Dick,

The Flash examples you mention run fine on my machine and I do not notice any significant sluggishness. On slower machines I do see some issues but I think the comparisons you are drawing are not exactly fair. The XUL examples I have seen are not exactly visually stunning. Also, currently XUL is very limited on what browsers it can be ran on.

I reviewed several of the examples at
http://www.xulplanet.com/tutorials/xulqa/ and they are all pretty simple UI's. Although they are more responsive they do not look anything like the examples that Flex provides.

If you look at how the flash player has changed over the last few versions you will see that performance has improved and the actionscript language has continued to evolve as well. Performance enhancements are high on the priority list for the Flash player and performance issues are not being ignored.

Brandon

>There have been several threads lately, about RIAs and PIAs --
>discussing Flex, Flash, XUL, Neuromancer (_javascript_ Remoting), desktop
>applications and XUL.
>
>I for one was/am critical of the Flash RIA as not being good enough --
>resulting in a heavy or sticky feel to the UI.
>
>I realize the advantages of Flash, especially its consistency and
>ubiquity -- I only wish it were more usable and more programmable.  
>Flex (at some cost) appears to address the programmability issue.  
>Also, there are hints that the next CFMX will include some Flash
>programming capabilities.
>
>That's all well and good, if Flash can deliver acceptable performance
>-- if it can't then I will look elsewhere for RIA solutions.
>
>I like MACR, Their products are good to excellent, their people are
>great -- I have had no negative experiences with either.
>
>That said, Flash still disappoints me -- I would like to see it
>succeed.  I would like to be able to use it.
>
>But there is something very wrong.
>
>Several days ago I mentioned that the Flex code explorer contained one
>of the slowest menu trees I'd ever seen.
>
>This is one of the apps that MACR uses to showcase the FlexFlash RIA.
>
>It just isn't fast enough to be usable, especially when you consider
>the amount of client resources it consumes to attain poor UI
>performance.  My particular criticism is with the expanding and
>collapsing of the folders -- where only client-side RIA processing is
>performed,
>
>Two days ago I learned about XUL.
>
>XUL looks like a possible RIA solution & that too, has been discussed
>in other threads.
>
>To prove to myself that the Flash Tree example was, indeed a poor
>performer, I decided to write a comparable tree menu in XUL.  All I
>really did was create (part of) the tree. Selecting an
>item/document/leaf node  doesn't do anything.  I nested the folders 3
>levels deep and duplicated the "Component Library" outer node enough
>times to get a tree larger than the code explorer. This is enough to
>compare expand/collapse with the Flash tree.
>
>  Here it is.  You need a current Mozilla or FireFox browser tto render
>XUL.
>
>      http://67.124.145.42/XUL/XULSampleTreeMenu.xul
>
>Put this up, side-by-side with the Flex Code Explorer at:
>
>     http://macromedia.com/software/flex/?promoid=home_prod_flex_111703#
>
>Expand enough  folders on each to get the window scrolling.  Then
>expand/collapse the top node.
>
>Why is Flash so slow?
>
>Thoughts!
>
>Dick
>
>P.S.  Display source and you will see the entire XUL program.
>
>P.P.S.  The XUL page may be slow to download -- The site is on my
>development machine which is quite busy -- JRUN/CFMX/MySQL,
>VirtualPC/WIN XP, 2 streamers, lots of browser windows & a few desktop
>apps
[Todays Threads] [This Message] [Subscription] [Fast Unsubscribe] [User Settings]

Reply via email to