Mozilla Tests Integrated Desktop Browser
HelloDotJPEG writes "Mozilla Labs, the organisation's experimental arm, has launched Prism for interested Windows users to try out. Prism is a piece of software which integrates web applications such as Gmail or Google Reader into the desktop. The program enables you to run multiple such sites as though they were local applications, each in their own dedicated browser window. The product isn't entirely new, but is an officially adopted and rebranded update to the Site-Specific Browser project WebRunner (not to be confused with XULRunner upon which it is built). From the site: 'Web developers don't have to target it separately, because any application that can run in a modern standards-compliant web browser can run in Prism. Prism is built on Firefox, so it supports rich internet technologies like HTML, JavaScript, CSS, and and runs on Windows, Mac OS X, and Linux. And while Prism focuses on how web apps can integrate into the desktop experience, we're also working to increase the capabilities of those apps by adding functionality to the Web itself, such as providing support for offline data storage and access to 3D graphics hardware.'"
"we're also working to increase the capabilities of those apps by adding functionality to the Web itself, such as providing support for offline data storage and access to 3D graphics hardware"
And thus it was so, that viruses became even more abundant, and 3D accelerated.
which is totally what she said
As always, the innovation over at Mozilla is incredible. After only months of intense development they managed to build an application that's like a browser except it's only a Gecko control in a window. No tabs, no anything. .NET's Browser Control.
I'm sure it would've taken years to build a similar application using
Dunno if i would ever use it. However if you were rolling out Googles web apps in an office enviroment then it might make life easier for the users. More of a desktop paradigm then a bunch of URLs.
I have excellent Karma and I am not afraid to Troll it.
Who would actually WANT something like this?
2 of the main reasons to run an application locally is so that you control your own data.....and don't have to look at ads. This looks like the worst of both worlds....right on your desktop.
I think it was almost ten years ago when Microsoft came out with active desktop and Netscape countered with something which was really a browser window taking up the whole screen and called a desktop.
I never saw either being used. Is this the same thing?
http://michaelsmith.id.au
Because Active Desktop caught on so well.......
You're doing it wrong.
"This is the dumbest thing I've heard since I started at Microsoft"
c++;
But does it support DNS?..
In Soviet Washington the swamp drains you.
Let me say goodbye to positive karma: Welcome back, dear Internet Explorer 3 days...
Mozilla head #1> Umm, MS copies our tabs in their so-called browser !
Mozilla head #2> Ok, let's make a version without tabs... and while we're at it, let's remove that pesky Back button - and we'll have a fix for the memory leak too !!!
Carefully crafted sig.
I don't care much about viruses, running Linux and all ... however XSS (cross-site scripting) is more of a concern. And site-specific browsers could be a good way to limit their reach, if they keep one set of cookies each.
Wrong again.
"This is the dumbest fucking idea I've heard since I've been at Microsoft"
"I think an etch-a-sketch with an ethernet port would beat IE7 in web standards compliance."
I'm sure it would've taken years to build a similar application using .NET's Browser Control.
.NET browser control and try to put something like this together in VisualStudio. And the result would suck because it's not the same thing. And that highlights a common problem with Windows developers: they don't think things through properly and instead take the obvious path that Microsoft has laid out for them.
.NET developers years to do this, and it still wouldn't work as well.
The site-specific browsers are full Mozilla browsers, they simply have some chrome removed.
But you are absolutely right that a Windows developer would likely take the
So, although it was meant as sarcasm, you are right: it would take a
This is useful for many users: it makes it much easier to migrate from desktop to web applications, and it is intrinsically easier for people to grasp "to get to your mail, click on the Mail icon" than "start the browser, go to your bookmarks, select...".
Also, if this is well executed, it provides a better level of isolation between web applications. Right now, it's pretty tricky trying to read mail for two or three GMail accounts (it would be less tricky if profiles weren't broken...), and if one web site locks up or slows down the browser, other web apps suffer as well. SSB can address those problems.
Am I missing something here?
How is this different to putting a URL shortcut on your desktop and having the browser window appear without an address bar?
What I would like to see is this implemented in such a way that the instances of these web-apps are separate from other XULRunner apps (e.g. if you also have Firefox running), and also isolated from one another.
The one advantage I can see here is keeping certain webpages open semi-permanently as applications (Gmail, Google Calendar, etc.). But it would be annoying if a crash of one app caused the others to close, too. Also, if this is designed properly it could be an advantage from a security standpoint. If your Gmail app is running separately from your web-browsing, there are fewer chances for cross-site scripting or other such exploits.
Since I frequently refer to Windows as "prison", I think "Prism" is a good name for a Windows app.
each time i try to install the program i get an error saying each file is corrupt. i tried to hitting ignore and letting it install anyway but when i run prism nothing happens. tried to redownload the install but the samething happens each time. any ideas?
