Adobe's Automanual Updater



  • What the smeg is wrong with Adobe, exactly?

    What was the sense in creating an automatic updater for Flash Player whose sole purpose is to nag you that Flash Player needs updating?

    Every minor release (11.1 to 11.2 … 11.6 to 11.7 and so on) requires manual user intervention. I can only guess that it's to drive you past their scumware download page — if they just needed your willing consent to perform a minor update (instead of a point update, which is absurd in itself), they could simply ask, instead of simply throwing their hands in the air and giving up.

    I will be most glad when Macromedia's wretched legacy lies rotting in some unmarked grave.



  • Java does the same thing so they can install their shitty browser toolbar. #conspiracytheory



  • With Java, you not only have the continued lack of compliance with UAC (unless that's been fixed recently), but an updater that doesn't even download the update in advance ready for you. Yeah thanks for wasting a load of my time, Oracle. So yes, they're probably more interested in your installation of the Ask toolbar than resolving the bugs.

    Another side to the conspiracy: the "no Ask toolbar for me!" checkbox doesn't have a clickable label, so if you click the label then "Next", you will still get the toolbar …



  • @Daniel Beardsmore said:

    With Java, you not only have the continued lack of compliance with UAC (unless that's been fixed recently), but an updater that doesn't even download the update in advance ready for you. Yeah thanks for wasting a load of my time, Oracle. So yes, they're probably more interested in your installation of the Ask toolbar than resolving the bugs.

    Another side to the conspiracy: the "no Ask toolbar for me!" checkbox doesn't have a clickable label, so if you click the label then "Next", you will still get the toolbar …

    Java is available as an RPM. Which is great, because I can put it into the automatic package updater. Except it doesn't have a repo, so there's only manual updates. And I have to uninstall it before updating because they included the version number in the package name.

    Yaaaaaaay



  • Something else that I've noticed is for a while now, one of the "Click here to update Java" dialogs that pops up (I can't remember which one it is off hand) take you to a website which doesn't resolve a redirect properly. It goes something like this:

    Click here to update Java

    Click here to agree to our TOS and download Java

    Get redirected to java.com/80:/lang/eng/download.jsp (or some such)

    Wait for that page to fail to load

    Manually delete the 80:/ from address bar

    Finally download/update



  • @blakeyrat said:

    Java does the same thing so they can install their shitty browser toolbar.

    Someone took care of that for you



  • @Daniel Beardsmore said:

    What the smeg is wrong with Adobe, exactly?
     

    I'm not going to embed a Rosie here.

     



  • @Ben L. said:

    Java is available as an RPM.
     

    Yeah.. but which one? There's the JDK, then Open-JDK then... I had issues trying to find which one was suitable. Even yum struggled to find the one I wanted.



  • @Cassidy said:

    @Ben L. said:

    Java is available as an RPM.
     

    Yeah.. but which one? There's the JDK, then Open-JDK then... I had issues trying to find which one was suitable. Even yum struggled to find the one I wanted.

    Oh, openjdk is set up properly.

    It's just that some programs check to see what vendor's Java you're running and get all confused when it's not Sun or Oracle.

    I'm talking about the java you get from java.com.



  • @Ben L. said:

    I'm talking about the java you get from java.com.
     

    Ahhh... that'll be the one I couldn't get working under Ubuntu and ended up with openjdk so I could install SQL Developer.



  • @Cassidy said:

    Filed under: Fuck you Oracle.

    QFT



  • @Daniel Beardsmore said:

    What the smeg is wrong with Adobe, exactly?

    Abridged answer; the fact that they're still in business.



    Flash Player isn't even the worst of their cockups. Have a little look at the licensing problems people get with their Creative Suite some time; licenses randomly expiring, reverting back to a trial version, being downgraded from a 'premium' to a 'regular', etc.

