Microsoft, invention and innovation

Discussion in 'NZ Computing' started by Brett Roberts, Oct 7, 2005.

  1. Brett Roberts, Oct 7, 2005
    #1
    1. Advertising

  2. Brett Roberts

    shannon Guest

    Brett Roberts wrote:
    > There are regular discussions on nz.comp regarding Microsoft and innovation
    > (or alleged lack thereof). This series of blog posts by John Carroll (a
    > Microsoft employee working in our IPTV group) are, IMHO, interesting reading
    > and more rational than some of the other commentary I've seen on the
    > subject:
    >
    > http://blogs.zdnet.com/carroll/?p=1495
    >
    > http://blogs.zdnet.com/carroll/?p=1496
    >
    > http://blogs.zdnet.com/carroll/?p=1497
    >
    >
    > Brett Roberts
    > Microsoft NZ
    >
    >


    Yum, astroturf.
    shannon, Oct 7, 2005
    #2
    1. Advertising

  3. Brett Roberts

    Steve H Guest

    On Fri, 7 Oct 2005 14:19:28 +1300, Brett Roberts wrote:

    > Brett Roberts
    > Microsoft NZ


    i hope you got your flame pants on because you are making too much sence
    and we all know the members of the oss circle jerk group dont like common
    sence.

    I for one agreee with what John is saying.

    ---------
    Steven H
    Steve H, Oct 7, 2005
    #3
  4. Brett Roberts

    Shane Guest

    On Fri, 07 Oct 2005 15:04:52 +1300, Steve H wrote:

    > On Fri, 7 Oct 2005 14:19:28 +1300, Brett Roberts wrote:
    >
    >> Brett Roberts
    >> Microsoft NZ

    >
    > i hope you got your flame pants on because you are making too much sence
    > and we all know the members of the oss circle jerk group dont like common
    > sence.
    >
    > I for one agreee with what John is saying.
    >
    > ---------
    > Steven H


    *lights up flame thrower mark 3 powered by Unix * :p
    --
    Hardware, n.: The parts of a computer system that can be kicked

    The best way to get the right answer on usenet is to post the wrong one.
    Shane, Oct 7, 2005
    #4
  5. Brett Roberts

    AD. Guest

    On Fri, 07 Oct 2005 14:19:28 +1300, Brett Roberts wrote:

    > http://blogs.zdnet.com/carroll/?p=1497


    I'm not one of those innovation deniers, but I had to take issue with the
    major thrust of this one.

    The praising of MS end to end consistency vs the way open source projects
    generally work is short sighted IMO. It seems a flagrant attempt at self
    serving vendor lock in, and an excuse to not bother with any open
    standards or attempts at interoperability.

    I would rather have a loosly couple set of applications that can all work
    together well. That way I'm free to simply change out any one part if it
    no longer meets my needs. If I've gone and locked myself into a tall stack
    of end to end integrated and tightly coupled products from one vendor, I
    just have to suffer in silence or go through huge disruptive expense
    changing the whole stack.

    --
    Cheers
    Anton
    AD., Oct 7, 2005
    #5
  6. Brett Roberts

    Steve H Guest

    On Fri, 07 Oct 2005 15:13:06 +1300, AD. wrote:

    > On Fri, 07 Oct 2005 14:19:28 +1300, Brett Roberts wrote:
    >
    >> http://blogs.zdnet.com/carroll/?p=1497

    >
    > I'm not one of those innovation deniers, but I had to take issue with the
    > major thrust of this one.
    >
    > The praising of MS end to end consistency vs the way open source projects
    > generally work is short sighted IMO.


    so you agree with the fact that the way open source project generally work
    (inconsistancy) is a bad thing.

    hell look at the lack of consistancy with how configeration infomation is
    stored on linux baised systems, all the diffrent windowing api's, all the
    diffrent system api's - it makes for developing on the platform hard as
    hell.

    i know windows isnt perfect, but at least there is a single store for
    configeration, there is a single API for how to get it, and there is a
    single windowing API.

    with the emmergence of .net the correct statement to make is 'a simple
    single api'.

    > It seems a flagrant attempt at self
    > serving vendor lock in, and an excuse to not bother with any open
    > standards or attempts at interoperability.


    ffs your contradicting yourself, the whole point of 'end to end
    consistency' is interoperability - how the hell could anything talk to
    everything else if it wasnt there.

    as for open standards, what do you think .net (at least C# as defined in
    ECMA) is all about, its not as if you have to pay for it so what is the big
    deal.

    > I would rather have a loosly couple set of applications that can all work
    > together well. That way I'm free to simply change out any one part if it
    > no longer meets my needs.


    so you want to be able to swap out any data, user, middle, busness, ws, web
    lyer

    in the world i live in its called n-Tier architecture and its considered
    normal to employ it.

    > If I've gone and locked myself into a tall stack
    > of end to end integrated and tightly coupled products from one vendor, I
    > just have to suffer in silence or go through huge disruptive expense
    > changing the whole stack.


    what bs, no instead you have to suffer the pain of hammering all the
    diffrent libary developers in order to get something fixed in your favorate
    oss program. thats if you know what libary is causing the problems.

    you have to bite your toung as some pimply faced 'card carrying oss circle
    jerk member' arrogantly flames you in private only to deny it in public.

    you have to wait years for something considered 'normal' on one platform to
    be supported and even then its a hack.

    thoes are my experiences with oss software that is 'actually' good and
    actually deserves the right of being called cross-platform. so god help the
    users of the other 90% of oss binary shit thats out there.

