Hacker News new | past | comments | ask | show | jobs | submit login
IE9 vs. Firefox 4 (done in canvas) (people.mozilla.com)
144 points by tomeast on Feb 16, 2011 | hide | past | favorite | 86 comments



I'm still not sure what the point of all this is. Any developer worth their salt already knows that IE9 is more or less a small step forward but still far short of what it needs to be and of what FF/Chrome have accomplished ages ago. Any non-technical user won't be reading this article and won't give a damn about IE vs. FF vs. Chrome, as long as their internets work.

Why keep up the pissing match? Who are they talking to?


The point is that Microsoft has been terrible at this since 1999.

They consistently do their own thing instead of going with standards, and they are always behind the curve. They seem to be trying to catch up with standards but have been unable to make any sort of impressive progress. To be honest, if they've put five years into it and are still so far behind, they need to fire some people.

Does anyone actually consider IE to be a valid contender for browsers? Personally, I gave up years ago.


> To be honest, if they've put five years into it and are still so far behind, they need to fire some people.

You are assuming Microsoft wants to adhere to standards. I find it more likely their strategy is to fragment the market because the day web applications become the norm, as opposed to desktop applications, they become OS-independent and Microsoft will have its main advantage (the Windows software ecosystem) negated.

They won't allow that.


I'm not sure MS are that dumb §. They know apps are moving to the web, and they want to be sure that Google and Apple don't control that experience.

They know IE9 is a bit crap, and they'll release an IE10 with proper history API, gradients, transforms and the rest when they can. They'll have a working web office someday too.

In the meantime they'll pimp IE9, the Office 'web apps' and other substandard ones to their customers until their kickass version 3 products come out.

§ Withstanding there are probably a few people in MS who really still love fat client.


> the day web applications become the norm,

That will never happen. Web apps have critical issues that can't be solved even by HTML8.


Care to explain those issues?


The issues are quite noticeable for someone that isn't heavily invested in web apps. Most of HN is - and that's why they prefer to ignore them.

But let's look at the market... Have web apps displaced desktop apps? No. Have they displaced mobile apps (as per HN mantra 90% of iPhone apps could be web apps)? Not even close.

Now why is this? It's because web apps offer a significantly poorer user experience across the board.

* with all the progress that's been made in performance, web apps are slower than native apps and will continue to be. There is only so much that you can offload from JS.

* web-apps don't have access to the platform's entire range of graphics capabilities, including standard widgets, animations and so on.

* web apps don't have access to special platform functionality such as GPS, sensors, camera, etc.

Furthermore, the biggest mobile players have all started to invest in native mobile apps to the detriment of web apps. Including Google. It's in their interest to lock-in users to their platform, even more so since Android became a major player.

These issues can be worked around eventually (maybe).

The core issues with web apps can't be worked around:

1. Web apps are using a text-transfer protocol with a pile of hacks on top to make it stateful. To this day engineers are still working on solving this problem (web sockets, web sql, etc). They can't solve it.

2. One does not own web apps, one merely rents them.


> Now why is this? It's because web apps offer a significantly poorer user experience across the board.

That is true today, but to say the issues can't never be solved is a bit of stretch:

- Speed issue: NaCl.

- Standard widgets: HTML5 already have input type=datetime, color picker, address book, and many more.

- Animation: WebGL.

- GPS: Geolocation API is available today on most browsers.

- Sensors: Accelerometer API is available today on some browsers.

Now I'm not saying that web app will certainly become the norm, but it's not impossible.


I think you're generally right that native apps are better than web apps. I have a few points to pick at, but don't take that as disagreement.

* web apps don't have access to special platform functionality such as GPS, sensors, camera, etc.*

Take a look at PhoneGap. This is definitely becoming less of an issue, and I wouldn't be surprised to see even more progress on it.

Including Google. It's in their interest to lock-in users to their platform, even more so since Android became a major player.

I can't speak for others, but Google's native apps are merely frontends to their web service. I regularly switch between the native and web versions of both GMail and Google Reader on my phone. There's slightly different feature sets (syncing/refreshing tends to be no issue on the web, but it's kind of weird on the native apps. The UI is better on the native apps. I can open links in background tabs for the web, but I have to switch apps for native, etc), but I could completely stop using the native apps with 0 issue.

