Not everyone loves the cloud: here's why

Not everyone loves the cloud: here's why

Summary: A step backwards, at least for personal computing, one observer fumes.


A lot of people are excited about the cloud, but not Josh Fruhlinger, as he rants over at Huffington Post.  A step backwards, he fumes, particularly as it relates to personal computing.

Security issues are one thing, but also consider the fact that going to the cloud also means going to the lowest common denominator in computing:

"What if your new computer has a faster processor and killer GPU that could make use of newer applications and workflows? Is logging into your 'old' cloud really pulling the most of of your new equipment's cycles? Ever notice, for instance, how your cable provider's set-top-box looks like it does a lot, but it's always hampered by clunky software that it runs on the cloud? Is that really the lowest-common-denominator performance we want out of our equipment?"

In his earlier days as a student working with a hiccuping campus computer system, Fruhlinger learned to back up everything he did on a floppy disk (anyone remember them?). "We've been here before, and I don't want to go back," he sniffs.

Anyone agree that we are taking a step backwards from a time when everything was available right off the drives of our own machines?

Topic: Cloud

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
  • RE: Not everyone loves the cloud: here's why

    Sort of. I have stated this time and time again the cloud has it's place in the computing world. However it is not a total catch all solution like the tech companys are trying to push. Email for instance is a good use of the cloud. But Google Amazon, Microsoft are all wanting your data and apps in there solutions. I think placing all of your eggs in one basket in insane and even if you use multiple providers management becomes an issue along with the fact that if it is a mission critical app to run your business processes.
  • Another dumbass question ...

    ... from a ZDNET blogger. I am growing tired of ZDNET posters who ask questions, or start threads, from positions which have been known for many years by even the newest to the IT industry. What I want is for ZDNET bloggers to start from the knowledge we have accumulated and the lessons learned, add some original thinking ... and move the debate forward.

    I do not want the tiresome 'M$ v MAC' flamebait, or the 'the network terminal is dead pronouncement'. And I don't want to hear that M$ recommends cloud computing or Steve Jobs recommends MAC's. I WANT TO MOVE FORWARD!

    As to Fruhlinger's expectation that his PC will be more powerful than the hardware used in cloud datacentres, well ... it will be when he does something trivial (like post to ZDNET) and it won't (when he requests a simulation that would have taken a week on a consumer PC).

    That will be the case ... if we ever discuss and solve the problems of architecture, vendor lock-in/costing , privacy and security ... instead of asking dumbass questions.
    • RE: Not everyone loves the cloud: here's why

      @johnfenjackson@... spot on. thanks.
  • RE: Not everyone loves the cloud: here's why

    No I won't move to the cloud, for a lot of reasons:

    1. The ISP's in my area have outages at least once a day, a PC from 10 years ago is more reliable than that. Note: I am in the Boston metropolitan area, which should have good service.

    2. I don't want to rely on a cloud vendor, worse yet I do business with some of the cloud vendors and I sure don't trust storing information from one vendor on another vendors cloud.

    3. Over the years I have used a number of on-line services, and been burned when they either close down or merge with another player.

    4. Finally, cloud security is still a joke. A PC with no virus proctection is more secure than many clouds.
  • Why the return to the glass palace

    In the beginning there were hot lumps. But then came mainframes. The mainframe era was characterized by centralized data centers where hardware was expensive, where the priests attending to the hardware were expensive, and where telecommunications costs -- although high -- were still low enough to make centralized hardware (with "remote job entry" via telecom) the near-universal answer.

    Then came microprocessors. These dramatically drove down the cost of CPU cycles, turning the economics of computing on its head. With microprocessors, it was cheaper to decentralize the hardware <i>and the data</i>, processing the data at remote locations (where the worker bees were) and submitting only crunched results via telecom to "headquarters."

    The whole thing with "client/server" was about minimizing telecommunications expense in a world where CPUs and small storage devices were cheap.

    The one ugly expense that was still there, and still pretty much centralized, were the people costs associated with keeping all this stuff running. Given their 'druthers, the IT people would still have preferred to keep the hardware and the storage in a centralized place. That was less about some Machiavellian notion of "control" than it was about making it easy for a small group of "priests" to minister to the hardware (and increasingly, the software).

