Slashdot Mirror
Firefox Is the First Browser To Pass the MathML Acid2 Test
An anonymous reader writes "Frédéric Wang, an engineer at the MathJax project, reports that the latest nightly build of Firefox now passes the MathML Acid2 test. Screenshots in his post show a comparison with the latest nightly Chrome Canary, and it's not pretty. He writes 'Google developers forked Webkit and decided to remove from Blink all the code (including MathML) on which they don't plan to work in the short term.'"
You don't need math.
I simply cant believe this...
MathML is a pretty important to allowing papers to be...
It's the worst HTML tag EVER.
I agree that it's surprising that Chrome seems to fail even on simplistic things. However, Chrome doesn't really feel like the kind of browser that goes for that sort of thing either. Given a choice between MathML and rounded corners (just as an example), I can well imagine that the latter would be far more popular, find wide-spread adoption, and be able to differentiate Chrome from other browsers.
The fairly limited set of publishers/users that would find MathML something that they'd have an absolute need form, seem to be using things that drop in an image of the MathML instead; stumbled across some sites in the past, equation in an image that replaced a piece of text that described the equation. No idea what that site used, but here's an example:
http://dlippman.imathas.com/asciimathtex/AMT.html
I can well imagine that supporting MathML does not exactly have a very high priority. Desktop sharing in Google Hangouts, albeit via a plugin at this time, on the other hand..
Seriously, where is anyone using MathML? All the major math sites (AMS, Math Overflow, ...) seem to be using MathJAX instead--and they work very well under Chrome.
Congratulations Mozilla! Still striving for standards means Firefox's job of keeping the others in check is just as important as ever.
Ask a web developer what they think about Chrome?
It is not all positive. It is buggy and has proprietary extensions similiar to something that sounded familiar in the past? Its javascript sometimes does not load on sites and its version of HTML 5 is differnent from others. HTML5test.com tests things that W3C implements a little differently or not at all.
Remember IE 6 was lean mean and standards compliant compared to the god awefull netscape 10 years ago too. Hard to believe in a place like slashdot to admit but if you go read slashdot history on the most discussed stories of all time "What keeps you on Windows from 2002" IE 6 is mentioned!
The switch to a new rendering engine is going to cause issues soon and many corporate oriented SVs and site makers will not be pleased.
http://saveie6.com/
when this exact argument was used against SVG? Now every browser has it, and the pain of using a javascript library shim for the holdouts is gone.
Not one that has been sucked through a jet engine then run over by a bad car analogy.
becasue LaTeX is easier to write than XML.
Webkit was caught patching to specifically pass the Acid3 test.
The real reason to improve MathML support is eBook readers that share the underlying rendering engines. Lots of textbook publishers want to use MathML, but without robust, reliable, visually appealing support, everybody has to do awful hacks with inline images or (hopefully) SVG instead.
Right now their only choice of rendering engines to truly show math symbols is Gecko. BTW, this isn't a new standard. Proprietary engines like Opera's Presto and Microsoft's Trident have had just as many years to implement this, and the eBook industry would've been just as happy to pay a license fee to use them.
There are critics of C++ that say the language is just pieces and parts hacked together. Even if that is true, mathematics takes the undisputed crown of bizarre hacked together symbols.
The symbols used in mathematics are unintelligible, inconsistent, don't even use a standard language character set and cannot be represented in a programming language.
These mathematical symbols either need to be modernized to come to a standardization or die.
The real answer is that they will die. And they should.
Because if it was made easier to understand then the aliens visiting us now would figure out how far behind them our tech is, and conquer us in 7 minutes.
That is rather impressive considering IE doesn't support MathML, and requires a plug in to get it working on IE 7 and 8. The plug-in has some trouble in IE 9 and 10 though, although they might have a beta working now. MS has no plans of adding MathML support and recommend using a MS program to export the equation as an image or using another program to export it as an SVG.
I love this browser is, let me to experience better results, such as google.com, kooyi.com, facebook.com
IE5 was lean and mean, and about as ideal as you could get for those days (security notwithstanding). The bloat began with IE6, and by then, Netscape was the better browser, feature- and resource-wise.
"If a nation expects to be ignorant and free in a state of civilization, it expects what never was and never will be."
Google's track record is that they're excellent at developing exciting new core tech, but mostly rubbish at making it rock solid or going beyond the basics. But they do have their hands in a LOT of projects, so that checks out. Shame Chrome is one of the better ones, but is still so flimsy, though. Sure, it gets massive points on HTML5 tests, but go a bit deeper under the hood and it's quite superficial support in the end. Mozilla's mostly caught up where it counts the most now, so it's going to be interesting to see what happens over the next few years. Slow and steady may just "win the race" (though I don't consider there to be much to win here)
When Chrome devs sneeze and accidentally create some flimsy new voice API that does things remotely anyway, the web gives them a shoulder rub.
When Chrome devs give up and fork their browser, that's a big thumbs up.
When Opera gives up the ghost and chooses Google's engine, it's "good job, Google!"
But when Mozilla listens to users and shrinks their memory usage to the point where Chrome can't even compete, "who cares"?
When Mozilla diligently catches up in Javascript performance, even overtaking Google with a clever stopgap solution for improving Javascript, "who cares"?
When Mozilla proves they've solidified a piece of web tech that many people already rely on, "who cares?"
