Over at ReadWriteWeb there is an article surrounding Adobe AIR focusing on the honeymoon period being over for the framework.
But recently, we’ve begun to question AIR’s longevity. Now don’t get us wrong – many of our favorite apps (TweetDeck, Tumbleweed, Yammer, etc.) are built using Adobe AIR. However, there’s no reason why these apps couldn’t just run in a browser instead…and that might even be a better place for them.
I personally think that the writer does have a point, some of these AIR apps could live in a browser and indeed many of them probably share the same codebase as a standard browser version.
There appears to be a love of all things Google Chrome in the article and comments, I like ReadWriteWeb but comparing Adobe AIR and Google Chrome with Gears is not entirely a like for like comparison.
It’s early days for this technology and yes it has been marketed towards the developers with web skills and not those who ultimately posses the money to create large AIR apps. The eBay desktop AIR application mentioned in the article is an example of where the design and UX can be enhanced by a RIDA (Rich Internet Desktop Application).
RIA’s are only just taking hold in large corporates where the management are people who remember Flash as a technology that kids used to build sites with long preloaders, skip intro links and hard to read text. This is not the case today with the Flash platform.
Adobe AIR performance needs to be addressed, there is now a general consensus of AIR is a resource hog. You only have to look at Twitter to see what the general public are saying about the runtime. Those developers in the know are aware of how to fine tune their AIR apps to maximise memory, especially when idle but it is the wider circle of people creating these apps that don’t optimise then AIR gets a bad name. This is not detracting from the fact that the performance of AIR needs to be much better and leaner. Having the ability to run multiple AIR apps that are running in the background sending notifications syncing data while the user is focused on another task while not hogging CPU and memory is vital.
It’s also my personal view that AIR should have two differing runtime types, widget and full blown app mode these would use different sandboxes and permissions. I know this has been mentioned before but how great would it be to simply click on a AIR widget link in a webpage and have it instantly appear as an AIR widget on your system. A fast seamless user experience without the usual security hassles because they run in a sandbox that is different to fullblown AIR apps.
These AIR widgets could then possibly talk to and launch main AIR apps and even have rich communication with say a Flash or Flex app inside a webpage for updating, retrieving data or similar.
AIR could also benefit from having GPS connectivity with a host system for location based data, yes I appreciate that this is a very complex thing to do and will probably not see the light of day for quite a while. If mobile apps have this functionality then why not AIR apps and widgets running on mobile and netbook pc’s.
Most people have got the AIR runtime because they installed Tweetdeck or another Twitter application, they might not even know they have it! AIR is much more than just Twitter apps just looking through RefreshingApps.com you see the diversity.
In response to the question of do RIA’s have a future or just better browsers, I say that RIA’s are definitely the way forward for certain applications and tasks. HTML 5 is an interesting step forward and with 3D CSS transformations coming (well available now in the nightly release of Webkit) we are going to see richer apps and sites running inside the browser without the Flash plugin. I welcome this BUT the teams at Adobe working on AIR and Flash need to keep ahead of the game, make things leaner and faster. If you can do basic 3D in a browser with very little resource overhead why would you need the Flash plugin?
AIR has so many business advantages over apps running in the browser. The offline capabilities are a prime example not just database access but ability to save and write files to the host machine with a custom file extension, comprehensive drag and drop, system tray and dock notifications the list goes on.
Adobe need to get AIR and ActionScript 3 onto mobile FAST. This is the killer for dev’s and businesses alike. Write once and deploy to set top boxes, mobiles, netbooks etc. is key. BUT it needs to be fast especially on mobile and portable devices, if they release without performance being top then Adobe risk the lifespan of the AIR runtime altogether.
IN the meantime let’s just keep on producing content and apps that people can use TODAY on the desktop with AIR.
[Caveat: I’m an Adobe Community Expert for AIR, however I’m not paid or employed by Adobe ]
