With the last manager leaving the area, I offered and recently became co-manager of the Central PA Flash User Group (CPFUG). Frankly, based on the last few meetings, CPFUG was dying, but I hope it’s something I can help reverse. I’m not much for managerial positions (especially with my time about to be soaked up by a new baby next month), but my time attending FlashCodersNY meetings made me appreciate the wealth of a dedicated user group community.

I have a few ideas that I hope can help spurn interest, but I also hope to reach out to the local Flash/Flex community and determine what it is that would make the meetings more valuable.

To start, we’re moving our meeting location–and considering a revolving meeting place. In the past, meetings have been held at JPL and PSECU. I don’t know about you, but by the end of the day, I’m not interested in sitting in another meeting room, and I think a casual setting’s just the right place.

Our next meeting is scheduled for 6-8pm today at Appalachian Brewing Company in Harrisburg (upstairs). We’re going to start a series of meetings covering “Actionscript 3.0 Animation: Making Things Move!” by Keith Peters. Attendance is free, though you’re encouraged to bring a copy of the book and a laptop with Flash installed (if you have either).

Also, feel free to bring Flash-related questions or suggestions you may have for group. Hope to see you there!

Designing my own portfolio has always been one of the most time-consuming projects because I, like any other designer, become my own harshest critic.

While I still intend to redo my company’s site in Flash, I decided in the meantime to redo my portfolio in an iPhone-friendly format by using only HTML, CSS, and Javascript (For those that might have come across my guest post on Fathom’s blog, I’m sure this must come off as an about-face). I didn’t really use HTML 5-specific tags like <header>, <footer>, etc., since my site needed to remain accessible, and I relied heavily on JQuery for the animation. Up until this point, I’ve read the complaints for and against both Flash and HTML 5, but this gave me the opportunity to guinea pig my own site and experience the issues firsthand.

Development Environment Differences

The first big difference I noticed was the work environment. JavaScript wasn’t really intended as an object-oriented language, and the work process felt much messier than building the site in classes (ala ActionScript). Furthermore, since many of the divs were hidden on page load, I felt that I wasn’t able to quickly preview elements during development. I ended up relying heavily on the Firefox Firebug plugin and going back-and-forth between the browser and the code.

Search Engine Accessibility

From a search engine standpoint, the differences between the two technologies feels somewhat of a wash. Both sites would have a page title and meta tags, so there’s no big advantage for either here. On the plus side for HTML, absolutely everything can be interpreted by all major search engines. Had I done the site in Flash, only Google would have been able to partially parse the SWF file. On the other hand, I couldn’t take advantage of separate URL paths or page titles like I could with Flash (with the addition of the SWFAddress). the only way I could have done this would be to force a page refresh–an interruption in the experience I really didn’t want to occur. Actually, Asual has the JQuery Address plug-in, which does the same thing as SWFAddress. Thanks @kweerious.

Font Options

The decision of HTML over Flash heavily impacted my font choices as well, since HTML pages are reliant upon client machines having the font installed. Flash, on the other hand, can simply embed the font. Granted, there’s alternatives, like SiFR and Cufón, however, I didn’t want to use Flash replacement text and don’t like the fact that Cufón doesn’t allow for text selection (neither of these options, by the way, seem to be intended for extensive font use).

Another option is to use the @font-face call with the Web Open Font Format (WOFF). This format places font files on the server for the HTML page to reference, and is currently the frontrunner for inclusion in the HTML 5 spec. WOFF is certainly the most elegant (see the recently-designed splash page I completed for metalledwith for example), however, because the font files are available on the server, the file can easily be downloaded illegally. Not wanting to infringe upon the license I have for Futura and not wanting to make all of the text into images, i simply used a safe system font (Trebuchet).

Browser Inconsistencies

A good front end designer ensures that their Website renders consistently across a variety of browsers. In the case of Flash, there’s really no need to test, as the plug-in ensures consistency across browsers and platforms. In my case, Internet Explorer protested to using transparent PNG files, so it required additional code. Furthermore, while the code I used to generate an image reflection is proven to work in IE7, it didn’t want to work with how I built the reflection into the JQuery Cycle plug-in. I’m sure there would have been a way to make it work, but frankly, I didn’t feel it was worth the time–besides, IE users should already be accustomed to limitations.