Most comments here now remind me of the whole "no wifi, less space than a nomad, lame"-comment about the iPod when it came out. These comments are completely missing the point.
The current problem is that our desktop is built up around the idea of local applications and that is all the current desktops are designed to handle. But nowadays people are using less and less local applications and more and more web applications (whether you like it or not), and all of these run in a separate layer through the web browser. At some point, if we aren't already there, many people will not use a single local application on their computer apart from their web browser.
At that point, the whole distinction between the web browser and the operating system becomes completely irrelevant and we approach stage where windows is just a collection of device drivers (quote Netscape, mid nineties?).
Currently, the operating system does a lot of great stuff for us with regards to the local applications, and it really needs to start doing the same with regards to web applications and the first step is to make web applications first class citizens on the desktop.
Finally, complain all you want about the privacy and security issues with web applications. Well founded as they may be, they will not change the fact that people are flocking to web applications.
Active Desktop was a bit lame and MS seemed to have no real concept of where they were going with it.It was also well before the age of "web applications" as opposed to web sites. Just because there may be similarities with that old concept doesn't make this stupid.
Alright, alright. But it is actually a little bit more than just a stand-alone firefox-window with some web-app in it.
Prism is built on Firefox
Prism is built on Gecko. Please correct this article, it doesn't give enough credit to Gecko and unfairly gives too much credit to the Firefox team, who sometimes make bad decisions with respect to their product.
Seamonkey is way better anyway.
I started using WebRunner a few weeks ago and like it. I can just leave it to run minimized and not accidentally close it when I close FireFox or accidentally surf off the tab. This seems practical in several situations, set it up to wait online for World Series tickets, keep a calendar open. Most practical would be for use as a corporate tool, I could run my corporate web based software. Since most of the uses involve typing I miss automatic spell check from FF.
That doesn't defeat the purpose of web applications AT ALL!
.NET and Mono, so it would be like... ... it would be like the application would be USING system resources, but not written specifically FOR that system. We'll call them.... ".NET APPLICATIONS". They could be even faster than AJAX! And you wouldn't even need to be on the internet to use them! Not to mention you could take advantage of existing GTK or WinForms calls in order to get the smoothest fastest interface.
:) :) :)
Wait -- this is brilliant. An application that has access to the system resources instead of simply running as a webpage? Why, we could just write our applications as lean, platform-independent C-libraries-- and and -- I GOT IT -- we could create brilliant interfaces by simply hammering in layers of C# on
Join me, Slashdot- let us develop "applications"!
But wait- what if they need to be 100% multiplatform. What if I am not smart enough to abstract my code into a library/interface model... perhaps Prizm is for me!
From the user angle, now I can run Gmail on my desktop, because running Apple Mail or Windows Mail and just using the pop server is not Web 2.0 enough for me. I need to have ultra-slow, buggy, server-locked AJAX technology to make my desktop applications seem like desktop applications even though they are not desktop applications but they're acting like them thanks to Mozilla!
Server >> Application >> Web-Interface >> Prizm > Application Web/Server
Now everything will run as slowly and inefficiently as firefox!
I work in a (biggish) bank and this would fit well with our current portfolio of applications.
For our web based applications, our users are used to working with multiple browser windows opened simultaneously, each for a different part of our system (e.g. separate browser window for our credit cards system, different browser windows for our treasury system, different browser windows for our customer information system etc).
We actually forbid the use of the "back" button, and where possible we disable it (it messes up our data integrity). We also hide the address bar.
Because we also have applications developed as native windows GUI applications, Prism would probably make our web applications blend in with our GUI applications, improving the look and feel of our system.
Definitely something to check out in the future; although I doubt if it will be worth the hassle of deploying it.
Data storage is not a new concept; it's part of the HTML5 specification (a.k.a. Web Apps 1.0) [Note: that URL seems to have some script issues...] and it is already implemented in the recent WebKit nightly builds.
This might be useful in that it frees the developer from having to decide whether build a web app or a desktop app. However, that benefit doesn't come free. Part of the advantage of web applications is that the software is always up to date. That advantage becomes harder to accommodate if the data is stored locally since the software provider must now deal with a problem it didn't have to before: backwards compatibility. The 3d support would allow some applications to become better but remember that Ajax applications are slow - very slow - and adding native support also means reduced ubiquity. Would this program allow for offline support? Google Gears in theory addresses this problem but hasn't been ported to all applications and would need to be used in combination. In short, I see this application as expanding opportunities for web applications but is far from displacing desktop applications. I view this application as closer to Java applets rather than active desktop. Java applets desktop applications to the web while Mozilla's approach brings web applications to the desktop. Java applets are slower to load in a web browser than Ajax pages while Java applications are faster to run on the desktop than Ajax applications. If our goal is to provide occasional desktop access to applications predominantly accessed from a web browser then this is a good approach. But if the goal is to make desktop applications with the benefits of web applications then as developers we're better off writing our applications in Java.,
I like the idea of setting a browser's window as a separate app for a given site, so that site's app can be distinguished more easily from the other (many) browsers open at a time, many of those others just "casual", while some are persistent through the day.
