Slashdot Mirror
Mozilla Rising ... As A Platform
ceswiedler writes "Salon is running a story about Mozilla's potential dominance as a platform for application development. They discuss the community development centering around Mozilla, and point out that its cross-plaform GUI environment is 'exactly the kind of thing Microsoft was trying to prevent when it launched its war against Netscape. It didn't want Netscape around, because Netscape was becoming a platform.' In what might be a Salon first, they even include a reference to a Slashdot comment by SkyShadow."
I wonder if this is Salon's attempt to /. Slashdot for all the times Slashdot has hammered Salon? ;)
"...you can steal my woman, but you ain't done nuthin' smart."
now only if salon would write an article about the comments posted on slashdot referring to the article on salon that referenced a slashdot comment. than, slashdot would have to post a story about the article on salon about the story on slashdot that arose from an article on salon that featured a slashdot comment...
sorry, its been a long day.
it seems that mozilla, as a whole, will evolve into a framework of reusable components that will transcend the browser application itself.
this will pose to be a problem for microsoft; why bother using microsoft components, which are bound to windows, when i can program across multiple platforms using mozilla components?
My vote is for SVG, even though the current support for it in Mozilla is pretty fragile [YMMV, I'm on 1.1 Linux].
With full support for SVG, Web applications could really take off in a big way (graphical and not just text interaction) that is unhindered by platform specific nonsense.
One big hitch though seems to be in rendering quality outline fonts. Everyone would love to have the precision of PostScript for determining exactly where text is located, how far it extends, etc, but there seems to be big players that are nervous about releasing outlines of their fonts and have punted about precise layout of fonts inside SVG, deferring to upper level CSS specifications and what not that permit layout decisions to change when we really need a web layout engine that doesn't change from platform to platform (and is free and open).
"Provided by the management for your protection."
This might work. I *love* Jabber just get a server that has a good ICQ gateway and you should be rocking.
d =9 71468490
http://www.jabbercentral.org/clients/view.php?i
Cypherpunks: Civil Liberty Through Complex Mathematics. Those who live by the sword die by the arrow.
I've been looking forawrd to the Mozilla Programiing book from O'Reilly coming out. According to their web site it is coming out this month. Conspiracy anyone?
I've played with Mozilla some. Java script with CSS is a powerful way to do UI development. The question is how are we going to build apps that
1) Havethe install flexibility of a website
2) Have access to the local hard drive.
One cool thing about Mozilla is that you can remote an XUL reference just like an html, and it will render. This means that you get a pretty huge toolbox of UI available for anyone browsing using mozilla. One development tactic might me to use a XUL interface for layout, and swap out the javascript file to have different behavoir if you want to process locally or remotely.
I'd love it if SVG got into the main branch. As I understand it, the reason it hasn't was due to Licensing Issue. The original is under LGPL and GPL, but Mozilla is also licensesd under the MPL. Not sure what the SVG authors view on the MPL is.
Open Source Identity Management: FreeIPA.org
I think that in order for it to really drive the nail in the coffin, it's going to need a niche market. Incredibly good functionality really isn't enough to make the average user go out of their way to get it. The future is likely in the ability to discover the niche application that makes it undeniably more useful -- then all it has to do is hang on for a couple of years (which is harder than it sounds...)
Law of Software Envelopment jwz edition
``Every program attempts to expand until it can read mail. Those programs which cannot so expand are replaced by ones which can.''
Microsoft's problem is that it tied IE to the underpinnings of Windows, which essentially means you have to keep IE around. Mozilla doesn't tie itself directly into the OS.
I'm not sure about the interaction, but I think it will be something like: [Kernel} ---> [Mozilla App Layer] ----> Application
Keep in mind that not every single application written for an OS will run through the Mozilla layer, only those apps written with the Mozilla framework would pass through the app layer.
"What kind of chip you got in there, a Dorito?" - Weird Al Yankovic
Sure, there is initial appeal to having your application look the same on all platforms. Who really wants to write the same application N times? However, cross-platform consistency isn't necessarily a good thing.