Performance

Performance-wise, I was a little surprised to see that the animations were sluggish on the iPad and iPhone in comparison to their desktop counterparts. Granted, JQuery isn’t the fastest JavaScript framework on the market, but I was expecting at least slightly better performance compared to how Flash performs on Android devices. Overall, I suppose it only reinforced my agreement with Apple that intense animations in Flash would yield poor performance on their mobile devices.

My biggest issue with the end product, though, is how the site downloads. I’ve seen it before–JavaScript-based sites can’t preload animations or elements well, and therefore animations start playing before all of the elements have downloaded. The reason for this is because the HTML page attempts to download all of the elements contained in it–even if they are set not to display.

I’m a firm believer that you don’t tax the user with downloading content they haven’t requested, so I tried implementing the LazyLoader plug-in to ease the site download. I honestly don’t have complete faith in this plugin, but fortunately, the site is relatively small. Regardless, Flash currently blows HTML out of the water with its programmatic ability to queue and download content without requiring a page refresh.

Conclusion

I’m sure that the development tools will continue to evolve and I’m not claiming that my implementation of the all the various plug-ins was constructed in perfect harmony (anyone out there, please correct me if there’s a better approach). In the end, this endeavor opened my eyes to some of the things JavaScript can replace Flash for, but HTML 5′s inability to create a length presentation (without a page refresh or complete asset download) makes me believe that Flash isn’t going anywhere for a long time.

Personally, I had a lot of fun at BarCamp Harrisburg 2, and got to meet a lot of people in-person for the first time, which was great (I hope to see everyone again next year!).

I noticed there were a lot of tagged mentions on my topic, and I feel like I need to set the record straight. Most of the tags said something along the lines of “HTML 5 – the death of Flash” when they should have said “HTML5 – The Death of Flash?”

The idea to present on this topic revolved around Apple’s decision to omit Flash from their mobile products. In some ways, I think it’s a good decision for the iPhone but childish for the iPad. While researching the topic, I was actually surprised to discover how much HTML5 actually could replace Flash. As a Flash designer/developer, I’d love to see a standardized video format hit the Web (though, unfortunately, it looks like OGG won’t be it.

In my presentation, I predicted that Adobe will either make Flash export to an HTML5 canvas or will build a new tool that does just that. Besides that, my conclusion was that Flash isn’t going anywhere for a long time. Some of the reasons:

• Better production tools (HTML5 will catch up)
• Bigger community (HTML5 will catch up)
• Ubiquity and consistency (HTML5 will catch up, but browsers will take years to meet standards consistently, plus the HTML5 spec won’t be finalized until 2022)
• Creativity – <canvas> doesn’t appear to support filters, or tie in effects with video like Flash does
• Video cue points – not built into H.264
• Fonts – there’s the Web Open Font Format and TypeKit, but neither are a part of the HTML5 spec. Me, personally, I’d rather use sIFR over images of custom fonts.
• Browser independence – I’m not talking about Adobe AIR (which, aside from Tweetdeck, seems silly due to the double install), but Flash runs on a variety of devices in standalone form and needn’t sit in a Web page
• Code privacy – perhaps not such a good thing, but compiled SWF files protect their authoring code from being viewed. A <canvas> element would be available via view source. It makes me wonder how long an HTML page will become with really complicated HTML5 canvas-based apps…

In the end, we all ultimately choose the best tool for the job, and my ultimate point was that if you look at the evolution of the Web, innovation isn’t standard.

There’s been a lot of talk regarding the future of Adobe Flash thanks to the blockade generated by Apple on the iPad and iPhone. It’s been covered extensively by the news media and blogs, but I figured as a Flash developer/designer, I’d chime in.

