Follow Slashdot stories on Twitter

 



Forgot your password?
typodupeerror
×
The Internet

Adobe Not Worried About the Future of Flash 328

An anonymous reader writes "Adobe company man John Dowdell isn't worried about the future of Flash. He writes in his company blog, 'There's really no "HTML vs Flash" war. There are sure people inciting to create such a war, and individual developers may have strong practical reasons to choose one technology over another, but at corporate levels that drive strategy, all delivery channels are important Adobe territory, whether SWF or HTML or video or documents or paper or ebook or e-mag or film or packaging or whatever. Adobe profits by making it easier for creatives to reach their audiences. We're on the verge of a disruptive change that, I think, will dwarf that of the World Wide Web fifteen years ago. It was great back then when any wealthy person with a workstation in a wired environment could easily reach any creative's webpage. With these cheaper devices we'll be reaching far more people, and with pocket devices we'll be reaching them throughout the day instead of just when "logged-on." The WWW was merely a pale precursor of the excitement we're going to see, I think.' It's interesting to note that he talks about the World Wide Web in the past tense. I find it instructive as to Adobe's perspective. Personally, I'm not worried about the future of Flash either. I don't think it has one."
This discussion has been archived. No new comments can be posted.

Adobe Not Worried About the Future of Flash

Comments Filter:
  • by Anonymous Coward on Monday March 29, 2010 @01:18PM (#31659290)

    The same way all-in-one computers with touch screens do: assume the mouse is still parked where you left it.

  • by Cicada7 ( 1051002 ) on Monday March 29, 2010 @01:32PM (#31659518)

    Sure they'll be competing with DreamWeaver or whoever...

    Just an FYI, DreamWeaver is an Adobe product too.

  • by LWATCDR ( 28044 ) on Monday March 29, 2010 @01:33PM (#31659526) Homepage Journal

    Because odds are you will have both a browser and Flash.
    So dropping Flash from you system will leave you with just the browser.
    Nobody that I know of just uses Flash without a browser. So by dropping flash you get rid of an attack vector. Now you only need to worry about your Browser and not your Browser and Flash.

  • by ShadowRangerRIT ( 1301549 ) on Monday March 29, 2010 @01:48PM (#31659732)
    You know, you can just hide the updates from those idiotic apps. I have one stupid app addicted friend whose wall I periodically check just so I can find the apps to hide from my news feed.
  • by hax0r_this ( 1073148 ) on Monday March 29, 2010 @02:07PM (#31659982)

    How does dropping flash for HTML5 remove an attack vector? It just replace one attack vector with another.

    Unless you're suggesting your browser would otherwise not support HTML5/Javascript, then you aren't replacing anything. Just dropping a third party plugin that is known to be buggy, non-standard and poorly maintained.

  • by Bigjeff5 ( 1143585 ) on Monday March 29, 2010 @02:14PM (#31660076)

    Imagine for a moment that at some time in the near future, Adobe has a new option on the menu "Export to HTML5".

    That's probably a very strong possibility, given that ActionScript is just an alternative JavaScript implementation.

    Sure they'll be competing with DreamWeaver or whoever

    Ahem, DreamWeaver is Adobe. :)

  • by Anonymous Coward on Monday March 29, 2010 @02:14PM (#31660080)

    I didn't even read that, but I could tell right away it could have been reduced to 3 sentences with no loss. Basic definition of TL;DR.

  • by icebraining ( 1313345 ) on Monday March 29, 2010 @02:27PM (#31660238) Homepage

    (...) Really, tell me when people start using open source to make professional quality games. (...)

    Sorry for the almost double-post, but:

    The [World of Goo] developers used many open-source technologies such as Simple DirectMedia Layer, Open Dynamics Engine for physics simulation, and TinyXML for configuration files. Subversion and Mantis Bug Tracker were used for work coordination.

  • by BoppreH ( 1520463 ) on Monday March 29, 2010 @03:35PM (#31661132)
    You are trying to convince me that OOP is not related to inheritance, polymorphism and encapsulation. Sure, you can do without it. But I don't care what Alan said about it, it's still a pretty damn important part of programming with classes and objects.

    And Flash is not JIT. You get some flash code, compiles it using Flash or Flex into a .swf file and then you send that file to the server. With Javascript, you get some code and just send it to the server. See the difference?

