Cole, I'm curious: What route would you suggest for learning native?
On Feb 11, 2014, at 2:11 PM, [email protected] wrote: > Grant, I was an early adopter of Appcelerator. I did a few user group > presentations on it. Over time, it got less stable with each new > release, iOS 5 was pretty bad. Is that still true generally, or are > you being careful to avoid certain parts? I did a few prototypes with > it, but could never get approval on the real thing. The Android > performance was noticeably off, and iOS was only so-so. What really > killed it was always it was short on the feature list, the lack of > support on our requirements. > > I never considered Appcelerator for games, because it's not really > built for that. Something like Cocos2D is, but now with the release of > SpriteKit, Cocos2D seems ancient and confusing. Technically, a hybrid > game engine like Unity or Unreal seems very appropriate for games. I > played with Unity, but didn't have the time it takes to do it for > real. I have to agree with Sean on Starling for games. I'm curious why > you you would not consider hybrid for games? > > Over time, I've gotten to the point where I'm spoiled by the full > native support. It's a blast, strong community, strong vendor support. > We don't ask ourselves "if" we can do something, but "do we want to > spend the time" to do it. I can produce an app in about the same > amount of time as a hybrid, with a ton more functionality. Android is > x3 time versus iOS though (cough, fragmentation, cough). No one > complains about crashing, because the stability is there. The > performance is there. Native is extremely rewarding, and is a solid > advantage on your resume. > > -- Cole > > Quoting Grant Shipley <[email protected]>: > >> I actually disagree with some of the things said here. I think a hybrid >> approach does make sense given the correct use case. Should you develop a >> game using a hybrid framework? Absolutely not. However, for certain >> productivity apps it can make a lot of sense. >> >> I have the most experience with Appcelerator Titanium and am recognized as >> a titan (http://www.appcelerator.com/blog/2011/01/titanium-titan-program/) >> Using a framework such as Titanium, which doesn't run in a web view btw, >> is just like any other language. You have to spend a lot of time >> understanding best practices for usage of the platform. Sure, there are a >> lot of crappy hybrid apps out there just as there are with any language, >> but there are some good ones as well. >> >> -- >> gs >> >> >> On Tue, Feb 11, 2014 at 10:25 AM, thin <[email protected]> wrote: >> >>> I'm totally on board with Timothy here. Sean and Cole, thanks for the >>> conversation and by all means keep it going, I've reading every word! >>> >>> Everyone else, you're strangely quiet. Is that to say that there are no >>> opinions/experiences worth sharing in the "hybrid" (or other) spaces? I'm >>> really interested in actual experiences with some of the other tools >>> available as well. Conveniences and hiccups (or support holes) would be >>> very interesting to hear about. >>> >>> Thanks again! >>> >>> >>> >>> On Tue, Feb 11, 2014 at 10:20 AM, Timothy Humphrey < >>> [email protected]> wrote: >>> >>>> Hey guys. >>>> >>>> FYI: I'm really enjoying this thread. Seeing each of your positions and >>>> watching each of you support your arguments. There are so many tools out >>>> there, that it's great to hear the nuts-and-bolts from the guys in the >>>> trenches on some of these things. >>>> >>>> Thanks. >>>> -Timothy >>>> >>>> On Feb 11, 2014, at 11:10 AM, Sean Thayne wrote: >>>> >>>>> You really should check out Apache Flex. I completely disagree that >>>> Apache >>>>> is a bad place for Flex. They have been killing it. The open source >>>>> community is thriving for flex. Downloads are going up significantly >>> ever >>>>> month. And they have achieved top level project status for a year now. >>>>> Fastest project to achieve it. >>>>> >>>>> Second, there are great cross platform systems that use native