The smartest dumb idea around

The smartest dumb idea around

Summary: every senior auditor I know is either losing sleep over his company's exposure to data theft arising from the use of laptops

TOPICS: Oracle

A few years ago some Sun people spent time trying to sell this idea to British Telecom and a half dozen other telcos. The spiel went something like this:


Let us provide both the Sun Ray for home use and the back end processing services, you guys provide the DSL services and do the billing.

That way, for a few pounds a month the customer gets a completely hassle free home computing solution: no need to worry about backups, viruses, worms, or learn anything about software installation and support. They turn it on, log in, and do whatever they want - email, web surfing, word processing - whatever.

You sell more DSL services, lock in your customers, and make a margin on our gear, our services, and our technologies.

More recently the word "solution" has become poisonous in IT related proposals and two new talking points have been added to the pitch:


  1. We can seamlessly connect that home user to a work environment like a Microsoft client-server set-up, securing both halves of the communication channel - from us to the Sun Ray, and from the user's office systems to us - while freeing the user from having to lug a notebook around and completely side stepping any risk of data loss due to notebook theft or damage.

    That works for the home user, but it also works for the business user. A subscriber can go anywhere there's a network connected Sun Ray and find his data -meaning his whole collaborative and software environment- ready and waiting for him at sign-on -with no risk of loss and nothing to carry.


  2. Sun Ray supports teleconferencing and VoIP applications. Go this route now, and you keep that customer as VoIP and video conferencing go mainstream.

There's a potential new one too: a handheld built using Sun Ray technology and radio based networking would be a better Blackberry - enabling the telco to bundle in mobile access without worrying about the patents held by other players like NTP and RIM.

The technology is proven. Sun did a careful pilot study for its own Sun Ray at home project, and the results have amply demonstrated both the technology and the combined security and convenience benefits.

As a result You'd think this idea would have been a big seller from day one -but it hasn't been. Instead, I think they've made it to third base a couple of times, but I'm not aware of any current or pending implementations.

So why not? Most of the people I know that have home computers think they're a hassle, feel that their network access provider over charges for relatively simple services, and complain to me (because they know I'm a Unix/Mac bigot) about the need to use Microsoft Word to work on office documents. Meanwhile every senior auditor I know is either losing sleep over his company's exposure to data theft arising from the use of laptops or off in a world of his own where reality does not impinge. So here's a proven solution to both sets of problems that nobody's offering to sell -and, meanwhile AOL still has an active customer base sending it a monthly fee for a lot less.

I don't know for sure what the deal breakers have been on this. I've heard that at least one telco's own data processing experts responded with a customer survey built along the lines of "You don't want to use a Sun Ray, do you?" while another deal is said to have collapsed over money because the telco demonstrated a rather over enthusiastic view of what its communications monopoly might be worth, but I don't see those explanations as sufficient.

What I think is that there have been two killer issues:

  1. first, it's just hard to sell a new idea to a telco, and particularly a new idea about home computing where the executives are less than knowledgeable about technology and the technologists they rely on for advice generally want to know nothing about Sun Rays;


  2. and, more importantly: "faster, better, cheaper" would make a great slogan for a technical Olympics, but there's no hot button issue there for either your average jaded telco technology advisor or his CFO boss.

What's needed is a better idea - something that would make a telco's senior management team sit up and pay attention.

So here's my suggestion to the people trying to make this kind of sale: go partner with one of the larger, more business visible, banks in whatever country you happen to be working in and have that bank approach the telco. That way you have bankers talking to CFOs on your behalf and the whole socio-tech thing doesn't get in the way of the deal.

So how do you get the bank on-side? Talk to their top people (not their technologists) about credit card fraud, the explosion in business to consumer internet sales, and the opportunity to step out ahead of competing card issuers.

The basic pitch is dirt simple: the home Sun Ray user does not need an expensive Java Card (few users actually do) and the card reader on the home Sun Ray can therefore be replaced with a credit card stripe reader.

That makes all credit card transactions originated on that device, "card present" transactions. Such transactions usually carry a discount rate (what the vendor pays the bank) on the order of 2% in the US and usually a bit higher elsewhere. In contrast most web purchases today qualify as card absent transactions for which the typical charge starts at 3%. Similarly per transaction fees start at around $0.20 in the US (usually higher elsewhere) for card present processes and around $0.30 for card absent transactions.

Individually the numbers are tiny - for example about a hundred million on Amazon's 8.5 billion in annual sales - but the opportunities they represent are important. For example, although transactions costs (and therefore losses to fraud) are built into the business framework now, the bank's default decision to pass the savings along will give it serious competitive advantages first in selling card processing services, and secondly in attracting and holding merchant accounts. Equally importantly, the overlap between the bank's existing customer base and that of the telcos involved will not be complete - meaning that the bank at least is likely to gain new customers while positioning itself as doing something effective about the trust barriers to e-commerce.

This is an idea in which everyone wins: the Sun Ray cost actually goes down a bit, the banks get new customers and new revenues, and the telco executives involved get a business reason to do what they should be doing: creating a steady new revenue stream for their companies by offering fully managed home Sun Ray services.

