One of the things that have thus far eluded IT has been a relatively simple straight forward approach to identity management that could please the two camps within IT that habitually go to war over who is in control of the issue. The two camps that routinely battle it out for control over identity management are the network administrators that feel that identity management is a security issue and application developers that feel that identity management is an access control issue that is crucial for them to control.
Unfortunately, both camps are right, and that in turn results in the usual bureaucratic malaise where everybody salutes the concept of federated approaches to identity management, where nothing actually gets done, but neither camp really wants to do all the hard work associated with creating such identity management schemes.
The only bit of good news to find its way into one of IT’s most glaring security issues is the arrival of a new company called Presence Technology that is delivering a new approach to identity management that promises to satisfy the needs of both camps within IT while still allowing them to get on with their day-to-day tasks with minimal interference. A startup company based in Salt Lake City, Presence Technology has launched a tool called PresenceID that consists of a database that keeps track of all the access rights associated with any given application. It does this by essentially scanning the meta data associated with access control in any given application and then storing that data in its database, thereby providing a central repository through which network administrators can manage identity management without requiring developers to change or rework their specific approach to access controls.
Scheduled to be available in January, Presence Technology is led by Novell veteran David Doane and has already signed up Identity Data Management Group as a reseller. But whether Presence Technology by itself can bridge the great identity management divide remains to be seen, but at least in some small measure somebody is trying to make some progress in a area that remains one of the great shames of IT as the lack of a unified approach to identity management not only drives up labor costs, but makes significant contributions to the exploits that security hackers will regularly exploit as applications rather than networks become the next big target.
There’s probably really no excuse for this not having happened earlier than now, but the good news is that encryption is finally on the cusp of going mainstream. After all, we have Microsoft getting ready to roll out BitLocker, and Seagate has come up with hard drives that have built-in encryption.
Of course, while those events are likely to get all the hoopla, it’s the inclusion of encryption as an option for Symantec’s Veritas NetBackup offering that’s likely to get the most accolades from the rank and file. Not that being able to encrypt backup files is anything new, but it usually meant buying third-party software at an additional expense and then mastering how that particular product worked. With the inclusion of encryption in Veritas NetBackup, it should prove to be a less expensive option that is easier for the rank and file IT staff to work with. More importantly, the Symantec sales forces and army of channel partners should make IT organizations a lot more conscious of the value of encryption during backup. That may not sound like a big deal, but as part of a lot of business continuity strategies, a surprising number of IT organizations have taken to taking backup tapes homes with them at night just in case something happens to the office overnight. At least that way, somebody would have ready access to the company’s latest data. Of course, that works really well until you create a security breach because the car the backup files are riding around in is stolen. Think that’s unlikely to happen? Well, check out the 12 months and $7 million of agony experienced by Providence Health & Services, which may just be wondering where Symantec and the rest of the industry were when they needed them.
The historic problem with a knowledge management systems is that they have always been canned systems created using a top-down methodology that never really reflects the changing dynamics of the business. So at best they usually wound up being expensive pieces of shelf ware that nobody every used because they were difficult to master and provided little additional value over running your own set of SQL reports.
But in a day and age where we have search engine technology that find almost everything and anything related to a given subject on the Web, it seems a pretty silly for us to be living in a digital age where we still don’t really know what we already know about any given situation.
Into this quagmire enters a new knowledge management application called OfficeEngine from HandySoft International to easily organize tasks and provide managers with relatively simple easy-to-use tools to query all the content within the enterprise for any information inside any given set of applications that might be related to the subject matter at hand. And just to make it easier to stay up on related current events, HandySoft signed a deal with Intellext to integrate Watson search engine technology within the OfficeEngine applications. What this means is that alongside OfficeEngine, users can have a separate client applet running that constantly searches for information on the Web that's related to the documents they are looking for.
You might well wonder what Microsoft and IBM/Lotus, among others, have been thinking about for the past few years because all the collaboration tools that most vendors have come up with at best over-engineered pieces of complexity that are more difficult to run than the actual worth they provide. And it's not like customers haven't been asking for this stuff because better collaboration and enterprise search tools are at the top of the wish list for most enterprise customers.
Granted, OfficeEngine is relatively unproven, but it's already got a track record in government sites such as the Department of Labor, Health and Human Services and an unnamed Federal law enforcement agency in addition to some usage at companies such as Samsung and LG Electronics.