UI. So >>>> your >>>>> most likely going to need to use a custom component set regardless. >>>>> Startling uses a UI framework called feathers. There are many 2D flat >>>>> business applications built on starling/feathers/air. The approach is >>>> very >>>>> similar to opengl. Only easier in my opinion. >>>>> >>>>> Another thing, Air supports native extensions. Allowing you to write in >>>>> objective c for ios and java. So any native functionality of UI >>> elements >>>>> you want to use. You can. >>>>> >>>>> Also air lets you control the chrome of the app. If you want it full >>>>> screen, you can, but it not forced to run in full screen. That's just a >>>>> setting. >>>>> On Feb 11, 2014 9:49 AM, <[email protected]> wrote: >>>>> >>>>>> Sean, using the Adobe game engine or Starling for a game is a >>> different >>>>>> animal. That's doing hardware acceleration, which definitely spanks >>>>>> Javascript. Yes, Adobe is still very serious about the gaming SDK, and >>>> they >>>>>> should. It's very good, you have a good point. If you're doing a game, >>>> that >>>>>> opens up a lot of other choices as well as AIR. Game engines take over >>>> the >>>>>> entire screen render. I was referring to non-gaming apps, where >>> current >>>> OS >>>>>> component and system feature support are important. >>>>>> >>>>>> The watchESPN app is a perfect example of a hybrid's hot and cold >>>> history >>>>>> at best. Let me quote a 4 star review, "App suddenly crashing is not >>>>>> happening." And a 3 star review, "It still crashes fairly often and >>>> there >>>>>> are some UI glitches...My biggest question is why it's currently >>> taking >>>> up >>>>>> almost 250MB of data." That's just in the latest 6 reviews on iOS. >>>> Google >>>>>> Play has a lot of the same issues, where each phone is either working >>>>>> great, or not working at all. I don't think ESPN has their mobile act >>>>>> together at all. They need to start over. My opinion. >>>>>> >>>>>> I'm not as excited about where Flex is with Apache. When Adobe was >>>>>> throwing serious money at development, yes. Considering Adobe's >>>> competition >>>>>> in the hybrid platform realm, the CES award was a given, at least for >>>> me. >>>>>> Currently, Adobe AIR is still the best, easily. Adobe pumped so much >>>>>> development into it, I don't see anyone coming close to that >>> engineering >>>>>> effort for a few years at least. Last I checked, FlexJS was two guys >>>>>> part-time, but that was two years ago. All good people for sure. I >>> just >>>>>> can't say it's the future of mobile. >>>>>> >>>>>> I'm a dedicated Adobe fan. I love what they are doing in HTML/CSS. I >>> use >>>>>> Adobe tools every day for web dev and app graphics. Just not as my app >>>>>> tool. Unless you are gaming, which uses very little of the actual OS, >>>> you >>>>>> are very limited. You're going to spend a lot of time explaining what >>>> you >>>>>> can't do, or can't do reliably. Those original mixed sentiments are >>>>>> justified and correct. I can't recommend cross platform for non-gaming >>>> apps. >>>>>> >>>>>> -- Cole >>>>>> >>>>>> Quoting Sean Thayne <[email protected]>: >>>>>> >>>>>> Cole, your correct that adobe stopped supporting "flash player" on >>>> mobile. >>>>>>> But Air and Flash Player are very different. Adobe IS still >>> supporting >>>>>>> Air. >>>>>>> They are also still adding awesome new features. That's one of the >>>> reasons >>>>>>> it won the CES award for best mobile framework this year. Adobe has >>>>>>> consistently held their stance that mobile AIR is the best cross >>>> platform >>>>>>> mobile system to date, and they work hard to continue to support the >>>>>>> latest. They are already one of the first frameworks to support >>> apples >>>> new >>>>>>> x code 5+ mandate. >>>>>>> >>>>>>> Another thing, many awesome companies use Air. ESPN, Angry Birds, >>> IHC, >>>>>>> etc, >>>>>>> etc. They all use the starling framework which has comparable speeds >>> to >>>>>>> native. The kind of performance that JavaScript apps will