    --------
    Steven H
    Steve H, Oct 7, 2005
    #6
  7. Brett Roberts

    AD. Guest

    On Fri, 07 Oct 2005 15:32:12 +1300, Steve H wrote:

    >> The praising of MS end to end consistency vs the way open source
    >> projects generally work is short sighted IMO.

    >
    > so you agree with the fact that the way open source project generally work
    > (inconsistancy) is a bad thing.
    > hell look at the lack of consistancy with how configeration infomation is
    > stored on linux baised systems, all the diffrent windowing api's, all the
    > diffrent system api's - it makes for developing on the platform hard as
    > hell.
    >
    > i know windows isnt perfect, but at least there is a single store for
    > configeration, there is a single API for how to get it, and there is a
    > single windowing API.


    Saying there is a single store for configuration (eg the registry) in
    Windows is just as accurate as saying there is a single store for
    configuration in Unix (eg the /etc directory). The actual information in
    the registry is no more consistent between applications than the
    information in the config files.

    A single API for accessing it? Sure on Unix its called POSIX system calls.

    A single Windowing API? For each version of Windows perhaps. That new one
    coming in Vista looks very different. And there are just as many windowing
    toolkits available for Windows as there is for X.

    >
    > with the emmergence of .net the correct statement to make is 'a simple
    > single api'.


    How many of these APIs has MS gone through in the last decade or so though?

    There is always some sort of new architecure just around the corner. Where
    is the consistency in that?

    There was DDE, OLE, OLE2, COM, COM+, DCOM etc DNA, .NET etc. For database
    access there was ODBC, RDO, DAO, ADO, OLEDB, ADO.NET etc. ASP to ASP.NET,
    VB6 to VB7 etc. And then there's all the new APIs coming along with Vista
    for just about everything. It's just continual churn.

    Open source might be inconsistent in space (ie between competing
    projects), but MS is inconsistent in time. What I mean is that once you've
    choosen a set of open source infrastucture it will stay consistent for a
    lot longer than the MS stuff will stay the same. Development in MS shops
    seems like an endless treadmill of updating your stuff to the API du jour.

    > ffs your contradicting yourself, the whole point of 'end to end
    > consistency' is interoperability - how the hell could anything talk to
    > everything else if it wasnt there.


    Where is this interoperability in practice then? It is a very small
    percentage of Windows apps that freely share data with each other.

    > as for open standards, what do you think .net (at least C# as defined in
    > ECMA) is all about, its not as if you have to pay for it so what is the
    > big deal.


    So? What does a language spec have to do with getting different apps to
    interoperate?

    >
    >> I would rather have a loosly couple set of applications that can all
    >> work together well. That way I'm free to simply change out any one part
    >> if it no longer meets my needs.

    >
    > so you want to be able to swap out any data, user, middle, busness, ws,
    > web lyer
    >
    > in the world i live in its called n-Tier architecture and its considered
    > normal to employ it.


    So in a typical end to end MS based n-Tier architecture with COM or .NET
    all the way through - how much scope is there to swap components in and
    out? You can't change the OS, you can't change the web server, and in lots
    of cases you can't change the browser either. You don't really get much of
    a look in with changing the app server either - it's part of the OS too.
    Changing the database ain't bad, but that usually because it isn't as
    tightly coupled into the stack with COM or .NET. The database in most
    cases is loosely coupled via SQL.

    >> If I've gone and locked myself into a tall stack of end to end
    >> integrated and tightly coupled products from one vendor, I just have to
    >> suffer in silence or go through huge disruptive expense changing the
    >> whole stack.

    >
    > what bs,


    How is that BS? What would your options be?

    > no instead you have to suffer the pain of hammering all the
    > diffrent libary developers in order to get something fixed in your
    > favorate oss program. thats if you know what libary is causing the
    > problems.
    >
    > you have to bite your toung as some pimply faced 'card carrying oss circle
    > jerk member' arrogantly flames you in private only to deny it in public.
    >
    > you have to wait years for something considered 'normal' on one platform
    > to be supported and even then its a hack.
    >
    > thoes are my experiences with oss software that is 'actually' good and
    > actually deserves the right of being called cross-platform. so god help
    > the users of the other 90% of oss binary shit thats out there.


    The software company I work has over the years been shifting various
    underlying libraries and infrastructure to those from open source
    projects. We have found the open source projects far more responsive and
    quicker to solve problems than the commercial alternatives we used to use.

    I don't know what attitude you've taken to deserve treatment like that,
    but in our experience that is far removed from the norm.

    --
    Cheers
    Anton
    AD., Oct 7, 2005
    #7
  8. Brett Roberts

    thingy Guest

    Brett Roberts wrote:
    > There are regular discussions on nz.comp regarding Microsoft and innovation
    > (or alleged lack thereof). This series of blog posts by John Carroll (a
    > Microsoft employee working in our IPTV group) are, IMHO, interesting reading
    > and more rational than some of the other commentary I've seen on the
    > subject:
    >
    > http://blogs.zdnet.com/carroll/?p=1495
    >
    > http://blogs.zdnet.com/carroll/?p=1496
    >
    > http://blogs.zdnet.com/carroll/?p=1497
    >
    >
    > Brett Roberts
    > Microsoft NZ
    >
    >


    hmmmm spamming.......

