SET SOAPBOX = ON

Summer Camp for Cloud Enthusiasts...how do you sit around a campfire in the cloud?

Hello Mudda, hello Fadda, here I am at CloudCamp Boston.  Camp is very entertaining and they say we'll have some fun if it stops raining.  (apologies to the late Alan Sherman)

Well, I just got back from CloudCamp Boston and it did rain on my way home.  Coincidence or not?  I stumbled on CloudCamp the other week while attending a Webinar that featured CloudCamp creator Dave Nielsen.  And since I'm interested in everything having to do with cloud computing, I jumped over to the CloudCamp Website (www.cloudcamp.com) and registered for their unconference in Boston along with 300 other CloudCampers.  Technically, CloudCamp Boston was in Cambridge at Microsoft's New England R&D Center, which is also the home of the Microsoft - Novell Interoperability Lab.

I was unfamiliar with the concept of an unconference so I consulted YouTube to see how Dave Nielsen described it.  I also viewed some previously recorded CloudCamp unconference sessions from CloudCamp Atlanta.  So with my on-line introduction to CloudCamp under my belt, I schlepped over to Cambridge on one of the hottest and muggiest days of the summer.  I arrived at Microsoft's New England R&D Center a few minutes before CloudCamp kicked off.  Just to be clear, CloudCamp Boston was not a Microsoft event although the company is a sponsor of CloudCamp and had representatives in attendance.

After welcoming the CloudCampers and making some acknowledgments,  John Treadway of CloudBzz and Judith Hurwitz, author of Cloud Computing for Dummies (October 2009) delivered an intro to cloud computing just to be sure we all knew why we were there.  Then each CloudCamp  Boston sponsor was allocated a 5-minute presentation, and I do mean 5-minute, to explain their angle on cloud computing.  Next, Dave and John outlined how an unconference works, including the self-selection of an unpanel to answer questions generated by the CloudCampers.

In an unconference you create the conference agenda, quickly, by proposing session topics, then you see who is interested by a show of hands.  The most popular cloud-related topics are then assigned meeting spaces and the topic proposers take nominal leadership roles in getting the discussions going. After several hours of attending unconference sessions, the CloudCampers reassemble in a large group and someone from each session provides highlights of their discussion.

CloudCamp Boston was provisioned with excellent food and drink to accompany the social networking (schmoozing) opportunities.  The time went by all too quickly and six hours after it began, CloudCamp Boston came to a close.  But don't worry if you missed it, there is already talk of bringing CloudCamp back to Boston in the fall, which seemed like a world away while walking back to the Kendall-MIT MBTA station in the subtropical night breeze.

My take-away from CloudCamp Boston was there are a lot of people excited about cloud computing and how it will reshape the IT landscape by turning it into a cloudscape over the next 10 years.  There was a palpable sense that the shift to cloud computing is already underway and that it is gaining serious momentum.  Hats off to the CloudCamp Boston organizers, to the sponsoring organizations and to Microsoft for hosting the event.

SET SOAPBOX = OFF

We've entered a long term economic "unwinding" but does IT get it?

Our economy is deep into its second year of recession, which is a prelude to a more long term "unwinding" of economic activity.  PC and server sales are down.  IT salaries are stagnant and budgets are being reduced.  Virtually every larger than life IT vendor (AMD, Cisco, Google, HP, IBM and Microsoft) has shed employees and profits as well.  Small and medium-sized businesses are locked out of credit markets.  And it ain't close to being over.  Commercial real estate mortgage defaults are on the rise and consumer credit card defaults are just around the corner.  In an economy that has accumulated mind-boggling private debt there is no economic recovery that is ever going to generate enough future wealth to pay it all back.

When the talking heads start bloviating about economic recovery, my response is recover to what?  A more of the same economic recovery isn't going to happen because the cheap energy and credit fueled bubble economy is over.  The best we can hope for is a gradual economic retreat or unwinding of the economy to a lower energy and material consumptive way of life. The population and resource constraints of living on a finite sphere are going to demand it.  But will IT get it?  Greening IT is a good starting place, but it is not the end of the journey.  IT needs to prepare for a decade of economic decline.

Since the future will not look like the past, what should IT be doing besides getting lean and mean?  Basically, IT needs to disappear into the business.  The era of putting users on hold and taking months to deploy applications is over.  No business can afford to wait for IT to do its job in the manner to which it has been accustomed.