Other than that, I strongly agree. I've never had to worry about my internet being slow or down to use a native app, and I know I can use them years later with no issue.


Irony: using a web app to suggest that web apps haven't displaced any native apps.


The fact that three people consider HN to be an application is nothing short of mind-boggling. I guess if this is the case, I've been doing webapps since before HN was even invented. We used to call them web sites back then though...


> 1. Web apps are using a text-transfer protocol with a pile of hacks on top to make it stateful. To this day engineers are still working on solving this problem (web sockets, web sql, etc). They can't solve it.

I am curious - why would the protocol need to be stateful? What are you trying to accomplish?

> 2. One does not own web apps, one merely rents them.

No. You can buy apps and host them on your own servers, on your own PCs.


2. One does not own web apps, one merely rents them.

One does not own native apps, one merely licenses them.


It's in their best interest, too, like for WP7 or Windows 8 for ARM. Otherwise, both WP7 and Windows ARM for tablets will be less appealing for consumers because they won't have enough apps.


Unless, of course, the whole Windows on ARM thing is just vaporware to show customers "look - you don't have to switch away from Windows when you switch to ARM" hopping this delays customers looking into alternatives in the x86 space.


Exactly... you, and every other serious developer. That IE is garbage is obvious to anybody who's done web development for more than a month. So why is Mozilla even bothering with these technical articles? IE lost to developers years ago, and IE9 won't change that.

I guess I'm just confused as to why this article, and the other similar ones from this past week, are even news.


> So why is Mozilla even bothering with these technical articles?

Because they need to educate the clueless who decide to make their internal web applications IE-dependent. Making an internal web app IE-only is one of the most common and crippling mistakes corporate IT makes.


Not news, but pressure and the fine art of telling the truth. It's to keep MS in check, transparent. In my opinion.


As a FF user, I wish they would adopt the same security improvements that IE has, instead of writing lame posts about canvases and shadows. i.e: automatically running under a restricted account and tab-as-process.


The point is that people listening to microsoft's PR will think that IE 9 is as modern as the next browser and that is clearly not true. As a web developer I want to use the new technologies and as long as there is a major share of IE users I just can't.


I listened to Microsoft's PR (which seemed to have some technical basis) and spent time asking MS about HTML5 history API out of genuine need, then wondering why there was no info, even when I asked MS directly. Ditto transforms.


Even if IE9 were in every way superior to Chrome 11 this would still be the case, and it wouldn't really be any the fault of MSFT.


Internet Explorer 9 Release Candidate was released and FF wanted to make a point that regardless of what M$ says, IE still sucks.


Any non-technical user won't be reading this article and won't give a damn about IE vs. FF vs. Chrome, as long as their internets work.

Not giving a damn isn't the same as not being affected by. IE slows down innovation and productivity on the web. Everyone's internets are worse for it, and it's a ball-and-chain everyone's paying to drag behind us.. whether or not everyone's aware of it.


Not all technically-inclined people or even developers do any web-related work at all. I've learned from this.


This is why Developers hate IE, we have to make code specifically for that browser. I guess if you truly hate Microsoft, you wouldn't code it and just tell the user to download Firefox.


> This is why Developers hate IE, we have to make code specifically for that browser.

Stop using -webkit, -moz, -o first.


Red herring. Vendor prefixes are part of the standards process.

Once the standard is finalized, you simply drop -moz and you're done. This is nowhere near the same league as IE CSS/JS fixes.


> What's missing in IE9? ... text-shadow ... CSS3 Gradients ... border-image ...

really? MS wont even support frikken text shadows? ffs ms. don't ship the fucking thing then, just take your time and get it right, please don't break everything again, my life is to short.

what did i do to you to make you hate me so :(


I feel 10 years younger reading this.


The craziest thing about it is they removed the old `filter` hacks for (bad, but hey) text shadows without replacing them with their modern CSS3 equivalent. (I should clarify: I mean in IE9’s IE9 mode, naturally.)


filter was renamed -ms-filter and the syntax was changed in IE8.