Each platform has its own quirks with how it should behave. For example, menus in Windows are expected to be static (that is, they stay visible after the user releases the mouse button), while Macintosh menus tend to be rubber-band (menu disappears when user releases mouse button). In Windows, a menu action simply happens while on Macintosh, the selected menu item flashes several times.
I could go on and on with the differences between the Windows and Macintosh platforms (to say nothing of UNIX!). The point is that an application that acts differently from every other program is an application that is harder to learn. Users are forced to keep two sets of expectations, which completely defeats the purpose of using a cross-platform GUI!
Yes, you can tweak the UI so that it looks more like the host operating system. This is a thin veneer, however, as the emperor's proverbial clothes come into view when the OS theme is changed.
It makes sense that the UI should be abstracted from the rest of the application, but XUL is not the answer.
Nathan
If you want to get an idea of what is possible, check out this tutorial.t u/
http://www.xulplanet.com/tutorials/xul
I played with it about a month back and was amazed at how easy it makes GUI development.
mozilla with xul/js allow you to build some interesting tools. But try building a simple front end tool that reads a RDF as a remote datasource. I have yet to see an online working example displayed in a tree.
/xul python search page. Quickly I checked the xul source to see if mark used remote RDF only to see the code commented out with a remark along the lines of, 'almost got going'. Marks example works ,but like the code I was working on it had to use a different approach.
While the responses on the mozilla newsgroups are excellent (with the actual netscape engineers responsible responding), the lack of consistant *complete working examples* is a pain.
I had to laugh when I stumbled upon Mark Hammonds site and found a mozilla
I just want to to use remote RDF feeds.
peterrenshaw ~ Another Scrappy Startup
This is where I do my little dance and feel special. Salon quotes me, *and* I get an article on the front page! Then I post this OT, worthless post and burn off my karma.
Every year during my review, I just pray the words "slashdot.org" aren't mentioned.
there already is a MozillaOS, it's called:
ByzantineOS it's bare bones Linux with Mozilla and sawfish. Boots and runs from a CDrom without touching the local harddrive. it's small...and I tried it on 2 machines, all I had to do was pick low or high res, get my connection "dhcpcd" , and start the GUI "startx" real slick once it loads you can remove the cd, and when you're done you don't 'shutdown' you just kill the power....and it's FAST.
ZillaVilla.com for Mozilla profile roaming.
Every program attempts to expand until it can read mail. Those programs which cannot so expand are replaced by ones which can
:) I think the ultimate stopping point of development on emacs is going to be when the emacs hackers sit down to make improvements in the program, and the program ends up responding, "I wouldn't do that if I were you, Dave"
Yep, that's one of those quasi-funny computer "laws" that actually has a very disheartening core of truth to it. Of course some programs such as emacs expanded until they could read mail and then kept going
Here's another one of those informal computer laws that's ha-ha funny...but serious:
Greenspun's Tenth Rule of Programming:
"Any sufficiently complicated C or Fortran program contains an ad-hoc informally-specified bug-ridden slow implementation of half of Common Lisp."
-- Truth goes out the door when rumor comes innuendo. -- Groucho Marx
I like XUL. I think it's a great idea and the implementation rocks. But most of all, it's simple. There are no DLLs, no IUnknown pointers or registry issues to deal with. Mozilla is a great browser, in many respects superior to IE, and in some inferior (my dream browser would be a combination of IE, Mozilla and Konqueror which runs on Windows, OSX and Linux. Oh well). But the difference it was designed from the sart to *be* a platform, where with IE platformitis was an afterthought.
But I disgress. The key here is going to be Mozilla's ability to gain critical mass with average developers in Windows for it to take off. I'm not talking about XPCOM hackers, I'm talking about the ones quoted in the Salon article. It will do Mozilla no good if it takes off in Linux, because Linux has no desktop presence to speak of, and it has a far greater variety of browsers that, while good for competition, also cause fragmentation.
I think Microsoft's response to this (if they do get to the point where they consider the Mozilla *platform* a threat) will be to essentially take IE and turn it into a .NET platform. If they can offer a platform to people writing C# and VB.NET and JScript.NET, they'll be all set - assuming the .NET thing does take off like they want to. Of course, one of the catalysts to .NET acceptance will be how many computers it happens to be installed in - imagine if anyone who wants to use the next version of IE has to download the .NET runtime?
