Gtk 3.2 Will Let You Run Applications In a Browser 184
An anonymous reader submits this intriguing tidbit: "Gtk+ 3.2 will let you run any application in a browser thanks to the new HTML5 gdk backend. That means you'll be able to run GIMP, Gedit, a video editor or whatever, remotely (or on the same computer), using a web browser. Just imagine the possibilities!" At this point, says the article, it's only possible with Firefox 4.
Standards people! (Score:2)
Re:Standards people! (Score:5, Informative)
Re: (Score:2)
Re: (Score:2)
Re: (Score:3)
Up-front disclosure: I've been involved with HTML5 since 2006, and worked for Opera Software since 2009, and (among other things) am QA for HTML/DOM/etc. implementation.
2022 is far from insane: CSS 2.1 has been under development since 1998, and is yet to reach REC (though hopefully will this year) â" and CSS 2.1 is just a minor update to CSS 2.
Yes, for most practical purposes the spec will be done long before it reaches REC, but writing a testsuite that is considered complete enough is massively-time c
Re: (Score:2)
Re: (Score:2)
IE9's support for newer features is poor - that's a fact. In none of my previous posts have I even mentioned Micro
GIMME A SHELL! (Score:3)
Cloud-enabled root privilege escalation, here I come!
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
>>>Protip: WebGL is not part of the HTML5 standard nor is it a W3C or WHATWG standard
Brain hurt.
Want to go back to simple HTML:
http://replay.waybackmachine.org/19961114151757/http://scifi.com/ [waybackmachine.org]
(look how fast it loads)
Re: (Score:2)
Want to go back to simple HTML:
http://replay.waybackmachine.org/19961114151757/http://scifi.com/ [waybackmachine.org]
(look how fast it loads)
It loads fast because it's not doing much. I write new HTML almost every day that performs just as well because it's not overly-complex.
Re: (Score:2)
Re: (Score:2)
Maybe Mozilla, Apple, Google, and Opera implement different subsets of the "standard" (HTML5 is not what I'd call a standard) and functionality this needs is implemented on
Re: (Score:2)
The idea with HTML5 seems to be that once browsers implement all these fancy proposed features, the resulting subset will be hammered into an actual standard.
At which point everyone's already implemented those features in their way so wtf do we even need the standard for?
Re: (Score:2)
These days much of the real value from standards is the process which creates them. The finished standard is almost just a diploma to put on the wall.
If we can find a superior way to get competitors working together, perhaps standards can just go away.
Re: (Score:2)
These days much of the real value from standards is the process which creates them. The finished standard is almost just a diploma to put on the wall.
If we can find a superior way to get competitors working together, perhaps standards can just go away.
So the idea of 'standards compliance' is not relevant anymore then.
Re: (Score:2)
It also requires Websockets, which isn't available in IE (yet?) and is disabled by default in Opera and Firefox.
Re: (Score:2)
The approach the IE team has started taking for unstable standards is to release prototypes independently of the main browser. For example, the websockets prototype is here: http://html5labs.interoperabilitybridges.com/prototypes/available-for-download/websockets [interopera...ridges.com]. My understanding is that there have been breaking changes in the websockets protocol spec fairly recently (to resolve security issues), so shipping an implementation for widespread use isn't a good idea until the spec stabilizes.
Re: (Score:2)
No you don't. Its a pure HTML5 thing. This will run, eventually, on an ipad for instance.
Re: (Score:2)
Offtopic but damn that was a cool demo. Thanks for sharing.
Why exactly? (Score:5, Interesting)
Why exactly do I want do this?
And more importantly, can I run firefox 4 in firefox 4 in firefox 4 in firefox 4?
Re:Why exactly? (Score:5, Funny)
You have answered your first question with your seconds question.
Re: (Score:2)
A migration path for simple apps (in either direction, from web to desktop or from desktop to web)?
Re: (Score:2)
Or using traditional application development tools to build a web app?
I like the Web, but I have to admit, GUI toolkits tend to be quite a bit better. I don't know how well this would actually work, but it would be nice to develop a web frontend using tools like glade [gnome.org] or QtDesigner [nokia.com] rather than what I do now with Haml [haml-lang.com] and jQuery [jquery.com].