Maybe someday Microsoft, IBM Lotus, or even Google might take a look at the HandySoft approach and ponder the possibility of actually delivering tools that adapt to the way people might want to organize their tasks versus building environments that ultimately force people to change the way they work to meet the ways the software works.
Cisco chief technology officer Charlie Giancarlo put forth an intriguing hypothesis on how the future of enterprise computing that claims that as Ethernet switches get faster they will emerge as a new bus architecture that links distributed processors, memory and storage units to create a virtual computing platform.
This approach would quite literally turn the network into the computer and Giancarlo argues that this would be a good thing because not only could we layer in more redundancy to help with disaster recovery scenarios, we would be able to scale systems in a true on-demand model that would make the concept of utility computing an everyday practice.
Driving this argument is an assumption that within the next few years Ethernet switch technology will be faster than any alternative bus technology and people will recognize the value of distributing various processor, memory and storage compute engines around the network. That of course would then require new approaches to operating systems that would be need to manage this new computing paradigm.
Based on some of the comments made by Cisco executives during its worldwide analyst conference this week, the company seems to think that it has a shot at developing the operating system needed for that new paradigm. Cisco is already working on creating a virtual layer of software above the network that will hide the complexity of the network from developers that are expected to write the next generation of IP-based applications. Furthermore, Cisco has also signaled its intention to embed more functionality, such as voice over IP and security, directly in the core processors of its routers and switches.
Giancarlo says elements of this new model of computing can already be seen today but that the complete transformation of enterprise computing as we know it today won't likely be completely realized until after 2010.
What’s interesting about this model is that is basically takes all the server and storage technology we have today and turns them into commodities because the only thing that can be easily replaced is the network and the applications that run on top of it. That might be interesting news for the denizens of Hewlett-Packard, Dell, IBM and EMC. More interestingly, it may lay the technical foundations for a wave of mergers and acquisitions that most people would have thought unlikely to happen.
The first letter in the abbreviation of CEO usually stands for cheerleader so it’s not unusual to see chief executives leading the cheering section, as happened today with John Chambers at Cisco’s worldwide analyst conference in San Jose, Calif.
But the recurring theme that Chambers returns to time and again is the link between innovation and productivity. Although he may be hard pressed to prove an definitive connection between the investments in IT and an increase in productivity, Chambers insists that the two are directly related because when there have been large investments in IT, gains in productivity have quickly followed.
And in more recent times, the decrease in IT investment in IT that began following 2001 is now resulting in a slowing of the rate of productivity per employee.
As a result, Chambers has become something of a one-man band trying to remind people that the investments made in technology in the 1990s led to major gains in productivity and that in the latter half of this decade customers need to replicate that investment to achieve the next level of gains in productivity.
Cynically, there is clearly a connection between making more investments in technology and the Cisco stock price, but to Chambers' credit, he has identified video as the next killer application on the Internet, which in turn will drive things like Gigabit Ethernet to the home as part of a general reassessment of the definition of the term broadband.
In other words, what we think of as broadband will essentially be the narrowband of tomorrow and it is that concept that is driving Cisco's overall strategies as applied to the company's data center, applications, mobile and consumer product strategies.
Chambers sees all this playing out between now and 2011, but what makes Chambers stand out beyond other CEOs of the day is that he's the only one banging a drum for unrestricted innovation. In contrast, just about every other major CEO is marketing the fact that there latest product is a slight improvement over previous renditions, which justifies the cost of the upgrade because it is easier to manage and run.
In short, most CEOs are long on selling products and very short on vision when it comes to articulating the real strategic value of next-generation technology.
Whether you agree with Chambers or not—and there are many who don’t because all they want from network vendors are big dumb pipes—the void that he is trying to fill is the total abdication of leadership by most technology company executives in the wake of previous economic downturn, stock market scandals and high profile CEO terminations.
But as the economy improves, the perception of technology as a series of commodity components that are not greater than the sum of its parts is a bad thing for the industry as whole. So just maybe Chambers is on to something, and other CEOs would do well to sit up and take note.
With the advent of the holidays upon us the time may have come to consider a novel idea that most people would have written off as absurd only a few months ago. But as the processing power of consumer PCs continues to outstrip their corporate counterparts, the idea of user-owned PCs finding their way into the workplace is not as far-fetched as it once seemed.
