Why Apple is betting on HTML 5: a web history - Page 5

Having read every one the the more than 160 posts in this article, all I have to say is "down with Flash" and bring on the the HTML 5 ads that can't be blocked.

Can anyone say 'Click to HTML 5'?

Long sentences are often essential for expressing complex thoughts, particularly when it's necessary to qualify a statement in some way; they are not examples of bad writing; quite the contrary in fact: they are often necessary for including sufficient detail to allow the reader to grasp the point the writer is trying to make.

Yes. It drags the article down. Daniel Eran Dilger is a talented writer and I usually enjoy his articles and I think there is a lot insight into them. But, the mentioning of Sarah Palin hit spot on. I hope he can do better.

Tom

All that said, may I suggest a copy of "Elements of Style"? Concessive brevity.

Omit Needless Words And Other Things That Don't Add To The Thoughts Being Expressed!

Joyce and Mervyn Peake were masters of The Long Sentence. Most others are not.

And this reads like a 2000 presidential campaign news release:

"He cited the Mosiac browser being developed at the American NCSA, which had been funded by congressman Al Gore..."
Really, all by himself? Or maybe by Congress as a whole (with other co-sponsors) - and with our money. But wait, I forgot, he did invent the internet......


Kitchen sink checklist arguments ("Look! We have this and this and this!) which ignore the functionality of the resulting experience have long been the refuge of the writers of, and especially the marketers of, bloated, dysfunctional software.

This can already be done - Just not in a standardized way. Its up to the flash developer to provide this functionality. Hulu already provides this feature with their current flash video player.

Why, oh why do you insist on calling is Mosiac?

Even in the graph it is spelled wrong.
It is named after a form of art, mosaic.

Then I hope they add it as it helps the content owners, the users and all the ISPs in between.



Its a typo. Its a gross oversight for the article, but in the forum such things are less significant. Usually its best just to shot a PM to the author or even make mention in the forum, but I doubt that they are insisting on spelling it wrong on purpose.

I think I'm understanding you correctly. What I mean to say is that Hulu is already doing this in their current flash video player. If you click on the "share" button next to any video you will see options to post to sites like facebook, digg, etc, email a url to someone, and send to someone in your Hulu friends list. At the bottom of that interface there is a "timeline" for you to set start and end time for the video you are sending.

I like the idea of HTML 5 and can't wait for every browser to have it implemented, but the idea that it will replace Flash and Silverlight I think is a bit far fetched.

First off, every browser has to render things exactly the same if its going to work. This on it's own seems a hard task, but even when they do someone at some point has to convince all the big multinational companies to spend millions upgrading their workers browsers.

Second, the IT world moves faster than people writing and implementing standards. Fine HTML 5 may have the latest video support. But thats only the currently latest. During the time I've been reading about HTML 5 Silverlights released version 2 and now 3! An HTML standard is never going to get passed the fact that plugins can get updated quicker.

Lastly there's just some things it wont be able to do. Flash and Silverlight both protect code and videos. With JavaScript anyone can steel your code as it runs on the client. Flash and Silverlight are both compiled though and thats a big deal. They both also protect your videos, even if HTML 5 does this to, at some point someone will make a workaround and it will be years to weight for HTML 6 to fix it. And finally Flash and Silverlight are more than just a browser plug in, they can run in a browser or direct on your machine as an app.

So I think while HTML 5 is going to improve the web its not going to kill off the plug ins that are currently providing the technology on a lot of big sites.

That is exactly what I am talking about. I didnt know that Hulu incorporated their Flash to do pull streams at user specific points like that. Thanks for the heads up. Now we just need the HTML5 video tags to include these elements so that this can be done without the need for Flash. This will save the host site from having to encode clips separately, save the ISP and host sites bandwidth from users playing short streams looking for the right starting point, and save the user the time and effort for searching for these start and end points in the stream.

Even yesterday I wanted to point out a part of Apples last WWDC event and had to go through the event to find the start time and then email the link with appropriate start and stop points. Now that I know that this is already incorporated in Hulus Flash I feel more confident that WHATWG will add it to HTML5 at some point.

Its not going to "replace Flash and Silverlight its going be used along side it, the way Silverlight is an alternative to Flash right now. Flash will still be dominate for a very long time, but HTML5 and its video tags will be gaining ground rather quickly in the coming years. That is all but certain.

Regarding security, that is one of the most important reasons why sites like Hulu use Flash, even though you can actually snag the stream, its not very easy to do. The proposed HTTP Live Streaming that is undergoing approval by the IETF as an internet standard has an option for encrypting streams so that sites like Hulu can still protect their content while offering a much more efficient delivery system.

Yes, because running a proxy to then connect MPlayer or VLC to capture packets on the port streaming into the client is via the browser [as it will be embedded] is just soo difficult to capture the content.

The security should be focused on making sure streams aren't corrupted with random bits to then pipe to i/o devices or ttys to then run as background processes, regardless if it's on Windows or Unix.

I agree that Flash has some life. I have never installed Silverlight, so I believe it is less ubiquitous.