To paint a picture of my stance, I will say that:

1. I hope open-source OGG format replaces Flash video. Flash has standardized a ubiquitous video format for the Web, and it’s wonderful evolution we can’t deny. Gone are the days of multiple encodings for Quicktime, Real Player, and Windows Media, and I doubt supervision excluding a for-profit company would have succeeded. That said, I think the Web has evolved, and the fact that HTML 5 has dropped OGG as the definitive format is a politically-driven mistake. We, as the creators of the Web, should be embracing an open video format.
2. I’m not an Adobe fanboy. Flash provides a format of expression unlike any other, and while I admit that Flash work is my bread-and-butter (and that I’m coincidentally wearing Adobe-branded apparel as I write this), I recognize Adobe is a company that answers to its shareholders. Frankly, their Open Screen Project and claims of Flash being an “open” platform register strictly as propaganda to me, and I felt the need to point out this stance because of the opinions I’ve encountered at some Adobe user groups.
3. I don’t believe HTML 5 is the Messiah. I support universal formats that ensure consistency between browsers. However, formats, while well-intentioned, don’t always reach the Web–they are ultimately dictated by the market. If that wasn’t the case, we’d all be discussing XHTML 2.0 instead of HTML 5 and be visiting .MOBI sites instead of using Webkit on our phones.
4. Flash isn’t going anywhere. Flash is used for more than just the Web–it’s great for standalone applications, and its games can’t be successfully emulated with HTML + JS. Furthermore, it allows for data without browser refreshing, and provides a more media-rich interactive experience. Regarding standards, see #3.
5. I want to see Flash Player on the iPad and the iPhone, but I believe this responsibility falls to Adobe.

I think something many in the Flash community fail to realize is that Flash Player doesn’t play well on mobile, and it’s not exclusive to Apple products. I know firsthand, as I considered purchasing a Nokia N900 as a result of Apple’s embargo. I checked out Flash Player 9.4 on the N900′s browser while stopping in at NYC’s Nokia Store, and the performance was nothing short of atrocious. YouTube videos were so choppy that they weren’t viewable, and simple Flash sites crashed the browser.

I should mention that the N900 is powered by the same ARM Cortex-A8 processor as the iPhone 3GS (the 3GS is underclocked to 600 Mhz). Sure, there are videos of Google’s Nexus One playing Flash Player 10.1, however, I wouldn’t be surprised if the Nexus One’s Snapdragon 1 GHz processor is concealing Flash Player’s inherent memory issues.

Reinforcement for this argument is Mozilla’s decision to drop Flash Player support for their mobile Firefox browser just prior to launch. I’m still amazed that this hasn’t garnered as much press as the iPad announcement.

This brings me back to the title of this topic–the Future of Adobe Flash.

Personally, I wonder if Adobe might have lost its way. Their community support is excellent, and I use their industry-standard tools every day, however, they’ve seemingly become obsessed with supporting a proprietary platform instead of focusing on building excellent tools, and their evangelists seem to point outward for blame rather than looking within. As I mentioned in #5, I want to see Flash Player arrive on all of Apple’s products, however, I think this falls upon Adobe improving the player’s performance. As for my predictions:

1. Flash Player goes truly open source and somehow gets included in the HTML 5 spec. Sadly, this is unlikely, as Adobe licenses proprietary video codecs for HE-AAC and h.264. More importantly, it’d likely affect Adobe’s bottom line.
2. Adobe publishes an API for Actionscript that supports gestures. Gestureworks is already providing this, and Adobe’s not far behind. Personally, I’m holding my breath for official Flash SWF support.
3. Adobe starts building applications that publish content for HTML 5 and mobile platforms.

Obviously, #1 and #2 are less predictions than #3, but I think it’d be refreshing for Adobe to begin exploring these new mobile outlets as opportunities to provide new software. Can you imagine a software that publishes iPhone, Blackberry, and MeeGo apps–without doing translations from Flash? Now that would truly be interesting . . .

This is one of those equations I always seem to forget, so I’m posting it here for my own selfish benefit (though it’s free for anyone to use).

It’s a range converter method that can be particularly useful for sliders, etc. (code is in ActionScript 3.0).

/**
* Translates a value between two numeric ranges.
* Usage example: trace(translateRange(0, 10, 0, 100, 5)); // returns 50
* @param a1 (Number) minimum value of original range
* @param a2 (Number) maximum value of original range
* @param b1 (Number) minimum value of new range
* @param b2 (Number) maximum value of new range
* @param num (Number) number from original range to translate
* @return Number Value of num in new range
*/
function translateRange(a1: Number, a2: Number, b1: Number, b2: Number, num:Number):Number {
var c:Number = ((num - a1) * (b2 -b1)/(a2 - a1)) + b1;
return c;
}


I figured I’d dedicate a post to this topic since I just wrapped up a rather painful learning experience with skinning Shopify (the Ruby on Rails-based shopping cart service) with a complete Flash frontend. Judging from the lack of documentation online, I’m willing to bet I’m one of the first crazy enough to take on such a challenge, and I’d hope to save some other poor developer some time.

