Tech Trends

Telcos Jump Into Netbook/Data Plan Business

Telcos are jumping into the netbook/data plan business with a vengeance. AT&T is offering netbooks for $99 with a two-year service contract at your local Radio Shack, with Verizon rumored to be next in line. (Several European vendors have also rolled out such offers.) Like with cell phones, the strategy is to subsidize the notebook (the razor) so customers will buy lots of data services (the razorblades.) One reporter points out this makes the actual cost $1,500 over the life of the contract, and with basic Internet access going for $50-60 a month per household, it may indeed not be a bargain. But for some customers, being bled by the month will be preferable to paying up-front for a $300 netbook or $500 notebook.

 No word yet on whether, or when, the telcos might take the much bigger step of offering hollowed-out netbooks (no processor, no disk drive, no operating system) and delivering the entire application stack over their 3G wireless broadband services. Earlier this month I reported on SIMtone, which hopes to provide the delivery platform telcos will use to push applications and data out to these ultra-small PCs. SIMtone predicts we’ll see the first such offerings by the end of the year. Me, I’ll hold onto my real notebook until my cell carrier stops dropping my calls…

Will Dumb Terminals Inherit the Earth?

In these days of hunkering down, cutting back and getting by it’s neat to talk to a venture-backed startup that’s out to change the world. The SIMtone Corp. wants to put the Internet “cloud” on steroids, feeding you everything from processing power to storage and even the user interface on your application from the cloud. Goodbye Microsoft and Intel, presumably, as you’ll need neither an operating system nor a processor – just a modern day version of the dumb terminals once used to access mainframe applications.

Oh, and who will be providing these incredibly rich, incredibly reliable cloud computing services to the starving masses? The telephone companies. Really.

As Chairman and CEO Mario Dal Canto sees it, today’s cloud computing environments require the use of proprietary protocols and expensive, complex PCs at the user end. As the price of bandwidth plummets, and ever-more-powerful virtualization makes it possible to effortlessly scale Internet-based application hosting, SIMtone says to move all – and they do mean ALL – the heavy lifting to the cloud.

The benefits for users? Imagine being able to upgrade your processor, your memory, the size of your hard drive or your application suite with the click of a mouse.  Imagine very, very low-cost mini notebooks like SIMtone’s own SNAPbook essentially given away with service plans the way cell phones are now. Imagine low-cost, maintenance-free computing for small business, education or low-income customers.

There are, of course, a few steps between here and the promised land. One is that wireless broadband is not yet as ubiquitous in the U.S. as in other countries (which is why the SIMtone model has more traction in other parts of the world.) Another is that nervous customers (read: me) are not yet ready to trust all their data, all their ability to communicate, all their ability to even write a note to a wireless connection.

Others are apparently bolder: SIMtone says it has ten “very strong engagements” to provide its services management infrastructure to telcos around the world, and that the first services will roll out later this year. For the telcos, he says, it’s a way to break out of the price wars that come with offering voice and data commodity services. (Several telcos have already announced cloud offerings, but nothing as dramatic as what SIMtone plans.) And it may be just the thing for IT customers short of capital budgets for new equipment, and consumers short on money and PC know-how.

For example: I know a fellow who, working with a wireless mouse from his easy chair, doesn’t know how to run a virus scan  but can find and watch any 60s and ‘70s sitcom on the 32” flat-panel hooked to his Vista desktop. Would he go for an easy-to-use, pay-by -the month, no-computer link to the Web? To quote a certain vice presidential candidate, you betcha.

Having taken the evening off to watch the Oscars, I hadn’t read a scary story about the economy in, oh, 14 hours or so. So I was relieved to get back into “read it and cringe” mode with a story from TechCrunch reporter Sarah Lacy about why the smart VCs are moving their investments from Silicon Valley to overseas haunts such as

China

,

India

and

Israel

.

 

Her basic take is that venture capital isn’t focused on high-tech per se, but on high growth, and emerging markets with rapidly rising incomes are more lucrative than mature markets such as information technology. This got me, as a marketing writer/consultant, wondering if the B2B marketing and PR business is about to go the way of America's auto business, hollowed out by foreign competition.

 

It’s definitely true that talent is global (as evidenced, again, by the Oscars) but the infrastructure to turn great ideas into profits isn’t quite so well distributed. Marketing is, in large part, a language skill, and language is local. (Just look at the English-language Web sites of some foreign-based IT vendors to see what I mean.) I would argue a lot of the marketing/PR infrastructure for B2B products and services will stay local, because understanding the year-in, year-out shifts in language, grammar and cultural references – even changes in mood like the current mania for frugal living in the U.S. – requires living locally.

 

As long as the U.S. remains one of the world’s largest markets, then, B2B startups will still have to spend a good deal of their marketing dollars here in the U.S. Or am I whistling past the graveyard?

