Microsoft Previews IE9 — HTML5, SVG, Fast JS 473
suraj.sun sends this excerpt from CNET on Microsoft's preview of IE9 in Las Vegas just now. "At its Mix 10 conference Tuesday, Microsoft gave programmers, Web developers, and the world at large a taste of things to come with its Web browser. Specifically, Microsoft released what it's calling the Internet Explorer 9 Platform Preview, a prototype designed to show off the company's effort to improve how the browser deals with the Web as it exists today and, as important, to add support for new Web technologies that are coming right now. Coming in the new version is support for new Web standards including plug-in-free video; better performance with graphics, text, and JavaSript by taking advantage of modern computing hardware. One big change in the JavaScript engine Hachamovitch is proud of is its multicore support. As soon as a Web page is loaded, Chakra assigns a processing core to the task of compiling JavaScript in the background into fast code written in the native language of the computer's processor." Microsoft didn't say what codec they were using for the HTML5 video demo, but the Technologizer says it's H.264.
Uphill Battle (Score:5, Insightful)
That's great and all, but Microsoft isn't competing with other browsers for market share, it's competing with its own older browsers. Anyone who knows anything about browsers is already using Firefox or Chrome or Opera, and anyone who knows nothing about browsers is using whatever came pre-installed on their computers:
IE6 if they're still on XP, Safari if they have a Mac, or IE 8 if they're running Windows 7.
Unless this is a mandatory upgrade to IE 8, it's not going to gain any ground.
And of course, the 30% of users still using IE6 will continue to do so until their computers die, or a techie relative replace it with Firefox.
Re:firefox is getting old (Score:5, Insightful)
It seems that even IE beat Firefox in Javascript performance now. Firefox sure has been slacking recently.
The chart you linked shows IE 9 and FF 3.7 more or less at a dead heat. So, even if this were an unfortunate turn of events, it's not as if IE 9 had a terrible lead.
But I'm not sure it's unfortunate. High performance javascript in what will likely be the world's most highly used browser for a while? Sounds pretty good to me.
Re:firefox is getting old (Score:5, Insightful)
All that skepticism aside tho, if this is the truth (that IE9 will be standards based --and push the performance envelope--) then MS may be on the road to redeeming themselves... But the question remains, how tight will it be to the OS? Would a simple security flaw give a bit of JS access to the kernel? Or are they going to significantly sandbox the JS, and try to do everything right (as opposed to just the rendering)... Only time will tell if IE will become a browser friendly to geeks and developers (although something tells me it won't)...
So... what is the catch? (Score:2, Insightful)
The demo looks good so far, but I know my MS, there is an angle. There always is. Some subtle way in which they screw it up. Royally. There must be. They have done it for over two decades. No ways after 8 major versions and several minor ones are they suddenly going to play nice.
Paranoid? It ain't paranoia if they are out to get you.
They seem to be really honest this time about following standards, admitting they are not there yet and that it is time they did... so where is the closed source proprietary crap you just know MS is going to insist on adding.
A while ago someone asked on a forum, what would it take to use a linux library for accessing MS services. And I said there was nothing they could do. No, opensourcing it wouldn't do it, because I know MS has in the past done that and then later added closed source extensions you couldn't get on anything but windows.
And before you mod me down, if Blair/Bush (in holland this doesn't apply, Bakellende is after 4 failed goverments still available for re-election) said they were sorry, they knew what they did wrong, know what to do know to fix everything and all they ask is for another chance, would you give it?
Lets face it, they knew since version 6 that they had created a beast that to this day and for years to come haunts them. And version 7 was a beast and version 8 was a beast. So, third time is a charm? This one won't be a beast? I remember when Windows 7 came out: "Oh wow, this is so good, it ain't as crappy as Vista, MS has finally got it." And now slowly the negative is getting out and SP1 is being launched in a rush to deal with all the issues that were overlooked before. IE7 and IE8 were hauled as great improvements on IE6, only for devs to then realize that they were still spending most of their time on getting sites to work with the crappy products of Microsoft.
Will IE9 be different? Will it finally have real dev tools? Will it finally respect standards? Will it finally not introduce a thousand new proprietary and conflicting features? Will it finally perform? Will it finally not have a security hole per day that goes unpatched for years? Will it finally render a page coded to standard correctly?
And will MS finally do something serious about forcing the upgrade of everyone to IE9? Like MS disabling access to all its own extra services to anyone with an obsolete browser and releasing IE9 for EVERY single windows version from 98 on so that everyone can finally switch?
I doubt it. Que the MS apologists who will claim this is finally it. They should be ready, they said it often enough before.
Holy shit (Score:3, Insightful)
I had to stare at the headline for like 5 seconds before it even parsed. It just didn't seem like a reasonable configuration of words.
Re:So... what is the catch? (Score:3, Insightful)
I know the catch. The catch is obvious. I know people who use IE 6.
IE 7 came out 4 years ago. IE 8 came out a year ago, not including the long public beta.
No matter how good IE 9 is, we'll all have to continue to support IE 6/7/8 for the next 6+ years. It doesn't matter if IE 9 was FireFox with a skin, the curse of IE will continue to haunt anyone doing web development for years.
Re:Nice try with ACID3, Microsoft (Score:1, Insightful)
I thought the fact that it got a 55/100 on the test meant that it failed acid.
Its not like they got 100/100 but just had that one stop in the middle and thus "ZOMG they're big liars! They cheated! It stopped in the middle! I saw it."
Re:Nice try with ACID3, Microsoft (Score:3, Insightful)
You've missed the overall point. This isn't even alpha quality software, it's in development. They aren't claiming they passed, they are just showing that they are making progress.
What you're doing is kinda like picking on a 2 year old for not having an expansive vocabulary.
Re:firefox is getting old (Score:2, Insightful)
IE has been pretty good with security with 7 and 8, IMHO. Coupled with DEP (which ships turned on in Windows 7), and the protected mode that the browser runs in (so if it does get hijacked, malicious software doesn't have access to the user's file or Registry, much less the system's) have given the browser a significant security boost.
This isn't to say that IE is perfect, but because it is the focal point of almost every single intel agency, botnet client maker, malware writer, and blackhat on the surface of the planet, it has shown to be able to withstand a lot of attacks.
My recommendation, and this applies to *all* web browsers: Use something like Privoxy. This will filter out one of the biggest sources of infection, and that is third party ad-servers serving up malicious code.
Re:Reopening tabs (Score:2, Insightful)
Which is better default behavior?
1. Open the browser as quickly as possible and let the user click the page they want from the history / most visited list (Safari, Chrome, Opera do this)
2. Open the browser and check all the plugins for updates, check to see if pages were open when the browser was last closed, stop loading, present a dialog asking the user if they want to load the browser (which is going to happen anyway regardless) or load the browser _and_ try to open N tabs simultaneously.
If you said 2 you are an imbecile.
Re:Nice try with ACID3, Microsoft (Score:3, Insightful)
I know they're not claiming they've passed. But you've assumed something pretty big: "hey are just showing that they are making progress". If they've only got to 55, and the process of reaching 55 does not fulfil the rest of the test (being smooth, namely) then it actually hasn't even got to 55. It may as well be at zero.
To be fair, you're also making a big assumption: that someone cares what you consider the score that an alpha browser achieves against a test it's not trying to pass is.
I mean, this is a site full of geek wankery, and I mean that in the most affectionate sense, but come on.
Re:Standard compliance? (Score:4, Insightful)
So with all of the nifty, new stuff they are finally compliant, right? I mean no more body {text-align: center;} instead of body { margin: 0px auto; } to center a fixed width layout, right?
Those are two different things. text-align: center centers stuff in a div. the margin: 0 auto you set to a div to center that block (the div) in its container. Even IE6 works correctly with this, so I don't know what the issue is here.
For those having box-model issues with IE6, you can easily fix this by using the HTML 4.01 Strict DTD, FYI.
Compare to the img element (Score:4, Insightful)
What good is a standard embedded video tag if there is no standard coded with which to play with it?
What good is a standard embedded image tag if there is no standard coded with which to play with it? Notice that HTML's definition of the <img> element [w3.org] doesn't require support for any specific image format.
Re:H.264 (Score:5, Insightful)
Firefox not playing h264 is a political decision (Score:1, Insightful)
Both Windows 7 and Mac OS X ship with h264 codecs preinstalled, and in the grand scheme of things, the (capped) $5M/year MPEG LA licensing fee would not really cripple Mozilla Corp (which gets $85M a year from Google for search box placement), so even if using built in h264 codecs is not an option for whatever reason, they could still ship ffmpeg.
Now let's assume they don't want to pay $5M. Even then there's an option which they deliberately declined to provide - have a plugin architecture in place which would allow third party codecs.
I'm not sure why they think Theora will win in the end, but at this point I'm fairly certain this isn't going to happen, no matter how hard Mozilla pushes Theora. With Chrome nibbling at Firefox's marketshare from one end, and IE9 offering h264 support on the other end, the lack of de facto compatible HTML5 video is a crippling disadvantage.
Re:Microsoft should stop (Score:4, Insightful)
As a web developer, I'm really glad that every version of IE has been more standards compliant than the last.
It would be nice if the everyone magically installed FireFox or Safari or Chrome, but that doesn't seem to have happened yet.
Our best hope for killing off older versions of IE is newer versions of IE and an automatic upgrade path.
Frankly 8 doesn't seem that bad to me. Most of my code just works with IE 8. I'm really excited about HTML 5 and SVG in IE 9.
Re:Reopening tabs (Score:4, Insightful)
What if there was an option to not check for updates and to not load the previously loaded tabs?
Re:H.264 (Score:3, Insightful)
However, anyone who uses an unlicensed h.264 decoder are still technically breaking the law (at least in the U.S.). I'd rather not have to break the law to watch a video online
I fully expect there to be licensed H.264 codecs for Linux being offered for a reasonable price, just like you can buy a licensed MP3 decoder or DVD player for Linux today.
Why so negative? (Score:1, Insightful)
They're freely admitting upfront, "hey, on this test, we're still doing badly, but we are working on improving. It's just not our focus."
Why should ACID3 be their focus? And even so, when did they mention it's not?
I think Microsoft working towards better ACID3 compliance is great news.
For a product the size of IE to make the changes needed to go from 22 to 55 in just a few months is incredible. This is regardless of who's working on it and I hope they get closer to 100% before IE9 is released.
I really don't get. HTML5 support, CSS3 and better Javascript performance and most of the posts on here are still complaining.
Personally, I'm just happy one more major browser is aggressively ( the standards haven't been ratified yet and are subject to change ) pursuing web standards.
Re:H.264 (Score:3, Insightful)
>neutral non-profit organization.
The MPEG WG and the MPEG IF may be non-profit (I don't see how they would qualify as neutral in any useful way), but that is irrelevant, as the MPEG LA is most definitely not nonprofit.
Re:Firefox not playing h264 is a political decisio (Score:5, Insightful)
It's not that they think Theora will win in the end. It's that they want some free standard to win in the end, and they know that won't happen if they (of all people) fold on H.264.
The money they'd have to pay for including it in their distribution isn't the issue. It's the fees people in future would have to pay for creating and distributing movies. They want the Web to be democratic, and that means everyone gets to contribute, whatever their financial means.
Re:Microsoft should stop (Score:1, Insightful)
Re:Agreed. (Score:3, Insightful)
If you don't want Active-X, disabled it.
Re:H.264 (Score:5, Insightful)
Or people who feel strongly about it will continue to use open formats and petition against software patents.
Good luck with that.
8 years ago, I've bought into Vorbis hype (coincidentally, it was about the time when I switched to Linux as primary desktop OS). My music collection was 100% Vorbis. I only bought players that could play it (e.g. iRiver).
