Slashdot Mirror

← Back to Stories (view on slashdot.org)

Get Ready for Rent-An-App

Posted by ryuzaki0 on from the stuff-to-read dept.

Baraka writes "Apparently MS is proposing a centralized, top-heavy system for delivering software applications in the future." It's kinda interesting: Web Applications in a way are kinda rent to own, and software licensing is so screwy that you don't really own it anyway. As irritating as it may sound, it would appear that application rental is coming... although not to my computer.

1 of 179 comments (clear)

  1. Balmer: The Network IS The Computer! See? TOLD YA! by maynard · · Score: 5
    This is basically what Balmer is saying, though I doubt he would cite the originator. What does Microsoft gain by this?

    Obviously a more controlled revenue stream, one which is predictable, and one which continues the advance of the Microsoft monopoly. You don't think you'll be running Windows and MS/Office off an X-Terminal, do you? No, it will be some proprietary protocol, encrypted, and damn difficult to reverse engineer (maybe even illegal to reverse engineer!). But the most important to Microsoft must be the controlled and even revenue stream such a system would foster.

    Microsoft is truly a house of cards right now. Here's a fascinating quote from last weeks Economist article on Share Options, Pg. 18, Aug 7th-13th, which illustrates this point:
    [Regarding overstated profits among high tech firms which hand out share options to their employees as a supposed salary benefit...] For instance, Microsoft, the world's most valuable company, declared a profit of $4.5 Billion in 1998; when the cost of options awarded that year, plus the change in value of outstanding options, is deducted, the first made a loss of $18 Billion, according to Smithers.
    Their stock is badly overvalued. They pay their employees poorly, over-work them, illegally hire so many temps that it's turned into a Seattle political scandal, and make up the difference with stock options for their core staff. This suggests that a sell off could be disastrous for the firm and it's employees.... hence it makes sense for Balmer and Gates to both sell off now (which they've been doing) and to look for a new revenue model which can assure higher profits in what will most certainly be a saturated market within ten to fifteen years.

    How should Free software and Open Source proponents fight this? We know how Microsoft is going to fight open protocols on the Net... by "de-commoditizing" (what an aberration of a word, clearly Vinod didn't have a dictionary by his side when he wrote Halloween I and II). And we know that they've already set the desktop standard with Win32/Windows. As long as they maintain control of the Office document standard through adding regular incompatibilities to thwart reverse engineers, they keep Windows installed on enough machines to make interopability with other systems a non-issue for the vast majority of users, and they succeed in keeping Win32 closed enough such that it can never be reverse engineered and re-implemented, they will continue to maintain their monopoly.

    The Justice Department may will the antitrust case, but by the time it's worked its way through the court systems, I'd argue that Win32 will have long since been rendered obsolete.

    This is why I think KDE, GNOME, GNUStep, and the like are a waste of time as far as attempting to take desktop sales away from Microsoft. Free developers may create a good MS-Office replacement like KOffice, or SIAG Office, but it won't ever read MS-Office documents properly just like Corel Office (Even under Windows!), ApplixWare, and StarOffice can't. Nor will a Freenix based desktop ever run exactly the same software (Wine may work well as a porting library, but I'm doubting its long term viability as a Win32 program loader -- MS will just change the underlying core OS enough to make they're new applications incompatible with Wine... it's a never ending game with no winners on the Wine side for this part of their project). Keyboard shortcuts are different, applications are different, they're all incompatible, and the current winner has a stake in maintaining this situation.

    So how do Freenix proponents win in this situation? In my opinion we can't win just be re-inventing the desktop wheel around X (or Berlin, for that matter). I think the Open Source community NEEDS a completely new approach, one which gives it the "killer app" advantage over Windows which will draw users not because of political issues over freedom, but because users will plain want said functionality.

    The next big revolution in user interface design is speech recognition incorporated directly into the systems interface with the computer. Something like what the Oxygen Project at the MIT media Laboratory is doing, with private funding I may add. (See this months Scientific American for a spread on the Oxygen project, the RAW CPU (a programable FPGA system), and their work with handhelds) I don't see any open source stuff from Oxygen... (does anyone know what their stance on Open Source is, and if they're currently accepting funds from Microsoft? If they are, who do you think will keep that codebase???) If you ask me, this is where Balmer wants to go. Everyone gets a handheld which is connected to the net via a radio/infrared networking, the system accepts speech input which is then passed to a server to interpret and resolve the problem, then passed back to the user in speech/video from the handheld. Such a system could be charged per minute, per query, or any of a number of other methods.

    This is what the free software community should be planning to implement. And most of the tools are already available... IBM's Viavoice is a good enough continuous speech recognition system -- though it's not free. It could be used as a module within GNU common lisp, which could then serve as a foundation for a new natural language systems interface based on an "expert system" which understands a simple enough grammer and converts this to UNIX commands, manipulating files, directories, and launching applications. If IBM never releases ViaVoice under an "Open Source" license, the fact that it's a module instead of being tightly integrated into the system means that Freeware developers could rewrite the recognition engine -- maybe with funding from the FSF, a university, or some other organization.

    Such a system would have to be tightly integrated with the desktop interface though, like common KDE and GNOME shortcuts, so that applications could know when to take speech input through their STDIN stream (like dictation) and know when you're attempting to give a command to either the application or the general operating system. Given GNOME's reliance on GUILE as it's "glue scripting language" it seems to me a bit closer toward integrating in a functional language with all its desktop applications, and thus being able to integrate natural language processing across the entire API suite. Just a guess though.

    I think this is where Freenix ought to go, and if it gets there before Microsoft, they will lose marketshare quickly. They may die a horrible death just because their financial situation is so precarious, but that won't do the Freenix community much good; they'll just file Chapter 11 and restructure -- with their monopoly intact. For the Freenix community to take the desktop we must provide an alternative which is easier to use, not just free.

    Of course, I may just be oversimplifying a very complex problem... Please feel free to resolve any problems of ignorance you may have noticed! :-)