 

 

Refine Your Cloud Pitch

Having recently ranted about the lack of clarity in how vendors describe their “cloud” computing offerings, it’s only reasonable to ask me how I would do it. Therefore, some top-level thoughts about how the technologies and  markets are breaking down, as of now.

 

Software as a Service (SaaS): The most recognizable and well-known variation on cloud computing, which includes thin-client-like offerings such as Salesforce.com. (The user interface runs on the user’s machine, the data and most of the back-end processing live on hosted servers in the cloud.) In this model, the “product” is an application showing up on a user’s desktop; the provider takes care of all the back end work, from physical hosting to server configuration, performance management and security.

 

Hosting As A Service: A variation on everything from traditional application outsourcing to co-location (multiple customers in one data center) to application service providers and even – anyone remember this? – time sharing. In this model, the service provider might provide anything from just power/cooling, space and physical security to providing physical or virtual servers to running and managing the entire application stack. Microsoft is among the latest players to jump into the application hosting market (with Exchange), while Google and Amazon, among others, offer the ability to create and manage your own virtual servers in their physical space.

Utilities As a Service Standard data center services such as backup, from vendors such as Symantec, MozyPro (EMC), Amazon, Carbonite and Nirvanix, or security (from vendors such as StillSecure and Purewire) offered over the Internet. Here, the product is not an application a user sees, but an underlying function required to keep those applications available.

 

Application Development/Middleware as a Service: No, this doesn’t mean an outsourcer writing apps or hooking them together for you. It means using a cloud service provider’s application programming interfaces to build your applications (and then host them with the provider) or use their messaging and communication interfaces and standards to link other applications. Salesforce.com claims to be one of the leaders in this space; one risk for customers is being locked into that vendor’s platform for future development and integration work.

 

Each of these four models (and I’m sure there are more) present different benefits, risks and disadvantages for customers, and give providers a chance to differentiate themselves and tell a compelling story. If you’re not being specific enough in describing your cloud offering, you’re missing an opportunity to stand out from the cloud also-rans.

Cloud Positioning Is, Well, Murky

Even before InfoWorld asked me to do a series of stories on cloud computing it was clear that “cloud” is high on the IT radar screen, either despite or –of because of – the economic downturn. It also helps that the definition of the “cloud” is, like its namesake, soft and fluffy enough to cover almost anything. Heck, even I'm a cloud customer now.)

 

 

When I learned to read network diagrams in the Stone Age, the “cloud” meant the Internet. It also meant you didn’t have to know the physical location of a resource (whether a Web page or your CRM app server) because the Web provides a logical address for it. Ergo, “cloud computing” means something 1) located on and delivered via the Internet and 2) whose physical location is unimportant, as long as an application, user or device can find it through a logical address.

But business being business, every vendor who could spell “cumulus” began pitching their stuff as cloud-enabled. Virtualization vendors are now “cloud” vendors since virtualization provides the logical paths to the physical resources. Anyone who provides security, or performance monitoring, or network bandwidth or even electricity, is now a cloud computing vendor since you need their stuff to access the Internet. We even now have “private clouds” made up of virtualized servers and storage that sit in a customer’s own data center. Uh, since when did “cloud” refer to a customer’s own WAN/LAN?

Needless to say, the editors I talk to are very, very skeptical of cloud pitches. If you do indeed provide (or help provide) apps or services over the Internet, explain precisely what you provide. Is it provisioning capabilities that help cloud service providers keep their own data centers running? Is it a specific application (like CRM or backup) you provide to business customers? Or is it an IT service (like security) you provide via the cloud?

If you’re playing in the service provider space – i.e., as an ISP or telecom providing backbone networks to the cloud providers – make that clear also. If your offering is several levels up the value chain from the end customer, it’s a tough sell to an IT book focusing on CIOs and other customer-level decision makers.

A great story, of course, is anything having to do with the downturn, and what some vendors claim is a “bottom-up” groundswell of thrifty business managers moving ultra-fast to cloud services for the cost savings, sometimes even bypassing (and annoying) IT. Do let me know if you have any such stories, as we in the IT trade press always love a fight, especially during a downturn.

Surrender to the Cloud

While the rest of New England was being slammed by an ice storm, I spent a merely rainy afternoon in

Boston

at a Salesforce.com customer event about – appropriately enough -- cloud computing.

 

Cloud computing, loosely defined, means an application where the processing and the storage of the data happens not on your own computer, but out in the Internet “cloud.” The “cloud” implies that you don’t know, nor do you necessarily care, whether this heavy lifting is being done in

Manchester

,

N.H.

or

Mumbai

,

India

, as long as you have an Internet connection.