However, whether they continue to be used, or not, is moot.


There are three reasons for this: mobile; mobile; mobile!


Within the next* 3-5 years the way we use use computers will change dramatically. There will be less need for desktops, laptops and even floortops.

The growth segment will be inexpensive, expendable/replaceable appliances-- the mobile. They will always be with us, and we will always be connected (cell, WiFi).

They will be lean/mean content creation/capture and delivery appliances.

The applications will be specialized for specific tasks. They will be easy to use, interrupt, then resume. Unnecessary bloat will be pared, as will performance and battery guzzling functions (Um, Flash anyone?).

It matters not, whether this or that web site uses a Flash or Silverlight browser plugin. If the site owners want to serve this content to mobiles (and they will); they will provide a direct, alternate, path which bypasses the browser altogether.

Look around, this revolution has already begun!


*

Mobile is growing so fast that even when Flash is still dominate on the desktop the shear number of mobile devices with real web browsers and apps that tie into efficient streaming video will, IMO, make developers think twice about using Flash as their medium. There is just too much change on too many fronts that Adobe can’t possibly fight with Flash. They will still be dominate the way IE is still dominate but like IE it will become the less desirable standard for developers. (This is my speculation, not a statement of fact)

As for Silverlight. It’s actually quite good for a MS product and I’ve heard that coding for it is pretty nice, too. The great thing about Silverlight is that it came out for all platforms at around the same time and it had HW acceleration. This seemed to put a fire under Adobe who later added it. Besides being another front that Adobe has to fight to maintain Flash, I don’t think that Adobe would have added the HW acceleration or H.264 when it did if not for Silverlight’s existence. (Again, this is my speculation, not a fact)

I just had a different thought. Is this better for developers or more of a headache. With phones like the iPhone growing in popularity and support for video in HTML 5 but no support for Flash or Sillverlight (more on Flash being an issue), what does a developer go for. At the moment its easy you just choose flash as its on over 90% of the PC's on the net. Adding flash video to a site is also relatively easy. Im far from an expert in flash and the other week it only took 3 hours to find out how you do it and get a player skinned as per design. Silverlight I would like to see take off as thats even easier as there's a Player .NET control that you just point at a wmv and it does the rest.

If popular phones don't support Flash though and until 90% of browsers people use are HTML 5 compliant we're just going to end up having to do browser recognition and present videos in both ways! Thats not easier, thats a hell of a lot more work!

Exactly!

I want to amplify the point of [single-purpose, limited-function] apps on the mobile device acessing web content directly (rather than through a browser).

On the iPhone, examples are: Maps, Weather, Stocks, YouTube, Google Earth and Google (search front-end to browser).


I have been experimenting with some music videos stored on my MobileMe site. I can access these through Mobile Safari on the iPhone.

it is slow and clumsy (pretty much unworkable on the iPhone):

-->(delay) Main Page-->(delay) Selected Video Page-->(delay) Play Video-->(delay) Return to Main Page...

...repeat the above for each video.

Most of the churning is web-page-to-web-page transition.

The browser gets in the way!

Certainly, a lot of the problem is because the web sites are [mostly] designed for large screen computers with lots of RAM (for caching pages) and high-bandwidth Internet connections.


For contrast, a speciality iPhone app accesses the MobileMe site. During startup, the app accesses a special file* that contains references to all the, text, image, and video content. After a few seconds, the app displays the "covers" of the videos in tabular or FlowCover** format. You can flip through the "covers", select one, and play it:

-->(no delay) FlowCover Select video-->(delay) Play Video-->(no delay) Return to FlowCover Page...

...repeat the above for each video.

There is no transitional delay! The actual video play is straight forward, and begins streaming after a few seconds-- just as it does in the browser (when you finally get to it).

It performs quite well even on 3G. This is especially notable because MobileMe is not known for its streaming capability


* the special file is roughly equivalent to the site's XML file without tags and unneeded information (uses about 9% of the bandwith).

** FlowCover is open source equivalent to Apple's CoverFlow. iPhone developers cannot use CoverFlow as it uses a private API/Framework.


I think the above illustrates how many users expect to access content on on a mobile device: get in quickly; get to where you want to be; do your thing; get out.




Mmm... you've piqued my interest... got any links to sites that exploit Silverlight.


*

Yeah, with video and audio integration in html 5 it will be even easier. The main reason FLASH is so prevalent and good right now for delivering video content is because its the only way to author once and play on virtually any platform or in any browser. Currently, if you encode h.264 mp4 videos, they will play in any flash player like youtube or hulu uses, and it will also play on the iPhone with quicktime. Right now that's the best combination.

The other video alternatives are in one way or another limited by browser or platform. Quicktime embedded in a web page the way hulu or youtube works, would be playable fine on macs, but would require a quicktime download in windows, and all your javascript to talk to the player would have to be quicktime specific. Windows media video embedded in a web page hulu or youtube style, would play fine on windows machines, but would require a flip4mac download to work on macs and play in the quicktime player. Then you would need to write javascript to talk to the quicktime player plugin AND javascript to talk to the windows media player plugin, and the code to determine which one to use based on the visitors platform and browser. None of these solutions take into account the few users on Linux.