    Blogs mean diddly squat, just another marketing avenue....

    regards

    Thing
    thingy, Oct 7, 2005
    #8
  9. Brett Roberts

    thingy Guest

    Steve H wrote:
    > On Fri, 7 Oct 2005 14:19:28 +1300, Brett Roberts wrote:
    >
    >
    >>Brett Roberts
    >>Microsoft NZ

    >
    >
    > i hope you got your flame pants on because you are making too much sence
    > and we all know the members of the oss circle jerk group dont like common
    > sence.


    yeah.....yeah.......

    Try and sort out the difference between marketing and reality. ie most
    stuff MS posts is spin doctor speak, so called factual papers, biased
    FUD, blogs by MS, more spin doctor speak by another avenue.

    > I for one agreee with what John is saying.


    That's cool, you are allowed to.

    ;]

    > ---------
    > Steven H


    While the blogs could well be genuine, given MS's apparent BORG like
    control it may well be it has been passed via 3 lawyers and PR
    first....and then finallt BG or SB.

    I think a fair number of press (eg www.theregister.co.uk) and readers
    have now got to the stage that MS lacks credibility in their eyes so
    anything coming from MS land is treated with suspicion. While there are
    true believers who may well take anything MS says as un-biased truth, at
    least for myself I look to prove what has been said, others seem to be
    doing such critical analysis as well (and not just MS, Sun, Oracle, CA
    as just examples).

    regards

    Thing
    thingy, Oct 7, 2005
    #9
  10. Brett Roberts

    Dumbkiwi Guest

    On Fri, 07 Oct 2005 14:19:28 +1300, Brett Roberts wrote:

    > There are regular discussions on nz.comp regarding Microsoft and innovation
    > (or alleged lack thereof). This series of blog posts by John Carroll (a
    > Microsoft employee working in our IPTV group) are, IMHO, interesting reading
    > and more rational than some of the other commentary I've seen on the
    > subject:
    >
    > http://blogs.zdnet.com/carroll/?p=1495
    >
    > http://blogs.zdnet.com/carroll/?p=1496
    >
    > http://blogs.zdnet.com/carroll/?p=1497
    >
    >
    > Brett Roberts
    > Microsoft NZ


    Brett,

    An interesting series - self serving, but interesting.

    The definition of innovation in the blog seems to be "incremental
    advancement of existing ideas and technologies". This is a fair enough
    definition. However, in my view this is a strength of the open-source
    world, not of the proprietary world. In the proprietary world you have to
    rebuild the base technology from scratch before you can increment, rather
    than re-use existing code. Furthermore, there is more scope in the
    open-source world to scrutinise and learn from existing code therefore
    speeding up the process of incremental improvement. Some would say that
    the ability to stand on the shoulders of giants is the very purpose of
    open source. So nothing new here.

    Another point made in the third blog is the innovation of providing end to
    end solutions. He challenges the ability of open-source to match that. I
    think this point has some validity in some circumstances. I also think
    the analysis fails in a number of others. One project where it fails I
    think is kde. This is a massive project with a huge number of
    geographically, and philosophically disparate contributors, but they have
    developed technologies for consistent and rapid development that is
    outstanding. The core technologies (for example kparts, kio slaves, file
    dialogs, consistent widgets, theming, dcop) are built into the core
    libraries and are available to all applications. Furthermore, and far
    more importantly than the prior point, is that unix programmers, as I
    understand it, generally write their applications with inter-application
    communication in mind - they are designed to take input from other
    applications, and output it to yet another - the idea of pipes and
    redirection. dcop is just an extension of that philosophy. So while
    developers may only work on a few isolated projects, generally, the
    applications are designed with inter-application communication and code
    re-use in mind, thus allowing those that follow to use the power of those
    "isolated" projects within their own applications. this is a unix
    philosphy (ie linking small simple applications together in new (and
    unanticipated ways) to create something powerful that is "end to end" for
    THAT user) that has been around for a long time. The ability for each
    user to build such an end-to-end solution for him/herself is a far more
    powerful, and prone to greater innovation, than a monolith
    developer dictating to its user-base a generic "end-to-end" solution for
    a perceived problem that only approximates the actual problem that each
    particular user has.

    I have to say the example of consistency that is given (ie end-to-end
    media framework), to me, smacks of vendor lock-in, and monopoly abuse.
    However, I suppose from the inside (of MS) it could seem like innovation,
    where such things are a business strategy, and the norm - take the
    Massachusetts situation as an example. Granted, Office provides an end to
    end solution, but if it's not interoperable outside the MS platform, then
    it will fail to meet what appear (at least to me, and a lot of others) to
    be sensible interoperability, and openness requirements.

    I'm sorry, Brett, but this just looks like astro-turfing, and doesn't
    actually advance MS's argument very far. MS through its development
    model is not doing anything more innovative than any other software
    development model, and in my view, open-source is far better positioned to
    innovate by incremental improvement, and to provide end-to-end
    solutions for each user's particular needs, faster than MS's monolithic
    proprietary approach.

