Google Chrome

I just noticed that you can 'pluck' tabs on to the screen, where they become semi-transparent page previews (or completely new windows if you let go). Pretty cool.
 
I know, but SVG isn't very usable for drawing either. That's why canvas was developed. And the only browser that natively supports SVG is IE, the rest use canvas.

Umm, no. Safari, Firefox, Chrome, and Opera support both Canvas and SVG. IE supports neither Canvas NOR SVG. The only thing IE supports is a crappy proprietary extension called VML (Vector Markup Language) In fact, lots of mobile browsers support the Tiny SVG subset.


SVG is perfectly suitable for what it does. It's a scenegraph based API. If you are developing an app, where you don't update the position of every object every frame and redraw all the geometry, then SVG is great. It's natively accelerated, powerful, and supports animation through SMIL, as well as event handling/picking. This is awesome for doing "desktop" style drawing apps like Powerpoint or Illustrator.

If you are doing something more like a game, then canvas corresponds more closely to immediate mode APIs like OpenGL or DirectDraw/GDI.

There's a time and place for each depending on requirements.
 
Umm, no. Safari, Firefox, Chrome, and Opera support both Canvas and SVG. IE supports neither Canvas NOR SVG. The only thing IE supports is a crappy proprietary extension called VML (Vector Markup Language) In fact, lots of mobile browsers support the Tiny SVG subset.


SVG is perfectly suitable for what it does. It's a scenegraph based API. If you are developing an app, where you don't update the position of every object every frame and redraw all the geometry, then SVG is great. It's natively accelerated, powerful, and supports animation through SMIL, as well as event handling/picking. This is awesome for doing "desktop" style drawing apps like Powerpoint or Illustrator.

If you are doing something more like a game, then canvas corresponds more closely to immediate mode APIs like OpenGL or DirectDraw/GDI.

There's a time and place for each depending on requirements.
Thanks for the correction. I'll look into it.

But what do you think about the rest of my criticism?
 
And it isn't so hard, they can take one of the methods and keywords used now, and actually make them work (like prototypes), or (even better:) create a few new ones on top of it that simply inplement types and inheritance.

That's what they are doing, adding syntactic sugar, but it doesn't address many of the problems with Javascript development-in-the-large like the lack of optional types, a package/module system, or early bound namespaces. There is a reason why these "died in committee", which was after careful analysis, it was found that those features would "break the web" For better or worse, browsers have evolved a well known order of evaluation for how they load and process script, and the proposed JS2 package/namespace systems would fundamentally break under dynamic script injection scenarios.

Yes, I'm waiting eagerly for one of those to become mainstream, especially GG. Although installing a pluging is a big no-no for most large companies, and so only partially useable.

Well, installing a new *browser* is equally a no no. I hear this bullshit all the time "we can't let employees install flash, blah blah" and then they turn around and have employees running insecure IE activeX shit on the LAN. Most large companies have rather insane policies for upgrading people's desktops.

no usable way to actually draw things like your own controls (canvas helps with the drawing, but you have to move controls on top, and no realistic way to use objects with events), and the totally awful DOM model (endlessly duplicated amounts of layers and stuff, but still no good hierarchy that actually makes sense, no actually usable general purpose areas -not even DIVs- where you can DRAW your own controls on, or simply put predefined stuff on in a predictable way).

The CANVAS is a DIV which you can draw controls onto if you want, you just have to do hit-detection yourself. SVG fully allows you to do exactly what you're asking, go look up Sun's Lively Kernel. http://research.sun.com/projects/lively/index.xhtml, this is all done with SVG.

And using solely CSS for positioning is a great pain when using VS.NET, the demanded development platform for most stuff.

Well, I feel sorry for you. Most Web 2.0 developers I know of don't waste time with Microsoft tools. If you're building an app with ASP.NET you're simply a dinosaur. Javascript is a real langauge, and you can write completely non-trivial apps in it using standard browser mechanisms without much trouble, as long as you use a nice library framework.

Look at what people are doing, like 280slides and Cappuccino, or ExtJS, or Google Apps. Look at the CAKE javascript canvas scenegraph. I would argue against trying to build more and more higher level APIs, and simply build a very fast Javascript VM, and some low level access to events, drawing, and the DOM, and do everything through libraries.

Even local storage today can be abstracted. Dojo Storage for example will automatically use whatever your browser supports, from Gears, to HTML5, to Flash Cookies.

I don't mean to be insulting, but you really need to get away from Microsoft environments and look at what is being done today on the web by the top players.
 
That's what they are doing, adding syntactic sugar, but it doesn't address many of the problems with Javascript development-in-the-large like the lack of optional types, a package/module system, or early bound namespaces. There is a reason why these "died in committee", which was after careful analysis, it was found that those features would "break the web" For better or worse, browsers have evolved a well known order of evaluation for how they load and process script, and the proposed JS2 package/namespace systems would fundamentally break under dynamic script injection scenarios.
Well, in that case, don't extend JavaScript, but offer a better alternative. I propose ObjectPascal. :)

Well, installing a new *browser* is equally a no no. I hear this bullshit all the time "we can't let employees install flash, blah blah" and then they turn around and have employees running insecure IE activeX shit on the LAN. Most large companies have rather insane policies for upgrading people's desktops.
Totally. Most still run IE6, for that matter, and ActiveX is enabled by default, as that's sanctioned by Microsoft...

It doesn't help, that just about all network administraters around here are MCSE certified, although Linux is taking big strides. (Mostly because of virtualization.)