What’s driving this concept is the simple fact that many users increasingly own notebooks that are significantly more powerful than the puny machines that their corporate IT departments routinely foist on them. Unfortunately, this usually means that users usually have to deal with two machines in their lives, which in turn forces upon them the choice of taking their corporate notebook on the road instead of their personal machine that includes all their favorite games and personal applications. Even if the machine in question is a desktop, most users given a choice would prefer to manage their corporate and personal applications, at least to some degree, on one machine.
The problem is that until recently there was not way to effectively isolate the corporate environment from the domain of personal applications. But now we have virtualization technology from not only the likes of VMware, but Sentillion and Citrix, which provide the tools necessary to completely isolate our professional and personal lives on the same machine even to the satisfaction of most compliance auditors.
This concept is not only attractive to at least a significant subset of the end user community, but chief financial officers are likely to be fond of the idea of reducing the capital budget by encouraging users to use their own hardware for work and then write some subset of the asset off on their personal taxes.
It might take a while for this idea to catch on, but even reactionary firms such as Gartner have found merit in the concept. And given the choice between the PC that an IT department sized exactly to just meet the needs of the corporate application workload and the more powerful machine users bought themselves, the latter option is going to win out almost every time.
You can’t help but wonder, given all the attention that Oracle is lavishing on building an applications empire to rival SAP, if the company has lost sight of all the things that have made Oracle the force to be reckoned with that it is today.
The heart and soul of Oracle is the database, which at its root function is intended to manage data. In a client/server construct, that was comparatively easy to do because all the data naturally found its way into the database. But in a world based on Web applications, the data is in the database, the file systems, the applications and hundreds of other nooks and crannies around the enterprise. In effect, the database is just one data container of many, which in turn is exerting price pressure on databases that is rapidly turning that class of technology into a commodity.
Oracle is doing its best to fight the commodity trend by concentrating on high-end transactional applications that need all the bells and whistles in the Oracle database, but its approach to distributed data management seems to be remarkably concentrated around the Oracle database that Oracle continues to view as the center of the universe.
In contrast, both IBM and Sybase are making compelling arguments that say IT organizations need a non-platform-centric approach to data management, while EMC is out arguing that ultimately the whole data management issue is going to return to the province of storage vendors.
And just to make matters more complex for Oracle, in addition to open-source rivals such as MySQL picking off customers that don’t need all the database bells and whistles, startup companies such as EnterpriseDB are making headway selling PostgreSQL databases that allow customers to run their existing Oracle applications on a database engine that is not only less expensive, but also easier to run. Oracle has responded with lower-cost versions of the Oracle database offering, but at the end of the day it’s always going to take a lot more time and energy to master Oracle than any of the open-source databases that, rather than being focused on all the bells and whistles, are just designed to give 80 percent of the customers the 20 percent of the features they really use.
We’re a long way away from seeing the imminent collapse of the Oracle database empire, but you can’t help but wonder if Oracle has seen the writing on the wall and has decided to move upstream to the application space, or if it is so focused on moving into the application space that it's forgotten to cover its base.
As Microsoft gears up to lay claim to the next generation of Web-based computing in what promises to be a titanic battle with Google, never has so much been riding on the fortunes of one largely unheralded product.
When you look past all the vague positioning statements from Microsoft, the one thing that stands out is how pivotal the Windows SharePoint Services technologies are going to be to Microsoft.
Originally developed as yet another portal product, SharePoint Services has evolved to essentially become a set of shared application services that provides the infrastructure needed to do any meaningful collaboration within any set of Microsoft applications. But the role of SharePoint doesn’t stop there. When you take a look underneath Office Live, the set of Web-based applications that Microsoft is rolling out to compete with GoogleDocs, what you find is that the underpinning of the service is a Web-based implementation of SharePoint services.
Over time, what Microsoft expects to happen is that people will want to seamlessly move across applications that reside locally on their client and applications that they access as service over the Web. And the glue that is going to provide that integration is none other than the SharePoint services that Microsoft expects people to implement at their corporate sites, which in turn will share metadata about the relationship between applications on the client and the applications in Office Live that are also being managed by SharePoint services.