Shopify tools

For starters, here’s a few of the tools Shopify provides for skinning:

• Shopify’s API, which allows you to communicate with the inventory, products, collections, and cart on Shopify.com
• Vision, a toolkit for creating a Shopify theme.

Working with Vision

If you're wondering how to access Shopify's API for your store--and your store only, check this link in your Shopify admin.

I  started with Vision, which is this handy application that you can run locally, modify Rails/HTML theme files, and then test in a browser. Once ready, you simply zip up the theme’s contents and upload it via your Shopify store’s browser-based admin. A nice setup, though I came across a few technicalities–

• As of this writing, Vision themes are limited a 40 MB size limit, and rightly so, since it’s a browser-based file upload. I happened to be working with a site with roughly 300 MB worth of photos, so I was forced to host the assets on a different site.
• Shopify’s admin allows you to individually edit theme files, but if you want to upload a replacement you have to first delete the existing file. This made it very time consuming, since SWF files can’t be edited in a text editor.
• Themes are broken down into three folders:
  1. assets – files associated with your theme. Think images, javascript, swf files, etc.
  2. layout – this contains the theme.liquid file, which is basically the wrapper in which all of the templates are loaded
  3. templates – this contains a variety of necessary pages for a shopping cart site, like a product page, cart page, etc.

  A few things to note here: Unfortunately, I learned that the assets folder can’t contain subfolders, which lead me to a rather messy site structure. Your Shopify site will require all of the default template files. Since I was essentially rebuilding these all in Flash, I embedded my main SWF in theme.liquid and HTML-commented the include calls in theme.liquid.

Getting ActionScript working with Shopify’s API

The key here is authentication, as described in Shopify’s API. The API spits out XML, which is easy enough to parse in ActionScript, however, since API calls look like this:

https://API_KEY:SOME_PASSWORD@some-shop.myshopify.com/admin/orders.xml

they shouldn’t be queried from within Flash directly, as this exposes the shop owner’s API Key and password. Instead, it’s safer to create a server-side script, open the XML, and then feed it to Flash.

Once I got past this, it was pretty straightforward, with one exception–I couldn’t find the cart contents in the API documentation. After doing some asking around, I discovered that the cart contents can be found at your-shop.myshopify.com/cart.js. Now, unlike the API, this list of products contained in the cart is not XML, but Javascript Object Notation (JSON). Sure, I could have taken the time to write a JSON parser for ActionScript 3, but thankfully the good people at Adobe have already written one. It’s in as3corelib, which is hosted on Google Code.

Security and crossdomain goodness

Anytime a SWF file attempts to access content from a URL other than the one it’s on it first checks for a crossdomain policy. This is basically an XML file named crossdomain.xml that is hosted on the root level of the server (You can catch the dry documentation on this here and here).

Fortunately, Shopify has crossdomain.xml files prepared, however, it’s still a bit tricky.

Since they don’t want anyone hacking their main site, there’s no policy file on *.shopify.com. Granted, your-shop.myshopify.com has a policy, but there’s no way to pull your theme files.

In the end, I explicitly referenced all of my SWF files from static.shopify.com and relied on the crossdomain policy file on your-shop.myshopify.com, which looks like this:

<cross-domain-policy>
<allow-access-from domain="*.shopify.com"/>
<allow-access-from domain="*.myshopify.com"/>
<allow-access-from domain="www.mydomain.com"/>
</cross-domain-policy>


So that’s it in a nutshell. If you’re out there using Flash as a complete skin for Shopify, drop a comment and let me know that I wasn’t alone!

I’ve recently been using the open-source AlivePDF AS3 library to satisfy a client’s request for printing a PDF from a Flash movie. It’s a handy library–and something that would be handy to have built into Flash  (then again, maybe Adobe Mars will take over).