I'm very skeptical, though -- there are ways the Web is currently better than many desktop apps. Even ignoring issues like bandwidth and performance, would this give me an app which
Re: (Score:2)
I really don't. I want that even less than I want ADF [oracle.com] and jDeveloper [oracle.com].
Partly because I want the result to not suck. Can I use those tools to develop a frontend which is essentially static HTML? Not actually static, use some templates and sugar to make it easier to build, but is it actually reasonable to develop a frontend which doesn't have to talk to the server for every little thing?
But mostly, I want to develop and deploy on open platforms. I can't count the number of times it's been helpful to read the s
Re:Why exactly? (Score:5, Funny)
No programs should ever migrate to the web.
What if they migrate back to the desktop every spring?
Re: (Score:2)
No programs should ever migrate to the web.
What if they migrate back to the desktop every spring?
Depends if it's an African or European program.
Re: (Score:2)
No programs should ever migrate to the web.
What if they migrate back to the desktop every spring?
Depends if it's an African or European program.
They could bring us coconuts!
Re: (Score:2)
I would run the emacs operating system inside firefox 4, and then load firefox 4 again within emacs. Using the inner firefox 4 I would run vim. This is finally a way to give emacs a decent text editor!
Re: (Score:2)
Why exactly do I want do this?
Simple really. This is for those of us who run applications remotely for one reason or another. I personally like being able to run GEdit remotely since my server doubles as my homework computer. If this would let me do the same thing from a web browser, so that I don't have to use a separate SSH application then I'm all for it.
Re: (Score:2)
Oh.. do you *really* need SSH? You couldn't maybe setup a web page control center that would enter the command on the server after you verify yourself? Pretty sure you don't *need* to ssh :)
Re: (Score:2)
I can think of lots of reasons I would like this.
I have a server at home.
I would like to edit something at home, from work. That's one trivial example.
Do you ever get tired of shooting down somebody else's idea just because it doesn't immediately fit whatever is in your head RIGHT NOW?
Give it a second. Stop to think before you call somebody else crap and maybe just bite your F'ing tongue.
Re: (Score:2)
I have a server at home.
I would like to edit something at home, from work. That's one trivial example.
Me too.
But I already have several remote access solutions. (Remote desktop services, vnc, etc) I'm not sure what advantage this give me.
I'm not stifling creativity. I'm just not sure what the point is.
Re: (Score:2)
There's a slight but incredible difference between
Why exactly do I want do this?
and "What could I use this for?"
It's like the difference between "That's crap" and "I think there's a better way."
You don't think so, but it comes off as block-headed.
It's negative and derogatory. I'm so sick of negative responses I can barely stand it.
Re: (Score:2)
incredible difference between "Why exactly do I want do this?" and "What could I use this for?"
I disagree that there is any difference at all. You might have read some sort of negativity there, but take a closer look with an open mind to the idea that there was none. There really was none.
It's like the difference between "That's crap" and "I think there's a better way."
I see the difference there, and understand what your complaint is. I just don't think my statement has the negativity you saw in it. Sure
Re: (Score:2)
But I already have several remote access solutions. (Remote desktop services, vnc, etc) I'm not sure what advantage this give me.
I'm not stifling creativity. I'm just not sure what the point is.
The point is that VNC/RDP-like functionality is now accessible to everyone, which is a good thing. When was the last time Six Pack Joe had a properly configured VNC client?
Re: (Score:3)
The point is that VNC/RDP-like functionality is now accessible to everyone, which is a good thing. When was the last time Six Pack Joe had a properly configured VNC client?
joe six pack would "Windows Remote Desktop" by clicking it in the start menu, and entering the ip address to connect to.
The reason most joe-sixpacks don't is because they need to enable the RDP server, tweak the firewall rules to let the incoming connection, and know a bit about tcpip to find their address and/or register dynamic-dns name
Re: (Score:2)
Jose Sixpack can easily connect to a remote server, regardless of who set it up. This means that this can be commercialized to provide remote services (i.e., teh cloud) to Joe Sixpack. Imagine being able to connect to a real Libre Office app instead of the severely limited Google Docs.
