Software, Fitness, and Gaming – Jesse Warden

Blog

  • Flash Lite 2 is a Presentation Layer

    It may seem like an obvious statement, but it’s not. Flash Player has been a presentation layer for years. The difference? Nowadays, even without a backend, you can create some all encompassing applications. Utilizing the stateful client aspect, you can parse data, and show that parsed data. You can really push visualizations driven by data.

    Flash Lite 2 has major limits to the above. For example, for and while loops; dangerous, will most likely crash. Think on that; could you live without for and while loops in your code? Your only recompense is using an onEnterFrame loop handler. A setInterval doesn’t count because it can fire multiple times in one frame, thus either overflowing the stack, or making your insanely low scriptTimeout limit to take too long. Can you do loops? Yes. In real world applications that you are used to in Flash? No. So far, the only surviving one I have is in re-drawing a List component’s children.

    Data parsing. Data parsing, like sifting through XML, or converting strings, uses a lot of the above. Additionally, it has “a lot of data”. A lot of anything pre-Flash Player 9 is bad. Since Flash Lite 2 is basically 7, you have to not only keep that in mind, but also that you have about 2 megs of RAM. This makes even GETTING the data into Flash a challenge. For example, I’m learning to parse XML in PHP currently because Flash can’t do it reliably. It’s a good excuse to learn, but… dammit, 4 years of study, no longer applicable. Yes, yes, knowledge transfer will help in PHP, no doubt, but still, frustrating I have these insane powers that are now useless.

    Communication. I spent 2 days messing around with Google’s Calendar API. The amount of header finagling, and HTTP Status response tweaking is insane. It certainly makes you appreciate companies like Yahoo who expose standard middle-tier API’s that you can call… and they… say it with me… “WORK!”. To be fair, there is this dude on the Google help forum who is a champ. I think he’s one of the dudes writing it too. Problem is, Flash & AS2, even with Fiddler, ServiceCapture, and Ethereal installed on my comp don’t fix the whack ass errors I’m getting. Instead, I search Google and find a plethora of PHP API’s for it. And after 10 minutes of tweaking someone else’s code into my own, walla, I’m up and running. I’m sure someone with something to prove could get it to work, but fuggit… Flash displays stuff well; let it do that. PHP is doing just fine getting data, handing Google happy headers, and sending the appropriate data to Flash. Since I used ARP, it wasn’t that big of a deal to change my Flash Lite AS2 code.

    Since Flash Player 9 can JIT to ARM (phone processors), I really question if the above really will become obselete or not assuming Flash Lite 3 is some sort of Flash Player 9 build (at least gets the new AVM if not Flash Player 8’s display capabilities). Since ActionScript compiled to machine code is just that more lightning fast, does it really remove my loop barriers? Can I continue using XML DOM parsing (which is faster than E4X) once AVM+ comes alone? Flash Player 8’s new garbage colletor I’m sure might help, but frankly I’ve been seriously impressed with Flash Lite 2’s ability to reclaim memory after I destroy stuff. As usual, my biggest bottleneck is the phone’s slow arse processor, and my next to nothing sequestered RAM food tablet for the day. Does insane speed fix these problems?

    I think not. 5 milliseconds vs. 15 seconds to use too much RAM is just a shorter time to use too much RAM. XML parsing would be nice to have back; I can rip out the data I need, and kill the XML reference, hoping that the GC removes it quickly. Even so, the amount of XML cannot be helped; some RSS / Atom feeds which are used in lieu of a lot of XML-RPC interfaces nowadays are pretty verbose. When a lot of these API’s are aimed at Java developers… well, I’m at the short end of the totem pole, and thus request the help of a middle-tier to do some of the heavy lifting; in this case, PHP parsing XML and making pretty headers.

    Anyway, summary: If anyone tells you Flash Lite 2 can be used to make apps, they are incorrect without including some other part that does the heavy lifting like some middle-tier, or if the SWF is embedded in a Java / C shell on the phone / device. My definition of app is an application you’d make on the desktop, only for the phone. For example, her majesty’s blackberry has a Java app that Google makes that allows you to check your GMail. It’s pretty slick, works, and is a great app. Nothing complicated compared the actual web interface of GMail.

    This is the 3rd time I’ve had both swords unsheathed, and charged full steam ahead, letting loose my nordish battle cry. All with a gung ho attitude at creating something cool with Flash Lite 2. Each time, I’ve hit a brick wall when it comes to data parsing and anything to do with loops.

    It’s all good, it’s still a very valid presentation layer, and middle-tiers abound with heavy lifting & parsing abilities. Besides, ARP works just fine in Flash Lite 2, and I’ve got enough UI components to be dangerous. My advice; plan in advance to have a backend along with your Flash Lite 2 front-end.