While trying to print out a screen capture of my SWF I came across a pesky runtime error. I’ve since posted to the AlivePDF open issues list, but I figured I’d post here (below) to help further help anyone searching for the issue.

What steps will reproduce the problem?

Basically, I’ve attempted to pass a parent of a DisplayObject or variable reference to a DisplayObject instance in my class. Examples:

myPDF.addImage(this);
myPDF.addImage(MovieClip.stage);
myPDF.addImage(MovieClip.parent);

What is the expected output? What do you see instead?

I wanted to take a screen capture instance of my entire SWF, but instead I’d get the following runtime error: ArgumentError: Error #2015 Invalid BitmapData.

The problem I found was on line 3190 of org.alivepdf.pdf.PDF.as. Using FlashTracer, I determined that the following values were being passed into the addImage function:

displayObject.width:44038179.6
displayObject.height:16513228.8

When I alternatively hard code the values, the PDF prints fine:
var bitmapDataBuffer:BitmapData = new BitmapData (1000, 600, false);

What version of the product are you using? On what operating system?

I’m using AlivePDF 0.1.4.8 with Flash CS3 and TextMate on MacOSX Leopard

I’m building an accordion menu for a current Actionscript 3.0 Flash project and have decided to use the … rest parameter in my accordion-building function to allow for a variable list length. The catch I’ve found is that while …rest is an array type, simply passing an array into the argument will treat it as a single String. An example:

function myFunction(... rest):void {
   trace(rest.length + ": " + rest);
   trace(rest[2]);
}
var myArray:Array = new Array("this","is","my", "list", "of", "arguments");
myFunction(myArray);
/*
OUTPUTS:
1: this,is,my,list,of,arguments
undefined
*/

You’ll notice that the length of the argument array is 1 and that querying a specific key beyond the argument length yields undefined.

If you want to instead submit the array as a series of arguments (and take advantage of …rest) do this instead:

function myFunction(... rest):void {
   trace(rest.length + ": " + rest);
   trace(rest[2]);
}
var myArray:Array = new Array("this","is","my", "list", "of", "arguments");
myFunction.apply(this, myArray);
/*
OUTPUTS:
6: this,is,my,list,of,arguments
my
*/

I figured it’s only fair to mention this after my last post on my Flash Lite exploration, but I recently upgraded my N95 with the 20.2.011 firmware update (thanks to this Engadget post) and discovered that a lot of the issues I experienced in deployment have been addressed. Upgrade issues aside (I feel fortunate to have discovered this release and was slightly annoyed that I had to reinstall my custom applications after the upgrade . . . it sure does make the iPhone process feel more streamlined!), this firmware update removed Flash Lite 2.0 and replaced it with Flash Lite 3.0. Also, the new OS interprets Web Runtime Widgets. Oh, and on an unrelated note, the camera is much more responsive and the direct Flickr upload is a nice touch.

Ever since I ordered my Nokia 6682 I told myself that I was going to build a Flash Lite application for it. I purposely ordered an AT&T plan to subsidize the $600 phone cost and even paid Macromedia the $10 download fee for Flash Lite. The funny thing about working for yourself, though (and perhaps it’s the industry in general), is that you never make time for personal projects–regardless of all the personal promises and financial stability to do so.

Two years later, Macromedia is now Adobe, The Flash Lite player is free and is often pre-installed on phones (thank goodness), the latest Flash Lite Player version is 3, and I’ve since given away my 6682 and replaced it with an N95 North American edition. Oh, and I finally made the time to play around.

I decided to build a simple Blackjack game. After reading about Flash Lite’s processing limitations, I chose something rather simple. I initially intended to release the code for free under a BSD license, but have discovered the hurdles involved with Flash Lite deployment (they were hurdles I always knew to be there, but sometimes one  just needs to learn firsthand). I also considered inviting artists to create artwork for the playing cards, although I’m not sure it’s so worthwhile considering how small the cards are and how intensive bitmap caching is.

So, this will serve as Part I of an ongoing series of my experience working with Flash Lite. By the end of it, I hope to create some helpful resources for someone and possibly proceed with releasing the software for free.

Oh, and If anyone reading this is interested in learning more or testing our the beta application (and happens to have an S60 device with Flash Lite 2.1 player), I’m happy to provide it.