Flash is a "magic bullet" in this regard because it solves all those problems. You encode once in FLV (flash video) format, and it will play on windows, mac, linux, and it will play in all the major browsers (including the dreaded IE 6) with one javascript API for all situations. Better yet, if you encode in h.264 mp4, then you can play in the flash plugin in any browser or any desktop platform, but this will also play in quicktime on iPhone (this is how youtube easily supports iPhone and AppleTV). Now if microsoft will stop screwing around and add support for h.264 mp4 in windows media player, then you can also add support for windows mobile platforms.

The point is, that Flash is the best current technology for delivering video on the web to as many platforms/browsers as possible. HTML 5 and the video integration everyone is talking about are great, but sadly, until microsoft implements them, they won't be widely adopted. This is because, like it or not, 90% or more of internet users are on windows, and more than half of them are using Internet Explorer. On top of that, something like half the IE users are using IE 6, which is a 9 year old browser. I worry that even if MS does embrace HTML 5 and the video extensions, it may be 10 more years before IE 6, 7, and 8 fade out enough that they don't have to be supported.

Microsoft is really damaging and holding back the web because they are slaves to their obligations to support old technology.

Mobile won't make developers think twice about using flash because they still have to support the desktop environment and the only way to reasonably do that is with flash.

Right now, developing for mobile devices is a different animal than developing for desktop platforms because of screen real-estate and user input devices. A site like hulu for example, would need a whole separate user interface for delivering its content to mobile devices. They wouldn't use flash to do that because of the diversity of devices and their capabilities, but this is not going to stop them from using flash on the desktop, because its still the best solution.

What would help, is if everyone could agree on a standard video format (not likely, I know), like h.264. It would be playable in web pages that use flash, and could be playable an any mobile device that supported it. Then developers could build simple mobile apps that use the same back-end infrastructure for video delivery that they use for their regular website which uses Flash for its front-end.

Back when I developed web sites, if you were smart, you used the lowest-common-denominator of features and tested on the major browsers on the major platforms. If you could run on 70% of these combinations, you were pretty much satisfied. You tended to avoid Flash (or other plugins) for anything except video, as they introduced another layer of complication.

Then along came WAP to support cell phones... it just wasn't worth the effort to accommodate the mobiles!

OK, that all changed with the iPhone. Now a mobile browser is practical, albeit Flash and other plugins are not.

So you are right, that Flash will continue for the desktop video environment, and that another means (maybe h264) is needed for the mobiles.

What is different, today, is: if you want to get the maximum eyes on your content, you you need to design for mobile first, then accommodate the desktop.

It's a switch of emphasis. Not only will plugins become less important, but web developers need to rethink how a site is designed so it supports the incidental way a mobile user will access and use the content.

*

I believe Flash will be a BIG part of mobile internet, maybe even bigger then on web. Why?

How many phone brands and models are out there which all have there own specifications, screensizes, etc.? Flash (Lite) is good, probably the best, to create applications that will work on all devices without porting, if they support Flash offcourse. But luckily that will keep on growing:
"According to Strategy Analytics, the number of Flash Lite shipped devices will reach 1 billion in 2009 and more than 2.5 billion by the end of 2010."

BUT you have to know how to do it...

Example: http://GamesPlaza.mobi . These games, well at least the ones created by BoosterMedia, can be played on all Flash Lite 2.0 handsets. This means Nokia, HTC, Sony Ericsson, Samsung, LG, touchscreen, non-touchscreen, etc. handsets.

I agree with you Peter the scope of Flash shouldn't be underestimated. By the same factor the scope of Silverlight shouldn't be underestimated.

Although Microsoft don't seem to have realized it yet, while Window Mobile ain't great, Silverlight could be the perfect technology to power a phone. Being part of the .NET framework it would also mean there would also be plenty of developers ready to build apps irrespective of if there desktop or web developers as its all the same language.

As good as Safari on the iPhone is, viewing normal websites isn't always perfect (e.g. its almost impossible to write this post on an iPhone as there's no way to see the form box all at once). With .NET and Silverlight though, a website can have a normal HTML front end for desktops and a silverlight front end for mobile all running off the same functional code. That in itself could be an amazing thing.

Another thing to question is why does the iPhone need iPhone apps as well as web apps? Silverlight can do everything and iPhone app can do and it can be used over the web. If HTML5 can replace Silverlight, doesn't that mean it should also replace iPhone apps?

Yet Safari on the iPhone/iPod touch have over 30% of the world wide mobile browser market share and well over 50% mobile browser market share in the US.

Other mobile devices such as android will support HTML5 (via Webkit), which will result in HTML5 being available in a higher percentage of the actual mobile web market than Flash light.

That was the original plan - developers wanted native tools, which lets them leverage code written for their desktop apps and allows for apps that can't be written in either HTML5 or Flash.