Apparently, Mozilla could be the best browser on earth and people would still scoff at it for not being Google's.
> 'Google developers forked Webkit and decided to remove from Blink all the code (including MathML) on which they don't plan to work in the short term.'"
Haha! What a bunch of losers. Psssst! What's he talking about?
Which is why the one who broke this story is a MathJax developer.
Why bloat the browser with something that 0.00000000000001% of pages need. Pages that need math symbols can use MathJax (http://www.mathjax.org/) or similar. No need to build this into the browser.
I don't think Google has given up on MathML though, they just are not supporting it until it reaches as usable and stable level in WebKit. Then they will port over that entire chunk of code.
This is a common thing for people doing forks because you don't really want to spend time folding in partially working code from the other guys that brings little benefit. Just wait until it works and do it all in one hit.
const int one = 65536; (Silvermoon, Texture.cs)
SJW, n: "Someone I don't like, and by the way I'm a fuckwit" - AC
Remember IE 6 was lean mean and standards compliant compared to the god awefull netscape 10 years ago too.
That's IE 6 great claim to fame. Compared to the bloated corpse of Netscape 4, it looked quite good.
But IE6 came out in the days of Netscape 6
Did firefox use some hardcoded font hack to pass an older acid (3?) test?
You can star issue 152430 to get support reenabled.
if you want MathMl enabled in Chrome click the star in Issue 152430 to register interest
.
On the other hand, it would be really nice if the Firefox developers fixed their proxy issues, and fixed the javascript engine choking on sites.
The problem with the testerone-induced rapid development cycle is that it apparently leads to a lot of bravado (we're better than Chrome") and little ongoing maintenance of browsing issues.
I come here to Slashdot to watch bickering basement dwellers argue about who knows more about some arcane technology, not do math. // or as the Brits would say, "maths"
Netscape 6 was based on an early beta version of Mozilla Suite, somewhere between the last milestone release (M18) and the first real beta (0.7). In fact, IIRC the Mozilla team retconned v0.6 to match what AOL pulled for Netscape 6.0.
The first production-ready version of Mozilla-based Netscape was 7, I think, which was based on Mozilla 1.x.
I don't think many people used Netscape 6.
Hail Eris, full of mischief...
E pluribus sanguinem
Oh boo hoo. They spent a few minutes or maybe (gasp!) hours perfecting their MathML support, so I'm going to point at that and cry.
Forget that the bulk of their effort is on other things, like for instance, spending the last two years modernizing their Javascript engine so that it now is about as performant as V8's, is still more standards compliant, and so forth.
What's important is that they took a tiny bit of time to fix MathML! That's not allowed!
2 years ago, Opera was one of the closest to getting the face right.
With the last version of Opera to use its own rendering engine, it's an abysmal failure.
Nice regression, guys. No wonder your boss said to throw your work away.
But how many of your average users are gonna be going to pages where MathML would be useful? if it was 3% I'd be amazed.
Again with the "if it's not useful to the 51% it's not useful to anyone" meme. For one thing, the Web was invented for use by academia. For another, pretty much everyone who goes to college or even high school ends up seeing an equation at some time.
I used a blinking tag in a non-annoying way: to emulate a DOS like cursor for my temporary landing page.
To simulate a blinking insertion point, you could have used a CSS animation.
What happens if you want to read some HTML containing formulas while offline?
Then make an HTML document with a MathML data island, and have a <script> element in the HTML document reference an offline copy of the JavaScript program that translates MathML to HTML+CSS.
Or get a MiFi.
The page @ https://eyeasme.com/Joe/MathML/older_MathML_browser_test.html
Displays correctly in FF 3.6.
FWIW, Chrome's result is identical to my default, pre-Chrome, Android 2.6.3 browser. This would make sense if Google had removed the code from Chrome rather than a half-assed version; this must be the default infinit incompetence look.
I take that back. Infinite incompetence would crash. Possibly also infecting the Internet.
(-1: Post disagrees with my already-settled worldview) is not a valid mod option.
The needs of the many outweigh the needs of the few
Not in all cases, as shown in historical examples of tyranny of the majority. It lets those with wealth get away with assaults on freedom because the majority don't feel like keeping themselves informed. Case in point: In the United States in the mid-twentieth century, the majority of European descent wanted to force "colored" people, mostly of African descent, to use a different drinking fountain. Was that an acceptable compromise to support "the needs of the many"?
there is NO point in adding yet more bloat and complexity to a browser, not to mention giving malware writers one more attack vector in said browser, all to support a teeny tiny niche that would probably just as well if not better supported by using a browser plug in to a real language like Java.
Flash and Java themselves have a track record of being "one more attack vector in said browser".
Being built-in to the browser gives the browser vendor better control over fixing any pwnage in a timely manner than having to rely on however many plugin vendors to do the same.
Same tired "bloat" arguments, valid against non-standard things. Don't waste time trying it on standards that belong there.
Webkit was caught patching to specifically pass the Acid3 test.
Firefox had better MathML support even back then for that old webpage. The MathML Acid1 test needs less implementation than MathML Acid2 test, which needs less then the MathML Acid3 test, which Firefox currently only scores 60/100 on.
Sorry to burst your conspiracy bubble.
Hey, I like to typeset music, What if I want my browser to parse musicXML? I need a plugin, deah!
It shows how spectacularly bad Chrome is at standards.