Topic: Oracle

Kick off your day with ZDNet's daily email newsletter. It's the freshest tech news and opinion, served hot. Get it.


Log in or register to join the discussion
  • Some thoughts

    1. The American Express "Blue" card IS a java card, and should work fine in the existing Sun Rays.

    2. The technology to make this work is not yet available! You need to prod the IEEE's butt to get Mobile WiMAX approved (almost 1yr since the "stationary" WiMAX standard was approved - which was about 2 years late). Using WiMAX with "meshing" networks would allow that SunRay portable to be constantly connected - at broadband speeds. So your network provider is the SAME at home and on the road - and your equipment all uses the same "parts" (economies of scale).

    3. What happened to the 1 million SunRays for China deal? Someone dropped the ball?

    4. The main problem with SunRay technology is that since a SunRay is a "dumb" box, there needs to be a "smart" box to make it all work. THAT means that to get your SunRay network working, you HAVE to RELY on SUN to provide servers/service. With a share price less than a McDonald's Happy Meal (TM), Sun doesn't look like the "right" player today. IBM could pull this off, but the cost would be very high. HP would be PERFECT - right price, big enough - but Hurd doesn't seem to have the vision of the future. HP and Sun should merge - give Sun control of the (RISC) server-side stuff, while HP keeps its strengths in printers and x86 stuff. Who ends up with Linux, I don't know - they BOTH suck at it. This would help Sun get respect, and help HP get OUT of the Itanic stuff . . .
    Roger Ramjet
    • What happened to the 1 million SunRays for China ? - see tomorrow

      Been reading my mind again? part of the answer is in
      tomorrow's blog...
    • An additional Point 5

      5. Who want's it?

      I'm only planning to accept this idea when hell freezes over. I don't want a Telco to tell me what I can and cannot do on MY system. Which brings me to point 6:

      6. How about performance. Probably goes something like this:

      -Put in your smartcard and enter your password to start the login. Probably takes about 5 minutes to load your desktop, even more when the internet is 'bussy'.
      -Select an application to start working. Takes another 2 minutes.
      -Load a file which you're working on. Takes, depending on size, another minute or more.

      My old 80086 PC is even faster!
      Arnout Groen
      • Not my experience

        Actual boot time for a Sun Ray is about 4 (four) seconds - of course telco's can (and will) screw this up, but the bandwidth needed isn't great so
        they might stretch it to 5-8 seconds.

        Actual performance tends to be better than PCs. Loading 1,000 emails, for example, is pretty much
        instant with a V890 backend and won't be much affected by the number of users until that gets well into the multiple hundreds. Again, this isn't terribly bandwidth limited and so relatively immune to telco mismanagement.

        Other functions are generally similar because you don't transmit the whole file. Load a 20MB Excel file and you transmit what you see - a couple of pages at a time: not a big deal, even at adsl rates.

        Sun's own experience with this has been pretty positive - despite having to deal with numerous telcos in the trial.
        • Sound and Video

          Ok - Can I play an audio file, a high quality (and high compression) video file, and use applications with significant amounts of bitmap type graphic content (like, say, browsing the web shopping for clothes) all at the same time?

          Any remotely modern PC can do this without any trouble. My wife does it all the time on her 4 year old thinkpad. I have no doubt that the processing part of the equation can be solved...but if you have a 1.5Mbit downstream 256kbit upstream connection I find it hard to believe that all that information could be streamed over it in an acceptable manner. 1.5Mbit is about enough for good video, assuming you have a processor to decompress it.
          • SunRay 1 couldn't

            I was thinking the very same thing when I purchased my SunRay1 setup (4x360Mhz 1Gb Sun AXMP server). SLOW AS MOLASSES! Couldn't do very much with my SunPCI stuff either (2 x SunPCI2 cards). Don't know if SunRay1G + faster server would do much better. (Oh yeah, all of this on a local 100Mb network).
            Roger Ramjet
          • SR1 wasn't great..

            but yours sounds like a set-up problem. Try it with Sol10/SR3 software and the 1g.
        • I'm not reading a lousy 1000 emails...

          I'm working with a 'tiny' SQL database.

          And guess what... Me and my colleages need to send 85.000 xml files (smallest quantity) of about 750kb to this database (approximately 61GB)on a regular bases, which were doing from our homes, because we can't do it during working hours.

          Because we have cut the link between the receceiving database (which by the way is under development) and the sending application, this means we have to load the messages to our local drives, before sending them to the database...

          Doing this in the way you described means a lot of frustration, especially when Telco's (can and they will) stretch up the bandwith, because of the money
          Arnout Groen
          • So you're not the target market

            What percentage of PC users look like you from a
            mail perspective? Virtually none, and those that do can (or at least should) have their own Unix workstations.
          • What about video?

            Videos are bigger that Arnout's files. And they're part of a "normal" business environment...