The disappearing IT function has a number of drivers.  Vastly reduced credit and borrowing will limit the ability of small and medium-sized businesses to run up-to-date computer rooms stocked with state-of-the-art servers and expensive to license proprietary software.  Small businesses will be financing their operations out of their accounts receivable and barter arrangements.  Rising energy costs will be back.  Yes, the price of energy is going to go up again with all that it entails, including higher energy costs for computer rooms, higher energy costs to light, heat and cool buildings.  Increases in state and local taxes are coming.  States and towns are not permitted to run budget deficits so taxes will increase in the short term. Eventually government at all levels will unwind due to an inability to raise tax revenues from businesses and citizens who are going to be a lot less wealthy.  

The new invisible IT will be born out of business necessity to not only do more with less but to do it faster with reduced capital costs and without creating new IT fiefdoms.  Well, if IT disappears into the business just where does it go?  The answer is to go outside during day. Look towards the sky and tell me what you are likely to see.  Did you see clouds?  If you did, then you have the answer.

It may be a coincidence that Amazon, Google, IBM, HP, Sun (Oracle) and a host of newcomers have all launched their cloud computing offerings within the past year just as our economy entered a long term recession because IT will be transformed within the infrastructure, applications and services of cloud computing environments.

The pay for what you use model of cloud computing combined with its rapidly expanding menu of computing services on demand is going to be irresistibly attractive to small and medium-sized businesses.  Businesses relocating their IT services to the cloud will need to install adequate Internet bandwidth with provisions for fail-over to another Internet Service Provider, but they can look forward to exchanging their costly computer rooms for cheaper wiring closets in which to locate their firewalls and routers.

Predictably, those with the most to lose when IT makes its move to the cloud will spread the usual fear, uncertainty and doubt (FUD) that cloud computing is not secure.  But if the truth be told, cloud computing is probably more secure than most computer rooms in small and medium-sized businesses.  But don't take my word for it.  Inquire with your prospective cloud computing providers with regard to their security practices and the audits they have submitted to in order to get their certifications for best practices and security.

The other bug-a-boo of cloud computing is the potential for service outages, which is not a good thing if your IT infrastructure, applications and data are located in the cloud.  In the past there have been temporary service outages at Amazon and several Software-as-a-Service (SaaS) providers.  So if the prospect of any service outage is unacceptable then you will need to work a bit smarter and spend more money to make sure your cloud computing environment is less subject to cloud provider glitches or a backhoe operator digging a hole in the wrong place.

OK, if IT disappears into the cloud, what will everyone in IT do?  Well, IT staff will still need to manage local routers and firewalls and premises-based end nodes like PCs, laptops, netbooks, thin clients and printers.  And they will need to manage the life cycle of their cloud-based servers although a fair portion of this work can and will be automated.  Developers may need to tailor existing applications for cloud computing.  Managers will need to oversee what is being done in the cloud and determine how well IT operations are meshing with the objectives of the business and the needs of customers who will have more to say about business processes and service than ever before.

The jump to cloud computing may not happen overnight for most small and medium-sized businesses.  However, IT staffers, developers and managers should be quick to embrace cloud computing or start thinking about a new line of work because it will happen with or without them. Resistance is futile...you will be assimilated.

For those of you interested in building your own cloud or a hybrid private-public cloud that is API compatible with Amazon, then take a look at the Eucalyptus open source project, which itself has just garnered $5.5M in funding to launch a company based on the software.  

SET SOAPBOX = OFF

 

 

Cloud Computing and the End of Capitalism as we know IT

Major shifts in computing can be influenced by economic events.  Take the collapse of financial capitalism last September.  The credit lockup that ensued doesn't look like it will end soon.  The high-flying Wall Street investment banks that survived the crash have had their wings clipped.  Risky banking and greedy investment practices have destroyed trust in the global financial system.  The federal government is attempting to "defilibrate" an economy that appears increasingly lifeless.  So what does an economy facing the material and energy limits to growth have to do with a shift in computing?  The answer is a lot.

Past computing shifts included the change from batch processing to interactive computing (1970s), from mainframe and mini computing to microprocessor computing (1980s), from proprietary networking to standards-based networking (late 1980s), from stand-alone computing to client-server computing (early 1990s), from private networks to the Internet and the World Wide Web (mid 1990s).  Each of these shifts changed the way we did computing.