 

That dependence on the Web is why, up to now, I didn’t get cloud computing. It drives me crazy when I’m traveling and can’t find a Wi-Fi hot spot. I also get annoyed every time a momentary glitch in my Wi-Fi connection interrupts a show I’m streaming to my notebook. Why would any anyone, I thought, trust an important application like CRM (customer relationship management.) to their Internet connection? My afternoon with Salesforce (whose CRM application is the most successful, example of cloud computing) mollified some of my fears, and more importantly showed me the true upside of the cloud.

 

I always knew that Salesforce touts the fact their customers don’t have to maintain their own servers and storage, or installs updates user by user since Salesforce handles all that in the cloud. What I learned the other day is just how successful Salesforce has been: 43 percent revenue growth for the quarter ending October 31^st and adoption by leading companies such as Dell.

 

The true power behind cloud computing, though, is as an application and interoperability platform. Salesforce says there are now 85,000 applications available on its AppExchange site written to run within or work with Salesforce. These range from applications to track vacation time to handle recruiting, project management and email marketing campaigns, among many others. Salesforce is extending its reach with alliances to other cloud powerhouses such as Amazon and Facebook, and is pushing its Force.com as a platform for Web applications.

 

As an old middleware reporter, I was also intrigued to hear how easy it is to move data among applications by using Salesforce.com as the “Rosetta stone” to translate among various applications. One telling statistic was that 55 percent of all transactions being done on Salesforce now go through its application programming interface (API). That means customers are not just entering customer records or sales updates through the off-the-shelf Salesforce user interface, but building applications atop Salesforce customized to their own business.

 

It also makes Salesforce (or any other cloud computing platform, such as GoogleApps) a more formidable competitor to Microsoft and its older computing model of applications based on, and running on, the user’s PC. Color me convinced enough to at least experiment with Salesforce as a platform for my sole proprietorship. I’ll report back on how well cloud computing worked for me, and how I’m doing with this phobia about being caught without an Internet connection. 

 

Storage Marketing Tips For the Downturn

I’ve been waiting to see just how sharply IT shops will cut spending in response to the just-declared recession. Judging from a recent newsletter published by the Wikibon on-line IT forum, things will get ugly.

 

According to co-founder David Vellante, Wikibon members reported that IT departments are facing cuts of between five and 30 – yes, 30 – percent, even while coping with ever faster-growing databases and storage volumes.

 

To the rescue rides heterogeneous virtualization – the ability to pool disk space from EMC, HP, Hitachi or other vendors so customers waste less disk and can easily move data from place to place as needed. Storage virtualization makes it easier to put different types of data on the storage device with the performance, and price tag, that data deserves. It also, says Vellante, helps reduce vendor lock-in because customers can buy storage capacity from whoever has the best price.

 

Cost-cutting plans from storage admins won’t come as much comfort to either storage vendors or their own staffs. Wikibon members are talking about pushing vendors hard for deeper discounts, renegotiating existing contracts, cutting unnecessary projects, reducing headcount where possible and focusing on only the most essential storage-related functions.

 

Nothing comes for free, of course. Virtualizing storage requires an up-front investment in not only products but in-house skills. Some of the cutbacks in staff and new equipment will also reduce the speed, reliability or up-time of storage some end users are used to. “Communications is the key,” Vellante says, as storage executives reset expectations that anyone in a non-critical part of the business “must experience somewhat lower service levels” as their employers hunker down to ride out the storm.

 

Hey, waiting for that customer record to load sure beats getting laid off. Now, some thoughts for storage vendors looking to sell into this environment.

 

Develop or resell consulting services to help customers decide which functions are essential and which can be suffer in a downturn. Consider bundling these services to differentiate commodity products.

 

Develop or resell, where possible, “canned” virtualization offerings that reduce the cost and risk of virtualizing storage environments by setting a fixed price for virtualizing a set amount of storage (or number of servers.)

 

As always, ROI calculators that help customers calculate the benefits of storage (or any other technology) are key. The more granular, and precise, the results delivered by these calculators the better.

 

Need help adjusting your marketing message for the recession? Call at 781 599-3262 or email at bob@scheierassociates.com

Would You Buy an OS From This Man?

When you watch a war movie and see those blinking lights, strings of numbers and (of course) moving target blip moving across the screen, did you ever wonder what operating system runs all those systems? One strong candidate would be Green Hills Software, a little-known 28-year old operating system vendor going public this month with what it claims is a new level of secure OS called Integrity.

 

 

Integrity is not, like many “hardened” operating systems a trimmed down and buttoned-up version of Windows or Linux. It is instead a proprietary, real-time operating system that runs directly on the microprocessor. Separate operating systems, or separate applications, can then run as distinct processes atop Integrity. The operating system is already being used, the company says, in the F35 joint strike fighter, the Airbus 380 and the Boeing 767. 