  • Debugging Tools for HTTP Headers with Flash?

    How are people debugging http traffic in Flash? I used to use HTTPLook, but switched to ServiceCapture because at the time, I was doing a lot of Flash Remoting and it nicely showed AMF traffic. Fiddler’s ok in that it shows better formatted headers, but I like ServiceCapture’s GUI better. Finally, I’ve got Wireshark the packet sniffer (which apparently is based on some sort of open source standard because out of the 3 I’ve tried today it’s GUI is almost identifcal). I can’t read what the heck Wireshark is outputting, but it’s helpful when you are debugging things for those who CAN read it.

    Bottom line, none of the above appear to be showing all of the headers Flash is sending. Using addRequestHeader for LoadVars in Flash Player 8 and 7 doesn’t seem to work in the above, at least for the few headers that I’ve tried that aren’t in the not allowed list.

    Beyond Charles, what are others using? What do you use to see what headers Flash is sending?

  • Game: Elder Scrolls Oblivion

    I rented Oblivion from the local video store at the behest of Jeff that I play it. It is a really fun game. It’s basically a role-playing game combined with a first-person shooter. It does have a 3rd person view as well. You can choose from a few different races, and your class & speciality skills after a first short adventure to get you acclimated. I chose a Wood Elf and created a custom class called “Swashbuckler”, although, I’m finding I much prefer the bow to the sword.

    While the game has an overarching goal, it’s extremely free form, staying true to the Elder Scroll roots. You can travel just about anywhere, take random quests, and explore locales at your own pace. While the pressure to complete an objective is what makes games fun and involving, Elder Scrolls mini-quests have that too. The graphics are pretty good on theXBox 360. I had to lower the brightness 1 notch to make the game realistic. Even so, if I don’t use a torch, I sometimes miss loot laying around. It’s interesting that the game actually supports pupil adjustment with lighting. Meaning, if you are in a dark cave with ambient lighting, you can see a short distance down the cave. If you suddenly light a torch, while you can see your immediate environs very clearly, but the distance at which you can see is greatly reduced as the light of your torch only extends so far and your pupils shrink, thus not allowing in enough light to see the darker environs. Combine that with your REAL eyes adjusting to the TV screen, and you get quite a unique experience while crawling around dungeons. I’ve been very tempted to light a torch, only to remember the last time I did, I blinded myself, and walked right into a goblin berserker. Very neat feature. I play a sneaking scout, but I’m sure a wizard might have it rough with all the flashy spells.

    I haven’t read the instruction booklet, so I may be missing some things, but I’m extremely critical of the games sneaking portion which just so happens to be my specialty in game. Games like Thief had this down to a science. You had a gem in the middle of the screen, and it’s glow represented how visible you were including the noise of your footsteps. There were varying levels of how hidden you were, and you could utilize shadowy areas to your advantage as well as objects to hide behind. Not so in Oblivion; it seems that the algorithm they used is extremely generic. You are basically hidden or not. The amount of stress, and thus the awesome rush you’d get in Thief or Splinter Cell is just not there in this game. If you decide to play a rogue like character, keep this in mind. Additionally, hiding behind objects doesn’t really appear to do anything. A minor consolation is that you CAN become hidden later on, even in the midst of a battle. In games like Thief and Might & Magic: Dark Messiah, if you were spotted, the jig was up and you did the classic fight or flight scenario. This game, however, is a lot like Splinter Cell in that if you move slowly and stay hidden in dark areas, enemies can lose track of you. For a hide in shadows guy like myself, this is great because I get a bonus to damage, even with my bow.

    There are a lot of realistic elements in the game like degrading quality of goods that you have to repair yourself or pay someone to do so, wandering around harvesting plants and stuff to make potions, the classic (the best feature of earlier Elder Scrolls) dress up. There are a bunch of outfits you can buy with bling to make your character look pimpin’. I’m not sure if this affects your Speechcraft skill, but regardless, neat feature.

    There are some silly things, like a high level Acrobat skill value can allow you to jump off the surface of the water. I think this is pushing suspension of disbelief too far, and they should keep that in the realm of magic. Additionally, you can dodge with a bow if you equip it, block with it, and then jump. Since keeping a heavy shield on hand for a bower kind of sucks, my guess is they did this to give archers a fighting chance in confined dungeons. Easier to do with an XBox controller than on a PC, I’ll give them that.