    Just some thoughts.
    Dumbkiwi, Oct 7, 2005
    #10
  11. Brett Roberts

    Peter Guest

    Brett Roberts wrote:
    > There are regular discussions on nz.comp regarding Microsoft and
    > innovation (or alleged lack thereof). This series of blog posts by John
    > Carroll (a Microsoft employee working in our IPTV group) are, IMHO,
    > interesting reading and more rational than some of the other commentary
    > I've seen on the subject:
    > http://blogs.zdnet.com/carroll/?p=1495
    > http://blogs.zdnet.com/carroll/?p=1496
    > http://blogs.zdnet.com/carroll/?p=1497


    At least he truthfully admits that software developments are built on
    earlier work by others. This is common to all academic, technical,
    artistic developments. Newton termed it "on the shoulders of giants".

    However, Carroll doesn't admit that MS is doing its best to prevent such
    innovative development. He ignores that MS uses a variety of means
    (patents, closed source software, EULA, legislation, DRM) to prevent and
    obstruct such development. MS does this for its own selfish reasons, and
    in doing so, sucks value from their customers and from society in general.

    Carroll also states there should be a standard that works everywhere. I
    agree. Standards are an established means of ensuring interoperability, and
    the concept was used in lots of fields before software came along. But
    this is another area where MS is doing their best to prevent innovation.
    They use many means to obstruct attempts by others at achieving
    interoperability with files produced by MS apps.


    Open source enables anyone and everyone to use, explore and build on the
    technology. Conversely, proprietary source proponents like MS seek to
    prevent anyone from understanding how their software works, and stop them
    modifying or reusing it, and they want to restrict its use to those who can
    afford to pay the inflated fee.

    MS is especially innovative and successful at their style of business, using
    unlawful, unethical and selfish methods to stifle markets, wipe out
    competitors and rip off consumers. Ultimately this is to the detriment of
    society and to technical progress.


    Peter
    Peter, Oct 7, 2005
    #11
  12. Brett Roberts

    Steven H Guest

    Hello AD.,

    warning: long as hell post

    > Saying there is a single store for configuration (eg the registry) in
    > Windows is just as accurate as saying there is a single store for
    > configuration in Unix (eg the /etc directory). The actual information
    > in the registry is no more consistent between applications than the
    > information in the config files.


    diffrent distro's store diffrent stuff it in diffrent places

    > A single API for accessing it? Sure on Unix its called POSIX system
    > calls.


    yea bad example, the point i was making that diffrent distro's store the
    same values in diffrent places. now if the 'POSTIX system calls' know exactly
    where a value is (independant of distro) then that is an achievement.

    you still need to know the path to the value on diffrent distro's

    > A single Windowing API? For each version of Windows perhaps.


    minor correction, you use the same api's created in 95 than you do in xp
    - CreateWindow[Ex], the mesage pump is still the same, in fact all the windowing
    api's that were in 95 are in XP.

    newer ones have been added through the years (renember when windows would
    'steal focus'), and api's have been changed.

    > That new
    > one coming in Vista looks very different.


    quite right, no more old fangled win32 api shyte to deal with - its all managed

    there will be a win32 'compatabilaty layer' so that older windows apps will
    still run in vista - this is Microsoft were talking about, who retain as
    much compatabilaty for stuff like this as possible. Not Apple who in their
    arrogance dont give much of a shit about older apps running on newer versions
    of their operating systems.

    > And there are just as many
    > windowing toolkits available for Windows as there is for X.


    you have 'wrappers' for win32 yes (mfc, and whatever 'object orientated wrapping'
    other langages provide) but there is only one windowing api - correct me
    if iam wrong (does happen) but doesnt mandrake have a diffrent windowing
    api than say some other desktop manager

    this is the lack of compatabilaty that does nothing but tie in developers
    to a specific windowing platform.

    >> with the emmergence of .net the correct statement to make is 'a
    >> simple single api'.
    >>

    > How many of these APIs has MS gone through in the last decade or so
    > though?


    where is the consistancy in 5 or so linux distributions that have managed
    to survive over the past say 10 years - how many breaking api changes

    in the past say 10 or so years microsoft has made very few major fuckups
    when it came to breaking api's

    >
    > There is always some sort of new architecure just around the corner.
    > Where is the consistency in that?


    there will always be some new architecture just around the corner, even if
    that corner is 2, 3 mabye 5 years away. its the job of a software developer
    to keep up with it all or loose their income.

    as for consistancy, if you havent noticed differing architectures over the
    past several years have had one main purpose, gradual abstraction (thus simplification)
    from the metal in order to speed up the development process and to allow
    more complex systems.

    > There was DDE, OLE, OLE2, COM, COM+, DCOM etc DNA, .NET etc. For
    > database access there was ODBC, RDO, DAO, ADO, OLEDB, ADO.NET etc. ASP
    > to ASP.NET, VB6 to VB7 etc.


    the world of software development is not static it is continually evolving,
    each of thoes 'standards' you list above had its place in the time it was
    in active use and no doubt within the next 5 or so years .net (more or less
    the current standard) will evolve into something else

    older technologies will still have their place - even if their just maintained.

    > And then there's all the new APIs coming
    > along with Vista for just about everything. It's just continual churn.


    winfs, new windowing system ... is that about it ?

    the windowing system simply *had* to change, its ugly (screen scrape), wastefull
    and its about damn time. winfs (bill gates baby) i dont really see its use
    apart from file system abstraction for all the knuckle dragers out there.

    > Open source might be inconsistent in space (ie between competing
    > projects), but MS is inconsistent in time. What I mean is that once
    > you've choosen a set of open source infrastucture it will stay
    > consistent for a lot longer than the MS stuff will stay the same.


