processes
Improve collaboration and information exchange
Increase innovation
Support our company's sustainability goals
1-Not at all important 2 3 4 5-Very important Don't know
IT is leaning out for growth at speed in 2010
Source: Enterprise And SMB Software Survey, North America And Europe, Q4 2009
Base: 1900 software decision-makers at North American and European enterprises and SMBs
“How important are each of the following business goals to your
internal IT organization when making software decisions?”
© 2010 Forrester Research, Inc. Reproduction Prohibited9 © 2009 Forrester Research, Inc. Reproduction Prohibited
0% 50% 100%
Update/modernize key legacy apps
Increase deployment/use of collaboration tech
Upgrade enterprise apps
Reduce number of major vendors we work with
Expand use of Agile software dev.
Increase use of SaaS/Cloud
Increase use of OSS
Implement "green" software tools
Consolidate or rationalize enterprise apps
1-Not at all important 2 3 4 5-Very important Don't know
Urgency to adopt OSS is fading…
Source: Enterprise And SMB Software Survey, North America And Europe, Q4 2009
Base: 1900 software decision-makers at North American and European enterprises and SMBs
“How important are each of the following business goals to your
internal IT organization when making software decisions?”
© 2010 Forrester Research, Inc. Reproduction Prohibited10 © 2009 Forrester Research, Inc. Reproduction Prohibited
…because it‟s now widely adopted…
Source: Forrester -Dr. Dobb‟s 2009 Developer Technographics Survey, Q3 2009
Base: 1,298 development pros at North American and European enterprises and SMBs
“Which of the OSS infrastructure tools have you included as part of your
development activities or deployed an application or software project to?“
(Select all that apply.)
3%
4%
7%
7%
10%
21%
22%
28%
45%
45%
46%
48%
57%
Portals or mash-up servers (e.g. Liferay, Dapper)
Business applications (Sugar CRM, Bravo)
Other, please specify
Business Intelligence tools (e.g. BIRT, Jasper Reports, Spago)
Content Management Systems (e.g. Alfresco, Drupal)
Have not used OSS as part of my development projects
Application frameworks (e.g. Spring, Rails, Zend)
Application servers (e.g. JBoss, Glassfish, Tomcat)
Web servers or networking components (e.g. Apache, Samba, Radius)
Databases (e.g. MySQL, PostgreSQL, SQLLite)
Development IDEs (e.g. Eclipse, NetBeans)
Operating systems (e.g. Red Hat Linux, Suse, OpenSolaris)
Programming languages (e.g. PHP, Ruby, Python, Java)
© 2010 Forrester Research, Inc. Reproduction Prohibited11 © 2009 Forrester Research, Inc. Reproduction Prohibited
…and management has caught up to developers
Source: Forrester -Dr. Dobb‟s 2009 Developer Technographics Survey, Q3 2009, Enterprise And SMB Software Survey,
North America And Europe, Q4 2009
Base: 1,298 development pros at North American and European enterprises and SMBs and 1,298
development pros at North American and European enterprises and SMBs
“Which of the OSS infrastructure tools have you included as part of your
development activities or deployed an application or software project to?“
(Select all that apply.)