    There are paying customers (paying way too much btw. because with Adobe's monopoly position they can pretty much charge whatever the fuck they want) being told by support to just uninstall, run a cleaner tool (that uninstalls additional stuff the uninstaller doesn't; WTF!) and then reinstall their product. Does the problem reoccur at a later time? You are advised to run through those steps again (ofcourse). Does it not solve the problem? They don't have a working solution for you; well, other than buying the new version of their software with another license and praying that one doesn't sponteneaously self-combust either...

    I had a problem like the above not too long ago with Flash Builder*, where every 30 days it would report a Creative Suite 6 license as expired, though Photoshop and other programs covered by the license would be fine. Eventually I got a hint out of an old post on a random web forum** to check the license verification logs.

    As it turned out, my particular version of Flash Builder was dual licensed for the Creative Suite 5.5 and Creative Suite 6 bundles. The license verification logs showed it starting with some type of collection phase, scanning for license types. Since I had a CS6 installation only, it would find licensing for CS6 and would report some default empty data for the CS5.5 license. Then it would start the actual verification phase. The CS 5.5 license data would be treated first and would ofcourse fail (since it was some bogus empty data only). Rather than continuing with the next set of license data (for CS6), the verification component would bail out entirely on the first failing license, leaving you with a license-state of 'expired' or 'invalid'. For bonus points; failing the startup license check puts Flash Builder into trial mode, but it does so silently, as the part of the Flash Builder UI that shows the Activated/Deactived state of the license does work correctly and reports the program as having an active license...

    The trick to solving all that? Activate CS6's license by starting Flash Builder; the logs would then reveal the CS6 license to have been positioned at the head of the queue, before the CS5.5 license and everything would 'work'. If you'd have started up Photoshop or other parts of the Creative Suite and had (re)activated the license through them instead, then 30 days later you'd be SoL again.

    I wonder if this fluke in their license verification would also explain the past problems some people would have with side-by-side installations of different Creative Suites, or would explain the problems with version upgrades...


    *) Don't ask; have some old Flash stuff my employer needs to continue to support for clients that demand fancy web stuff works in IE8 and below as well, bla bla... web development drama... yada yada...

    **) Adobe's own community forums are ofcourse about as helpful as you'd expect them to be for these kind of things; i.e. , not at all.



  • @mikeTheLiar said:

    Something else that I've noticed is for a while now, one of the "Click here to update Java" dialogs that pops up (I can't remember which one it is off hand) take you to a website which doesn't resolve a redirect properly. It goes something like this:

    Click here to update Java

    Click here to agree to our TOS and download Java

    Get redirected to java.com/80:/lang/eng/download.jsp (or some such)

    Wait for that page to fail to load

    Manually delete the 80:/ from address bar

    Finally download/update

    Bitch please. Try using the 64bit version. Upon update, the "auto" updater will link you to the web installer. The web installer will only install the 32bit version. To get the 64bit one, you have to navigate to "other systems and versions" or something and re-download the offline installer. Maybe TRWTF is Java software that has a version that can only be run in 64bit Java... why is that even a thing?



  • @derula said:

    Supposedly 64 bit Java gives better performance

    Wha--how would that even make sense?

    All 64 bit does is make your pointers twice as big.



  •  I suspect that all theFlash 'security issues' are just excuses to trick people into installing Chrome, so they get money fom Google.



  • @Ben L. said:

    @derula said:
    Supposedly 64 bit Java gives better performance

    Wha--how would that even make sense?

    All 64 bit does is make your pointers twice as big.

    A 64-bit CPU can fetch twice as much data at once through the data bus. Quite how much of an advantage this really is, I don't know — I imagine that it's great with SIMD though.

    Of course, since so much software is still 32-bit …



  • I could have sworn that they added an option to just install the updates for you a while back. My complaint is that it keeps reseting that option each install.



  • @Ben L. said:

    @derula said:
    Supposedly 64 bit Java gives better performance

    Wha--how would that even make sense?

    All 64 bit does is make your pointers twice as big.

     

     Since all X86_64-architecture supporting CPUs also support at least SSE2 (probably SSE3) and other assembler instructions, you could probably speed up a lot of low-level code?