    > Development in MS shops seems like an endless treadmill of updating
    > your stuff to the API du jour.


    stuff written in older langages does tend to stay in older langages unless
    there is a 'actual need' for it to be re-written. why? because re-writing
    already established code (however crusty) is expensive as hell.

    also what tends to happen is that projects tend to be 'converted' to newer
    api's and *over time* the older logic is enhanced to use the facilatys that
    the new api provides.

    that way you can push out new releases as you go, while maintaining old (proven)
    code & adding new features.

    >> ffs your contradicting yourself, the whole point of 'end to end
    >> consistency' is interoperability - how the hell could anything talk
    >> to everything else if it wasnt there.
    >>

    > Where is this interoperability in practice then? It is a very small
    > percentage of Windows apps that freely share data with each other.


    mostly because there is little need for it.

    consumer type apps are *generally* input, output (word-processing, email,
    spreadsheets) and dont require a high level of interaction between them.
    on rare-cases when interoperability between them is required it can be done.

    interoperability is mainly usefull in buisness applications when one 'service'
    needs the 'service' of another, they 'convey infomation' in a common format
    (say xml) and thus achieve interoperability. as long as the services are
    written in such a way that little 'provider specific' logic is required (say
    xml serialized .net datasets - a bitch for interoperability) you can have
    diffrent services written on diffrent architectures talk to one another.

    as for interoperability in file formats, I have always thaught the likes
    of govt, public agencies should use an independant (non binary) format even
    if it means 'missing out' on provider specific fungus.

    >> as for open standards, what do you think .net (at least C# as defined
    >> in ECMA) is all about, its not as if you have to pay for it so what
    >> is the big deal.
    >>

    > So? What does a language spec have to do with getting different apps
    > to interoperate?


    i forget the point i was trying to make (if i renember i will come back to
    it)

    >> in the world i live in its called n-Tier architecture and its
    >> considered normal to employ it.
    >>

    > So in a typical end to end MS based n-Tier architecture with COM or
    > .NET all the way through - how much scope is there to swap components
    > in and out?


    depends on how much abstraction you code into it, just like any other platform.

    > You can't change the OS,


    ..net works on *nix - there are (and always will) be specific issues for provider
    (microsoft.* namespace) specific logic as it has to be re-coded, the likes
    of Mono are doing just that and have been reasonably successfull in doing so.

    if a buisness is willing to put up with the risk of using a non-supported
    os then they can go crazy, if/when microsoft do choose to open up their provider
    specific namespaces for the likes of *nix then it is their decision to do so

    > you can't change the web server,


    ahem mod_aspnet or whatever it is called, again if you want to use it go
    for it.

    but as we all know buisness dont tend to go for this sort of thing, they
    like to settle for a providers 'total package' that way they get support.

    > and in lots of cases you can't change the browser either.


    bs, if your wanting your web-app to be 'cross-browser' then do what all web-devs
    do and code for it !

    > You don't
    > really get much of a look in with changing the app server either -
    > it's part of the OS too.


    app server ???

    > Changing the database ain't bad, but that
    > usually because it isn't as tightly coupled into the stack with COM or
    > .NET.


    personally i prefer to code against System.Data.SqlDB, but if i really wanted
    to get wanky i could code against system.data.oledb and have its many abstraction
    layers talk to a shyteload more db's

    but most people would rather use provider specific code over sqldb because
    as you would guess provider specific logic is always going to be faster than
    a set of highly abstracted libarys.

    >>> If I've gone and locked myself into a tall stack of end to end
    >>> integrated and tightly coupled products from one vendor, I just have
    >>> to suffer in silence or go through huge disruptive expense changing
    >>> the whole stack.
    >>>

    >> what bs,
    >>

    > How is that BS? What would your options be?


    bs mabye too strong: 'not quite agree' would be better (yes i know i said it)

    if you are using a 'tall stack of end to end integrated and tightly coupled
    products from one vendor' the likelyhood is that you can leverage that vendors
    support through support contracts & the many developers who specalize in
    that vendors products.

    that works for any 'end to end' provider

    > The software company I work has over the years been shifting various
    > underlying libraries and infrastructure to those from open source
    > projects. We have found the open source projects far more responsive
    > and quicker to solve problems than the commercial alternatives we used
    > to use.


    i suppose if you guys shifted to oss infrastructure then it must have gotten
    bad

    tell me, are thoes oss projects *nix or windows orientated (as in what platform
    do they target their development for). i do have to confess oss projects
    that claim the title of 'cross platform' have their work cut out for them
    because it is a lot more complex than specalizing in one platform.

    > I don't know what attitude you've taken to deserve treatment like
    > that, but in our experience that is far removed from the norm.


    in my experience its typically 'card carrying oss circle jerk member' not
    project developers who put people off even dealing with them


    ---------
    Steven H
    Steven H, Oct 7, 2005
    #12
  13. Brett Roberts

    Rob J Guest

    In article <>, says...

    > Carroll also states there should be a standard that works everywhere. I
    > agree. Standards are an established means of ensuring interoperability, and
    > the concept was used in lots of fields before software came along. But
    > this is another area where MS is doing their best to prevent innovation.
    > They use many means to obstruct attempts by others at achieving
    > interoperability with files produced by MS apps.


