Mad Dog 21/21: The Maltese Logon
Published: December 12, 2011
by Hesh Wiener
If your apps all work through a web browser or are written in agnostic Java, relax. Don't bother reading this. But if Windows grips some of your apps, as it does at most IBM i shops, there are things going on you had better watch. Until recently, Windows dominated client environments. Not anymore. ARM-based devices running iOS (the Apple operating system, not the IBM midrange one) or Android or BlackBerry OS outsell X86 machines running Windows. Windows on ARM won't close the gap. Windows 9 might run mainly on cloud servers with island clients a second thought. When you say Wintel, people are going to think about a Swiss guy whose kid wears an apple and a Depend.
IBM has been aware of this for a long time. For more than two years, Big Blue's services group has offered virtual Windows desktops that let a user run on a thin client, an obsolete Windows machine, a Linux box, or just about anything else: DaaS (Desktop as a Service). But IBM has not had the smarts or gumption to get corporate users to accept the fact that a client only has to be, well, a client. The end user's machine doesn't have to be a full or nearly full replica of, for example, the Windows server at the other end of the wire. But maybe things are starting to change inside IBM the way they have in Malta.
A Maltese outfit called 2X thinks it can beef up Windows servers (including servers working with all the bigtime virtualization schemes) so they can reliably and securely deliver Win apps or whole Win desktops to thin clients . . . and do it for peanuts. Basically, 2X provides alternatives to XenDesktop from Citrix Systems or View from VMware that the vendor claims are better and cost a lot less. Which clients work with the 2X server? For starters, the ones running the 2X client, available for Android, iOS, Windows, Linux, Mac OS and, if you want to try something new, a little Linux disto the Maltese outfit built that's called 2X OS. (2X OS will run on a zombie X86 box from the Windows XP generation or earlier as well as a netbook or an ultra-light barebones laptop.)
The client apps are free! Last time we looked, 2X charges only for the server and it has the kind of pricing that growing companies are bound do love: It is priced per system not per core or per CPW or per Oracle whim. (It's slightly more complicated than this. There are actually three 2X products: a simple thin client server, a richer applications server and a load balancing program that works with 2X or Citrix servers.)
All 2X clients will converse with Windows servers running stock Microsoft Terminal Server, too, and with Citrix or other Windows sharing offerings that cooperate with RDP clients. What 2X has recognized is that its customers (including freeloaders like me who only pick up client side code) might want to use one client side package or one of a set of consistent client side packages to talk to 2X servers, Citrix servers, Terminal Services hosts, and other RDP-supporting server.
In IT, few things happen in isolation. More often, developments come in waves. So it's not surprising that there are other offerings designed to displace fully featured Windows clients and replace them with server-based technologies. To pick another example, I spotted a software management package built from the ground up with DaaS in mind. I'm talking about some code from Desktone that is apparently good enough to get the attention of IBM, which now uses it for its own DaaS offering. Desktone is one of a number of software vendors whose products attempt to make it pretty easy to manage the delivery of apps to end users. There's a lot of permissions stuff involved and quite a bit of management record-keeping. Desktone and others trying to build virtual desktop management packages find customers among the IT managers who think babysitting dozens or hundreds of PCs is a nightmare. That turns out to be a pretty big sales target, mainly because managing lots of PCs is undeniably difficult. Whether supporting the same number of end users with virtual desktops is actually easier is an open question. But client management is such a pain that quite a few shops seem willing to hear the proposition of server-side DaaS vendors of software, services or both.
Of course DaaS has an obvious limitation: The end user must have Internet connectivity. No Internet, no desktop. So DaaS isn't for everyone everywhere. And there are aspects of DaaS security and privacy that unquestionably discourage some prospective users. On the other hand, DaaS has phenomenal appeal among people who like green screen computing because it can give glass house specialists very precise control over the end user experience. It's not so easy to keep Windows users from messing around or goofing off. One big difference between green screens and current client devices is that a thin client can provide a gorgeous display along with a lively and effective end user experience while still remaining within pretty strict confines. It's probably not possible to make work fun but a high quality presentation layer goes a lot way in that direction, and it doesn't take a whole Wintel box to deliver first class presentation. Just take a look at some of the popular phone and tablet apps or some of the possibilities in HTML5 a presentation coding scheme that works on browsers for phones, tablets, skinny computers, and fat workstations.
It also turns out that cloud desktops, which some outfits are trying to peddle at $90 per user per month are also available for half that amount or less. Basically, companies with at least a couple dozen DaaS end users ought to be able to get virtual desktops that include productivity software and a nice complement of other goodies for $25 to $50 per seat per month. And if they just want to deliver DaaS themselves, an X86 box that can run Windows made multi-user by 2X, Citrix, VMware, Microsoft Terminal Server or, alternatively, a bunch of virtual instances of Windows (or Ubuntu if that's your favorite client environment).
Now this DaaS idea is really hot when it comes to talking about stuff but nobody, not even the software vendors that sail on Windows's vast sea of users, is sure who wants to set up their own DaaS box and who just wants to get Windows by wire. In fact, Microsoft itself seems to be at sea, not because it hasn't studied its customers but because nobody, not even the big shots in Redmond, seem to be able to get good answers.
In the end, the industry's strategy might be shaped by a visionary not by the focus groups or industry watcher narcissists (those wankers) glowing with glory while proclaiming the obvious. It won't take a Steve Jobs, or at least it better not because we've had the only one and we aren't going to get another. But it will take somebody with considerable insight and the wit to express it well.
Glass house folk may find the rapidly evolving mobile device and thin client userverse unsettling, but actually it could turn out to be a cultural development that will make IT a lot easier for the people at server central. It could also turn out to be a very nice business not only for IBM (if it can get smart enough fast enough) but also for other major players in IT such as Hewlett-Packard, Dell, Oracle, Microsoft, Amazon, Apple, and Google. What the industry will eventually offer is not another dull idea, like Windows over the phone, but presentation with pizzazz, a way to affordably place a superior presentation layer in front of legacy as well as brand new applications. Call it Web 3.0 or something equally obnoxious if you have to, but it is important to recognize that presentation is already a very important aspect of computing and it will only become more so in the future.
The company that is probably the most desperate to participate in a new world of presentation (and possibly the least aware that it ought to be doing this) is Yahoo. This was one of the first dot-com companies to see that search as it was done in the old days was only a first step. But after trying to come up with what we call a portal, Yahoo lost its way. If it can get good geolocation services out of its marriage-of-necessity partner Microsoft it might be able to figure out how to combine cloud services and localization in a way that lets ordinary businesses add top notch presentation to their sales and support arsenals.
IBM is very well equipped to step in here, but it's been a long time since it has had a vision that includes stuff for the little guy. IBM's idea of small business is an organization with a thousand seats and certainly no more than ten thousand. Still, IBM has ample talent, technology, capital, and worldwide reach. It would be a shame to see it miss the chance to be number one in presentation services or at least a contender in a race with, say, Google and Amazon, for the chance to provide the business technology where the transactions will take place. For the most part, the companies using IBM servers and services would be happy to buy presentation from Big Blue, but if not they will surely buy it elsewhere. These days, companies have to be the ones for whom the smartphone rings . . . or they will be the ones for whom the bell tolls.
Post this story to del.icio.us
Post this story to Digg
Post this story to Slashdot