3%
4%
7%
7%
10%
22%
28%
45%
45%
46%
48%
57%
4%
7%
6%
6%
12%
13%
35%
58%
58%
24%
61%
55%
Portals or mash-up servers (e.g. Liferay, Dapper)
Business applications (Sugar CRM, Bravo)
Other, please specify
Business Intelligence tools (e.g. BIRT, Jasper Reports, Spago)
Content Management Systems (e.g. Alfresco, Drupal)
Application frameworks (e.g. Spring, Rails, Zend)
Application servers (e.g. JBoss, Glassfish, Tomcat)
Web servers or networking components (e.g. Apache, …
Databases (e.g. MySQL, PostgreSQL, SQLLite)
Development IDEs (e.g. Eclipse, NetBeans)
Operating systems (e.g. Red Hat Linux, Suse, OpenSolaris)
Programming languages (e.g. PHP, Ruby, Python, Java)
© 2010 Forrester Research, Inc. Reproduction Prohibited12
11%
15%
17%
65%64%
8%
16%
46%
33%
68%
60%
Portals or mash-up servers (e.g. Liferay, Dapper)
Business applications (Sugar CRM, Bravo)
Business Intelligence tools (e.g. BIRT, Jasper Reports, Spago)
Content Management Systems (e.g. Alfresco, Drupal)
Application frameworks (e.g. Spring, Rails, Zend)
Application servers (e.g. JBoss, Glassfish, Tomcat)
Web servers or networking components (e.g. Apache, Samba, Radius)
Databases (e.g. MySQL, PostgreSQL, SQLLite)
Development IDEs (e.g. Eclipse, NetBeans)
Operating systems (e.g. Red Hat Linux, Suse, OpenSolaris)
Programming languages (e.g. PHP, Ruby, Python, Java)
20000+
5000-19999
1000-4999
500-999
100-499
20-99
OSS adoption by company size
“Which of the OSS infrastructure tools have you included as part of your development
activities or deployed an application or software project to? “(Select all that apply.)
Source: Enterprise And SMB Software Survey, North America And Europe, Q4 2009
Base: 1900 software decision-makers at North American and European enterprises and SMBs
© 2010 Forrester Research, Inc. Reproduction Prohibited13
8%
12%
9%
18%
10%
41%
66%
71%
30%
63%
68%
4%
6%
6%
11%
14%
36%
60%
56%
23%
63%
55%
Portals or mash-up servers (e.g. Liferay, Dapper)
Business applications (Sugar CRM, Bravo)
Business Intelligence tools (e.g. BIRT, Jasper Reports, Spago)
Content Management Systems (e.g. Alfresco, Drupal)
Application frameworks (e.g. Spring, Rails, Zend)
Application servers (e.g. JBoss, Glassfish, Tomcat)
Web servers or networking components (e.g. Apache, Samba, Radius)
Databases (e.g. MySQL, PostgreSQL, SQLLite)
Development IDEs (e.g. Eclipse, NetBeans)
Operating systems (e.g. Red Hat Linux, Suse, OpenSolaris)
Programming languages (e.g. PHP, Ruby, Python, Java)
UK
Canada
US
Germany
France
US adoption is closing the gap with Europe
“Which of the OSS infrastructure tools have you included as part of your development
activities or deployed an application or software project to? “(Select all that apply.)
Source: Enterprise And SMB Software Survey, North America And Europe, Q4 2009
Base: 1900 software decision-makers at North American and European enterprises and SMBs
© 2010 Forrester Research, Inc. Reproduction Prohibited14
8%
13%
13%
50%
73%
41%
76%
70%
20%
71%
21%
44%
Portals or mash-up servers (e.g. Liferay, Dapper)
Business applications (Sugar CRM, Bravo)
Business Intelligence tools (e.g. BIRT, Jasper Reports, Spago)
Content Management Systems (e.g. Alfresco, Drupal)
Application frameworks (e.g. Spring, Rails, Zend)
Application servers (e.g. JBoss, Glassfish, Tomcat)
Web servers or networking components (e.g. Apache, Samba, Radius)
Databases (e.g. MySQL, PostgreSQL, SQLLite)
Development IDEs (e.g. Eclipse, NetBeans)
Operating systems (e.g. Red Hat Linux, Suse, OpenSolaris)
Programming languages (e.g. PHP, Ruby, Python, Java)
Retail/Wholesale
Manufacturing
Business Svc/Const.
Public Sector/Govt.
Fin Svcs.
Media, Ent. Leisure
Utilties/Telco
Telco, MEL & Fin Svc lead industry adoption
“Which of the OSS infrastructure tools have you included as part of your development
activities or deployed an application or software project to? “(Select all that apply.)
