News for hackers. Stuff that matters.
Re:HTML5, Flash and Silverlight. Which one is dead? by zdw on Monday November 15, 2010 @11:17AM (#1906751)
Flash should have died years ago.

I do work with blind people, and a Flash site is a smack in the face to them. They most commonly access the web the same way Google would - as text.

Yes, you can make Flash that supports accessibility, but in practice nobody does this - they really don't care about anyone using their only available alternate method of viewing the web.

Re:HTML5, Flash and Silverlight. Which one is dead? by poppysan on Monday November 15, 2010 @11:58AM (#1906955)
Flash is accessible -- unless devs dont use it properly. My problem with the flash debate is that terrible developers are the root of nearly every issue, not the language...
Re:HTML5, Flash and Silverlight. Which one is dead? by kj12345 on Monday November 15, 2010 @12:20PM (#1907037)
Yeah the whole issue is being talked about in a very bizarre way. If Flash were really just dying, no one would be talking about it (see JavaFX). If Flash/Flex were simply not accessible, there would be a push to get better accessibility support. (For the record Flex, at least, does have fairly solid accessibility support: http://www.adobe.com/accessibility/products/flex/overview.ht... )

Instead what I think is going on is that there's a big pushback against Adobe for failing to develop and open up the platform, and this anger at Adobe is being worded as analysis.

Re:HTML5, Flash and Silverlight. Which one is dead? by watty on Monday November 15, 2010 @11:42AM (#1906874)
I'm not even aware of how to support blind accessibility in Flash but if it does have adequate support, how can you blame the platform and not the developers? It's like blaming Flash for making advertisements too easy to create.
Re:HTML5, Flash and Silverlight. Which one is dead? by zdw on Monday November 15, 2010 @12:27PM (#1907060)
To use an analogy, let's say you had two programming languages:

- Language A has very poor support for unit testing and requires that you put a lot of extra effort in to do it

- Language B supports unit testing out of the box, and supports automatically generating stub code for the tests.

All things being equal, which setup do you think will lead to more unit testing, and by extension a higher quality/more maintainable product?

(I use unit testing here as it's something that is often unseen but useful, much like accessibility)

Tools are not solely to blame for poor craftsmanship, but they can greatly contribute to it.