I know. But neither works in IE9 standards mode.


> don't ship the fucking thing then

I thought IE9 was still a work in progress?


The IE9 Release Candidate has shipped. They are done adding features.


Ok if we are going to discus this all over again then it helps to first read the older thread at http://news.ycombinator.com/item?id=2221831


The funny thing is, while as a web developer I hate with all my guts fighting IE standards(?), I also hate how Firefox (yes even version 4) is a memory hog. It's unbearable to work with it consuming 700mb to 1Gb of memory, with just a few tabs and firebug on.

Chrome(webkit) on the other side, wins both of these two combined any time of day.

Mozilla, you have come a long away, you created history when you came out with Firefox, you destroyed IE user base back in the days, and in Europe you are now the number one top browser. However, please stop comparing yourself to other browsers and work on your own issues.

Point being is, I love FF (although I'm not using it lately), but I'm hating how cocky they are becoming.


Do you have steps to reproduce the huge RAM consumption by any chance (which sites do you have open, which OS)? I'm very interested in solving these kinds of performance issues in the product.

And for the record, I wish Paul Rouget hadn't posted this. IE9 is a great browser - they do some things better than us, even (text-overflow, SVG acceleration via D2D, out-of-process tabs), and we should recognize this.


I understand that people like netbook/nettops etc, but ram is really cheap. At 10$ / Gig range there is little reason to have less than 8GB and 16GB while probably overkill is still fairly inexpensive.

http://www.newegg.com/Product/Product.aspx?Item=N82E16820231...


"Algorithms are for people who don't know how to buy RAM." - no idea who wrote this


"The secret to happiness is running software on a better computer than the dev who created it." -me


that just shows how bad developers are out there


My experience is the reverse. I recently stopped using Chrome and switched back to Firefox 3.6 because Chrome was using over 3 gigabytes of memory with about 40 tabs open. (40 tabs sounds like a lot until you realize that no single workspace of my 10 active ones had more than 10 tabs).

With Firefox I've had over 100 tabs open simultaneously.

Love/hate with Firebug, though. I can't work without it, but it does eat a lot of memory and slow things down.


I just created a user to say to the site's author:

Your site isn't HTML5 compliant, bro.

JUST SAYIN'.

http://validator.w3.org/check?uri=http%3A%2F%2Fpeople.mozill...


Kind of reminds me of when Scott McNealy was so obsessed with Microsoft that he didn't seem to notice Sun's real competition sneaking up behind him.


it reminds me of the last 10 years of my life wasting countless days/weeks cleaning up after msie 6/7/8.


Well, at least IE9 is somewhat up to date. HTML5 is still a working draft that no-one implements fully. Everyone implements a certain subset of it and I don't quite see how one subset is supposed to be superior to another subset as long as their is no one who implements it fully. Gee, at least Microsoft is making an effort!

And by the way, an uncluttered, fast-starting browser that implements a somewhat smaller subset of HTML5 than a messy, slow-as-molasses browser might have something going for it, too. Just sayin'


There is a common subset that most browsers have supported for a long time. I think the benefit of supporting that subset is pretty obvious, and I don't see how you can make the argument that a subset of a subset is better than the larger subset.


What's the deal with all of this high school bitchfesting going on lately between Google, Microsoft and Mozilla? My opinion of each of them drops every time they involve themselves in this crap.

How about you all stfu, work on your respective software and let your products do the talking?


Seems to render well in Chrome. ;)


We really need to stop measuring "compatibility" numerically. Its dishonest and encourages engineering decisions to be made instead of policy ones, when the data is purely the result of someone else's policy decision.

Not all things have equal weight; choosing which things to call an element, which ones to merge, which ones to split is always subjective.

Valid discussions are where we fully state our chosen assumptions, not ones where we then make fake data out of them.


Does the page render correctly in IE9? :P


No, all the graphs are missing and the specified fonts aren't loaded.


Because IE is still missing something as fundamental as data uris, and the font is specified that way in the CSS.


The horrible flash-of-unstyled-content suggests that this is not a good use of canvas. :P


The flash is because the styles are defined at the bottom of the document (not really sure why...)