     



  • @henke37 said:

    I could have sworn that they added an option to just install the updates for you a while back. My complaint is that it keeps reseting that option each install.

    It installs point updates (11.1.1, 11.1.2, 11.1.3 etc) automatically, allegedly (not a lot of use if a) you hibernate and thus have the browser running 99.9% of the time, or b) your PC is switched off any time that you're not logged in and running the browser).

    However, Adobe regularly put out minor updates (11.1, 11.2, 11.3 etc) which the updater won't download or install.



  • @Ben L. said:

    All 64 bit does is make your pointers twice as big.
    Come on, who wouldn't love to make his pointer twice as big?



  • @Anonymouse said:

    @Ben L. said:

    All 64 bit does is make your pointers twice as big.
    Come on, who wouldn't love to make his pointer twice as big?

     

     

    We need more jokes about big dongles and forking

     



  • Bitch please. Try using the 64bit version. Upon update, the "auto" updater will link you to the web installer. The web installer will only install the 32bit version. To get the 64bit one, you have to navigate to "other systems and versions" or something and re-download the offline installer. Maybe TRWTF is Java software that has a version that can only be run in 64bit Java... why is that even a thing?

    I can explain that - the java program hooks up to a 64bit native dll. If the java process isn't 64bit, it can't load 64bit dlls.



  • @Helix said:

    We need more jokes about big dongles and forking
    Purple dildos aren't enough for you?



  • @mikeTheLiar said:

    @Helix said:
    We need more jokes about big dongles and forking
    Purple dildos aren't enough for you?

    Actually blue ones are better. Then you can call it "Big Blue" after IBM and get literally fucked by IBM, which makes it a lot more interesting that doing it with something called "Supple Purple". Or so I've heard at least.



  • @mikeTheLiar said:

    @Helix said:
    We need more jokes about big dongles and forking
    Purple dildos aren't enough for you?

    [url=http://www.gizmag.com/lovepalz-iphone-teledildonics-app/24254/]Teledildonics[/url].



  • @Ben L. said:

    All 64 bit does is make your pointers twice as big.

    Going to long mode enables a bunch of other stuff as well. Code compiled for long mode will usually run faster than the same source code compiled for legacy mode and run on the same hardware, even though all those twice-as-big pointers do occupy more RAM.



  • @Daniel Beardsmore said:

    A 64-bit CPU can fetch twice as much data at once through the data bus.

    Compared to older hardware, sure. But neither the way an x86 processor is wired to the outside world nor the sizes of its cache lines depend on its operating mode (long vs. legacy); 32-bit code running on a 64-bit-capable x86 still gets the full benefit of its wide data paths.



  • You know, you can fix the twice-large pointers problem by having a global void* which holds where your heap begins and pass around 32-bit pointer offsets instead of pointers. Example code:

    #include <stdlib.h>
    #include <stdio.h>
    #include <stdint.h>

    #define POINTER(offset) (MEMSTART + (ptrdiff_t)offset)
    #define OFFSET(pointer) (uint32_t)(pointer - MEMSTART)

    static const uint32_t BAD_OFFSET = 0xFFFFFFFF;

    /* symbol defined in standard lib, points to data segment in memory /
    extern char end;
    void
    MEMSTART;

    const char* MESSAGE = "Hello World!";

    void printstr(uint32_t str) {
    printf("%s\n", POINTER(str));
    }

    /* returns BAD_OFFSET on error /
    uint32_t omalloc(size_t size) {
    void
    pointer = malloc(size);
    if (pointer) {
    return OFFSET(pointer);
    } else {
    return BAD_OFFSET;
    }
    }

    int main(int argc, char** argv) {
    MEMSTART = &end;
    uint32_t str = omalloc(strlen(MESSAGE));
    memmove(POINTER(str), MESSAGE, strlen(MESSAGE));
    printstr(str);
    free(POINTER(str));
    return 0;
    }

    </stdint.h></stdio.h></stdlib.h>



  • In your e-mail, that came through as "#include #include #include #define …" and I wondered what sort of awful hack this was … ;-)



  • @Daniel Beardsmore said:

    In your e-mail, that came through as "#include #include #include #define …" and I wondered what sort of awful hack this was … ;-)

     

    And of course now I can't edit it. Well just imagine theres stdlib.h, stdio.h and stdint.h in there. It's the most awful sort of hack.

