Topics for this podcast:
*Hadoop v1.0 and year ahead
*Oracle-Cloudera deal for more Hadoop
*Oracle’s ‘Sun spot’ with Solaris
*Open Source M&A outlook for 2012
*Our new MySQL/NoSQL/NewSQL survey
iTunes or direct download (28:49, 4.9MB)
Previously, I’ve called out years for non-desktop Linux in 2008, Linux in both the low and high-ends of the market in 2009, ‘hidden’ Linux in 2010 and last year, cloud computing in 2011. For 2012, I see continued growth, prevalence, innovation and impact from Linux, thus leading to a 2012 that is dominated by Linux.
I expect to see nothing but continued strength for Linux and open source in cloud computing in 2012. The cloud continues to be the biggest disruptor and opportunity for Linux providers. 2012 got off to an interesting start with Microsoft’s efforts to support for Linux on Azure, which highlights just how pervasive Linux has become in cloud computing. As detail in our special report on The Changing Linux Landscape, we also expect Linux to continue to be the basis for most offerings in IaaS and particularly PaaS, which is burgeoning across open source languages and frameworks as well as verticals and enterprise customers. Its popularity among enterprise and other developers will also bolster Linux and open source software in 2012.
We can certainly expect to see Linux continue its domination in supercomputing and the Top 500 Supercomputer List, where Linux continues to grow its share above 90% while others, such as Microsoft, Apple and BSD, fall off of the list.
I also expect Linux will grow its presence and impact on the wider, more mainstream server market, where Red Hat and SUSE continue to benefit from Unix migration, particularly from Solaris. Our analysis with survey data from 451 Research division TheInfoPro shows server spending for databases and data warehousing favoring Red Hat with Linux over Oracle with either Linux or Solaris. Out of more than 165 server professionals interviewed by TIP, 67% are planning to spend more with Red Hat on database/data-warehousing, and only 6% plan to spend less. The positive figures for Red Hat mirror negative spending intentions for Oracle, with 55% planning to spend less and only 9% planning to spend more. Spending continues to decline strongly for all of the primary Unix providers in the study, which in addition to Oracle includes IBM and Hewlett-Packard.
We may also see further expansion for Red Hat, which may be eyeing key acquisitions, and other Linux and open source vendors as they continue building their channels and wade more into midmarket and SMB customers.
In smartphones and mobile software, I also expect Linux will do quite well in 2012 with continued Android strength, diminished FUD and possibly an open source boost from a newly-open sourced WebOS. We also see Ubuntu arriving on the mobile and converged device scene, including ‘concept’ appearance at CES.
We’re also likely to see Linux in automobiles, health care and other electronics even more in 2012, though you may never hear Linux or open source. Don’t be fooled though, Linux is expanding its already impressive, wide presence and 2012 looks to be another year of significant gains.
There has been no shortage of reaction to HP’s move to make the Linux-based WebOS open source software. Below, I offer some of my thoughts on the meaning for the different players affected.
*What’s it mean for WebOS?
Moving WebOS to open source was best option for HP. It retains some value in the software depending on its involvement. It is also the best fate for the code, rather then being sold or simmered to its IP and patent value or even used as another weapon in the ongoing mobile software patent wars. Still, the move comes amid huge developer and consumer uncertainty for WebOS. Nevertheless, at least WebOS was already in the market with a compelling products, the Palm the Pre, in the modern smartphone market. WebOS will hopefully have a faster path to open source than Symbian since the former is based on Linux. I still think the greatest opportunity for WebOS may be in serving as an open alternative in the market, particularly after Android has proven to handset makers, wireless carriers, OEMs and others that a Linux-based, open source mobile OS can succeed in the market and provide profit for multiple parties. Furthering this opportunity, WebOS may be even more attractive to these key vendors, channel players and other stakeholders who are tired of the IP and patent stress and expense around Android. Of course, Android was not under patent or IP attack until it was successful in the market and the same may be the case for WebOS, though we think its IP roots and history in touch and smartphone technology are less complex in terms of origin and ownership.
*What’s it mean for competitors?
For Apple, an open source WebOS means more market pressure and open pressure, more competition for developers and a real danger WebOS hooks into the Android ecosystem. WebOS may also be harder to attack from a patent and IP standpoint since it is older and more singular in ownership (Palm and now HP). Other factors include HP’s own formidable patent portfolio and the perception of Apple as a patent aggressor, which would be reinforced if it attacked WebOS the way it has gone after Android.