Microsoft hasn’t come right out and said all this because all the layers of management in Microsoft that have to line up the company’s strategy and related marketing efforts have created a “slowly I turn, step by step” approach to the market that does homage to the Three Stooges, who popularized this concept in the movie called "Gents Without Cents" way back in 1944.
Eventually, Microsoft will pull its act together enough to remind people that rather than the future being all about software-as-a-service, what we’ll really see is a convergence of client and server computing that a few years from now we’ll all sit back and wonder what all the fuss was about in the first place. But in the meantime, grab some popcorn because the new Redmond adventures of Moe, Larry and Curly is going to be one of the best shows in town for years to come.
There’s nothing quite like the arrival of a new platform to get the competitive juices of a vendor going, and no where does that seem to be more apparent these days than at Hewlett-Packard.
A rejuvenated HP appears to be just spoiling for a fight with Dell as it relates to the transition to Vista because from HP’s perspective the migration skills required to get customers over to Vista play a lot more toward its strengths than to what HP executives derisively refer to as Dell’s “mass duplicator” approach to the market.
HP executives say the company’s core strength among business customers lies in its people, processes and powerful machines that business customers are going to demand more of as they struggle to move up to Vista. In contrast, they argue that Dell’s missteps in terms of customer service and product quality will mean that customers are more likely to trust HP to get them through the migration to Vista than Dell.
If true, this means that in terms of corporate sales of Vista at least, we could see a two-tier rollout of Vista, with HP taking the lion’s share of the early adopters in 2007 while Dell waits for the bulk of the market to shift over to Vista in 2008.
HP, of course, needs to turn its vision of its place in the Vista hierarchy into reality in order to gain as much market share altitude over Dell in 2007 as possible. And this of course assumes that Dell does little to nothing to improve its battered customer service reputation in 2007. Given that it’s highly unlikely that Dell will do nothing on that front, the odds are good that whatever advantage HP thinks it has in terms of Vista will probably not be as great or as long as they think.
But in the meantime, we should probably expect both sides to start throwing some sharp elbows in early 2007 because the only time there are really major shifts in market share is either when one company really fouls up or there is a major platform shift. Both companies have had their share of foul-ups in the last two years, so that must mean we’re in for an old-fashioned street fight for Vista market share that may determine HP's and Dell's relative position to each other well into the next decade.
Every now and again several new concepts go mainstream enough to pull along another concept that many people might have considered a little too esoteric to have mainstream appeal.
Three concepts that have grabbed everybody’s attention as of late are server consolidation, multi-core processors and virtualization. While each of these represent distinct enterprise computing disciplines, any discussion about any one of these topics usually gets extended to include the other two.
The connection among them is the hope that virtualization technologies will some day lead to higher utilization rates on multi-core servers that are housed in consolidated data centers.
The problem with this grand vision is the execution of the strategy requires mastering a range of virtualization software and automated server provisioning tools at a time when the easiest approach to capacity planning is to just throw another $1,000 to $2,000 server at whatever application problem come up. But as tempting as this may sound, it does inevitably lead to server sprawl that ultimately creates higher systems and network management issues.
Fortunately, it appears that there’s at least one server company, Azul Systems, that has decided to wrap its intellectual capital around all these issues to provide an appliance that dynamically provides virtual compute cycles for any application. Alas, the Azul appliance platforms today are specifically tied to Java applications, but the implications of the approach should be self-evident to any server manufacturer. Instead of forcing customers to choose between tightly-coupled multi-processor servers that are expensive, in addition to being difficult to deploy and manage, and inexpensive racks of servers that are easy to deploy but expensive to manage, IT organizations can opt for a virtual compute engine appliance that lets any Java applications dynamically scale.
The implications of that ability for utility computing model are pretty profound. And with the addition of support for multi-core processors, the Azul appliances should be able to scale to fit just about any size Java application.
In essence what Azul has done is pulled together the concepts of server consolidation, multi-core processors and virtualization into a single commercial entity that is a whole lot more approachable than all the piece parts being rolled out by multiple server companies. Such companies seem to be more focused on these concepts as a way help them sell a lot more higher end machines than solving the actual practical processing problems that customers face every day.
Fortunately, practicality has a way of winning out over marketing and packaging schemes eventually. So don’t be surprised if you see mainstream server companies start rolling out compute engines similar to the Azul model over the next couple of years once they figure out the problem that customers actually want them to solve versus the one manufacturers find it convenient to solve for today. In fact, lawsuits are already flying in this space so stayed tuned because you know there will be a lot more to come.