    Now comes Act III, in which communications bandwidth becomes so cheap that the centralized/decentralized equation flips again, this time favoring big honking machines in giant data centers presided over by a small group of very highly skilled people. In this world, the "worker bees" lose custody of the data because moving it around is practically free. So now it goes into "the cloud," which is a way to avoid saying, "a big centralized data center in a glass palace."

    Some people observing this have concluded that this also means thin clients on the worker bee desktops. It probably should, but not yet. Bandwidth is cheap, but not so cheap that we can ship high-def video from The Cloud to every desktop on the planet.
    Robert Hahn
  • RE: Not everyone loves the cloud: here's why

    Heres something interesting to think about; if we look 10-15 years in the future, what do we think the world will be like as it relates to what we call cloud computing now (note, all the technologies (except for the XML processing language) are available now); <br><br>10-15 years from now the personal computer, as we know it, wont exist as it will be embedded in all parts of our daily lives. With RFID that is attached to us in some significant way (internally embedded, worn externally, etc.) these unseen computing resources (call them personal processing networks or P2N) will know who and where we are at all times (private and public spaces). There will be significant processing power locally (via home/apt/vehicle modules) that access various display/sensory communication devices as we encounter them. There will also be processing done individually in the form of watches, phones, game handhelds, etc. (call that Mobile P2PN or MP2N) that will be used for when we are in transition between P2N processing nodes (towers, modules, etc.) and will maintain the processing conversation so that we never notice any degradation or change in service. Most if not all of these MP2N and P2Ns will be accessed via voice or gesture but will also understand text and visual recognition (face/body/gait/movement, etc) and will be explicitly tailored to our individual requirements and consistent across all of themno matter where we are in the world.<br><br>Significant modular devices (SMD) will likely be associated with where we live (i.e. our home location/address for nationality purposes) but the processing that occurs will always be directly associated to us (via RFID). Its likely that the SMD will be supplied by the various telco carriers/cable companies as the set top box (STB) from them evolves into SMD (theyve captured at least the living room, so far) and the mobile modular devices (M2D) i.e. the phones and TVs and Bluetooth devices, etc. will be used to interact with us depending on the content and where we are in relation to those M2Ds. Simply put, the interactive experience will follow us throughout our day and be seamless as we transition from one device to another, from home to work, from work to play<br><br>Large infrastructure providers will evolve their services to include storage of our life code which is a synchronous, portable and globally partitioned object/relational database of our lives including everything about us (shopping habits, food preferences, financial and health records, etc.). This will allow us to interact with all of the various day-to-day physical service providers (i.e. Grocery and department stores, movie theatres, mechanics, dry cleaners, etc.) in a completely digital manner and in most cases automatically based on our previous behaviors.<br><br>There will grow up a huge broker middleman tier (think credit card processing services now) that basically stitch together all the various services we request (explicitly and implicitly) as we go about living. There will be some form of government collaboration if not outright management of the underlying infrastructure (hello Big Brother) that these brokers ride on top of in order to maintain personal privacy guarantees and public access of MP2Ns. It may be that the telcos/cable companies do the brokering but I doubt it as these storage and broker services (and federal support infrastructures) will be cross boundary (org, city, state and country) and necessarily built on open standard-based interaction/interface layers (i.e. no proprietary payoff) which will likely be an XML-based communication protocol that can also encapsulate processing languages to enable the seamless transition between nodes. <br><br>Of course, this is just one guys opinion
    • RE: Not everyone loves the cloud: here's why

      Dream on ...
      The telcos are already salivating at the prospect of being able to add niggling frustrations so that they can then bypass them for another fee.
      They drool at the prospect of selling you a service enhancement then charging a larger rental for its use, and a per occasion usage fee whenever you exceed some petty limit.
      And the reality will be that nobody enjoys the all singing, all dancing experience, but everybody wants it.
  • definitely

    PC+Cloud = Terminal+Mainframe.

    You are at the mercy of the guy at the end of the wire.