     



  • @Daniel Beardsmore said:

    What the smeg is wrong with Adobe, exactly?

    What was the sense in creating an automatic updater for Flash Player whose sole purpose is to nag you that Flash Player needs updating?

    Every minor release (11.1 to 11.2 … 11.6 to 11.7 and so on) requires manual user intervention. I can only guess that it's to drive you past their scumware download page — if they just needed your willing consent to perform a minor update (instead of a point update, which is absurd in itself), they could simply ask, instead of simply throwing their hands in the air and giving up.

    I will be most glad when Macromedia's wretched legacy lies rotting in some unmarked grave.

    ...and why do Adobe's installers have the arrogance to set System-Modal on all their dialogues? Pisses me of when a shitty installer thinks that I absolutely must see it count to 100. If I want to hide your piece of crap while it installs yet another point release, then that should be my choice.



  • @Ben L. said:

    @derula said:
    Supposedly 64 bit Java gives better performance

    Wha--how would that even make sense?

    All 64 bit does is make your pointers twice as big.

     

    http://stackoverflow.com/questions/607322/what-are-the-advantages-of-a-64-bit-processor

    Bigger address space.  More CPU registers. You can safely assume the existence of SSE2. Page-level NX memory protection to guard against buffer overruns and pointer corruption.

     



  • @Mo6eB said:

    @Daniel Beardsmore said:

    In your e-mail, that came through as "#include #include #include #define …" and I wondered what sort of awful hack this was … ;-)

     

    And of course now I can't edit it. Well just imagine theres setjmp.h, com.oracle.java.h and php_use_vars.h in there. It's the most awful sort of hack.

     



  • @mikeTheLiar said:

    Something else that I've noticed is for a while now, one of the "Click here to update Java" dialogs that pops up (I can't remember which one it is off hand) take you to a website which doesn't resolve a redirect properly. It goes something like this:

    Click here to update Java

    Click here to agree to our TOS and download Java

    Get redirected to java.com/80:/lang/eng/download.jsp (or some such)

    Wait for that page to fail to load

    Manually delete the 80:/ from address bar

    Finally download/update

    Java's weird URL is more like https://something.java.com:80/somethingelse, which includes TRWTF, trying to speak HTTPS to port 80, which causes some strange network error.

    Android and Minecraft are sadly the only two things keeping Java on life support with the general population, like how Silverlight is exclusively Netflix now.

    Flash's shit is done as soon as Autodesk or someone releases a vector-based animation studio and markets it a as a vector-based animation studio.



  • @MiffTheFox said:

    Android and Minecraft are sadly the only two things keeping Java on life support with the general population, like how Silverlight is exclusively Netflix now.
     

    The problem with that statement is that "Android" isn't a product, like Minecraft and Netflix are, but a platform, and the fastest growing platform out there to boot.  Which means that Java's not "on life support" and it's not going away anytime soon; it's thriving.  (And I don't even like Java, or most of the concepts behind it.  But you have to face the facts.)

     



  • @MiffTheFox said:

    Flash's shit is done as soon as Autodesk or someone releases a vector-based animation studio and markets it a as a vector-based animation studio.

    Basically we're waiting for enough copies of IE to disappear that we can all move to native SVG (which you can manipulate with JS).

    It might be a while.



  • @Daniel Beardsmore said:

    we can call move to native SVG
     

    noooooooooooooooooooooo



  • @dhromed said:

    @Daniel Beardsmore said:

    we can call move to native SVG
     

    noooooooooooooooooooooo

    It can't be worse than Flash!