    The last cool thing is that this game takes great advantage of the XBox Live Marketplace. There are some extra levels you can purchase (some were free for Christmas). In particular, I bought horse armor for $2.50 (200 points) and I don’t even own the game! Her majesty and I were looking at what you could buy, and the light elven armor looked awesome on the horse. Since horses are expensive (1000 gold) your first set of armor is free and helps protect your horse. Additional armor is 500 gold thenceforth from the stable near the main Imperial town. While the horses eyes do look spooky, it’s a pretty neat addition, the fact you buy whole games and accessories via the marketplace.

    I used to be a big critic of Valve’s Steam which is their endeavor to get a gaming platform, complete with their Source engine, to allow developers to build games, deploy them, and get paid without the insane publishing / delivery costs typical in boxed packages you’d get at Best Buy for example. This really frustrated me because Steam felt extremely low quality, even today years after it’s initial inception and I would prefer Valve spend their time on creating games vs. platforms. However, after purchasing Half-Life 2: Episode 1, and then Might & Magic: Dark Messiah based on the recomendation of Johnathan Miranda… I can see where they are going and have enjoyed both purchases immensely. Therefore, I think XBox Live really has something special here. I think they overcharge like crazy, though, and their naming scheme based on your name being synonymous with your account is a flawed system. ICQ’s best feature was they gave you a unique number. You could then use whatever name you wanted. XBox Live actually charges you to change your name, as well as changing your picture, and a variety of other personalization’s which I think will hurt them in the long run. Regardless, overall, it’s really cool to be able to choose things I want to purchase for an existing game; really really cool!

  • Consulting in 2006

    2006, the year of consulting. I hooked up with Universal Mind, a consulting firm and Adobe partner in February after 6 months of flirting. I did 3 main gigs, learned a lot from them, and felt like I’ve grown. Additionally, I’ve found more focus on where I want to take my career. I know what I want, and am working on finding out how to get there. I really want to get more into product development as opposed to service work.

    If there is one thing this year taught me, it’s that Flash isn’t dead. I was under the false impression that I’d be able to do full time Flex development on larger scoped projects. Quite the contrary. My year has been full of both. This should hopefully strike a chord for Flash designers and developers in 2007; your skills are in high demand. Charge for it, and recognize you have valuable contributions to make in larger Flex development projects. Furthermore, I don’t see this changing for at least 2 years. A lot of the software development influx from enterprise companies as well as smaller firms do not have the multimedia and/or design talent for the type of Rich Internet Applications they wish to build. They will be looking for designers familiar with Flash & Flash Developers familiar with software development practices to help bridge the chasm, fill in the gaps, and warn of any pitfalls based on copious Flash Player knowledge. Some of the egotistical ones will think they can do this without designers and/or Flash devs, quickly realize they have talent gaps, and then reach the same conclusion. This will happen in the WPF/E community as well, so keep your eyes open for early adopter opportunities there as well.

    If anyone tells you we are not in another internet bubble, they are full of shit. A lot of people have approached me, and colleagues of mine this year with start-up & product ideas that have no chance to make money. Most are either burning cash from VC’s that isn’t theirs, and they have no accountability for failure. Others just don’t know they have no chance to make money… but most are the former. The risks I guess to them are worth it for they are looking for the big payouts. Looking at history, it took about 2 years for the nastiness to hit the fan. I’d assume things would go quicker, and thus my predictions for 2007 are a lot harsher than most. How many of the following do you think can pay their bandwidth bills next year with revenue generated?

    I’m a programmer, not a businessman. I specialize in writing code, not making money. Therefore, maybe I’m being to harsh, but frankly, I’ve seen this before; a lot of us have. We know how it ends. Either way, in the end, the consumer (contributor?) is winning so it’s all good.

    My mouth has gotten me in trouble 3 times this year. 3 different people I look up to and respect have politely suggested I learn tact noting it’ll improve my public image and help my career. I feel I am most effective at being brutally honest since I see a lot of people not speaking up when they should. Uphill battle.

    Another thing this year taught me is that I’m still wrong about 2002. Back then, I had said Flash Development would continue to be a seller’s market till mid-2004. By then, the talent pool would be saturated, and contractor rates would fall, as would the influx of work. That hasn’t happened, and doesn’t look like it’ll end at least until 2008. I read a few websites this year that said companies were keeping tight IT budgets. Not sure what companies those were because the amount of work this year was just like the last 3; crazy. The difference is, I’ve done next to nothing to promote myself or my services this year, and yet I’ve nearly tripled the amount of requests for work and talent this year compared to 2003 and 2004 as well as steadily increasing my rates. I DO however believe my CTO, Tom Link, that the Flex Market will do what the .NET one did, and get saturated. The reason I was wrong with Flash is that the barrier of entry to Flash Development never lowered. That’s why Macromedia made Flex in the first place. The ease at which entire engineering departments of companies, not just various individuals of programmers, have taken to Flex is scary as is the enthusiasm. While it’s great to see SWF finally become en masse a de facto standard for deploying applications to the web, it’ll make it harder to justify rates and certain processes as more and more people get into Flex development by the end of 2008.

    Talent was still a big problem this year, and I see it getting worse next year as many of us continue to be stretched thin. The 2 worst contributors are designers and information architects next to qualified Flash Developers. A lot of designers currently either aren’t interested, or just don’t know, that they can contribute to Flex development and Flash Lite development. Flexdevs, at least right now, are programmers. Programmers don’t design. Furthermore, a lot of programming teams I’ve been on do not have design advocates. Meaning, they don’t care about design. Designers need to get someone on those teams (like me!) that are design advocates, and thus demand the correct processes be implemented between teams. On the Flash Lite front, Flash Lite 2 can take it’s time; there is still plenty of Flash Lite 1.1 work cropping up that is best suited from designers even with next to no scripting knowledge. Most have the same problem I do; we don’t know, nor care to know, about phones. Cell phones have reached a near alpha geek status. They are so different from the computer industry right now, that it really takes an early adopter mentality to get in the door. This is NOT a requirement, however. The club has no cover charge, it’s just down an alley on a hidden street. Employers don’t realize they can hire the same designers that tried to pass themselves off as Flash Developers 3 years ago for Flash Lite 1.1 work.

    IA’s are the worst. Not only are they still debating about their titles & place in the world, but their discipline is grossly under-appreciated and misunderstood by the industry as a whole. Yet, they are the keystone to make most of these projects work, and they aren’t put on these projects. Information Architect, Interaction Designer, User Experience Head… whatever. There aren’t enough of you, and employers don’t see why in the nine hells they should pay for you. I know why, but I’m not a salesman. Frankly, it’s not my industry either; I’m just helping out because I know the ramifications if I don’t. Not sure impact in 2007, but I DO know I’m going to start putting my foot down next year on future projects more forcefully, demanding an IA in some form; it’s starting to get ridiculous. The cop-out small companies use is “the designer does both”. If you see the red flags, come crying to me, I’ll buy you a beer, and together we can pray it’ll be better in 2008… or start a bar fight.

    I’ve seen a few big companies putting forth massive cash budgets towards Apollo endeavors. If you’ve read my blog at all this year, I’ve mentioned Apollo like once. I’m kind of over Flash on the desktop. Yes yes, I know it’s HTML,CSS, JavaScript, AJAX, and PDF too, but I don’t use those technologies, and most of my jobs have no need for what the desktop provides. Regardless, there are a ton of people who aren’t me, who do use those technologies, and who need those features. As such, designers & IA’s will have valid contributions here. If there is an Apollo project going on in your company in 2007, and your a designer that isn’t on it, raise hell. Same goes for Flex.

    Next to my trip to Europe and something else to be left unmentioned till early 2007, one of the greatest experiences I had this year was getting to buy dinner for some of the main dudes who made my career in this industry. There were countless others including some other “main dudes” both from Macromedia / Adobe, as well as non-employees but if you were at MAX 2006, did YOU meet everyone there AND take them out to dinner? Didn’t think so. Shown here, I got to hang with Peter Hall of Flashcoder fame amongst other things, Nigel Pegg who wrote a lot of the early Flash MX & Flash MX 2004 components as well as one of the dudes who wrote Breeze with Peldi and a bunch of others, Steven Sacks who has helped keep me sane, and Chafic Kazoun of authority on Flash coding standards fame amongst other things. It’s pretty surreal; after you’ve met these dudes and become friends with ’em, you no longer fawn over them in person. You can still put them on pedestals, but you don’t act like it. You just talk as equals about cool shit, like code, business, and life… and take the piss out of ’em on occasion. Anyway, was an honor to buy dinner for these dudes, glad I got the opportunity.

    Flash / Flex Rock Stars

    Beyond finding product work in 2007 (finding & doing are 2 different things, hehe!) and demanding more IA’s on projects, I have a bunch of things planned. Speaking at the 360Flex conference as well as WebDU in March. Furthermore, I’ve been doing a lot of Flex 2 & Flashcom in my spare time, so hope to share more code on that front as well as the logistical challenges involved in deploying a service based product. Finally, I plan to release my open source Flash Lite 2 component framework written in ActionScript 2 called “Shuriken” by the end of March.

    The wine is making my fingers go numb… hey did I miss New Years? Crap, it’s 3:12am! Oh well, happy new year, suckaz!!!