Canvas is only used for the charts.


I placed far too much trust in the title of the story, alas.


Ugh. What's with the caps lock? This is incredibly hard to read (for me, at least). It just looks offensive.

This is going to be far less effective than a nice, polite feature comparison would be.


Why would I care what platforms are supported if I only use one? Also you could probably the numbers reversed if you asked Microsoft.


seeing all this energy wasted trying to convince people firefox 4 is better than ie 9 sort of solidifies my belief that chrome reigns supreme


Not saying I agree or disagree, merely: "consider the following"

Chrome: 10.0.648.45 dev

  HTML5 score: 293  FF4: 255
Compatibility differences from [1]:

( "-" == in other category )

         Chrome 10          FF 4
  Partial:
     CSS Text-Overflow
     Datalist Element        - 
     WAI-ARIA                - 
     SVG for HTML            -
     SVG filters             -
       -                    h264 video
       -                    Web notifications
       -                    SVG fonts
       -                    Web SQL
       -                    WebSocket
       -                    HTML5 Form
  Complete:
      Server-sent DOM events
      CSS Text-Stroke
      Progress & Meter
      h264 video             -
      CSS masks
      Web notifications      -
      SVG fonts              -
      Web SQL                -
      CSS Animation
      WebSocket              -
      HTML5 Form             -
                            Form Validation
       -                    Datalist Element
                            calc() in CSS
                            Animated PNG
       -                    WAI-ARIA
       -                    SVG for HTML
       -                    SVG filters
  Summary:
        87%                 81%
And WTF are they doing with that table[2] (disable "detailed tables")? It doesn't render correctly on anything I've got (OSX).

[1]: http://caniuse.com/#agents=firefox,chrome&eras=near,unof... [2]: http://caniuse.com/#agents=firefox,chrome&eras=near,unof...


Given how instructive the source code is, I think it serves as a nice demo. Even moreso when you consider that this is (by all appearances, anyway) a personal project and not an orchestrated marketing campaign.


IE 9 is a valid contender, especially outside of the tech crowd. It's also about to be released, so it makes sense for marketing to focus on it.


True, but it's not the contender they should be focusing on.

It's like Apple comparing iOS to Symbian and lauding how great iOS is, when really they should be watching Android.


I work for Mozilla, and we are watching Chrome. It's just not the right time to campaign against Chrome. Chrome is only eating away from IE's market share, and IE 9 could attract a lot of non-techie users.

Adopting a process-per-tab model is currently being implemented (https://wiki.mozilla.org/Electrolysis), and I imagine once that's done we will be discussing Chrome a lot more.


"Chrome is only eating away from IE's market share" Really? I'd love to see some sources on that, because I know of dozens of people who've switched from Firefox to Chrome.

Obviously I wasn't suggesting you weren't watching Chrome at all, but it's good to hear you're making the right decisions in competing with Chrome (primarily the reduction of Firefox's inherent clunkiness and slowness).


Firefox has stayed around the 30% mark of browser share, and there's an obvious connection between the steady IE drop and Chrome gain.

http://gs.statcounter.com/

There's been several articles about this:

http://www.google.com/#hl=en&sugexp=egsisas&xhr=t...


I switched from Mozilla to Chrome... Mozilla is just waaay too slow. The only thing I missed were the extensions, but Chrome is getting better on those as well.

Based on my experience, Mozilla has a lot of reasons to be afraid of Chrome.


Have you tried Firefox 4 Beta?

http://www.mozilla.com/en-US/firefox/beta/

We are nearing the end of our Firefox 3 cycle (Firefox 4 is going to be released next month), and that's where our significant speed improvements have been implemented in the last 2 years. It's about even with Chrome speed-wise.


I've been using it for the last week or so and its a huge step forward, if you are an IE8 user, 9 is huge step forward...

Its actually good enough to never have to worry about moving to a different browser (If its what you are using).


Not sure why their marketing affects the actual merits of Firefox or Chrome, but. I agree that Firefox is trying a bit too hard (though this appears to simply be one user). Additionally, Chrome is still faster and has less ui-chrome that gets in my way. I also find the extensions to be more cleanly implemented still than Firefox 4's.