    Not for me!
  • RE: Not everyone loves the cloud: here's why

    In one word why I have no love: Sony
    put all my personal data on the cloud right for the taking.
  • RE: Not everyone loves the cloud: here's why

    It's a question of expectations. Of optimism and pessimism. An optimist will usually always be disappointed, while a pessimist will often get a nice surprise ;-)

    You need to be aware that the cloud will fail. If it's not virii, your software, your hardware, your router, your ISP, net traffic, it'll be the cloud vendor. This is not really a problem if you plan for it or don't use the cloud for mission critical applications.

    To use an example that has complex interactions take World of Warcraft. Yes a lot of work is done on the desktop, but the game needs the cloud to function so Blizzard needs to have servers running 24/7 for its 13 million subscribers. They have the money (subscriptions), the experience and the technology, but no-one really expects that the servers will be error free with no connection problems 24/7. I think their service is great - especially since they are providing far more than the database lookups that most "cloud" software is actually displaying - but in the end it's a game and if I can't get on at a particular time, it may be a pain. but it's not the end of the world or my business. The same can't be said if you rely entirely on cloud services for your business or critical activities.

    Don't put all your eggs in one basket is a very old saying, but that's because it's good advice ;-)
  • Cloud is absolutely useless

    - if you want to time-shift your TV shows. The bandwidth to allow every person in the town, much less the sate, nation, or world, to view a different show at the same time does not, and will not exist for the foreseeable future.

    - in major disasters. You lose the power nodes in your area, then you lose the transmission nodes too. You may lose the physical sites for the transmission nodes. Local computing, with local backup generators, will allow local use. As long as you're not buried by a pyroclastic cloud, vaporized by a terrorist nuke, had the building shaken into rubble, or been swept away by storm surge or tsunami, you can still operate on location without needed constant connection to the Cloud.

    - if you want to minimize attack routes to your data or applications. There are no cloud solutions that won't increase the vulnerability of your data to interception or corruption. If you have any connection to the internet, you will be hacked someday. But if all your data and applications are ON the internet, that day is coming sooner than later.

    - for maintaining your actual in-house network and workstations. SOMEBODY has to be either on site, or readily available to keep your stuff connected and working in the cloud. If you have more than a dozen nodal points in your facility, you're going to need on-site I.T. people. So where's the cloud benefit then?
    • RE: Not everyone loves the cloud: here's why

      "Time-shift your TV shows". This is a pre-cloud concept. If all TV is available on-demand, there is no scheduled time from which to shift it.

      There certainly is enough bandwidth for everyone in the world to watch a different show at the same time. One problem is that so much bandwidth (terrestrial, sattelite, OTA) is currently dedicated to broadcasting. Re-purpose that bandwidth for on-demand use to see amazing flexibility. However, the latter will happen very slowly since the industry is driven by the needs/fears of the advertiser; the desires/preferences of the consumer come a very distant third or fourth.

  • RE: Not everyone loves the cloud: here's why

    Yeah, just what I want to do... upload my most private data to some nameless, faceless server farm with no real accountability, with the only thing protecting me is some lame privacy policy that is unenforceable anyway.


    This "cloud" thing isn't new. There were these things called "servers" before this "cloud" buzzword came to be. Maybe you've heard of them? The difference then was that you had some measure of control over your data and the protection of that data.

    A "cloud" simply means that you don't.

    Anyone who falls for this "cloud" business deserves what they get. Just ask Sony. How's their cloud doing? Oh, and Amazon. That seems to be working out pretty well for them. And now Microsoft is betting the farm on "the cloud". That should tell you something. ;)
  • RE: Not everyone loves the cloud: here's why

    P.S. Here's a lesson from an old IT guy to any of you noobs out there... once something in IT becomes a "buzzword", RUN AWAY. Run away FAST. Run away FAR.

    It's a bubble waiting to burst. In IT, a "buzzword" is just like a fad... it's always temporary and is almost always created by people trying to make a quick buck on something that is unsustainable.

    Besides, we all know it's really about "Web 2.0" anyway, right kids? ;)