The current computing shift is to virtualization of computing workloads, data storage and applications in cloud computing infrastructures.  The new computing shift is also unstoppable.  Why?  Because in a economic decline that is going to re-define the way we live and work on an energy and material basis, the capital, maintenance, operation and real estate costs of premises-based computing workloads, data storage and applications will eventually become too expensive for most businesses and organizations.

When will the shift to cloud computing architectures dominate the IT landscape?  A reasonable answer is that it will take the better part of a decade from where we are today.  Amazon, Google, IBM, Microsoft and others are already or will soon be offering cloud services for computing workloads, data storage and applications.  The "pay for what you use" model seems to be established in cloud computing, and that could change over time.

Cross-cloud compatibility will become essential.  It is the "holy grail" of cloud computing.  And towards that end an IBM research center in San Jose recently moved a computing workload around the Earth from one IBM cloud center to another and then from IBM's cloud to a private cloud.  IBM may have been the first to demonstrate this particular cloud computing capability and it is an important one.

The financial and technical advantages of cloud computing will look attractive to businesses and organizations faced with cutting their premises-based IT costs.  Paying for what you use in the cloud does not require large capital investments, which could be important in an economy with reduced credit availability and borrowing.  The size of corporate and branch offices can be reduced as fewer people will need to commute to them in order to do their jobs.  Technical advantages like the automated movement of computing workloads and data could help avoid the consequences of local natural disasters that frequently cause businesses to fail.  Development projects can be migrated from cloud to cloud depending on where the developers are located.  Data duplication to different geographically located cloud computing centers will offer even the smallest businesses a high level of data protection and redundancy.

It is interesting to note that a few years ago many people anticipated the arrival of what became known as Software as a Service (SaaS) and today you can subscribe to any number of SaaS products from vendors providing office suites, sales and customer management systems, financial accounting systems, voice and messaging systems all delivered as a service for which the customer pays by the month or annually.  Cloud computing leverages the development of SaaS and creates a "mashup" of sorts with the virtualization of complete IT infrastructures running across a computing grid that can span the entire Earth.

Cloud computing will function well in a world economy more bound than ever before by material and energy constraints.  If you are not paying attention you need to begin actively investigating how you can shift some or all of your IT infrastructure into the cloud.

SET SOAPBOX = OFF

Year End Grab Bag...Open Source, Microsoft, Cloud Computing and SCO!

There were a number of new Linux distribution releases to help ring out the old year. Fedora 10, openSUSE 11.1 and Ubuntu 8.10 all arrived in time to celebrate the holiday season. Note that Ubuntu 8.10 is not a "Long Term Support" or LTS release, so brace yourselves for another Ubuntu release around April 2009.  Ubuntu users who need LTS should stick with Ubuntu 8.04.   Fedora and openSUSE users are not offered  long term support releases.  However, most personal users or enthusiasts don't usually mind being on the "bleeding edge" of a new release.  Note that Novell, which sponsors openSUSE, does offer long term support for subscribers to SUSE Linux Enterprise Server 10 and Desktop 10.  Look for Novell to release SUSE Linux Enterprise Server 11 and Desktop 11 around the middle of 2009.

Windows Vista, which has been around for almost two years, seems to have become the "Rodney Dangerfield" of operating systems because it can't get any respect.  Businesses have largely avoided it and most IT professionals seem to be either bored or annoyed with it.  Only Microsoft fans and the buyers of home computers have embraced it.  The former because there is no end to their love of Windows and the latter because it came already installed on their computers so they are stuck with it unless they want to remove it and install Linux.

While Microsoft is currently preparing Vista SP2 for release in 2009, Microsoft business customers running Windows XP are holding out for Windows 7, which should be ready in 2010.  Although Windows 7 is built on the Vista code base, it will be modified in ways to make it more attractive to business and professional users.  The question is just how much pent up demand will there be for Windows 7 by the time 2010 rolls along?  With Microsoft Windows now running on less than 90% of PCs, could we be witnessing the beginning of the end of the Windows era?  Will Windows 7 be too late to stop the decline of the Windows monopoly?

2009 could be the year of the cloud because a critical mass of vendors, developers, consultants and customers are starting to turn to cloud computing to deliver on demand computing, storage, applications and development services.  Cloud computing heavy weights like Amazon and Google are revving up their application and storage engines, while Ray Ozzie and company prepare Azure, Microsoft's cloud computing platform, for release sometime in 2009.  And with a second tier of cloud computing players and tool makers getting busy, you can bet that things are going to get more interesting in 2009, so pay attention to what is going on in the cloud.