The CANVAS is a DIV which you can draw controls onto if you want, you just have to do hit-detection yourself. SVG fully allows you to do exactly what you're asking, go look up Sun's Lively Kernel. http://research.sun.com/projects/lively/index.xhtml, this is all done with SVG.
I'll check it out. It sounds like what I want.

Well, I feel sorry for you. Most Web 2.0 developers I know of don't waste time with Microsoft tools. If you're building an app with ASP.NET you're simply a dinosaur. Javascript is a real langauge, and you can write completely non-trivial apps in it using standard browser mechanisms without much trouble, as long as you use a nice library framework.

Look at what people are doing, like 280slides and Cappuccino, or ExtJS, or Google Apps. Look at the CAKE javascript canvas scenegraph. I would argue against trying to build more and more higher level APIs, and simply build a very fast Javascript VM, and some low level access to events, drawing, and the DOM, and do everything through libraries.

Even local storage today can be abstracted. Dojo Storage for example will automatically use whatever your browser supports, from Gears, to HTML5, to Flash Cookies.

I don't mean to be insulting, but you really need to get away from Microsoft environments and look at what is being done today on the web by the top players.
Tell me about it. I've been building my own framework like that multiple times, but one of my managers is totally sure that all our customers only want Microsoft products, and C# ASP.NET applications. That's why he always starts telling them that that's what we're going to use up front.

And when I build my own stuff, I don't have the required knowledge of .NET, as there are much better default .NET methods I should use instead.

He really doesn't like our Delphi developments as well, even for industrial W32 use (and you really don't want to use .NET for that, as I experienced the hard way). Happily, my other manager shares my view on the matter.

And I'm going too fast, I generally have to explain him nowadays that I passed the reason he is shooting it down some time ago.

But I love my job anyway, even while the pay is average.
 
Home PC SunSpider JavaScript Benchmarks

Intel Q6600 @ 3.42Ghz, 4GB DDR2, Vista 64bit, Verizon FIOS 50mb/20mb connection:

Google Chrome: 0.2.149.27: 1543.4ms +/- 0.8%

Firefox 3.0.1:
2784.6ms +/- 1.7%

Apple Safari 3.1.2:
3333.4ms +/- 0.1%

Opera 9.5.2:
3952.0ms +/- 0.7%

Internet Explorer 8 Beta 2:
7471.4ms +/- 0.2%
 
There was already an update this week-end. I have not seen any release notes, but at least it fixes the 'formatc:%' bug and the url mouse-over crash pointed previously.
 
Using Chrome as my main browser over the weekend, I've begun to find some of details I strongly dislike.

* GIF performance is terrible - going to a "GIF thread" pushes my CPU usage to ~70%, and the GIFs don't animate at full speed.
* quick searching at this stage feels unwieldly. In Firefox, you drop your search-engine down, pop in the term and go. In google you either click the new tab box, and see it on your front page (if that works) or type the shortcut into the address bar ("ebay computer table") and then scroll down the correct choice. Either way is more unwieldy than I'm used to
* the download window working "by tab" is a bit of a pain, although there's full "downloads" tab available, I haven't worked out the easy way to see the progress of all my downloads without tab-scouring just yet. Firefox's ctrl+j is easy, and I want any new browser to be just as easy.

Incognito is great though!
 
So, does Chrome update silently in the background? I no longer crash due to the bad urls, but I don't remember ever being asked about updating.
 
I updated manually from the About window (where it says 'update available, click here to install'.

One difference I notice is that Flash seems to work properly now. Eurogamer now runs as expected, and performance seems to be pretty good - my laptop stays silent, cpu use is never over 25%.
 
I updated manually from the About window (where it says 'update available, click here to install'.

One difference I notice is that Flash seems to work properly now. Eurogamer now runs as expected, and performance seems to be pretty good - my laptop stays silent, cpu use is never over 25%.

I just loaded Chrome on my PC at work, went to the about tab and it says chrome has been updated and to restart it. Was there an option when installing to get it to update automatically? I don't mind if it does, but it would be good to at least get notifications.
 
So far I think I'm going to keep using it. I like the UI better than Firefox's, it has so many nice little touches, and I like the fact it's basically just a shared title-bar and a fat easy to read address bar without any of the other guff (firefox's ui takes up more room than Chromes even before the tabs start to display). And it's, in my experience, more robust and faster feeling so far.

They just need to put smooth scrolling in and I'm completely happy for what I do.

* quick searching at this stage feels unwieldly. In Firefox, you drop your search-engine down, pop in the term and go. In google you either click the new tab box, and see it on your front page (if that works) or type the shortcut into the address bar ("ebay computer table") and then scroll down the correct choice. Either way is more unwieldy than I'm used to

You can also, once you've searched at a site, just hit tab to turn the address box into a search box for the selected site. So for me at the moment to search ebay I just type "e" then tab and then enter the search string.

Which I really like.

I haven't worked out how the browser works out which sites are searchable though. If I use rpgwatch's frontpage search, for example, it won't let me tab-search from the address bar afterwards.
 
I did not bother to read the 5 page thread because it just seems too long for a browser related discussion but anyone not bothered the least bit with the EULA and how all the content you view Google and store it? So if some person goes online and does some banking then Google can cache that information?
 
I did not bother to read the 5 page thread because it just seems too long for a browser related discussion but anyone not bothered the least bit with the EULA and how all the content you view Google and store it? So if some person goes online and does some banking then Google can cache that information?

They already changed it. They copied it from some other EULA, apparently.
 
Back
Top