Even though today marks the official launch of the business edition of the Vista operating system from Microsoft, it already feels like we're all suffering a bit of Vista fatigue largely because people have been talking about it for the past five years.
And while Vista itself represents a step forward, it's hard to get too excited about it in the absence of killer applications. What's in there today is nice, but true to form it really is just another iteration of an operating system where Microsoft has once again bundled a variety of existing third-party utilities into the operating system.
What's far more interesting are the things to come on top of Vista, such as the tightly integrated VOIP (voice over IP) application, the new video conferencing tools and the multimedia authoring tools that Microsoft CEO Steve Ballmer promised would be out next year as part of a bid to dramatically lower costs to use these types of applications. What's driving these applications is a much improved TCP/IP stack, a much improved graphics engine and, in the not too distant future, support for high-definition video.
But for now all we've got is a much improved Office environment on top of Vista that makes it easier to work with photos and spreadsheets inside a document, and unfortunately, that by itself is not going to be enough to over the upgrade inertia that is driven by warranties and depreciation cycles.
If you're due for a major upgrade today then moving to Vista when Microsoft rolls out the first service pack in 2007 is probably a good idea. But for everybody else, Vista is going to be a tough argument to make until there is a body of killer applications available that are too much to resist. And that’s not likely to happen until late next year at best.
In 2007 there is a high probability that IT budgets are going to be adjusted to reflect rising electricity costs, which in short means that a lot of IT organizations might have less to spend next year because they need to help offset rising electricity costs emanating from the data center to that 3-year-old printer sitting in some remote corner of the organization.
Most people routinely accept the fact that rising demand for oil coupled with Mideast politics is largely responsible for rising energy costs that in turn drive up the costs of electricity. But a recent study that is at the heart of a lawsuit concerning the cost of electricity may have long-range repercussions across a nationwide electric power purchasing system that defies anyone's notions of a free market system.
The lawsuit, filed by Robert Weishaar on behalf of a range of industrial companies that consume large amounts of electricity, alleges that the market set of by the Federal Energy Regulatory Committee is little more than a government-sanctioned oligarchy. The heart of the suit stems from the fact that the way electricity pricing is set on any given day is through an auction system where whatever power company bids the highest for electricity sets the price for everybody else. So basically, the power company with the least amount of purchasing power sets the price of electricity for the entire nation. Clearly, Adam Smith must be rolling in his grave over this.
The lawsuit cites academic studies from Carnegie-Mellon, Cornell and George Mason to back up its claims that basically concludes that wholesale electricity markets combined with regulated utilities defy any pricing benefits that might actually be derived from unfettered competition.
You might think that given the amount of capital that is literally being burned up every day the federal government might actually be aiding the plaintiffs rather than being the defendant in this case, but as usual with all federal issues government inertia sets the policy. But this isn’t necessarily because any elected official wants it this way. It’s because most elected officials can’t buy a clue when it comes to understanding market dynamics unless somebody makes it about votes back home or campaign dollars to line their coffers. So if you really want to help fight the battle to rein in electricity costs, the best thing you can do is call your CFO and legal teams and get them to add their financial and political support to the suit that is being brought forward against the FERC. On top of that, they might also remind computer manufacturers that rather than solely trying to profit from a bad situation exacerbated by stupidity in order to sell you more energy-efficient upgrades, they might actually want to look after the interests of their customers by lending their support to the suit as well.
Clearly, there are a lot of unnatural market forces being brought to bear on the nation’s energy supply, but until these issues move from the back room to the court room most people are going to be completely in the dark as to why we continue to burn money simply because a few people have a vested interest in a system that works against the best interests of the vast majority of the businesses and people in this country.
As the fourth quarter capital budget season is upon us, one of the things that IT finds frustrating is how a perfectly good machine that was bought just over a year ago can become as slow as molasses despite that fact that no new applications have been added to its load.
The reason this happens is most often attributable to applet pollution, the process by which users continually add small amounts of code to their systems usually by downloading some widget from a Web site that promises to perform some task for them. That piece of code can be as large as something like iTunes or as small as a cookie, but whatever it is they add enough load over time to affect the performance of even the most robust of machines.