The one thing I dislike about Chrome is that I still find it buggy at times on my mac. Buggy in strange ways and especially with the inspector/debugger. That being said, I still use it as my main browser. Going back to FF annoys me when there are two input boxes (address bar and search) and that the tabs don't close under each other (for really convenient tab closing). Other than that I'm sort of indifferent.


I'd love to hear more about those bugs you're getting. Feel free to bug (hah!) me on twitter @paul_irish or report straight to http://new.crbug.com


I get a heck of a lot more bugs (almost 2x) with Firebug on FF4 than with Chrome's inspector, though that still means I have to restart both almost a dozen times (no, really, that many) per day while building a web page.

A recent one: an input field would be spontaneously filled with a "1" if (and only if) Firebug was active. Persisted across application re-starts, but not through a reboot.


I have to agree.

I was an avid FF user for many years. I shied away from Chrome for quite some time, until this December. A fresh FF install with no addons was painfully unresponsive, given 'too many tabs'. Those same tabs in Chrome chugged along.

I hope FF can get back on track(relative to Chrome:P) and reduce some of the bloat.


these are all my thoughts as well. i didn't mean to suggest their marketing affects/suggests anything really. i just can't see who they're targeting with the posts. was there anyone who needed convincing that ie has less features? how does this benefit them?

in my completely unqualified opinion, i'd have guessed they'd be better off persuading the growing wave of users who are switching from firefox to chrome.


all this tells me is that firefox is scared of IE. which I guess is a good thing.


They have nothing to worry about. Take this from someone who is going bald at a faster rate because I needed to assist in porting an chrome-only app over to support IE9. FF, Safari, Opera were no problem, money was all spent on IE


I think it's exactly the opposite. The FF team has come out and set the agenda: that IE is not compliant. Microsoft came back with a ham-fisted "nuh-uh, we are not!", which is playing directly into the narrative Mozilla wants.

Terrible PR management on behalf of MS, here.


I really don't get why Mozilla posted this, other then to say to the 'tech' crowd "we're better then IE" but they already know that.

If, Mozilla really want to impress the 'tech' crowd they should show how they compare to Chrome because, mainstream users ONLY care about being able to view the websites they want to visit which is why most of them stick to Internet Explorer which came installed on their PC's (despite new legislation saying Microsoft had to suggest other browsers too) as they 'trust' that it will display the websites they want to see.


Your mistake is mistaking the personal blog of someone on Mozilla's marketing team for an official Mozilla position, I think...

If you want to know why _Paul_ posted this, you should probably ask him.


Microsoft is pitching IE 9 as, basically, "Hey guys, we finally caught up! We now offer an awesome HTML5alicious browsing experience, so y'all can stop switching to Firefox and come back over here!" This is Mozilla's return salvo. It's about keeping their message out there.


The way it is appearing is that Mozilla are scared of IE especially when you consider where the product has evolved from IE6.

Additionally, it needs to be said that whilst IE9 may be lacking some HTML5 and CSS3 features that Firefox 4 supports that HTML5 standards won't be completed until 2014, and Mircosoft will have its IE product fully supporting HTML5 by then.

I'm a firefox user but I still believe that Mozilla had no reason to do this, other than that they are realizing that IE has become a decent competitor again and they can't just pump out a slightly better browser to convince the crowd to convert. They've made themselves look a little stupid especially with regards to standards which won't be completed until 2014, and you can probably expect Google Chrome team to release a Chrome Vs. FF4 comparing its standards etc.


I think Mozilla is worried that people will accept IE 9. If Microsoft's pitch takes hold in people's minds, the push for Firefox will lose steam.

And all of this "HTML5 standards won't be completed until 2014" talk is kind of a semantic game. That fallacious concept is why Ian Hickson is trying to move away from the unrealistic notion of a "totally finished standard." In practice, most of the stuff Firefox implements is already nailed down. IE is unquestionably lagging.


Anyone else find this infographic hard to read?

Chrome 10.


Not sure what you are trying to say. I’m in Chrome 9 and it looks pretty great.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: