It would be good if the flash player we had worked as well as pre 1.2

So, having flash 9.4 means that we cant see flash 10 products at all? or we will be able to see them poorly?

Theoretically all you'd need to do is get the libflashplayer.so from android, do an ldd on it and see what libraries it uses and then create symlinks to the maemo equivalents. Unless there are ABI differences, it 'should' work.

Of course that's in theory. I do know from looking at the N900's libflashplayer.so before, it is linked to a lot of Xorg's libs. Android doesn't run Xorg, from what I recall it's just a framebuffer.

That would make 'porting' it a bit difficult, though not necessarily impossible.

I still say Nokia or Adobe should get punched in the sack for not releasing it, or at least telling us if they are planning on it or why specifically they are going to stiff us on it.

slaapliedje

Someone correct me if I'm wrong, but it depends on the developer of the site. A site developer can include code their site so that it checks for the version of the flash player the user is using and decide whether to allow them to view the Flash content or not.

At the company I work for we develop e-learning modules for BIG corporations, Universities and Government, and we use Flash for a quiz at the end of a module.

We are publishing for Flash v6 because the ******ed IT departments at those places don't update their software.
In most of those places they still use IE 5 or 6!!!
Even at IBM, yes IBM Australia is one of our clients. Sometimes they don't have Flash player installed at all...

What happens if you don't have at least Flash Player 6 installed is that you get an "update your flash player" warning and it will not play.

Not if you're coding to certain versions of the Flex SDK or Flex IDE (now known as Flash Builder 4). At the moment, I'm pushed to FP 10 because of my tools.

If it were Flash's IDE, I can force the output (depending on features I use, ActionScript level) to be most any version of Flash. But if I choose AS3, it's FP 8 and higher (usually). And so forth.

I hope some drops a piano on you for making a quiz require the monstrously bloated and slow flash plugin

http://talk.maemo.org/showthread.php?t=49392

As opposed to what?

As opposed to a down pillow, or more generally anything lighter/less-injurious than a piano?


I presume he meant html forms, with whatever timekeeping requirements displayed client-side and enforced server-side, and of course answer verification server-side.


I had an experience with an online driver-ed program (to knock a few lead-foot points off my state record, back when I had a pickup), which I shall proceed to relate:

Due to a state requirement of N hours instruction for a "certified driver education program", each lesson page had a JS countdown timer before it would make the link for the next page visible. As I read faster than average, I had time to look over the source while I was waiting, and about two pages in got curious enough to try the link before time-out... and it worked, whereas I'd expected it to (leniently) redirect me to the original page, or possibly restart the whole chapter as a disincentive to tampering.

So, I merrily skipped through with no more timeouts. When it came to the quizzes, I had to wonder -- they couldn't be doing that client-side, too, could they? Turns out, they actually didn't, not quite. The quizzes were fairly straightforward forms, which the server checked and generated a score page.

But the test page itself was dynamically generated with randomized questions from a pool; obviously, they couldn't apply a simple "ADBBC..." answer key, but needed some way to keep track of which answer was correct for each question in that test. They cleverly solved this challenge by dropping a "question list" hidden field in the generated test page, and the server would verify that the radio-button selected for each question matched the answer-key for the corresponding question, according to the list submitted by the client.

A likely vulnerability, of course, but not real bad, yet -- surely it'd throw out obviously-faked question-lists, so until you'd been through it once and sold an answer-key, not much use... then I saw a quiz where it randomly selected the same question twice.

Having a hard time believing it, and a decent web browser, I edited the page to fill the question list with the first entry, and checked the same answer (the correct one for the first question, obviously) to every question -- and got a perfect score! Basically, it'll let you gamble as many questions as you like on one you know the answer for. I played the rest of the questions by the book (trust me, I didn't need to cheat -- my problem was with not caring how fast I drove on open roads, not on driving drunk, road rage, not knowing the rules of the road, or the dozen other things they tried to teach me -- and it was faster and easier to play it straight), but was left, frankly, flabbergasted by how sloppy they were.

There's a saying "never trust the client", which that on-line testing company utterly disregarded -- they just assumed nobody could/would tamper with the execution of javascript in their browser. And that makes me wonder how much the appeal of Flash is that it saves the developer from having to dodge all those pitfalls, since nobody could/would tamper with the execution of actionscript in their flash plugin.

Now in your case, there may well have been other reasons -- after all, not all quizzes are simple multiple-choice with only an overall time limit; it's quite conceivable you needed a more complex platform, and I don't pretend to know enough details to judge that. (And I'm not claiming you left it as vulnerable to someone with a "compromised" flash player as the above is to someone with Opera; it's just as possible (and just as much work) to get it right with flash as with plain html.)

But yeah, unless there's a real good reason, I'd be annoyed at needing flash for a quiz, too. Maybe not annoyed enough to start waving pianos around, but annoyed.