Why cloud computing now?  After all, some aspects of cloud computing, like SaaS (Software as a Service) have been around for years.  Well, with the economy in a severe recession, businesses are looking for ways to reduce IT operations costs and IT capital costs.  So what does this have to do with cloud computing?  It is going to be 10% to 90% cheaper for business to move their "premises" applications to the cloud.  And by moving their "premises" servers and storage devices into the cloud their maintenance and operating costs will be reduced along with their IT capital costs for replacing those "premises" servers and storage devices.

Who will be putting all those applications and servers into the cloud?  The smart guys are saying small and medium sized will be the initial adopters.  They tend to run less complicated IT infrastructures and their applications tend to be more standardized.  IT is headed into the cloud and 2009 will be a good year for a lot of small businesses to figure out how it can use the cloud to deliver IT services while reducing IT costs.  And a more prolonged economic recession could hasten business IT migrations to cloud computing infrastructures.

Finally, will we ever see the end of the long-running SCO saga?  In July 2008, the U.S. District Court in Utah handed down its "Findings of Fact and Conclusions of Law and Order" which upheld most of the August 2007 judgment that declared Novell the copyright owner of Unix and UnixWare.  In November 2008, the court rendered its Final Judgment.  Case Closed.  But an appeal by SCO, which is currently in bankruptcy, could delay the final outcome.  Who knows when it will really be over but 2009 could be a good bet.

SET SOAPBOX = OFF

Network Security Also Means Watching Your Own Back...

This past July the City of San Francisco's chief network engineer was arrested and charged with multiple counts of computer tampering after refusing to divulge the passwords to critical routers on the city's wide area network.  (He subsequently did reveal the passwords to the Mayor of San Francisco.)  There was no destruction of city data nor was there any attempt to extort money from the city.  The network engineer in question is still being held in jail with bond set at $5M.  The trial has yet to begin and several motions for bail reduction have been denied.

There is a lot we don't know yet about this episode but the obvious question is what were the city's IT managers thinking when they allowed one network engineer to claim ownership to the "security keys" of the city's network?

Even if this network engineer was a Cisco Certified Internet Expert (CCIE), which is the "gold standard" in Cisco certifications, it stands to reason that you do not let one person gain exclusive control to important network devices.

The passwords and configurations of the city's routers are not the personal property of any network engineer no matter what they think they are doing.  And it was the City of San Francisco's lack of appropriate IT policies and procedures that contributed to the creation of this security problem.

So, what can be done to prevent a trusted network engineer from going rogue?  Well, regular IT management communication with the network engineer would have been a good start.  A network engineer working in isolation with little management contact could easily imagine that he is the only one smart enough to do the job right.  And having policies and procedures regarding securing network device passwords and proper storage of network device configurations would have been a good thing to have in place.

In the vacuum created by the lack of of such policies and procedures, this network engineer decided that he knew best how to protect his network from those who he likely deemed to be less competent or potentially destructive.

Once again, IT management failed.  Most network engineers take pride in their work, but when pride leads to jealous guarding of information and mistrust of IT management you wind up with a serious personnel and security problem.

Whatever this particular network engineer feared may have been completely justified in his own mind, but the fact that he was able to unilaterally take action to secure his network the way he wanted to is an indictment of the IT management of the City of San Francisco.

Should this network engineer be punished?  Only if he is found guilt of the charges filed against him, but IT management is equally culpable for allowing this security abuse to exist in the first place.  Heads need to roll in order to establish that there was incompetence in the IT management of the city's network.  The city's IT administration also needs to demonstrate to other network engineers that appropriate IT policies and procedures are going to be designed and implemented to prevent further abuse.

The City of San Francisco, which is short drive from Silicon Valley, was publicly embarrassed by the news this event generated and may have to spend up to $1M to rectify the security problems exposed by the actions of this network engineer.  It is also a good story of what not to do with talented and trusted employees.

Do you have IT policies and procedures in place to prevent a similar situation from happening in your organization?

For those interested in the outcome of the trial and additional news related to this security lapse, read the blog posts of my former co-worker, Paul Venezia, which are published at http://weblog.infoworld.com/venezia.

SET SOAPBOX = OFF

Microsoft and Novell make summertime news...

It has been a pretty slow summer for open-source news, but Microsoft and Novell have been in the news recently and what they are doing does affect the open-source community.