Fast forward to today... only about 10% of my music is still in Vorbis, and I still have trouble with that (e.g. my car won't play it, so I have to recode). I'm afraid that MP3 has won, and AAC is picking up from there.
And that was with Vorbis, which was actually technically better than MP3 in many aspects (better compression, extensible meta-information with proper Unicode support etc). And Theora is technically inferior to H.264...
Still, good luck.
Re:plug-in-free video? (Score:3, Insightful)
MS just got slapped by a fine in excess of 5 billion dollars in EU for anti-competitive practices. If you were an MS executive, would you seriously be willing to do something that would piss off the very same people who came up with that fine, and who are likely to apply some punitive multipliers for repeated offense?
If it comes right up to the line the EU drew in the sand, dances on it, leans a little, but never really goes over... yes.
A Generation Behind (Score:2, Insightful)
This is no different than when IE8 was released and IE finally supported CSS 2.1 when all the other browser vendors had.
Webkit, specifically Safari, has been leading the way in CSS innovation & Javascript performance with each release with Chrome slightly behind. Firefox & Opera seem to be battling it out for third place and IE, of course is always an entire generation behind.
Re:H.264 (Score:3, Insightful)
Re:H.264 (Score:2, Insightful)
The point is not H264 vs Theora advantages. The point is the difference between a "standard" and a "proprietary extension".
If your definition of a web "standard" is a specification for a patented technology that acts as a barrier to entry to everyone except the select few handful then H264 is your choice.
However, if that's the case, then I do not agree with you. The reason for a WWW "standard" is that it should be free to all to make, distribute and render however they see fit regardless of the OS, platform, hardware, software, day of the week, or weather in Japan. That's the spirit of a true world-wide-web "standard".
That's also part of the reason why some are upset about HTML5 video over Flash. Adobe/Macromedia Flash plugin is a proprietary extension to the browser. There is no secret, or ifs or buts about it. Users are aware they don't have flash on their iPhones. If they don't have it on their computer, they go and download the plugin. With the distribution of H264 in their browsers, Apple, MS, Google and alike are contending (whether implicitly or explicitly) that they are implementing some kind of web "standard" in their HTML5 video. They are trying to blur the line between a true "standard" and a proprietary extension by confusing general public who doesn't know the details (and by general public I mean semi-savvy people who at least know what a web browser or Internet or a website is).
Maybe the next thing will be a proprietary patented Javascript or CSS additions that will be pushed out as an HTML6 "standard" that will only be reasonably available to the select few with deep pockets; that can then sprinkle the licensed goodness to all for "free" or low cost for certain uses, on certain OSes, on certain browsers. Sure that could cover most of the browser market, but is that right?
Screw all that! We should want a web "standard" to be available to everyone equally without any licenses or fees. We should want "standards" that spur technology innovation on the web, not a legal web of patents and licenses that are kept hostage by a handful of corporations. In fact, if those same companies are so keen on defining a true web video "standard" why not invest a pocket change out of $10s of billions cash they each have in their bank accounts and help improve performance of Theora, or release a specification that's truly free for all? The answer is obvious - it's because that's not their intention - their intent is to hold hostage the innovation with software patents.
So, sure go ahead and argue how H264 is the best thing since sliced bread. Maybe it is, maybe not, but that's missing the point. The point is it is either a true "standard" OR it's a patented proprietary extension that tries to deceive users and pass itself as the new HTML5 web video "standard". Well, I know which one it is and based on your post, so do you.
Nonsense and nonsense. (Score:3, Insightful)
Nonsense.