          • Video works reasonably well

            In short bursts, at least. YOu get network degradation when lots of people using the same wires get involved concurrently -i.e. 30 teleconference users in a 100BaseT world. That actually doesn't really apply to a home DSL link - assuming the service is fast enough at all, it's not shared with anyone and so should not degrade.

            Cable users, however, suffer from this because their neighbours share the bandwidth).

            Servers, and telco switches, on the other hand...can be underpowered and often are -but that's not an artifact of the technology and not a necessary consequence of a telco/sun deal.
      • "My old 80086 PC is even faster!"

        But your new Vista PC won't be!
    • Thick or Thin

      Just another example of the same old argument - thick or thin? Sun and many others long for the days of centralized computing - as in leasing time on a mainframe and using a dumb terminals for access. Everyone in Telco read, "The Rise of the Stupid Network", where in the intellegence is placed at the edge and not the center. They didn't get it then and I doubt they have yet but they do understand how much it costs to support devices connected to their network.
  • What's mine is mine.

    What's AT&T's is AT&T's.

    The expression, "I'm from the government and I'm here to help you." is not used to show how to gain immediate trust.

    The expression, "I'm from AT&T, and I'm here to take over your computer." would probably have a similar effect.

    And at a Telco, the expression, "I'm from Sun and I'm here to take over computing from all your customers." might provide an extraordinary diversion on an otherwise ordinary day.
    Anton Philidor
    • Partly agree, re businesses

      I feel the same way about my personal and business data and the ability to handle it

      I have to admit that a partial answer (especially for businesses) would be to use at least two such services. Ideally via different cables / radio etc. With some sort of automatic data mirroring.

      Failing that automatically and immediately have all data mirrored on a totally separately owned and managed web site, thus accessible from a normal PC etc. from anywhere.

      In any case have an intelligent hard drive / flash drive or two plugged in beside the Sun Ray.

      It is all too easy for a supplier to mess up and then claim that you haven?t paid them / you are not who you say you are / you abused their terms / you said you were quitting etc. etc.

      Paul - I?m not saying that there are not many merits to discuss, and it might work well for a lot of people, especially casual home users who mainly just wanted email, and a bit of browsing.

      But I?m a little surprised to hear the idea from someone who has just argued against outsourcing??!!

      No need to discuss the "incompetence" issues, but don?t so many of the security-type issues also apply to it?

      E.g. No Axe?s point about how hard someone else would fight legally to prevent access to one?s data.

      Back to the idea of selling it via banks perhaps? - we could be offered the chance to buy various types of and levels of insurance!
      • Yes, there are issues with out-sourcing here

        I'm thinking of this as a processing service for home users - and a connectivity service for businesses. Does grandma need to own her files? probably not and so this model gets her a working solution for which she pays a monthly and needs no expertise.

        Does a business need to own its files? you bet. SO this service acts to connect the remote user to systems run by (at) that business and the user relies on the company's own IT people to maintain the data.

        Now, I think there is a slippery slope here leading to an ASP model for remote business appplications. Some people will want to sell this, others to buy it; but not me. To me a system like this should do what it's good at: dial tone quality home e-communications: web, mail, voice, dollars -and nothing else.
        • Even grandma appreciates her privacy.

          And younger generations often have things on their harddrives they'd prefer not to have recorded and available off-site.

          Google has become identified with this problem because that company has been working to increase revenues with targeted ads. But trust for the local Telco or Yahoo is no greater.

          For many people off-site mail addresses are provided when the inquirer is not trusted.

          And that's leaving out the issue of modifying software. What happens to use of third party software to modify operation of general purpose software?

          Basic functionality is insufficient for most people, at work or at home. That's why standard issue software is insufficient.
          Anton Philidor
          • Privacy, security, hassle, redundancy

            Anton you're right re privacy, for many people

            But Granny should already worry about her phone, email etc. communications being spied on. This proposal should leave her more secure from "unofficial" hackers. If she doesn't have concerns about government etc. monitoring, this might well suit her.

            Privacy is, for good reasons or bad, a concern of many. But worrying and hassle and limitations due to hacking, malware, rootkits, updates, etc. are a strong counterbalancing concern, and these things also often themselves have privacy implications.

            Many parents would be happy for their children to be restricted to a service that the children are wary of, re privacy, knowing that it was indeed recorded and available off-site.

            And there are many households which include a sufficient number of users to make Paul's suggestion a useful part of the story. For greater privacy and / or more than basic functionality they can use additional boxes and connections.

            Paul - re business, for remote users to dial in, fine. Your diagram shows VPN. Presumably very strong encryption could now be provided by some sort of modern "different code every time" gizmo which banks are moving to.

            And no harm in firms having one or more of these things, (suitably controlled as regards access to the main stuff) around as an alternative in an emergency, and for use by guests and employees in their own time.

            What a balancing act it all is!
    • More important, court ordered discloser.

      So assume you have written something "private" that you want to remain private. As we are seeing with Google, all it takes is a court order and your privacy is pretty much meaningless.

      I see the lawyers having a field day with this.
  • You've got your title wrong

    Its the dumbest dumb idea