Source: Enterprise And SMB Software Survey, North America And Europe, Q4 2009
Base: 1900 software decision-makers at North American and European enterprises and SMBs
© 2010 Forrester Research, Inc. Reproduction Prohibited15
At the operating system level - development
62%
7%
14%
3%
3%
3%
1%
2%
1%
1%
54%
7%
17%
4%
3%
2%
2%
3%
0%
1%
71%
5%
4%
3%
2%
1%
5%
1%
3%
5%
Windows
Mac OSX
Linux-Ubuntu
Linux-Fedora
Linux-SUSE
Linux-Debian
Linux-RHEL
Linux-Other
Solaris
Other
Eclipse - 2009
Eclipse - 2010
Dr. Dobbs
“What is the primary operating system you use for development” (Choose one)
Eclipse 2009 – 26%
Eclipse 2010 – 30%
Dr. Dobbs – 16%
Base: Base: 1481 application development professionals, 1948 application development
professionals, 1298 application development professional
Source: 2009 Eclipse Community Survey, Q2 2009, 2010 Eclipse Community Survey, Q2 2010, Dr. Dobbs Developer
Technographics Q3,2009
© 2010 Forrester Research, Inc. Reproduction Prohibited16
At the operating system level - deployment
37%
3%
11%
3%
5%
6%
9%
5%
5%
9%
36%
2%
11%
3%
4%
8%
8%
5%
3%
7%
57%
1%
4%
2%
2%
1%
9%
5%
4%
11%
Windows
Mac OSX
Linux-Ubuntu
Linux-Fedora
Linux-SUSE
Linux-Debian
Linux-RHEL
Linux-Other
Solaris
Other
Eclipse - 2009
Eclipse - 2010
Dr. Dobbs
“What is the primary operating system you use for deployment (Choose one)
Eclipse - 2009 – 39%
Eclipse – 2010 – 40%
Dr. Dobbs – 23%
Base: Base: 1481 application development professionals, 1948 application development
professionals, 1298 application development professional
Source: 2009 Eclipse Community Survey, Q2 2009, 2010 Eclipse Community Survey, Q2 2010, Dr. Dobbs Developer
Technographics Q3,2009
© 2010 Forrester Research, Inc. Reproduction Prohibited17
*Source: 2010 Eclipse Community Survey
† Source: Forrester -Dr. Dobbs 2009 Developer Technographics, Q3 2009
At the app server level
30%
9%
5%
3%
3%
3%
1%
1%
27%
3%
5%
29%
20%
14%
14%
4%
2%
3%
1%
7%
0%
6%
Apache Tomcat
Red Hat JBoss
IBM Websphere
Oracle Weblogic
Sun Glassfish
Jetty
Oracle AS
SAP Netweaver
None
Don't Know
Other
Eclipse*
Dr. Dobbs†
“What is the primary app server you typically use for deployed applications?” (Choose one.)
*Base: 1948 app dev professionals building server apps and programming in Java
† Base: 218 app dev professionals building server apps and programming in Java at least 50% of the time
© 2010 Forrester Research, Inc. Reproduction Prohibited18
At the DBMS level
28%
19%
7%
10%
4%
1%
13%
6%
15%
17%
32%
5%
4%
2%
18%
7%
MySQL
Oracle
SQLServer
PostgreSQL
DB2
Sybase
None
Other
Eclipse*
Dr. Dobbs†
“What is the primary app server you typically use for deployed applications?” (Choose one.)
*Source: 2010 Eclipse Community Survey
† Source: Forrester -Dr. Dobbs 2009 Developer Technographics, Q3 2009
*Base: 1948 app dev professionals building server apps and programming in Java
† Base: 218 app dev professionals building server apps and programming in Java at least 50% of the time
© 2010 Forrester Research, Inc. Reproduction Prohibited19
*Base: 1,948 application development professionals
† Base: 1,298 application development professionals
At the SCM level
52%
11%
1%
3%
1%
3%
6%
3%
5%
1%
3%
32%
13%
12%
6%
9%
5%
2%
1%
8%
0%
13%
Subversion
CVS
Microsoft VSS
IBM Rational ClearCase
Microsoft TFS
Perforce
Git/GitHub
Mercurial
None
Don't Know
Other
Eclipse*
Dr. Dobbs†
“What is the primary SCM system you typically use?” (Choose one.)
*Source: 2009 Eclipse Community Survey
† Source: Forrester -Dr. Dobb‟s 2009 Developer Technographics, Q3 2009
© 2010 Forrester Research, Inc. Reproduction Prohibited20
Why developers drive adoption of OSS
Source: February 2, 2009, “Best Practices: Improve Development Effectiveness Through Strategic Adoption Of Open Source”
Forrester report
The software “iron triangle”
1. Cost
2. Speed
3. Integration
© 2010 Forrester Research, Inc. Reproduction Prohibited21
Why developers drive adoption of OSS
Source: February 2, 2009, “Best Practices: Improve Development Effectiveness Through Strategic Adoption Of Open Source”
Forrester report
The software “iron triangle”
1. Cost
2. Speed
3. Integration
© 2010 Forrester Research, Inc. Reproduction Prohibited22
Adoption paths are shifting toward devs
Developers resist technology that
doesn‟t meet their needs
Traditional financial controls are of
limited value
LOBs defends dev teams that
produce value
Management is willing to yield when
a “win-win” results
The path from developer to customer
is getting shorter
Developer productivity is no longer
the problem
CIO
ADEA I&OPMO
BAs QAArch Dev
More than ever: Developers can
block – or significantly aid the
adoption of software!
© 2010 Forrester Research, Inc. Reproduction Prohibited23
But executives are starting
to reassert control!
A shift in decision makers affects adoption
Time
RevenueIBM, Oracle, SAP, and HP
Red Hat, Atlassian,
SpringSource, and Adobe
Skimmers
PenetratorsMicrosoft
Executives
Developers
Who‟s in
charge?
© 2010 Forrester Research, Inc. Reproduction Prohibited24
What it means
Economic uncertainty and cost reduction was driving OSS adoption, now
were shifting toward speed and innovation
Continental Europeans lead in adoption but the US is catching up
Very small and very large organizations have “more time than money”
As OSS business models evolve, concerns over viability and IP are
receding
Developer adoption of Linux continues to trend upward, and a focus on UX
pays off
The enterprise software market and the OSS market are set for an
inevitable collision, and then a convergence
To win you must drive adoption and affirmation through developers, and
purchases through management
© 2010 Forrester Research, Inc. Reproduction Prohibited25
Market Trends And OSS Evolution
Examples And Best Practices
Agenda
© 2010 Forrester Research, Inc. Reproduction Prohibited26
Case Study 1: Leading fin services firm
Provides asset management, investment banking, private
banking, treasury and securities, and commercial banking.
Using Linux since 2001, OSS now moving up the stack.
Using OSS app server keeps big ISVs honest.
Need good developers – not average ones.
Used OSS as an opportunity to refashion dev processes.
Now looking at BI, using a lot of Spring
Need to manage the support „fear factor‟
Dev community very happy, I&O is mixed
Result: Per project software costs savings range from 30%
to 80%. For every $10 they put in, they get $4 back.
© 2010 Forrester Research, Inc. Reproduction Prohibited27
Case Study 2: Global fin services firm
Market leader in securities, asset management and credit
services
Replacing proprietary ECM with Alfresco – in production since 2007
Extensibility and flexibility due to OSS and APIs
“First major firm to do Linux as a strategic play”
Still evaluate commercial option – but license costs are increasingly
untenable
Have created their own fork, and work with Alfresco to manage
change delivery back to trunk
Result: Reduced upgrade costs compared to commercial
solutions by high 6 figures
© 2010 Forrester Research, Inc. Reproduction Prohibited28
Case Study 3: Sabre
Leading provider of technology and marketing services for
the travel industry
Rewrote z-based reservation app using scale-out arch with Linux,
Tomcat, FUSE – now running at peak volumes on 6 M tpd
Need to make investments in architectural competency
Acquisition policy “OS first, buy next, build last”.