But a separate app seems like a lot of overhead for a narrow solution. When I have a dozen Firefox windows open now, even though only half are persistent through the day, they all share the common resources. A separate app for each of the half-dozen persistent sites/apps means a lot of redundant app resources, in an already fat app (Firefox). I like the idea of securing each app from the other by separating processes, but that could be achieved with just proper programming the single app instance.
What is a better combo of all these features is just adding "window racks" to Firefox (and properly programming cross-site separation). I want to rack up several windows into a virtual rack that opens and closes all the windows together. So I can open a multiwindow session, with different apps in them, positioning/sizing the windows in the right arrangement for those apps in that session. I want to assign a "bookmark" (or app launcher) to that rack, assign a name to the rack that's displayed on each window, and probably an icon, too. And a rule that opens any clicked link in a new browser rack. That way the GUI is "partitioned" into my sessions for easily keeping them organized, both onscreen and in my mind when starting "apps".
Really this is a feature of the window manager (like X) that should be used by all apps. Because what I really want is a window rack of multiple apps I use all at once. I want to click a button and get my email, browser and spreadsheet apps launched at once, positioned properly on the screen. I could write a wrapper script right now to do that, and maybe even a GUI tool to populate "rack" scripts, but I want extra GUI support that makes all the windows in the rack clearly associated, maybe by a colored window border, common icon, or even (togglable) connecting lines. And I want the rack to enforce IPC permissions, so I don't even have to copy/drag/paste through my whole desktop, but rather just pop fields and objects among targets in the rack. IPC authorization among only "racked" (and otherwise explicitly associated) apps would make the whole OS more secure.
Window racks give me what I want. I think Prism is a long step down the wrong road.
--
make install -not war
Let's look at the facts. Mozilla is a highly profitable organization. You would think that Mozilla could afford to spend at least a little money on hiring Thunderbird developers. Yet in reality Mozilla has done the opposite: they have completely abandoned Thunderbird.
Why? Because of money.
The vast majority of Mozilla's income comes from Google. One of Google's main products is Gmail. Thunderbird competes with Gmail. So it makes sense that Google wants Thunderbird dead. Of course, they're not going to announce their intentions in a press release, but in reality that's exactly what's going on. Announcements like this one only make their plan more obvious than before.
This kind of anti-competitive behavior is exactly why most Slashdot readers hate Microsoft. Why is Google getting a free pass here?
I am all for this project and with the new features in FireFox 3 to help make this a better product for the desktop I am looking forward to using this in future products.
The main item I am looking into is that will each new instance be shielded from the other instances where they will not share cookies and other information?
I just tried it. (On gaiaonline.com, if you must know.) Horrid. Why?
Websites are designed to be used in a browser. Removing all the controls and stuff makes them hard to navigate, and the lack of tabs is quite a pain as well. I've seen that it has settings to add the navigation/etc back in, but then... Isn't it back to just being a browser? The biggest problem with this is that sites aren't designed for it.
In fact, I just tried it in GMail (on our domain) as well, and other than the fact that it's in a seperate process from firefox and they shouldn't crash each other, it's crap. If you click any of the links at the top, they open in browser. (I assume this will still be the case if you can 'installed' the 'app' for those as well.) None of the firefox extensions are available. Speed (of course) isn't improved.
And the worst part? Making it happen for a new site means creating a file, zipping it, and launching it separately. I could write a script that will quickly do that for me from a URL, but I shouldn't have to.
Unless I'm extensively working with a site that tends to crash the browser, I doubt I'll get much use out of this.
"If you make people think they're thinking, they'll love you; But if you really make them think, they'll hate you." - DM
To all you "This is stupid because I can do something similar and didn't get the credit for inventing anything"-sayers.
...... browsers, not appbrowsers, but webpage browsers. As webapps resemple local apps more and more, it makes sence to make them behave like normal apps.
The point of this is not "to make a browser without tabs" but to isolate a webapp, making it resemble a "conventional" app more.
Why not keep webapps in the browser? I don't know, why aren't we making local apps run in the browser, I mean, even though word doesn't use all web content, I'm sure you could get it to render in a IE or firefox tab. Then we could have all apps local or webbased, or as most are now: a mixture (Even word gets content off the internet when you run it "localy" with internet access).
The point is simply that browsers are meant to be
Ofcause this has been possible for a long time, but what this tries to do is simple follow the trends, and bring webapps to the local app space in a convinient and simple way.
It'll be funny if the wave of making webbased clones of local apps, ends up being swallowed by the wave of bringing webapps into the local app space.
Gmail competes with Thunderbird. Huh? What's the market share of Thunderbird again?
Oh, and Google just made IMAP an option on Gmail accounts. Thus allowing Thunderbird to "compete" more effectively against Google.
As I see it, this is basically making the browser serve a role similar to an XWindows server.
For great justice.
I don't want to kick off a flame war here, but upon reading some comments, the situation looks like a repetition of the good old browser wars.
/. constantly (MSFT) or sometimes (GOOG) frowns upon for bad competitive behaviour. What's your guess? Does this have the potential to become part 2 of the browser wars? Will the Firefox fanbase realize what Mozilla has become and switch to something else? KHTML? Something new, based on WebKit? Something completely new? Opera? Opera made GPL? Won't anybody care about Mozilla's creeping corporateness and be led right into the Google's den or will those ties break and Mozilla become less googly again? Discuss.
On one hand, there's Microsoft. With IE7, I hope we can all agree upon, they have fixed lots of very bad bugs it had. It's still a rather big ugly mess, but progress seems to be going in the right direction. With a bit of goodwill, one might even imagine them to produce a secure and usable browser with v8 or so. On the other hand, there's the ex-underdog Mozilla. Right after their Firefox started to take off, they managed to get quite a deal running with Google, who aren't evil. They also began "improving" CSS (moz-opacity and co., anyone?) and I think even html (not quite sure about this) in their Browser. Just like MSFT did in the browser wars. Now it seems they're attempting their shot at the web application game. They're trying to spread the use of XUL and apparently Prism will containt various "improvements" to agreed-upon standards.
Correct me if I'm wrong, but to me it seems the ambitious Mozilla project has evolved from a great OSS project to something that's not too different from companies
...why I'd want to pretend GMail was a dedicated email application running on my desktop, when I could just...run a dedicated email application on my desktop?
This is a move toward building a reasonable environment for web apps, not "typical websites". A web app wants full control of user interaction, in the same way that a client-server app or desktop app wants full control. Think of the classic problem: "I logged in to my online banking site, did some stuff, didn't log out, navigated to a few other sites, and walked away from the PC. Somebody came along and used the Back button to back into my online banking site, where I was still logged in...." What's wrong with this scenario (other than a brain-dead user)? The fact that the user could navigate away without logging out of the app or closing the window (with the effect of logging out). That's the role that Prism plays.
Think about your Father, Mother, clueless brother. This is much more user friendly than an in-browser view. Ofc, we all understand that it's not really any different, but explain that to your sister-in-law.
So PRISM is basically the "Open Link in New Window" function? How long did it take them to reinvent that?
I like Open Source software and Mozilla as much as the next guy, but doesn't it make sense to have your embedded controls be tightly integrated with the Operating System?
Why would you want your embedded controls to be tightly integrated with your OS? There's no reason for an HTML window to need tight OS integration. It's another web browser that's susceptible to all the issues that the core HTML engine is. It wouldn't necessarily be subject to the full browser's interface bugs, but it's got the same core so it would share those vulns. Ideally, you wouldn't want any integration with the OS.
I'd rather not need to have both IE and Gecko loaded into memory whenever I run Winamp.This acts as a full replacement for the IE control. If you have some apps calling one and some apps calling the other, yes, both will be loaded into memory. However, if all apps call only Gecko, then only Gecko will be loaded into memory. Excluding behind-the-scenes OS-IE integration that causes (parts of) IE to be loaded, of course; the apps themselves will only load one or the other.
Does it run threads? No? Don't call us, we'll call you.
GLaDOS for President 2016! "Well here we are again. It's always such a pleasure." -- GLaDOS, 2011
" we're also working to increase the capabilities of those apps by adding functionality to the Web itself, such as providing support for offline data storage and access to 3D graphics hardware.'""
forest for the trees.
They all run as a separate instance. Prism is based on WebRunner which you can try right now. Each runs as its own process so one screwing up doesn't screw up the others.
That said, WebRunner itself is extremely glitchy - maybe it's just behind the current Firefox code, but running things like Gmail and Meebo in it are major annoyances - above all, the cursor is never visible, in Gmail you often can't write a new email (only Reply to people), in Meebo closing a window causes the cursor to freak out and start selecting everything... pretty annoying. Prism will hopefully keep up with Firefox instead of doing its own thing like WebRunner seems to.