    What is innovative about producing software that clones an existing
    application and opens its files?

    > Open source enables anyone and everyone to use, explore and build on the
    > technology. Conversely, proprietary source proponents like MS seek to
    > prevent anyone from understanding how their software works, and stop them
    > modifying or reusing it, and they want to restrict its use to those who can
    > afford to pay the inflated fee.


    There are many companies developing top calibre closed-source products
    worldwide, including right here in New Zealand. Not every one of these
    products is mass-market like Linux or OpenOffice. For example, Jade
    Corporation produces the Jade application development environment. The
    cost of buying Jade software for use in a business is small change
    compared to the total costs of project development, in which much more
    money will be paid in consultancy and developers' fees.

    Don't cite IBM - they make a lot more money from the other parts of
    their vast business activities and can afford to lose money on open
    source development, just as MS could afford to lose money on Internet
    Explorer.

    > MS is especially innovative and successful at their style of business, using
    > unlawful, unethical and selfish methods to stifle markets, wipe out
    > competitors and rip off consumers. Ultimately this is to the detriment of
    > society and to technical progress.


    So you claim. But in fact there is nothing about closed-source
    development that inherently suggests this is the case.
    Rob J, Oct 7, 2005
    #13
  14. In article <>,
    Rob J <> wrote:

    >What is innovative about producing software that clones an existing
    >application and opens its files?


    You mean, like how Internet Explorer started out by copying the NCSA
    Mosaic code? And how MS-DOS originated from Q-DOS, which wasn't
    developed by Microsoft at all?
    Lawrence D'Oliveiro, Oct 7, 2005
    #14
  15. Brett Roberts

    Dumbkiwi Guest

    On Fri, 07 Oct 2005 09:11:36 +0000, Steven H wrote:

    > Hello AD.,
    >
    > warning: long as hell post
    >
    >> Saying there is a single store for configuration (eg the registry) in
    >> Windows is just as accurate as saying there is a single store for
    >> configuration in Unix (eg the /etc directory). The actual information
    >> in the registry is no more consistent between applications than the
    >> information in the config files.

    >
    > diffrent distro's store diffrent stuff it in diffrent places
    >


    Within certain parameters - yes. However, almost all configuration data
    that matters is in /etc. However, the only person this seems to concern,
    is you.

    >> A single API for accessing it? Sure on Unix its called POSIX system
    >> calls.

    >
    > yea bad example, the point i was making that diffrent distro's store the
    > same values in diffrent places. now if the 'POSTIX system calls' know
    > exactly where a value is (independant of distro) then that is an
    > achievement.


    The storage of configuration data has nothing to do with an api - or at
    least my understanding of an api.
    >
    > you still need to know the path to the value on diffrent distro's
    >


    Why? Please explain.

    >> A single Windowing API? For each version of Windows perhaps.

    >
    > minor correction, you use the same api's created in 95 than you do in xp
    > - CreateWindow[Ex], the mesage pump is still the same, in fact all the
    > windowing api's that were in 95 are in XP.
    >
    > newer ones have been added through the years (renember when windows
    > would 'steal focus'), and api's have been changed.
    >
    >> That new
    >> one coming in Vista looks very different.

    >
    > quite right, no more old fangled win32 api shyte to deal with - its all
    > managed


    So although the api's are consistent, they're "shyte", and need changing
    - huh?

    >
    > there will be a win32 'compatabilaty layer' so that older windows apps
    > will still run in vista - this is Microsoft were talking about, who
    > retain as much compatabilaty for stuff like this as possible. Not Apple
    > who in their arrogance dont give much of a shit about older apps running
    > on newer versions of their operating systems.
    >
    >> And there are just as many
    >> windowing toolkits available for Windows as there is for X.

    >
    > you have 'wrappers' for win32 yes (mfc, and whatever 'object orientated
    > wrapping' other langages provide) but there is only one windowing api -
    > correct me if iam wrong (does happen) but doesnt mandrake have a
    > diffrent windowing api than say some other desktop manager


    You are wrong. There is no such thing as a desktop manager - there are
    window managers, desktop environments and display managers and then there
    is X. They all do different things - some have an api, and some don't.
    Then there are distributions (Mandrake - now called Mandriva), which have
    nothing to do with api's - they simply package software, which already has
    an api. This is where I stopped reading. Clearly your understanding of
    linux and api's is almost non-existent. To then start calling people
    "circle jerk members" simply backs up my view.

    <snip>
    >
    > ---------
    > Steven H
    Dumbkiwi, Oct 7, 2005
    #15
  16. Brett Roberts

    Peter Guest

    Rob J wrote:
    > In article <>, says...
    >> Carroll also states there should be a standard that works everywhere. I
    >> agree. Standards are an established means of ensuring interoperability,
    >> and
    >> the concept was used in lots of fields before software came along. But
    >> this is another area where MS is doing their best to prevent innovation.
    >> They use many means to obstruct attempts by others at achieving
    >> interoperability with files produced by MS apps.

    >
    > What is innovative about producing software that clones an existing
    > application and opens its files?


    That wasn't the statement.
    Carroll explained that innovation involves building on earlier works.
    Standards enable interoperability, which hence enables others to build on
    or extend the technology. MS seeks to obstruct interoperability, which
    therefore precludes this type of innovation or development.