If you think about it, connection via RDP is already commercialized (i.e., Amazon EC2) but it still requires the user to set up a client. Being able to connect via
Re: (Score:2)
I'm not stifling creativity. I'm just not sure what the point is.
And everyone knows rock attained perfection in 1974. You're messing with his Zen thing, man.
Re: (Score:2)
just because something can be done doesn't mean it should be. in this case, a 'cloud' centric computing future does NOT look like this utopia you speak of solely out of YOUR convenience.
Re: (Score:2)
You can already, though I haven't tried it on 4.
Just use the following URL:
Do it in a new tab and that tab will act like a new Firefox window (in a tab) with its own tabs and your plugins.
It works because Firefox is really just a XUL runtime...
Re: (Score:2)
Well, it looks to me like one of those situations where you apply the DRY principle to your design, and suddenly realize that the uncoupling of assumptions has created some surprising, oddball ways of using your code. Whether or not actually doing those things is sensible, you've got to code up some demonstrations just to show how successful your refactoring is.
Another way of saying this is that some people ask, "Why should the system be designed to make this possible?" Others ask, "Why should the system
Re: (Score:2)
Emacs is a nice OS, but it lacks a decent editor...
If you're going to start holy wars, you gotta get it right. ;)
Re: (Score:2)
Emacs is a nice OS, but it lacks a decent browser.....
An OS with cooperative multitasking didn't cut it in 2000, and certainly doesn't cut it now.
Re: (Score:2)
we have to go deeper.
That's what she said.
Just imagine the possibilities!" (Score:5, Funny)
Can't... I'm too busy imagining the performance...
Re: (Score:2)
And we reinvent X (Score:3)
X over HTTP?
If you thought remote X was slow, imagine the performance.
Re: (Score:2)
Save your breath. Inventing new standards and then running emulations of old standards over the new standard, or emulating the new standard over the old standard, is the bread and butter of 90% of the programming workforce. The more the new standards suck, actually, the better -- more work to do. It's a jobs program, I tell ya. And so much less challenging then actually learning the math needed to make something truly innovative.
That aside, this particular hack is actually potentially useful going forwar
Re: (Score:2)
still a bloated mess. X11/vnc/ms remote desktop et al are far superior for this than some stupid html/gtk wrapper. what will they think of next? oh wait, I know, they'll use some trendy half-interpreted half compiled language to code this in, increasing the bloat and slowness by multiple factors.. yuck. the stacks the industry is building these days are disgustingly obtuse... and all for the sake of moving execution control away from the consumer (so he can be charged over and over).
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
In terms of effect, what's the difference?
Running applications remotely with the display local? That's what X *does*.
this opens a possibility (Score:4, Funny)
this opens a possibility that even the modern, newest computers, with gigabytes of RAM and multiple processors/cores won't be enough to do a basic thing without nearly hanging the entire machine.
Re: (Score:3, Funny)
Whats hardware innovation if you can't make ludicrously inefficient use of it?
FTFY (Score:2)
This seems like the same sort of thing that gets Windows into trouble all the time (Flash in Excel for example).
Honest question. (Score:2)
Re: (Score:2)
Some reasons that come to mind:
1) Your'e not at home/wheever the app is running. You're someplace else and wanr to run it remotely. Think home security applications, programming your DVR, etc).
2) You can't run it regularly because you don't have access to the computer it's running on. Maybe it's running in the cloud.
3) This is just another way of building highly interactive web pages. Beats the crap out of AJAX. Google could rewrite GMail as a linux app rather than using web technologies, and you'd not know
Re: (Score:2)
Why would I want to run something like Gimp in a browser, when I can just run it regularly?
Because if you embed anything in a browser it automatically becomes a "web app" and you can charge for it, hold the data hostage, raise the prices whenevery you want, and yank access if the users complain.
Search your feelings. You know it to be true.