For Android, it may finally get a dose of its own open medicine, feeling the pressure of another Linux-based, open source mobile OS that is familiar to many developers, compatible with newer smartphone technologies and appealing to handset makers and other key OEMs. However, WebOS is also a validation of Android, which paved the path for mobile Linux and open source to finally break through beyond geeks to reach a mass consumer audience.
As for other proprietary players such as Microsoft and RIM, another open source rival is bad news. It presents another open source option and potentially serious competition on developers, applications, devices, carriers and consumers. An open source WebOS may also make Android, in effect, more open with faster, easier access to code for both Android and WebOS compete. This could make it even harder for these older, proprietary players to get developer or consumer mind share that is already slipping.
*What’s it mean for open source? Really, there is no downside for open source except that it will be viewed as a form of software cemetery if WebOS is not developed or delivered to market. HP’s WebOS move does give open source greater prominence in mobile software. Again, it is a validation of Android, which is Linux-based and open source, and shows that we haven’t seen the last of mobile Linux and open source software in Android.
Topics for this podcast:
*Continuent extends MySQL replication to Oracle Database
*CFEngine updates server automation software
*Devops moving mainstream
*Neo Technology integrates with Spring
*451 CAOS report from Hadoop World
iTunes or direct download (26:56, 4.6MB)
Topics for this podcast:
*Opscode Chef extends to Windows for more enterprise devops
*Black Duck continues growth, gains new funding
*Cloudant expands NoSQL database focus, customers
*New open source Web server and vendor Nginx arrives
*The downside of Microsoft’s Android dollars
iTunes or direct download (27:35, 4.7MB)
Topics for this podcast:
*Cloud M&A potential around OpenStack
*Oracle’s commercial extensions for MySQL
*Puppet Labs rolls out Enterprise 2.0, hosts PuppetConf
*Basho bolsters Riak distributed data store in NoSQL race
*Our latest special CAOS report, ‘The Changing Linux Landscape’
iTunes or direct download (25:59, 4.4MB)
Engine Yard acquires Orchestra. Red Hat considers NoSQL move. And more.
# Engine Yard announced a definitive agreement to acquire Orchestra, bringing PHP expertise to the Engine Yard platform.
# Red Hat’s CEO indicated the company is interested in a NoSQL or Hadoop acquisition.
# Gluster announced Apache Hadoop compatibility in the next GlusterFS release.
# Microsoft signed an agreement with China Standard Software Co (CS2C) to support CS2C NeoKylin Linux Server running on Microsoft’s Hyper-V.
# Mitchell Baker kicked off a discussion on Mozilla’s future.
# Hortonworks announced that the next generation of Apache Hadoop MapReduce has been merged to the Apache Hadoop mainline
# Rapid7 offered a $100,000 investment fund to support the development of up to seven promising open source projects in the security industry.
# ReadWrite Enterprise examined the changing Linux landscape, covering Jay Lyman’s recent LinuxCon session.
# CloudBees announced a partnership with MongoHQ, a provider of cloud-based MongoDB data hosting and services.
# Twitter announced Bootstrap, a front-end toolkit for rapidly developing web applications.
# The UK Cabinet Office began building its open source strategy on proprietary software.
I wrote last year about the way Google’s Android mobile operating system was serving as a more open alternative to Apple’s iOS, but not so open that it didn’t leave opportunity for an even more open alternative.
Given that we continue to see software patent-based attacks on Android, as well as swirling FUD around coverage of the attacks and never ending suits and settlements and courtroom developments, it is clear it will be a long time before any of this legal business is ever close to settled, unless ended by settlements first, which is likely.
However, I’m more interested in the technology in the meantime. I also think it’s interesting to see, if not a ‘more open’ alternative emerging, at least another, ’somewhat open’ option in the tablet market, this being HP’s WebOS. It’s interesting that WebOS evolved from Palm, which HP acquired in March 2010 for $1.4 billion. Though Apple’s iPad is still the clear leader in tablets, it is interesting to see continuing signs that what happened in smartphones (where iPhone led and Android quickly caught up and then passed iOS) may be happening in tablets. There is also still the possibility that tablets may play out like netbooks, with wild popularity followed by a fade in favor of more traditional PCs for traditional PC needs. It is interesting to note that Google’s Eric Schmidt recently commented on the continued utility of PCs, which will remain key to professionals, consumers, and also developers, largely because of the tactical keyboard. What is most likely is continued convergence, and it will be interesting to see what ties emerge between WebOS and PCs as computer hardware giant HP rolls out the OS in tablets and smartphones.
We also see other signs that new, open entrants may be mixing things up in the mobile and converged device market, such as word of a possible Android and iOS competitor from Mozilla. There is yet another project that is already a factor in netbooks, other mobile devices and the burgeoning IT market of automobile information and entertainment systems, MeeGo, which is also open source. Even Research in Motion’s Playbook is based on the QNX operating system, for which source code was made available by its previous owner to make it more like the open source Linux OS, which was attracting developers and interesting customers.
We believed there was a fairly prominent place for open source software, open source operating systems and general openness in mobile software when we wrote our report, Mobility Matters three years ago, but we would have never guessed that the openness of this software would be so significant in two respects: defense from patent and other intellectual property attacks; the market power of open source, which draws in not only developers, but manufacturers and other third-parties. We’ve seen the speed and strength at which a project and community such as Android can grow. Will we now begin to see other alternatives that are even more open emerge as top choices among developers, hardware companies, wireless players and consumers? Never before have those alternatives really existed in the mobile software world, so it’s good at least to see the possibility is there.
The Apache Software Foundation’s latest statement on the Java Community Process highlights continued dissatisfaction and dissent from Oracle’s stewardship and involvement in open source software.
This comes after some ups and downs for Oracle and its oversight of Java and other open source software that was previously under the auspices of Sun Microsystems. Oracle started off on a rough path when it sued Google over its implementation of Java in Android without preemptively or clearly stating that it was not attacking open source. At about the same time, it let OpenSolaris die a slow, somewhat confusing death. Oracle won a point when IBM came out with its support in favor of the JCP and OpenJDK over Apache Harmony, and this contributes to the adversarial positioning between Oracle and the Apache Software Foundation. However, Oracle has also seen an erosion of open source support and confidence as OpenOffice.org developers have migrated away from Oracle, many to contribute to the new Libre Office project.
Oracle’s moves illustrate the company’s lack of complete understanding of open source and the value of open source software communities. While it appreciates and leverages open source as an effective, efficient software development approach, it does not truly see the value of providing software to a community and attaining benefits of efficiency, reach and innovation as a result. This is not to say that supporting an open source software community will automatically translate into commercial and community success (not the case with Symbian, for example), but Oracle does not appear to support community as a priority in its proprietary and admittedly successful software strategy.
MySQL can be an example of Oracle doing things right with open source, though we may see similar dissatisfaction and defection as Oracle moves further toward commercialization and further away from free, community software. Still, Oracle at least showed it could continue and contribute and support a successful open source project in the case of MySQL. The same may not be said for OpenSolaris, OpenOffice.org or, increasingly it appears, Java.
We continue to see more evidence of the themes we discuss in our latest CAOS special report, Seeding the Clouds, which examines the open source software used in cloud computing, the vendors backing open source, the cloud providers using it and the impact on the industry.
First, as usual, we are seeing consistencies between our own research — which indicates open source is a huge part of today’s cloud computing offerings from major providers like Amazon, Google, Rackspace, Terremark and VMware — and that of code analysis and management vendor Black Duck. In its analysis of code that runs the cloud, Black Duck also found a preponderance of open source pieces, in many cases the same projects we profile in our report.
Indeed, open source software is an important part of the infrastructure, data and application layers of today’s cloud computing stacks with significant use of Linux, open source hypervisors KVM and Xen, open source data technologies such as MySQL, PostgreSQL, Hadoop, NoSQL and memcached and open source languages such as Java, PHP, Python and Ruby on Rails.
There will be plenty of users and customers content to use non-open source options that serve as the defacto standards, but we do see a move to higher-level, production and mission critical use, which represents continued commercial opportunity for open source and other vendors.
One of the more subtle effects of all this open source in the cloud, as covered in Seeding the Clouds, is the impact on discussions, debates and downright fights in the market. There is much scrutiny on claims of being open, technical aspects of open and what ‘open cloud’ means. A prime example is the Twisticuffs that have gone on between Simon Crosby of XenSource and Citrix, discussing OpenCloud and the response from Open Cloud Initiative co-founder Sam Johnston, who claims this is misuse of the open label.
We already saw open source playing a role in the discussions and debates about open clouds, open APIs and open data, and this latest confrontation is evidence that role continues to be significant. We still wonder though about the question of open enough as we contemplate openness in the clouds.