First, let's consider Microsoft, which handed a $100K check to the Apache Software Foundation (ASF) to become a Platinum sponsor.  Now $100K is chump change to Microsoft, so what does this latest "make nice" with the open-source community mean?  It could mean that Microsoft understands that open-source software is now established as a model for software development and they need to show some respect.  After all, the Apache Software Foundation is well-known and respected in the open-source community, so it would pay for Microsoft to be on speaking terms with the ASF.

Those with more suspicious minds and longer memories see Microsoft inserting its tentacles into the ASF in order to promote their own self-serving interests, which many still believe revolve around derailing or co-opting the open-source movement.  Less suspicious minds see Microsoft as needing friends in the open-source community and willing to pay money to get them.  Well, money is what Microsoft has tons of, but what it doesn't have is much good will or trust in the open-source community.  And as the Beatles told us back in the 60s, "money can't buy me love".

Second, let's consider Microsoft's recent announcement to invest another $100M in Novell to support their November 2006 agreement to cooperate on technical, licensing and marketing matters relating to improved interoperability between Microsoft Windows and Novell SUSE Linux.  Now $100M is not exactly chump change for Microsoft, so what benefits are flowing from this nearly two-year old agreement that warrants a second cash infusion by Microsoft?

Well, Novell has gotten a tidy sum of cash from Microsoft since the initial agreement was signed.  Mutual customers of Microsoft and Novell have received assurances that they are both cooperating to improve the interoperability of their products.  The open-source community, however, split on the value and meaning of the agreement.  The fanatics denounced it and swore never to use Novell SUSE Linux.  Jeremy Allison, Samba project director, bailed out of Novell in protest and landed at Google.  Novell channel partners generally thought it was a good thing because it diffused a lot of the Novell is good and Microsoft is bad attitudes that have been festering over the years.  Ray Noorda probably rolled over in his grave given his dislike for Gates and Ballmer, but then again, Ray was also a proponent of cooperating with your competition.

On the nuts and bolts side, their agreement has had some tangible results.  Microsoft and Novell collaboration on server virtualization has shown results just as Microsoft launched its Hyper-V virtualization hypervisor.  This should give Microsoft and Novell a leg up in the management of each vendor's virtual server environments.  Novell SUSE Linux Enterprise Server 10 SP2 with Xen can run Microsoft Windows Server 2008 as a virtual machine.

The International Standards Organization's (ISO) recent and controversial approval of Microsoft's OOXML as a standard document format coupled with Novell's release of a ODF - OOXML document translator in March 2007, was another good thing coming out of their agreement.   But some industry pundits want to see Microsoft do more in this  area.  Their preferred state is complete, transparent and fully functional document exchange, including macros, between Microsoft Office and OpenOffice documents.  This doesn't seem to be an unreasonable request and if Microsoft has half an interest in protecting its Office franchise they need to step up and make this happen because Novell cannot do it alone.  Translators are one thing, but full document exchange interoperability is something else.

And in September 2007, Microsoft and Novell opened their interoperability lab in Cambridge, Massachusetts.  The lab is staffed by Microsoft and Novell engineers and is a physical manifestation of their agreement to cooperate in technical areas related to virtualization management and better Active Directory - eDirectory integration.

The $100M in new money Microsoft is pledging to their agreement will be used to buy SUSE Linux Enterprise Server support subscriptions from Novell.  Novell has stated that the existing agreement has significantly improved the uptake of SUSE Linux in Microsoft Windows accounts.  Their agreement makes Novell SUSE Linux the safe Linux server play for Microsoft Windows shops that need or want to deploy Linux servers.  Any interoperability problems encountered will get the attention of both Microsoft and Novell.

So what is left out?  Some people would prefer that Microsoft work more closely with the Mono project.  Mono is an open-source project sponsored by Novell that runs Microsoft .NET applications on Linux.  This should be a no-brainer for Microsoft as it would promote the broader use of .NET applications and give .NET developers a cross-platform audience for their code development efforts.  And best of all, both Microsoft and Novell would support it.

In the end, the success of the Microsoft - Novell agreement will be based on their mutual willingness to do what needs to be done to improve the use of each others products in mixed network environments.  Microsoft and Novell customers need to let their reps know that it is important to continue to deliver the interoperability goods.  Their November 2006 agreement runs through 2011, so let's hope for continued progress.