On the surface, this looks like it performs the same function as hypervisors like VMware’s ESX or Microsoft’s Hyper-V. But it’s very different because Integrity has a much smaller code base than such hypervisors, and was built from the ground up to enforce user-controllable limits on how much memory or processor resources a given OS or application can use, says Green Hills CTO Dave Kleidermacher.

Rather than doing virtualization through the hypervisor, in the Integrity world virtualization is another service running on the guest OS. One sweet spot for this capability would obviously be virtualized servers where the failure of the entire server would crash multiple applications. Another use would be installing Integrity on a corporate notebook, where sensitive corporate data and applications would live in one partition, with Web browsers and email clients that link to the Internet live in another. (While Green Hills hasn’t yet announced pricing for Integrity, it is talking with OEMs about pre-installing it for such use cases.)

Green Hills’ proof of its claims is its use in (pardon the cliché) mission-critical environments, and its recent award of (Evaluation Assurance Level) EAL6+, which came after National Security Agency hackers couldn’t hack the OS even after two years of trying and access to all the source code and documentation. This raises the bar for computer security, says Kleidermacher, since everyday OSes such as Windows XP, Solaris, Linux and VMware are rated at only EAL4+.

This is a potentially compelling technology, especially if it requires as few memory or computing resources as Green Hills claims. Its greatest challenge is explaining and proving how Integrity works its magic, since most of the “secret sauce” is down at the code level and the only proof is the EAL6+ certification, which to most users will sound more like a Navy fighter jet than a benefit they can sell to their chief financial officer.

 

Symantec Offers Low-Cost Help From the Cloud

Timing is everything. Symantec probably didn’t have a financial meltdown in mind when it began developing its latest managed service offerings last year. But their rollout this week sure looks timely.

 

Its Veritas Operations Services is a “software as a service” offering – i.e., provided through the Internet "cloud" -- that automatically checks a customer’s data center configuration for any known problems, such as unsupported hardware, that could make the environment unstable. The first offerings under this umbrella are Veritas Installation Assessment Service, a Web-based service that (the company says) “simplifies version upgrades and new installations by validating the complete stack prior to starting the installation” of Veritas Storage Foundation and Veritas Cluster Server.

 

These offerings are agentless, which is good news because no IT manager wants to load any unnecessary third-party software on their systems. On the downside, being agentless means the service won’t find any changes made since the initial check.

 

What’s makes this so timely is that Symantec is charging very little for these services. They’re free for anyone with an existing maintenance agreement, and any other customer gets indefinite access for a one-time fee of $500 per server.

 

This is somewhat like the “phone home” software built into fault-tolerance, high-availability server hardware. It also gets more mileage out of the reams of trouble-shooting and trouble logs Veritas is gathering in any case from its customers. I don’t know if any other data center vendors are offering a similar service, but it’s a cost-effective way to help users get the most out of the software they already have --and help ensure they’ll want an upgrade the next time around.

AutoVirt: A Simple Story for Uncertain Times

Having just finished a story on storage migration, I know there are a ton of problems – and a ton of tools – around the seemingly simple process of moving data from one storage system to another.

 

However, there are still niches that haven’t picked over by the major vendors yet. AutoVirt (debuting this week) goes after one such niche: Live migration of the millions – or is it trillions? – of files produced by Microsoft Office productivity applications such as Word, Powerpoint and Excel.

 

AutoVirt is built atop the current mechanism within the common Internet File System (CIFS) that tells the machine or application requesting a specific file share the physical address of that share. AutoVirt specifically doesn’t migrate large databases such as Oracle, or database-like application servers such as Microsoft Exchange. Nor, does it go after the Unix market, whose underlying Network File System doesn’t accommodate their virtualization scheme.  

 

Rather, it focuses (for now) on Windows file servers operating within mid-size companies with “one domain, one data center, one all powerful IT function,” says founder Klavs Landberg, a 30-year veteran of Datapoint, Alliant and Sun, among others.

.

AutoVirt tries to build scalability into the pricing model, pricing by the number of shares under management, with pricing beginning at around $10,000. It’s also starting modestly, building policy-based migration for one-time migrations rather than for the fancier, information life-cycle management that would require ongoing automatic storage migrations – something most admins aren’t ready to allow.

 

AutoVirt is launching in stormy economic times, but its ground-up approach (tackling common migration issues rather than esoteric high-end problems) and straightforward pricing may serve it well. Another strong point, which other tech vendors would do well to emulate, is that it explains itself in big bold print on the first page of its introductory slide deck: “The AutoVirt File Virtualization platform is a native extension of Windows.  The killer application is automated, transparent migration of online data.” That’s the kind of good, crisp messaging any vendor needs, in good times or bad.