Re: (Score:2)
Can someone clarify what this does? (Score:2)
So the app is generating an HTML5 based web page that you serve to the remote browser? I'm not up-to speed on HTML5, so how does this handle the application "pushing" screen updates to the browser?
Re: (Score:2)
WebSockets: allows any web application to open a direct connection to a server (by default following the same origin restrictions) and it works only in Firefox because it is a nice hack/experiment and it is not targeting to be released on GTK 3.2. For those asking, it only works on Firefox NOW because the developer is just experimenting at the moment, using disabled by default WebSockets implementation in Firefox 4 (disabled on Opera too until the spec is more mature).
Not really running in a browser (Score:4, Interesting)
More accurately, GTK 3.2 will let you take a GTK program running /outside/ the browser and, assuming it does not use anything X11-specific, forward drawing as gzipped data: URIs to your browser which then assembles things in a element. It's basically a poor reimplementation of a VNC that only works in GTK. Significantly more interesting would be a GTK that draws with PPAPI and runs in NaCl, which would allow you to develop a web app using GTK, deploy it on the web, and run it (safely) within your browser.
Re: (Score:2)
More accurately, GTK 3.2 will let you take a GTK program running /outside/ the browser and, assuming it does not use anything X11-specific, forward drawing as gzipped data: URIs to your browser which then assembles things in a element. It's basically a poor reimplementation of a VNC that only works in GTK. Significantly more interesting would be a GTK that draws with PPAPI and runs in NaCl, which would allow you to develop a web app using GTK, deploy it on the web, and run it (safely) within your browser.
Two thumbs up for being knowledgeable on the subject.
Re: (Score:2)
Yeah, someone doesn't understand the difference between a GUI toolkit and a browser. I doubt Gtk+ 3.2 includes a C->HTML5 transcoder.
Anyone want to write a Gaussian Blur filter in ECMAScript, and run it on a four-million-pixel, 4-channel raster image?
Re: (Score:3)
Anyone want to write a Gaussian Blur filter in ECMAScript, and run it on a four-million-pixel, 4-channel raster image?
That's kind of doable now with (what is colloquially referred to as) HTML5. I know you're referring to the atrocity of running the actual convolution with browser JavaScript engines, but as it stands, you can specify the convolution filter in ECMAScript and pass it off to WebGL. The early part of this video has a pretty cool demo.
http://www.google.com/events/io/2010/sessions/gwt-html5.html [google.com]
Re: (Score:3)
It's totally doable *right now* in Flash.
http://www.adobe.com/devnet/flex/articles/pixel_bender_basics_flex_air.html [adobe.com]
Re: (Score:2)
> Anyone want to write a Gaussian Blur filter in
> ECMAScript, and run it on a four-million-pixel,
> 4-channel raster image?
So like https://bug495499.bugzilla.mozilla.org/attachment.cgi?id=380489 [mozilla.org] more or less, but doing gaussian blur instead of desaturate? The image there is a 4,096,000 pixel image (2560x1600); the desaturate filter takes 250ms on my machine in Firefox 4 (and that includes getting the bits out of the image and into JS and then back into image form).
Of course it's a bit slower in oth
Re: (Score:2)
Significantly more interesting would be a GTK that draws with PPAPI and runs in NaCl, which would allow you to develop a web app using GTK, deploy it on the web, and run it (safely) within your browser.
I agree that running a GTK app entirely clientside would be very interesting. There is another way to do that: Compile a GTK app from C to JavaScript using Emscripten [emscripten.org]. (I wrote Emscripten, sorry to plug my own project, but I'd be thrilled if someone used it to do something like this!)
screen (Score:2)
I would really enjoy having a "screen" tool for my GUIs sessions.
I recognize this idea... (Score:2)
Re: (Score:2)
WRONG, it is a web based VNC like utility, done at the GUI toolkit level. the apps runs on the server. Probably not useful for many people, but a modern replacement of the X protocol that can be made efficient at the toolkit instead of sending bitmaps all time like VNC do, and web enabled is something cool for remote administration for example (if done securely)
SAAS (Score:2)
Sounds good and all, but this just sounds like it will enable more "Software as a Service [bluehome.net]", or at the least make a quick and dirty kludgey GUI. I would rather see it done better myself. But I guess to each his own, unfortunately.