    >> Open source enables anyone and everyone to use, explore and build on the
    >> technology. Conversely, proprietary source proponents like MS seek to
    >> prevent anyone from understanding how their software works, and stop them
    >> modifying or reusing it, and they want to restrict its use to those who
    >> can afford to pay the inflated fee.

    >
    > There are many companies developing top calibre closed-source products
    > worldwide, including right here in New Zealand. Not every one of these
    > products is mass-market like Linux or OpenOffice. For example, Jade
    > Corporation produces the Jade application development environment. The
    > cost of buying Jade software for use in a business is small change
    > compared to the total costs of project development, in which much more
    > money will be paid in consultancy and developers' fees.


    The fact remains; the closed source approach inhibits innovation /
    development by others.


    > Don't cite IBM - they make a lot more money from the other parts of
    > their vast business activities and can afford to lose money on open
    > source development, just as MS could afford to lose money on Internet
    > Explorer.


    IBM claim their Linux related operations are profitable. There are other
    companies operating profitably on open source products. How does this
    relate to the MS vs innovation debate?


    Peter
    Peter, Oct 7, 2005
    #16
  17. Brett Roberts

    thingy Guest

    Dumbkiwi wrote:
    > On Fri, 07 Oct 2005 14:19:28 +1300, Brett Roberts wrote:
    >
    >
    >>There are regular discussions on nz.comp regarding Microsoft and innovation
    >>(or alleged lack thereof). This series of blog posts by John Carroll (a
    >>Microsoft employee working in our IPTV group) are, IMHO, interesting reading
    >>and more rational than some of the other commentary I've seen on the
    >>subject:
    >>
    >>http://blogs.zdnet.com/carroll/?p=1495
    >>
    >>http://blogs.zdnet.com/carroll/?p=1496
    >>
    >>http://blogs.zdnet.com/carroll/?p=1497
    >>
    >>
    >>Brett Roberts
    >>Microsoft NZ

    >
    >
    > Brett,
    >
    > An interesting series - self serving, but interesting.
    >
    > The definition of innovation in the blog seems to be "incremental
    > advancement of existing ideas and technologies". This is a fair enough
    > definition. However, in my view this is a strength of the open-source
    > world, not of the proprietary world. In the proprietary world you have to
    > rebuild the base technology from scratch before you can increment, rather
    > than re-use existing code. Furthermore, there is more scope in the
    > open-source world to scrutinise and learn from existing code therefore
    > speeding up the process of incremental improvement. Some would say that
    > the ability to stand on the shoulders of giants is the very purpose of
    > open source. So nothing new here.
    >
    > Another point made in the third blog is the innovation of providing end to
    > end solutions. He challenges the ability of open-source to match that. I
    > think this point has some validity in some circumstances. I also think
    > the analysis fails in a number of others. One project where it fails I
    > think is kde. This is a massive project with a huge number of
    > geographically, and philosophically disparate contributors, but they have
    > developed technologies for consistent and rapid development that is
    > outstanding. The core technologies (for example kparts, kio slaves, file
    > dialogs, consistent widgets, theming, dcop) are built into the core
    > libraries and are available to all applications. Furthermore, and far
    > more importantly than the prior point, is that unix programmers, as I
    > understand it, generally write their applications with inter-application
    > communication in mind - they are designed to take input from other
    > applications, and output it to yet another - the idea of pipes and
    > redirection. dcop is just an extension of that philosophy. So while
    > developers may only work on a few isolated projects, generally, the
    > applications are designed with inter-application communication and code
    > re-use in mind, thus allowing those that follow to use the power of those
    > "isolated" projects within their own applications. this is a unix
    > philosphy (ie linking small simple applications together in new (and
    > unanticipated ways) to create something powerful that is "end to end" for
    > THAT user) that has been around for a long time. The ability for each
    > user to build such an end-to-end solution for him/herself is a far more
    > powerful, and prone to greater innovation, than a monolith
    > developer dictating to its user-base a generic "end-to-end" solution for
    > a perceived problem that only approximates the actual problem that each
    > particular user has.
    >
    > I have to say the example of consistency that is given (ie end-to-end
    > media framework), to me, smacks of vendor lock-in, and monopoly abuse.
    > However, I suppose from the inside (of MS) it could seem like innovation,
    > where such things are a business strategy, and the norm - take the
    > Massachusetts situation as an example. Granted, Office provides an end to
    > end solution, but if it's not interoperable outside the MS platform, then
    > it will fail to meet what appear (at least to me, and a lot of others) to
    > be sensible interoperability, and openness requirements.
    >
    > I'm sorry, Brett, but this just looks like astro-turfing, and doesn't
    > actually advance MS's argument very far. MS through its development
    > model is not doing anything more innovative than any other software
    > development model, and in my view, open-source is far better positioned to
    > innovate by incremental improvement, and to provide end-to-end
    > solutions for each user's particular needs, faster than MS's monolithic
    > proprietary approach.
    >
    > Just some thoughts.
    >
    >


    Very well put.

    regards

    Thing
    thingy, Oct 7, 2005
    #17
  18. Brett Roberts

    thingy Guest

    Steven H wrote:
    > Hello AD.,
    >
    > warning: long as hell post
    >
    >> Saying there is a single store for configuration (eg the registry) in
    >> Windows is just as accurate as saying there is a single store for
    >> configuration in Unix (eg the /etc directory). The actual information
    >> in the registry is no more consistent between applications than the
    >> information in the config files.