Took three tries to get it right
Anticipate higher costs of training, risks of proliferation
Always buy support
Result: Saving millions over multiple years vs. mainframe
environment.
© 2010 Forrester Research, Inc. Reproduction Prohibited29
Case Study 4: Landmark Graphics
Landmark Graphics supplies software to Oil and Gas
industry across a broad variety of application areas
OSS Steward monitors policy compliance
Prioritize standardization
Restructured release process
– Use BDS to monitor compliance
– PM assumes responsibility for OSS
– Remediate if/as violations are found
Contributing back in limited cases
Result: Rapid adoption of the latest models and technologies,
with accurate identification of OSS dependencies
© 2010 Forrester Research, Inc. Reproduction Prohibited30
Case Study 5: Reliant Security
Reliant sells PCI compliant in-store systems that include
many OSS subsystems.
Set a clear policy for OSS use
Tuned acquisition policies
– OSS first mandate
– Prioritized “ilities”
– Loosely coupled design
Adjusted dev processes
– OSS use identified at design
– Developer on the hook for provenance
Result: Significant customer savings (7 figures in large
deployments) over commercial alternatives
© 2010 Forrester Research, Inc. Reproduction Prohibited31
Case Study 6: European Airline
Global carrier created by merger of two large carriers
Set a goal for tactical cost savings in operations
Replaced Solaris and AIX as base OS supporting SAP
Rolled out a combined SAP implementation on Linux
Result: Savings of close to E 1 million.
© 2010 Forrester Research, Inc. Reproduction Prohibited32
10 Best Practices From Enterprise OSS Adopters
1. Appoint an OSS steward
2. Create a comprehensible policy
3. Frontload acquisition processes
4. Require project leaders to identify OSS dependencies
5. Use EA to regulate exploitation and maintenance
6. Trust teams - but verify with code-scanning utilities
7. Maintain a repository of preapproved OSS components
8. Don't dwell on processes and artifacts; focus on outcomes
9. Don't expect perfection, and plan for remediation
10. Set a contribution policy – it will happen over time anyway
© 2010 Forrester Research, Inc. Reproduction Prohibited33 © 2009 Forrester Research, Inc. Reproduction Prohibited
Companies may be slow to contribute, but not devs…
Source: Forrester -Dr. Dobb‟s 2009 Developer Technographics Survey, Q3 2009
Base: 914 development pros at North American and European enterprises and SMBs
“Please select any statements that describe you as a
developer.” (Select all that apply.)
60%
39%
21%
20%
15%
12%
5%
Engaged in developing on 1 or more side projects outside business
Attend local user group meetings
Published 1 or more articles about software dev
Contributor to 1 or more OSS projects
Hold 1 or more patents
Recognized by 1 or more ISV programs
Written 1 or more books
© 2010 Forrester Research, Inc. Reproduction Prohibited34
What it means – Looking ahead
Open source adoption is moving from “don‟t ask, don‟t tell” into a period of
strategic adoption
Business critical OSS systems exist; they work and they scale
IT is still learning about different OSS business models – help them
understand
It‟s time to focus on the secondary virtues of OSS, speed, flexibility and
engagement
Most companies aren‟t averse to paying for value fairly delivered
Contributions from IT developers are coming, but not through official
channels – need to work to assuage their concerns over risk
It‟s not really about the Cathedral OR the Bazaar – it‟s more about a mixed
source combination
© 2009 Forrester Research, Inc. Reproduction Prohibited
Thank you
Jeffrey Hammond
+1 978.226.8886
Twitter: jhammond
www.forrester.com