  • @Daniel Beardsmore said:

    Filed under: Shirley not



  • Protip: Ninite.com solves all your installer problems.



  • @Daniel Beardsmore said:

    @MiffTheFox said:

    Flash's shit is done as soon as Autodesk or someone releases a vector-based animation studio and markets it a as a vector-based animation studio.

    Basically we're waiting for enough copies of IE to disappear that we can all move to native SVG (which you can manipulate with JS).

    It might be a while.

     

     

    Have you actually tried doing that? Or worked with any of the tools associated?

     



  • @Daniel Beardsmore said:

    @MiffTheFox said:

    Flash's shit is done as soon as Autodesk or someone releases a vector-based animation studio and markets it a as a vector-based animation studio.

    Basically we're waiting for enough copies of IE to disappear that we can all move to native SVG (which you can manipulate with JS).

    It might be a while.

    Wow, you fell right into the same trap as Adobe. Pretty much any western 2D animated series from the past decade and a half is animated in Flash.

    If you go to Williams Street or Studio B and tell them to switch to SVG and JavaScript you'll just get funny looks. Flash should be for artists who draw on a canvas and apply tweens, not coders. That's why I said vector-based animation software, not "SVG and JavaScript".

    Btw, dynamic SVG is dead. Look into the canvas tag.



  • @Daniel Beardsmore said:

    @dhromed said:

    @Daniel Beardsmore said:

    we can call move to native SVG
     

    noooooooooooooooooooooo

    It can't be worse than Flash!

     

    It's XML.  Of course it can.

     



  • @derari said:

    Protip: Ninite.com solves all your installer problems.


    Unless you want to install into a non-default directory.



  • Supposing canvas has all the same vector graphic features of SVG, what would make it superior? Naturally this would have to include the ability to store an object collection together with support for translation, rotation etc, and of course all the usual stroke styles, line markers, gradient fills etc. NIH dictates that people would always choose to re-implement SVG from scratch in canvas instead of using actual SVG, together with some sort of WTFey file format based on Mork for sending over asset collections via AJAX.

    I'm looking forward to a fire2k fireyrant on SVG!

    (I have actually seen a private portal system animate SVG (just a pie chart that spins and expands out into view on load) — that came as a real surprise.)



  • @Daniel Beardsmore said:

    Supposing canvas has all the same vector graphic features of SVG, what would make it superior? Naturally this would have to include the ability to store an object collection together with support for translation, rotation etc, and of course all the usual stroke styles, line markers, gradient fills etc. NIH dictates that people would always choose to re-implement SVG from scratch in canvas instead of using actual SVG, together with some sort of WTFey file format based on Mork for sending over asset collections via AJAX.

     

     Or you could pipe Canvas INTO SVG: http://svgkit.sourceforge.net/SVGCanvas.html

     



  • I want my rant! :P



  • I'm not saying canvas is superior. I'm saying it won.

    Also your argument is weakened by claiming that IE doesn't support it. IE supports both Canvas and SVG. No, the browser you have to wait to die is Android 2.x, which will die once people stop buying those shitty shovelware phones running it. (Making Android open source was a huge mistake, although it might have been better if they used the GPLv3 instead of the practically public domain Apache license.)

    However, since I don't develop for browsers besides Chrome anymore, I do use SVG in one context: SVG assets sent over the wire instead of rending them out to PNG.



  • @MiffTheFox said:

    Also your argument is weakened by claiming that IE doesn't support it.

    I am sure that a lot of people here would love to use the same version of Internet Explorer that you are ;-)



  • @Daniel Beardsmore said:

    @MiffTheFox said:

    Also your argument is weakened by claiming that IE doesn't support it.

    I am sure that a lot of people here would love to use the same version of Internet Explorer that you are ;-)

    And yet nobody accepted that same argument about me using Firefox 3.5 until about the era of Firefox 10.


Log in to reply
 

Looks like your connection to What the Daily WTF? was lost, please wait while we try to reconnect.