Oh, if you want to know more about Microsoft - Novell interoperability you should visit the site at http://www.moreinterop.com.

SET SOAPBOX = OFF

A Declining Economy Creates Opportunities for Open-Source

In his recently published book titled Reinventing Collapse, author Dmitry Orlov compares and contrasts the collapse of the former Soviet Union with the observable signs of a declining United States.  Orlov's wide-ranging analysis touches on the information technology business where he makes several insightful comments.

Orlov states that the United States initially dominated the information technology arena during the free-wheeling era of the 1970s and 1980s that produced a number of significant random and accidental successes.  These storied successes included:  Apple, 3Com, Compaq, Microsoft, Novell, Lotus, Sun, WordPerfect, Borland, Ashton-Tate and Cisco.  But along the way the United States engaged in a strategy to "kill the golden goose" by enforcing intellectual property laws, allowing patents on software and perpetrating a fraud called "enterprise software."

In the Internet age of the 1990s it was no longer possible to enforce intellectual property laws governing information in the form of computer code or movies, since it could easily be copied and distributed.  Orlov argues the the only people still paying to rent this information are Americans who are afraid of being prosecuted.  And with the dubious extension of patents to software, the United States has now limited the type of speech software developers are free to engage in unless they are employed by large corporations who have amassed a war chest of software patents, which they threaten to use against potential lawsuits by competitors or anyone they feel threatened by. 

Orlov criticizes bureaucratized corporate software development, which has placed multiple layers of project managers, systems architects, marketing flacks, sales weasels and corporate management on top of software developers who must cobble together pieces of purchased and licensed code that takes an army of software consultants to get working, if it ever works at all.  The end result is a "solution" that is expensive to license and implement and not open to inspection and modification by the people who have to use it. Bugs are fixed on the vendor's time-table and are frequently late in coming or not adequately addressed.

Under such a system Orlov claims it is no wonder that the best developers now create their software outside of this system as open-source projects.  And unless software patents are done away with, the commercial software business in the United States is in danger of disappearing as open-source garners a larger and more receptive audience.

It is easy to see how open-source software will improve and thrive in a declining economy. Businesses are under immediate and long-term pressure to reduce IT costs, which they can do by replacing expensive proprietary software with open-source alternatives.

Ditto for federal, state and local governments whose storage of public documents in vendor-controlled proprietary formats will be challenged because these formats make public documents more expensive to access when they require the use of a vendor's proprietary applications. 

Public K-12 educational institutions face serious budgetary constraints in a declining economy.  Taxpayers will demand that public schools use free and open-source software alternatives to more expensive proprietary software that completely stifles the educational benefit of being able to take software apart, learn how it works, modify it and give copies of it to your friends.

The good news in this scenario is the demand for people capable of supporting and modifying open-source software will be on the rise.  So the sooner the K-12 public education establishment gets going with open-source software adoption, the sooner their graduates will have opportunities to enter gainful employment creating and supporting open-source software.  A side benefit could be increased student interest in attending higher education programs in computer science.

Open-source software will be increasingly valuable in a declining economy.  That said, can there be any doubt that the only software development model that will thrive in a declining economy will be one based on open-source software?

SET SOAPBOX = OFF

Is Microsoft Windows still Relevant?

There has been a lot of talk recently about whether or not Microsoft Windows has a future.  Some of the conversation turns on the perceived lack of value in Windows Vista.   Others contend that a  "heavy" operating system, like Windows, is not needed by most corporate computer users.  And industry pundits have criticized Microsoft for not having re-designed Windows Vista using a modular architecture.

So just how valuable is it to have a Windows operating system in order to take advantage of today's network-based Web applications and services?  The answer is, not very.  The size and complexity of Windows is working against it.  And the ability of Microsoft to keep schlepping its legacy application baggage along has reached the breaking point. Windows Vista offers the corporate Windows user no compelling reasons to go through the expense and pain of upgrading from Windows XP.  And in these tougher economic times, the prospect of deploying a Windows Vista upgrade is a non-starter when Linux and open source software alternatives can do the job for less.

Microsoft XP will be 7 years old this fall.  Seven years is a long time for any Windows operating system to maintain its currency.  Windows 95 and 98 were current for shorter times.  Ditto for Windows 2000 and Windows 2003.  Only Windows NT maintained its currency for the same reason Windows XP has survived.  Both of them managed to find a sweet spot in corporate computing environments where it seemed just right.

