5 ms·
Client adoption isn't a problem any more, IE is declining to around 30% on most popular sites. The largest issue with killing flash is the legacy: video players
by pippy 14y ago
Client adoption isn't a problem any more, IE is declining to around 30% on most popular sites. The largest issue with killing flash is the legacy: video players, games, even carousels use it.
If omegle was released in the next months, it likely wouldn't use flash.
- Quarrelsome 14y agoFibs. The whole reason I'm using flash in my startup (and it is a major component atm) is because of IE share (and the relative immaturity of HTML5). Are you really telling me that you want to deny your service to a fresh install of windows? While many people appreciate the benefit of better browsers there is still a significant portion that don't and hopping aboard this wagon prior to full adoption means that you exclude them or present roadblocks to them. I have the added benefit of targeting (in part) the older generation with the software I produce making all this excitement muted without cross-browser consensus.
- analog 14y agoOut of interest, and not trying to be snarky, but do you have a solution in place for mobile as well? If it we me and the choice was between supporting older browsers or supporting mobile, mobile would win out.
- shimms 14y agoI think it depends strongly on your target market. If you're building B2B enterprise software used exclusively on PCs on a desk, in an office, with no remote access, the mobile might be less important.
- analog 14y agoIt's still a tricky choice though. By going with Flash you're tying yourself into a dying technology which may not be such a hot choice if you're building a business [1] I'd still try to go with js/html and shim away as best I could for older IEs. [1]http://swombat.com/2011/12/7/picking-technologies http://swombat.com/2011/12/7/picking-technologies
- chii 14y agothats a very narrow market me thinks? also, the sort of enterprise you speak of would certainly not be purchasing software like these - they'd stick with the trusted and known big name vendors. As they say, nobody ever got fired for using IBM (or microsoft).
- analog 14y agoIt's not impossible to get your foot in the enterprise door, I've built several solutions for one particular client. If you've got a contact who trusts you then you could certainly pitch solutions to them, and the prices you can charge are a lot higher. Do you want to work in an environment where your target browser is IE6 though (not even joking).
- Quarrelsome 14y agoNot yet mobile is the last crowd we're supporting. Our application is people that are going to die. You get me?
- Quarrelsome 14y agois for people*. and yes, I'm putting it bluntly ;)
- analog 14y ago"people that are going to die" so it's aimed at.. everyone then? ;)
- tracker1 14y agoI work for a company whose target audience tends to be men over 65 years old... And we've seen mobile devices (specifically tablets) grow to nearly 17% of our traffic in the past two years.. I'd say mobile is critical, even more for seniors.
- zapdrive 14y ago"Are you really telling me that you want to deny your service to a fresh install of windows?" Does flash come pre-installed in Windows now? The last windows I used was XP, and it didn't come pre-installed in XP.
- etanol 14y agoWell, there is still one use case where IE is indispensable: http://www.bourkedesign.com/wp-content/uploads/2010/10/ie-poster.jpg http://www.bourkedesign.com/wp-content/uploads/2010/10/ie-po...
- eurleif 14y agoIE is only at 13% on Omegle. But I'd still hesitate to tell those users to get lost.
- zapdrive 14y agoLol... yea don't do that. But instead ask them politely :)
- gz5 14y agoJust detect IE users and prompt them to download a plug-in. Meanwhile, voice and video enable [your app] for the other 70% of your users. Can argue that WebRTC is not really about universal calling regardless...and yet still a game-changer: http://nextblitz.com/blog/webrtc-analysis/ http://nextblitz.com/blog/webrtc-analysis/