    >
    >
    > diffrent distro's store diffrent stuff it in diffrent places


    and MS does not?

    Becides which virtually all configuration files are stored under /etc/.

    I have just started to use SUSE after years on RH and Debian and the
    layout is similar enough and by using Linux I have gained enough "under
    the hood knowledge" not to have big issues.

    regards

    Thing
    thingy, Oct 7, 2005
    #18
  19. Brett Roberts

    shannon Guest

    Dumbkiwi wrote:


    >
    > You are wrong. There is no such thing as a desktop manager - there are
    > window managers, desktop environments and display managers and then there
    > is X. They all do different things - some have an api, and some don't.
    > Then there are distributions (Mandrake - now called Mandriva), which have
    > nothing to do with api's - they simply package software, which already has
    > an api. This is where I stopped reading. Clearly your understanding of
    > linux and api's is almost non-existent. To then start calling people
    > "circle jerk members" simply backs up my view.
    >
    > <snip>


    Yup
    both sides of the endless pointless debate have their embarrassing members.
    shannon, Oct 7, 2005
    #19
  20. Brett Roberts

    Allistar Guest

    Peter wrote:

    > Rob J wrote:
    >> In article <>, says...
    >>> Carroll also states there should be a standard that works everywhere. I
    >>> agree. Standards are an established means of ensuring interoperability,
    >>> and
    >>> the concept was used in lots of fields before software came along. But
    >>> this is another area where MS is doing their best to prevent innovation.
    >>> They use many means to obstruct attempts by others at achieving
    >>> interoperability with files produced by MS apps.

    >>
    >> What is innovative about producing software that clones an existing
    >> application and opens its files?

    >
    > That wasn't the statement.
    > Carroll explained that innovation involves building on earlier works.
    > Standards enable interoperability, which hence enables others to build on
    > or extend the technology. MS seeks to obstruct interoperability, which
    > therefore precludes this type of innovation or development.
    >
    >>> Open source enables anyone and everyone to use, explore and build on the
    >>> technology. Conversely, proprietary source proponents like MS seek to
    >>> prevent anyone from understanding how their software works, and stop
    >>> them modifying or reusing it, and they want to restrict its use to those
    >>> who can afford to pay the inflated fee.

    >>
    >> There are many companies developing top calibre closed-source products
    >> worldwide, including right here in New Zealand. Not every one of these
    >> products is mass-market like Linux or OpenOffice. For example, Jade
    >> Corporation produces the Jade application development environment. The
    >> cost of buying Jade software for use in a business is small change
    >> compared to the total costs of project development, in which much more
    >> money will be paid in consultancy and developers' fees.

    >
    > The fact remains; the closed source approach inhibits innovation /
    > development by others.


    I disagree with this.

    As a software developer - someone who develops on a contract basis and also
    sells my own products - I would have very little incentive to develop my
    own products if I gave the source code away. Giving the source code away
    greatly limits my ability to make a decent return of the ideas and effort
    that have gone into the development. What would be my incentive?

    Open source works great for "infrustructure" type applications - things
    everyone uses - such as operating systems, office suites, browsers etc.

    Open source is great for reducing costs for applications you businesss needs
    to use - it's not great if you are in the business of selling software for
    a profit. And not all software needs expensive support contracts, saying
    that opensource revenue should come from supporting the software doesn't
    help.

    Caveat: I use Gentoo linux only at home and at work and 95% of the software
    I use is open source. I am an open source advocate, and avoid proprietary
    solutions where I can. I don't, however, hold the view that closed source
    software inhibits innovation.

    Software patents, on the other hand, greatly inhibit innovation. Abusing a
    monopoly also inhibits innovation as it can be used to prevent competitors
    from entering the market.

    >> Don't cite IBM - they make a lot more money from the other parts of
    >> their vast business activities and can afford to lose money on open
    >> source development, just as MS could afford to lose money on Internet
    >> Explorer.

    >
    > IBM claim their Linux related operations are profitable. There are other
    > companies operating profitably on open source products. How does this
    > relate to the MS vs innovation debate?
    >
    >
    > Peter


    Allistar.
    Allistar, Oct 7, 2005
    #20
    1. Advertising

Want to reply to this thread or ask your own question?

It takes just 2 minutes to sign up (and it's free!). Just click the sign up button to choose a username and then you can ask your own questions on the forum.
Similar Threads
  1. Guest

    A tad OT: Invention Submission?

    Guest, Jul 13, 2003, in forum: Computer Support
    Replies:
    2
    Views:
    574
    =?ISO-8859-1?Q?R=F4g=EAr?=
    Jul 13, 2003
  2. Peter Jones
    Replies:
    4
    Views:
    451
    Andre
    Jul 13, 2003
  3. Peter

    Microsoft "innovation"

    Peter, Aug 21, 2004, in forum: NZ Computing
    Replies:
    25
    Views:
    617
    Peter
    Aug 24, 2004
  4. Peter

    Microsoft's idea of innovation

    Peter, Nov 27, 2006, in forum: NZ Computing
    Replies:
    18
    Views:
    548
    BrianM
    Dec 1, 2006
  5. - Bobb -

    OT: " Microsoft, and Innovation"

    - Bobb -, Aug 29, 2010, in forum: Computer Information
    Replies:
    1
    Views:
    468
    Ragnar
    Aug 29, 2010
Loading...

Share This Page