But even if Windows XP was just right, it is showing its age and faces a questionable future after Microsoft discontinues its availability on new PCs after June.  Emerging desktop virtualization technologies may give Windows XP a reprieve from retirement, but just how long are those Windows applications running on Windows XP going to be needed?  Maybe longer than most people think, but then again, maybe not.

Today, the environment for delivering applications and data has shifted to the network and its servers and away from the desktop.  In modern computing environments a Windows desktop becomes just another delivery appendage to a network's ability to present users with access to applications and data.   Windows terminal servers and Windows desktop virtualization can extend the usefulness of Windows a bit longer because these technologies also reduce the management costs of Windows and allow organizations to contain their Windows and Windows application licensing expenses.

In the world of network computing, the future belongs to Linux and open source software.  Linux can be stripped down and embedded into just about any electronic device.  Novell just released SUSE Linux JeOS (Just enough Operating System) to developers who want to embed their applications in a minimal Linux environment.  Red Hat is also working along similar lines. 

For its part, Microsoft produced several embedded Windows versions over the years, but they missed the boat on the current movement of embedding just enough of the operating system with an application.  Microsoft appears to have been too conservative and too focused on its "heavy" operating systems to take notice of what was happening around them.  Score another win for the flexible Linux OS and the open source community.

For over 25 years Microsoft has been a successful marketer of expensive proprietary software.  And over that time we know that nothing can fail like success.  Another thing we know is Microsoft Windows has past the peak of its power and influence.  We have reached "peak Windows" as more adaptable Linux-based open source technologies along with the Apple Mac begin to erode Microsoft's Windows empire.

The irrelevance of Windows is on the horizon, but it isn't here yet. A lot may depend on what Microsoft does or doesn't do with Windows 7.  Those who don't care or can't afford to wait to find out are deploying Linux and open source software alternatives now.

SET SOAPBOX = OFF

   

Open Source Moves Uptown

Linux was born in the early 1990s but the term "open source" wasn't coined until 1998.  And over the past ten years the open source development model has steadily gained momentum. Today, many commercial sponsors of open source project have seen the value of their projects rise in the marketplace.

Acquisitions of companies based on open source projects include:  JBoss, which was acquired by Red Hat ($350M).  SiteScape, which was acquired by Novell (undisclosed).  Scalix, which was acquired by Xandros (undisclosed).  Zimbra, which was acquired by Yahoo! ($350M).  Xensource, which was acquired by Citrix ($500M).  MySQL, which was acquired by Sun ($1B).  The fact that large sums of money are being spent to acquire the commercial sponsors of open source projects is proof that software development momentum is shifting to open source and away from proprietary software development.

A movement as powerful as open source was destined to reach high levels of value and interest.  The open source model is world-wide and brings together the talents of many developers who work on a project not because they want to get rich, but because they want to do it.  In other words, they have found an "itch" that they want to scratch.   Open source software by its very nature can be fixed or modified by anyone with the skill to do it.  See something wrong...fix it.  Want to add some feature to a project...submit it.  Want to take a project in another direction...fork it.  Open source is about the freedom to freely use software that is valuable to you. 

Closed source or proprietary software development typically serves corporate interests, which are usually motivated by profit.  Closed source software is about  keeping the source code secret and making you pay for a license to use it.  You cannot take it apart.  You cannot fix it.  You cannot copy it and you cannot give it away to someone else.

So far the acquisition of commercial sponsors of open source projects  has been made by companies committed to open source software development.  But the big bucks exchanged for some of these high profile project sponsors has generated concern among the purists.  Should we be worried that this is happening?  Probably not too much because even if the commercial project sponsor ceases business operations, the maintainers and contributors of the project can organize themselves to continue the project on their own or fork it if they decide to take a different direction.

Open source projects cannot be "orphaned" unless everyone involved with the project stops working on it.  With over 130,000 open source projects registered on SourceForge.net you can bet there are a lot of projects that have received very little attention for years, but there they are waiting for someone who wants to continue scratching the "itch" that got the project going in the first place.  You can't say the same thing about proprietary software where products can be abandoned leaving users with no legal access to the program source code.

When it comes to selecting software for your organization, it is now important to weigh the value of choosing open source software.  After all, which software development model offers you more freedom and security?  The correct answer is open source.

SET SOAPBOX = OFF

 

Super NOS Redux...now that SCO is dead will Novell open UnixWare code to Linux?

