Public clouds just don’t cut it when it comes to mission critical applications that form the backbone of your business. Security concerns, especially when you’re dealing with sensitive customer data like building security drawings, could also put a damper on any cloud computing strategy.

You could try to build a “private cloud“, though the term could be misleading, depending on your definition of cloud computing.

In May, Amazon Web Services senior vice president Andy Jassy noted that private clouds usually incur “very high fixed cost and lack the benefits of the cloud. Companies still own all the capital expenditure, data centers, servers; it’s not pay as you go and it’s not truly elastic on the company level because you still own and manage the infrastructure by yourself.”

In a move that stunned many in the tech circles, Oracle pulled a trigger on Google with a lawsuit alleging that the Android platform and devices infringe one or more Java patents and copyright.

Oracle claims that Google has been aware of Sun’s patent portfolio, including the patents at issue, since the middle of this decade, when Google hired certain former Sun Java engineers.

The lawsuit leads one to question the intentions of Oracle’s latest lawsuit and its commitment to open source. Java has been covered by the open source General Public License since 2006, way before Oracle’s acquisition of Sun Microsystems was completed in January 2010.

In a turn of affairs, SAP announced this week that it would not contest the liability of TomorrowNow for downloading proprietary, copyrighted software products and other confidential materials used by Oracle’s support organization.

In 2007, Oracle filed a lawsuit against TomorrowNow, a now defunct SAP subsidiary that offered maintenance and support services for Oracle software at a much lower cost than that provided by Oracle. SAP had said then that it will aggressively defend the claims made in the lawsuit.

On Thursday, SAP said that it will accept financial responsibility for any judgment awarded against TomorrowNow, despite the fact that SAP was not involved in TomorrowNow’s service operations and did not engage in any of the copying or downloading alleged in Oracle’s complaint.

Two bungling engineers and a faulty cable brought down Singapore’s biggest bank DBS — all of the ATMs, internet banking — for about 7 hours last month on 5th July.

Or so that was the narrative painted by the Straits Times two days ago on Thursday that it was all due to human error. (There’s a far lengthier version in the printed Straits Times version than the gimped version online). The big headline inside the paper on page four was this: “It was definitely a human error”.

Really? Is that the best narrative that explained why the system crashed that day? Everything was due to “human error”, and two “bungling” IBM engineers were to blame?

If Singapore’s biggest bank could so easily be brought down by “human errors”, then I find it genuinely shocking. Surely IBM’s 10-year S$1.2 billion outsourcing contract — about S$120 million per year to maintain the IT infrastructure — details a stringent process for disaster recovery?

Doesn’t DBS and IBM have SLAs that spell out how IT failures should be recovered from, with a detailed escalation process? And seriously, a single misplugged cable can bring down your entire storage system? I don’t buy this at all. You’re not talking about a start-up servicing a bank; you’re talking about a maintenance contract deal worth millions.

Thus, the main point is not about “human error” — a totally wrongheaded slant that ST took, in my opinion — but the fact that DBS’ business process screwed up along the way. Yes, human error may have started this, but the recovery process screwed up and failed to kick in.

The blow-by-blow account of how engineers triggered this failure is not interesting. What’s interesting would have been how and why DBS’s disaster recovery process failed to kick in.

Hitachi celebrates 100 years as a company this year, and I was invited to Japan to attend their annual uValue event, where they had some fanfare for this occasion. This year the event runs from the 22nd to 23rd July at the Tokyo International Forum (near Roppongi).

One thing I learnt from the uValue event is that Hitachi is a sprawling technology conglomerate that spans many divisions: from IT and telcos, to transport, to consumer devices.

Yes, I know they are a high tech company that builds computer and consumer electronics, but I didn’t know they also build train systems, elevators and nuclear power plants.

The company generates US $100 billion in annual revenues and it’s the third largest high-tech company in the world in terms of revenues, behind Samsung and HP.

Starhub has quietly introduced a “redirection service” that takes MaxOnline broadband customers to a Yahoo search page if you happen to enter an invalid URL in your browser. Prior to this, you would typically get an error message from your browser telling you that the invalid domain’s server cannot be found.