    I, again, don't care what the engine will use at the end. That's semantics, we would be arguing all the way down to the hardware. The important part is: to get your AS code running, you must put it through a compiler. This is important because it reveals syntax errors, for example, includes the currently present classes around it in the file and correct general simple programming mistakes ("this class doesn't have a foo property", "this variable is the wrong type", etc).
  • by shutdown -p now ( 807394 ) on Monday March 29, 2010 @04:45PM (#31662004) Journal

    I'm not trying to be snide here, but perhaps you should read up on OOP core concepts and features [wikipedia.org] which include classes, inheritence, abstraction, encapsulation, polymorphism, and decoupling.

    Did you even read the section of the article that you've linked to? I mean, it immediately starts with:

    "Not all of these concepts are to be found in all object-oriented programming languages, and so object-oriented programming that uses classes is called sometimes class-based programming. In particular, prototype-based programming does not typically use classes. As a result, a significantly different yet analogous terminology is used to define the concepts of object and instance."

    Guess what kind of an object-oriented language JavaScript is?..

  • by shutdown -p now ( 807394 ) on Monday March 29, 2010 @04:59PM (#31662180) Journal

    Allow me to chime in here.

    Object-oriented programming isn't "defined" as anything, because there's no universally accepted single definition of OOP. There is a bunch of abstract definitions that claim to be ones, but they contradict each other on some points.

    So, the only way to define OOP at the moment is to look at the features of the languages that are considered OOP by majority consensus. For example:

    1. Encapsulation - arguable; present in most OOP languages, but not impenetrable in many (e.g. Python, which just provides auto-prefixing; or CLOS, in which members can be accessed if fully qualified).

    2. Abstraction - not a feature in any way unique to OO, or even significantly tied to OO concepts.

    3. Runtime polymorphism (qualification is needed here because the word has different meaning in other language families - c.f. FP parametric polymorphism) - seems to be present in all OO languages without exception.

    4. Inheritance - not present at all in classless OO languages, and not even in some class-oriented ones. Can be replaced with delegation (which is what prototype-based OO effectively does), or mix-ins, with no harm to other OO concepts.

    So, the only definite characteristic of OO languages seems to be runtime polymorphism; that is, the ability to specify that, for a function/method call F(x1, x2, ...), the actual function being called is selected at run-time based on actual types of one or more arguments (a typical syntax in single-dispatch languages, such as x1.F(x2, ...), is only different visually and not semantically).

    There is, actually, one other important distinction. OO languages, as the name implies, are about objects. And the way objects are different from values is that objects have an inherent object identity. That is, even if you have several objects with no associated data (i.e. no fields), those objects are still different from each other, and that difference is observable (e.g. operator == on object references in Java exposes it). And if you have several objects with associated data, and that data is equal value-for-value, the objects are still different.

    If you look at OO languages, they all have something along those lines. Most expose identity explicitly in form of object references. In C++, object identity is its address, which is why it is guaranteed that any complete object (as opposed to a base class subobject) has a non-zero size - this ensures than no two objects can have the same address, and therefore the same identity.

  • by TheRaven64 ( 641858 ) on Monday March 29, 2010 @07:16PM (#31663804) Journal

    Classes are irrelevant to OOP. OOP is about simple models of special-purpose computer (objects) communicating via message passing. Everything else is incidental. Self is a pure object oriented language (everything is an object) but did not contain classes. It is possible to automatically translate code between Self and Smalltalk, yet Smalltalk has classes.

    JavaScript uses the Self model with a few small modifications. And, yes, Flash is JIT compiled. You compile the ActionScript to bytecode. Tamarin then JIT compiles traces (you can find some interesting papers on how this works - it's a very nice strategy). JavaScript needs to be parsed in the browser too, but if your compiler is spending more time on parsing and semantic analysis than it spends on code generation, then either it's not generating good code or your front end needs a complete rewrite.

    Whether you run static analysis tools on your code is completely irrelevant.

  • by smclean ( 521851 ) on Monday March 29, 2010 @07:18PM (#31663832) Homepage

    You will note that IE6 is still the most popular browser on the market

    No I will not [w3schools.com].

I've noticed several design suggestions in your code.

Working...