Back in 1993 you may recall that after Novell acquired Unix Systems Labs from AT&T there was a lot of speculation about whether Novell would create a "Super NOS" by engineering a NetWare-Unix hybrid OS.  Well, it didn't happen.  Instead of doing that, Novell turned out an excellent Unix implementation on Intel processors called UnixWare.  Novell wound up with both NetWare and UnixWare to offer its customers.  Ironically, Novell wound up in a similar situation in 2004 after acquiring SUSE Linux.  This time the plan was to replace NetWare with Linux due to NetWare's declining fortunes in the OS marketplace.

Unfortunately, UnixWare met with only modest commercial success.  Even though Novell now owned Unix, it was not regarded as a Unix company like Sun.  So in a moment of questionable decision making, Novell sold the marketing* rights for Unix and UnixWare to the Santa Cruz Operation (SCO).  At the time, SCO was regarded as the "Rodney Dangerfield" (I don't get no respect!) of Unix on Intel processors.  At least SCO had a reputation for doing Unix on Intel processors, which was more than Novell had at the time.  This decision, however, would come back to haunt Novell in 2003.

By the mid-1990s the open-source Linux OS was emerging as a challenger to both Unix and Windows for a share of the server OS market.  Former Novell CEO Ray Noorda funded Caldera, which created its own Linux distribution aimed at the business customer using a reseller channel distribution model.  Along the way, Caldera eventually acquired SCO and renamed itself The SCO Group after abruptly dropping out of the Linux distribution business.  New management then decided to branch out into the intellectual property licensing business.  Next thing you know it is 2003 and The SCO Group files a lawsuit against IBM for $5 billion claiming IBM allowed Unix code from their AIX implementation to be "donated" to Linux.  There was wide-spread suspicion that Microsoft had arranged the financial backing SCO would need to litigate this lawsuit, which turned out to be true.

Novell responded to The SCO Group v IBM lawsuit by claiming it still owned the Unix and UnixWare copyrights because it never sold them to SCO...just the marketing* rights.  The SCO Group then sued Novell for defamation.  Novell said we'll see you in court.  Fast forward to August 2007 when a Federal District Court judge in Utah ruled that Novell, not The SCO Group, owns the copyrights to Unix and UnixWare.  The SCO Group filed for Chapter 11 bankruptcy protection one month later.  End of story?  Not yet.

By way of Shickshinny (a town in northeastern Pennsylvania) we finally get around to what Novell should have done with UnixWare back in the mid-1990s.  Novell should have opened up the UnixWare code base to Linux kernel developers.  Why?  Because Novell was going nowhere fast with UnixWare, NetWare was beginning to lose out to Windows NT, and Linux was just reaching  take-off velocity in the OS  market.   For  a company looking for the "next big thing" in OS software, this should have been a clue about what to do.  But nothing fails like success and Novell management was still enamored with the success of NetWare despite the growing competition from Microsoft.  Too bad that Novell had lost Ray Noorda's insight into the importance of Linux.

Today, there should be no reason why Novell, which is also the distributor of SUSE Linux, would not want Linux to take advantage of technical advances "locked away" in UnixWare.  Rather than "fork" Linux development by keeping it just for SUSE, Novell should open-source UnixWare to Linux kernel developers and see if we can't finally get the "Super NOS" that was rumored about many years ago.

As an addendum to this post, Steven J. Vaughan-Nichols (Linux-Watch, January 4, 2008) has been thinking along a similar line when he writes:  "What will Novell do with Unix?  I don't know, but what I'd like them to do is to open-source as much of the code as they can.  There's still some goodness left in Unix that hasn't been duplicated in Linux."  He goes on to say that when Ransome Love, former CEO of Caldera and one-time Novell executive, wanted to open-source Unix he discovered that the Unix code contained copyrighted code from other companies.  Steven concludes by writing:  "Since Novell is a Linux company, it make perfect sense to me if they were to cherry-pick Unix for its best code and release it to the public".  I assume by "release it to the public" he means under the GPL.  You can read the full Steven J. Vaughan-Nichols article at http://www.linux-watch.com/news/NS6243163353.html

Well, there you have it.  We both agree that Novell could make a valuable contribution to Linux kernel development by releasing some of the Unix/UnixWare code to the open-source community.  And as soon as the wooden stake is pounded into SCO's heart, maybe that is just what Novell will do.  Let's hope so.

SET SOAPBOX = OFF