never beat. >>>>>>> >>>>>>> Lastly the key framework "Adobe Flex" is now owned and maintained by >>>>>>> Apache. The key adobe inventor Alex Harui is still fulltime on flex. >>> As >>>>>>> well as a bunch of awesome apache guys. It's very well supported at >>>> this >>>>>>> point. Their dev/user mailing lists are extremely active (100 emails >>>> each >>>>>>> per day). They are also hard at work on a new system called FlexJS >>> that >>>>>>> will allow you to use common libraries between your Air and >>>>>>> HTML/JAVASCRIPT/CSS apps. They are using Google closures on this end. >>>>>>> Which >>>>>>> is a very awesome choice. >>>>>>> On Feb 11, 2014 1:15 AM, <[email protected]> wrote: >>>>>>> >>>>>>> Hi Everyone, >>>>>>>> >>>>>>>> I'm going to be a bit blunt, but I want to share my experience doing >>>> apps >>>>>>>> full-time for 7 years. >>>>>>>> >>>>>>>> As an Adobe User Group Manager, and now an Adobe Community >>>> Professional, >>>>>>>> I >>>>>>>> think I can speak on what Adobe is doing and recommending. Flash >>>>>>>> Professional is focused on producing HTML 5 content, especially >>>> Canvas. >>>>>>>> Which I think is smart. Flash is no longer focused on using AIR to >>>>>>>> produce >>>>>>>> native/hybrid apps since Adobe stopped supporting a Flash mobile >>>> plugin. >>>>>>>> I >>>>>>>> think it is wise to take the hint. Speaking for myself, I would not >>>>>>>> recommend AIR development for mobile. >>>>>>>> >>>>>>>> PhoneGap is a product that has Adobe?s focus. This is a previously >>>> open >>>>>>>> source project, and it is a hybrid, using HTML, CSS, and Javascript. >>>>>>>> Speaking for myself, it?s a crash-fest. I would not consider ANY >>>> hybrid >>>>>>>> solution. I?ve tried all the multi-platform frameworks and tools, >>> from >>>>>>>> Appcelerator to Embarcadero. I?m done. No more. Here?s why you >>> should >>>> go >>>>>>>> native, and native only for apps: >>>>>>>> >>>>>>>> 1) Wrapped-Browser rendering leaks memory. It will blow up on you. >>>> It?s a >>>>>>>> fact. It?s not up for dispute. Just ask the likes of Facebook and >>>>>>>> LinkedIn, >>>>>>>> the loudest supporters of this approach. Everyone that matters has >>>>>>>> abandoned it, code in the dumpster. They started over. >>>>>>>> >>>>>>>> 2) The biggest myth in the mobile world, by far, is that hybrids >>> save >>>>>>>> time >>>>>>>> and money. They don?t! Hybrids fail, and they fail huge. Hybrid >>>>>>>> development >>>>>>>> is becoming synonymous with developers that just take your money and >>>>>>>> leave >>>>>>>> you with a bad app. Ouch. >>>>>>>> >>>>>>>> 3) OS support by these tools is late, incomplete, or completely >>>> missing. >>>>>>>> Xaramin is a great example of a tool that has all these checkmarks >>> on >>>>>>>> their >>>>>>>> site of what they support, like video. But when you read the fine >>>> print, >>>>>>>> 95% of video properties and methods are NOT supported. If you are >>>> lucky, >>>>>>>> you?ll get a general subset of cross-platform-common-denominator >>>>>>>> support, >>>>>>>> a year late. You aren?t getting everything in Android and iOS, not >>>> even >>>>>>>> close. >>>>>>>> >>>>>>>> 4) Quality matters. Bad apps get deleted. This is exponentially true >>>> for >>>>>>>> iOS users. Like it or not, the native competition is plentiful and >>>>>>>> fierce. >>>>>>>> Do not bring a butter knife to a gunfight. Hybrid apps are >>> notoriously >>>>>>>> slow, and leave a very large footprint. Even if your app is >>> marginal, >>>>>>>> once >>>>>>>> the user is looking for bloated apps to delete to make more room, >>> your >>>>>>>> hybrid app is sitting at the top of the size list. >>>>>>>> >>>>>>>> The reality of the hybrid cross-platform solution is you?ll be >>>> unstable, >>>>>>>> low-featured, old, slow, and bloated. There?s no free lunch. There?s >>>> no >>>>>>>> shortcut. Don?t pretend you can play with the big dogs in an app >>>> store. >>>>>>>> It?s mobile career suicide. If you want to develop apps, do it >>> right, >>>> go >>>>>>>> native. >>>>>>>> >>>>>>>> If you want to