Alas, in corporate environments we rarely see robust machines. What we usually see are machines that are just barely sized to run the official application load as defined by the company, so the sensitivity of these machines to applet pollution is high. Unfortunately, most IT organizations do very little to police these systems so the poor user that is left complaining about sluggish system performance usually doesn’t have the tools or the expertise to go find all the random bits of software that are choking off their system resources.
Of course, there are tools available to deal with this problem, ranging from austere solutions such as those offered by Persystent that automatically wipe any unsanctioned software off a machine that doesn’t have express permissions to run or the spyware and compliance management tools offered by Shavlik that make it easier for IT organizations to police applet pollution on their systems.
The only real problem is that IT organizations have to have the funds and the will to deploy these types of products in the name of forestalling unnecessary system upgrades driven by unwary users.
And yes, you could argue that what we need to do is change the behavior of the user, but the truth of the matter is that even with years of therapy and counseling most users are never going to change their behavior.
It’s a harsh world out there on the world wide web where every day there are people actively conspiring to steal limited system resources and network bandwidth. You can either sit back and suck it up or build fences to protect what's yours. After all, whether in cyber-space or the real world, good fences still make for good neighbors.
Every now and again somebody talks about a future breakthrough that sounds pretty cool, only to discover that somebody else is already moving in the same direction but on a slightly different path.
The somebody else in this case is Sentillion, which until recently was focused almost exclusively on health care applications where they allowed health care organizations to create virtual environments on remote PCs. In its most common usage, health care organizations would deploy Sentillion in order to partition a PC in a doctor's office for a specific set of remote applications, thereby creating a sandbox environment that is isolated from the rest of the files and applications that a doctor might have on their PC.
Now Sentillion is looking for channel partners to take it's virtual PC software out to the broader marketplace where it envisions that it could be used to create a relatively simple approach to disaster recovery, software distribution, compliance, and perhaps most intriguing, make it easy for users to use their own typically higher quality PCs in place of corporate PCs by creating virtual environments on their PCs that are isolated from the rest of their applications.
This latter capability has been a long-held dream of most financial people that are growing tired of buying PCs for workers that have already invested in higher-quality, under-utilized machines at home.
The Sentillion software works as a piece of software that is downloaded from a host server and resides on the local PC. This stands in stark contrast to a developing server-centric approach to this same problem that Citrix has embarked on under the guise of an effort known as Project Kent.
Both efforts have their merits and drawbacks based on the uniqueness of any given customer’s IT setups. But the Sentillion approach has the benefit of being available today while the fruits of Project Kent are not expected to arrive until 2007.
Whichever path is chosen, both approaches have merit as viable ways of delivering applications to users in the event of a disaster. And given the fact that disaster recovery is always on the top of the IT project list, it's time IT organizations focused in the event of an emergency on getting people access to their applications that run the business rather than just trying to get systems up and running.
Given the general movement to free software as a service offerings in the consumer space that will be funded by advertising, it was only a matter of time before somebody came up with a free suite of security tools that is similarly funded by advertising.
CyberDefender has launched a security suite that includes tools that fight spyware, viruses, spam and phishing attacks and even intrusion detection software that are all linked back to an early alert center that tracks threats encountered by users and then passes alerts on to other users of the products.
Initially aimed at consumers, CyberDefender has ambitions to extend this service into the small and midsize business space. Naturally, it's logical to conclude that Microsoft and Google are going to see similar opportunities to add security software to their Office Live and GoogleDocs offerings, which should result in just about every consumer being given free security software in the name of the collective good sometime by early 2008.
This brings up another interesting issue that is starting to develop for IT managers. CyberDefender is only one manifestation of the free security software phenomenon. Everywhere you turn it seems like some software or hardware vendor is now bundling in security software, particularly anti-virus software engines.
Ultimately, this could mean that IT organizations will find themselves trying to manage an update to all kinds of security software products that have been embedded in all manner of products.
Of course, the logical thing to do is to turn off all the security software that is coming into the organization via being embedded in some other product. But this is a time-consuming task and could even have negative implications because there could be special security features in those embedded tools that the application or hardware is counting on.
Whatever happens, it’s becoming clearer that we’re about to be inundated with security software. None of it will probably be any easier to manage, but at least the cost of acquiring security software in the first place will increasingly be mitigated by vendors delivering security software as part of the ante for doing business with them in the first place.