According to a Starhub FAQ list, the service does not track individual Internet usage patterns. The service “simply redirects queries to non-existing domain names to a useful search results page instead of a cryptic error message page or browser-defined page”. Users can also choose to opt out of the service at a preferences page.

What Starhub is effectively doing is DNS hijacking, a controversial practice that has led to security breaches by hackers. Some ISPs in the United States have had their users open to cross-site scripting attacks due to lax Web programming techniques by some search partners.

In 2003, Verisign also directed users to paid search results, a move which led to an investigation by the Internet Corporation for Assigned Names and Numbers (ICANN).

Among its findings, ICANN noted that DNS redirection “disturbed a set of existing services that had been functioning satisfactorily. Names that were mistyped, had lapsed, had been registered but not delegated, or had never been registered in DNS were resolved as if they existed.

“As a consequence, certain e-mail systems, spam filters and other services failed resulting in direct and indirect costs to third parties, either in the form of increased network charges for some classes of users, a reduction in performance, or the creation of work required to compensate for the consequent failure”.

In the past, when I went to one of IBM’s Smarter Planet events I found it hard to write something and distill the message for readers.

At the back of my brain was always this burning question: Just what is IBM selling here? I have difficulty connecting their really big picture green IT story to what they do as a technology company.

Let me set the context and take a short detour to explain why. I’m better known as a technology journalist-blogger hybrid, but I worked for a very brief time at the Centre for Liveable Cities in Singapore. In that short stint I gained an appreciation of the complex problems facing cities.

Different cities face different problems, depending on how developed a city is. A developed megalopolis like New York or Tokyo will face vastly different challenges than say, Hanoi, Vietnam or Sao Paulo, Brazil. And this is only one one aspect of a city. Culturally, economically, politically, every city is different and will have different issues.

From July 13, Microsoft will end its support for Windows 2000 and XP Service Pack 2, according to a media statement Monday.

Specifically, users of these two operating systems will not receive security updates or non-security hot fixes, free or paid assisted support, the option to engage Microsoft’s product development resources as well as
updates to online content (KB articles, etc.)

Microsoft customers are “highly encouraged” to migrate to the latest supported service pack which is the latest and most secure version of their product. Staying on a supported service pack is the only way to ensure continued access to security updates and the ability to escalate support issues within Microsoft.

To ensure that customers understand exactly what end of support means, Microsoft has put together a list of questions customers might ask about end of support. In addition, Microsoft has guidance for customers to understand what version of Windows they’re using today and how to upgrade their PCs to the version that’s right for them.

Launched in February 2000, Windows 2000 promises to build on the security and stability of its predecessor, Windows NT. Targeted largely at business users, Windows 2000 came in various flavours including Windows 2000 Professional for enterprise desktops and Windows 2000 Server, Advanced Server and Datacenter Server on corporate servers.

Networking giant Cisco Systems unveiled its vision of smart, connected cities of the future during a media event at the Shanghai World Expo today.

In a showcase of interest to urban planners striving to build sustainable and environmentally friendly cities, Cisco demonstrated several network enabled applications in areas such as health care, security and transportation.

Telemedicine, for instance, allows patients to measure blood oxygen levels and send that information to their doctors while urban transportation management systems allow city officials and transportation service providers to monitor real time traffic conditions and divert traffic if necessary.

Underpinning these technologies are ubiquitous smart networks and data centers that process and serve up critical information in real time.

Every good story needs a hero and a villain.

In Salesforce.com’s case, that villain is Microsoft, who is the poster boy for enterprise software.

Or so that is the narrative that Salesforce.com CEO Marc Benioff wanted to keep forefront and centre during his excellent presentation at the Cloudforce Tour 2 event in Singapore today.

It is a story of how traditional on-premise software behemoths — e.g. Microsoft, Oracle, SAP, IBM — are going to have their lunch eaten by nimbler Internet companies who started from the cloud, i.e. the Salesforce.coms, Amazons and Googles of this world.