First of all, Microsoft does *not* have "nothing to worry about" with H.264. Just because it pays tribute (er, licenses patents) from one organization does *NOT* mean that it's protected from all other organizations. In fact, once you demonstrate that you're willing to pay to one organization, others will start to show up to get some money too. For an analogy, look at the history of the Vikings; once people started paying tribute, the odds of looting parties showing up INCREASED. And we don't have to just use analogies; look at the recent history of sound codecs, specifically MP3. Microsoft paid big$ tribute for MP3, but Alcatel-Lucent sued Microsoft and won a record-breaking $1.52 billion in damages via a jury verdict. Now it's true that Microsoft got lucky in that one; in the MP3 case, a judge reversed the jury, a highly unusual event. If a judge hadn't reversed it, Microsoft would have paid $1.52 billion in additional damages for something it had ALREADY PAID LICENSE FEES for. And even so, Microsoft spent a FORTUNE in court on MP3, a codec that it was already paying license fees for. So it appears that "licensed" codecs have a HIGHER risk, not a lower risk, historically speaking. Wikipedia has more about the MP3 patent stuff [wikipedia.org].
Second of all, there's already been a lot of money and research spent to make sure that Ogg Theora is free of patent issues. Few things in life are "conclusively proven"; let's use realistic measures. The evidence, in this case, is really strong that Ogg is safe. Strictly speaking, it's not that Ogg Theora is patent-free, it's that all known required patents have been released under and irrevocable free license [theora.org]. That is actually a stronger legal position than simply "not knowing of any patents"... here we have a granted patent, which is then released. The Ogg folks spent $ to do their own legal searches, too, something standards bodies emphatically do NOT do, giving you additional protection. Most companies that claim that "Ogg has unknown patent issues" are basically flinging FUD; it's mainly a protest claimed by companies who have a vested interest (a kickback) from the patent licenses. In particular, it's my understanding that Apple *makes* money from the H.264 patents. So unsurprisingly, Apple works to lock everyone else into the patents they partly control, and actively works to *prevent* the use of open standards for codecs. But you don't need to buy into that.
Sure, it's always possible that there are unknown submarine patents, but submarine patents are risk to all codecs, including H.264; that is not specifically a risk to Ogg Theora [bluishcoder.co.nz]. Indeed, H.264 is MORE dangerous. Because H.264 was developed in an environment where patents were permitted (for shame, ISO), and there was no *requirement* for an external patent search (ISO doesn't require it), there was an incentive to patent everything, both by the participating parties and by external parties. There have been a number of court cases about MP3, but none about Vorbis, which shows that once you let patents into a standards process, things can get really bad.
Someday, someone may find a patent problem with Ogg Theora, but this is highly unlikely. In contrast, we have hideous patent problems with H.264, today. Why worry about Ogg, when there's a wolf already in tent? We need to dump H.264 (with its KNOWN problems) and switch to Ogg (which has NO known problems). First step: Get the browsers to support Ogg Theora. Then websites can more rationally use the format. It's better for Microsoft's customers: They can then easily use an open standard. It's also better for Microsoft: If more people use an open standard, they won't be as beholden to the H.264 licensors and will reduce the risk of me-too lawsuits like that of Alcatel-Lucent.
Re:Reopening tabs (Score:4, Insightful)
Let me rephrase then.
What is the better default behavior given that there is an option to alter said behavior if you don't like it?
The default behavior attempts to keep Firefox as up to date as possible. It also tries to recover from the browser crashing or some other misfortune. If I need to go in a hurry, I can close the browser and it will reopen where I left off.
Your argument seems to be that people are imbeciles if they don't have the same priorities you do. I don't subscribe to that point of view.
Re:H.264 (Score:3, Insightful)
Re:H.264 (Score:3, Insightful)
However, right now any implementation of H.264 in the core of firefox is not going to happen.
No-one is asking for that. In fact, you yourself go on to say...
Maybe it will be possible to have a pluggable video decoder for Firefox for the HTML5 Video tag so you can hook up your own solutions. That might solve the issue for everyone.
It would have solved the issue for everyone. The problem is that Mozilla explicitly refuses [mozillazine.org] to do that for ideological reasons! They don't want to give users freedom of choice, if that freedom may lead them to choosing "unfree" codecs.
(note also that most claimed technical problems with DirectShow in that blog post are pure FUD)
In fact, there already is a patch [mozilla.org] to enable GStreamer support for video codecs, but so far it's only been accepted for Fennec, not for mainline desktop Firefox.