Dream the dream of days gone by... (Score:2)
Oh how wonderful to see this technology foretold so long ago [michaelv.org].
Hmmm, there used to be the same thing for Window Maker, but I can't find it anymore.
QDB says it best (Score:2)
<Pomax> 20 years from now, someone is going to have the radical idea to give users access to the underlying OS, rather than to the browser API, and he will be heralded a revolutionary.
<Pomax> All manner of programming languages will pop up that work outside "the browser", giving access to "offline" applications, storing files in "user space", even perhaps running in something called "kernel mode".
<Pomax> It'll be a brave new world.
<Mirell> It's scary that's be
Re: (Score:2)
You can expect the APIs on this kind of thing to become standardized in a few years.
Re: (Score:2)
...and then Apple will abandon the idea of forcing people to write all phone apps as a web app, and then Android will feature an NDK designed to make it as easy as possible to get your code running right under the kernel... ...except it didn't take two years, let alone 20, it was more like two months. So that actually never came to pass and that particular wag was amusing but nothing more.
Hurrah \o/ (Score:2)
Really? (Score:2)
Re: (Score:2)
Why wouldn't GTK apps under *BSD/MacOS/Windows work?
Re: (Score:2)
Re: (Score:2)
wormholes.
Re: (Score:2, Funny)
They get rid of minimize/maximize buttons to stay away from Windowsisms, yet they want to replicate what is basically ActiveX integration?
WTF, GNOME?
Re: (Score:3)
Re: (Score:2)
> Didn't we do this in the eighties with X-Terminals?
Always wondered why nobody just did an X Server as a browser plugin. Stick to the old 2D X drawing primitives and it is actually fairly fast vs posting bitmaps for everything like modern Qt and GTK does. It is a well known and documented interface and would allow pretty advanced apps.
Re: (Score:2)
http://www.google.com/search?q=X+Server+as+a+browser+plugin [google.com]
Seems someone did do it.
Re: (Score:2)
Making it a browser plugin is not fundamentally better than making it a separate app, as you still need to download and install it. Now, writing one in pure JS (with HTML5 canvas and WebSockets) would be impressive.
Re: (Score:2)
http://www.jcraft.com/wiredx/ [jcraft.com]
Re: (Score:2)
Re: (Score:3)
At least the implementation will be better.
You are joking, right?
Re: (Score:2)
Re: (Score:2)
This amount of bloat is insane.
The old joke was, {all programs will expand until they can read usenet}
Now it seems to be, {all programs will expand until they can run a browser inside them}
In both situations, the sane response is, "Quit adding bloat, and instead give me reliability, speed, and efficiency!!!"
Re: (Score:2)
I'm going to start presuming that any user ID over 2e6 is posting that link.
Re: (Score:2)
It's Slasher again. This guy's an elite troll. At least he's a nice refreshing break from all the noob/lame trolls, it's good to see somebody putting some thought and effort into it.
Re: (Score:2)
kristopeit seems to be back too.
Re: (Score:2)
why on earth does the OpenShot package in Ubuntu depend on libgfortran3
Presumably because they use linear algebra routines for something. Check to see if it also relies on LAPACK or BLAS, or a package which relies on either of those two. Or, perhaps there is some other maths package which it uses.
Re: (Score:2)
You can already run that one with CGI, if your goal is simply to make your server commit suicide. No need for a GTK wrapper (which you forgot to provide in any case). I think, perhaps, you don't understand that the apps are running on the server and only using the browser for display and input.
Re: (Score:2)
VNC is more or less just a video stream with input controls, while GTK+ 3.2 is an interface toolkit. This not only uses less bandwidth for the typical "remote desktop" scenario, but it also can be used for entirely different purposes, such as a web application or an interface to a remote application. They're two totally different things.
Re: (Score:2)
I remember that abomination. We looked at it, shuddered, and decided that *that* was going to be seriously banned in our company. We never regretted that.