do mobile, but you want to stick to HTML, there?s >>> good >>>>>>>> news. Jump hard into responsive and the frameworks, and make a great >>>>>>>> mobile-enabled site. Adobe has awesome tools for that also. >>>> Angular.js is >>>>>>>> cool. There?s very good money in that work right now. That?s >>>> definitely >>>>>>>> doing it right. Very hipster. >>>>>>>> >>>>>>>> -- Cole Joplin >>>>>>>> >>>>>>>> Quoting thin <[email protected]>: >>>>>>>> >>>>>>>> Thanks Sean, I thought of you with this question for sure (and >>> assumed >>>>>>>> you >>>>>>>> >>>>>>>>> would tout Air) :) In fact, I pitched Air to some people around >>> here >>>> and >>>>>>>>> got some mixed sentiments with some of the complaints being >>> bad/slow >>>>>>>>> updates/support and claims that Air is being abandoned by Adobe >>>> (which I >>>>>>>>> have not way of verifying). >>>>>>>>> >>>>>>>>> What about PhoneGap+Steroids (by AppGyver)? Anyone out there have >>> any >>>>>>>>> pro/con experiences to share there? >>>>>>>>> >>>>>>>>> >>>>>>>>> >>>>>>>>> On Mon, Feb 10, 2014 at 1:41 PM, Sean Thayne <[email protected]> >>>> wrote: >>>>>>>>> >>>>>>>>> AIR, haha, but really, it won the 2013 mobile framework of the year >>>> at >>>>>>>>> >>>>>>>>>> CES >>>>>>>>>> last month. >>>>>>>>>> >>>>>>>>>> ~Sean >>>>>>>>>> >>>>>>>>>> <http://www.skyseek.com> >>>>>>>>>> class *Sean_Thayne* >>>>>>>>>> extends Developer { >>>>>>>>>> public $skype = "sthayne23"; >>>>>>>>>> public $gTalk = "[email protected]"; >>>>>>>>>> public $url = "www.skyseek.com"; >>>>>>>>>> } >>>>>>>>>> >>>>>>>>>> >>>>>>>>>> On Mon, Feb 10, 2014 at 1:27 PM, thin <[email protected]> >>> wrote: >>>>>>>>>> >>>>>>>>>> Heya all, >>>>>>>>>> >>>>>>>>>>> >>>>>>>>>>> This may be a little off topic, but I wanted to ask about >>> peoples' >>>>>>>>>>> experiences with multi-platform mobile development and which >>> tools >>>> to >>>>>>>>>>> praise or avoid, pros/cons of various tools, etc. >>>>>>>>>>> >>>>>>>>>>> >>>>>>>>>>> Everyone's combined experiences would be really awesome to hear >>>> about, >>>>>>>>>>> so >>>>>>>>>>> feel free to speak up about Titanium, PhoneGap, AppGyver, Sencha, >>>>>>>>>>> Rhomobile >>>>>>>>>>> and beyond! >>>>>>>>>>> >>>>>>>>>>> What I'm trying to do is to make a good pros/cons list of the >>> most >>>>>>>>>>> common >>>>>>>>>>> and/or current and/or robust offerings out there. >>>>>>>>>>> >>>>>>>>>>> Thanks tons! >>>>>>>>>>> >>>>>>>>>>> _______________________________________________ >>>>>>>>>>> >>>>>>>>>>> UPHPU mailing list >>>>>>>>>>> [email protected] >>>>>>>>>>> http://uphpu.org/mailman/listinfo/uphpu >>>>>>>>>>> IRC: #uphpu on irc.freenode.net >>>>>>>>>>> >>>>>>>>>>> >>>>>>>>>>> >>>>>>>>>> >>>>>>>>>> _______________________________________________ >>>>>>>>> >>>>>>>>> UPHPU mailing list >>>>>>>>> [email protected] >>>>>>>>> http://uphpu.org/mailman/listinfo/uphpu >>>>>>>>> IRC: #uphpu on irc.freenode.net >>>>>>>>> >>>>>>>>> >>>>>>>>> >>>>>>>> >>>>>>> >>>>>> >>>>> >>>>> _______________________________________________ >>>>> >>>>> UPHPU mailing list >>>>> [email protected] >>>>> http://uphpu.org/mailman/listinfo/uphpu >>>>> IRC: #uphpu on irc.freenode.net >>>> >>>> >>>> _______________________________________________ >>>> >>>> UPHPU mailing list >>>> [email protected] >>>> http://uphpu.org/mailman/listinfo/uphpu >>>> IRC: #uphpu on irc.freenode.net >>>> >>> >>> _______________________________________________ >>> >>> UPHPU mailing list >>> [email protected] >>> http://uphpu.org/mailman/listinfo/uphpu >>> IRC: #uphpu on irc.freenode.net >>> >> >> _______________________________________________ >> >> UPHPU mailing list >> [email protected] >> http://uphpu.org/mailman/listinfo/uphpu >> IRC: #uphpu on irc.freenode.net >> > > > _______________________________________________ > > UPHPU mailing list > [email protected] > http://uphpu.org/mailman/listinfo/uphpu > IRC: #uphpu on irc.freenode.net _______________________________________________ UPHPU mailing list [email protected] http://uphpu.org/mailman/listinfo/uphpu IRC: #uphpu on irc.freenode.net