Still, Mozilla has the upper hand because it's off on the race and Microsoft is standing in the starting line wondering what the futz is going on and why are all these geeks cheering?
(it exists. I think this is it)
Download the adblocker.xpi file (Shift+click to download). When you download the adblocker.xpi file in Netscape 7, it will add .txt to the filename
(adblocker.xpi.txt). Before saving the file, remove .txt
from the filename and save the file to disk. Then in Netscape 7 click
File | Open to install.
In Netscape 7 click Edit | Preferences | Advanced - Scripts & Windows to unselect or select the Open unrequested windows
So close and yet so far from the world's perfect ID number
When the absolute best things going for Mozilla for developers is its array of integrated development tools. Mozilla's DOM Inspector and JavaScript debugger are absolute heaven after coding for IE and MS's poor excuse for a browser development environment.
The DOM Inspector lets you interactively walk through the DOM of a page viewing each containers attributes and children. You can interactively change values and appearance. You can turn on the 'blink' feature to temporarily 'blink' whatever element you are selecting in the DOM. You can also view all CSS elements on the page and inspect how they are cascading. And lots more. Wow!
The JavaScript debugger is everything we have come to expect in a 'standard' development environment... but it is for JavaScript. Set breakpoints.. set watches.. step through code.. evaluate javascript in context.. change code on the fly..
And included in the JavaScript debugger app is JavaScript profiling! Turn it on and play with the page.. then save the results to a number of different formats. You get an excellent breakdown of what code was executed and for how long, how many calls were made, where the execution time was spent etc etc.. just like you would expect from a Profiler. Now I can definitively show how much overhead comes with using DynAPI!
And all of this built into the browser! I think from the development standpoint alone, it will boost productivity by an order of magnitude. Takes out so much of the guesswork that usually goes along with front-end development.
I think Microsoft should be afraid. Very afraid. Mozilla is what browsers should have been 5 years ago. I've now switched my development environment to developing under Mozilla and then testing IE later for any quirks. The dev time is radically decreased.
Cross platform is irrelevant when you have 95% of the market. That's like saying that a business isn't viable unless it gets 100% of potential customers. That's ridiculous. Ever notice that there isn't just one gas station in a city? How about restaurants?
Hell, I own my own retail store, and I get about 60% of all potential customers in my area, and that's more than enough. Why would I bother watering down my product line to pull in everybody else? It'd ruin my business. One size doesn't fit all in ANY product.
Non-commercial developers? What does that have to do with anything? I write apps all the time for myself using IE. What's your point? There aren't any restrictions if you write a shrink-wrapped app that grabs a few IE objects. You don't have to license IE. You just specify that IE is required to be able to use your product. And with a nearly 100% saturation on the Windows platforms, which have a 95% desktop saturation, that's not a problem.
I can speak on that. WSS tags your browser with a cookie containing, among other things, the number of times you've visited a site and the last time you visited that site. The WSS server software can take that and determine whether you're a unique visitor to the site that day and/or forever. The browser figures are based not on total hits, but on unique browser instances as determined by the cookies. So it doesn't matter if you hit the site once or 1000 times, you're one unique visitor for purposes of counting browser types.
Note that identifying a particular browser instance isn't needed with cookies, the fact that WSS's servers got that particular cookie automatically does all the work that unique IDs would have been needed for. Of course, it also means that if you block third-party cookies WSS can't keep track of the timestamps and counts and so can't include you in the statistics.
Disclaimer: I write WSS's front-end software, the stuff the browser actually talks to. Take this as you will.
Greenspun's Tenth Rule of Programming:
"Any sufficiently complicated C or Fortran program contains an ad-hoc informally-specified bug-ridden slow implementation of half of Common Lisp."
"Including Common Lisp."
- Robert Morris
(I love this one -- I found it on Graham's webpage, you know, the one developing the 'arc' programming language.)
-Billy
Now you tell me ! I've had a page that's been downloading for 7 years. Now I can finally stop it and reboot !
Skyshadow:
Why use Netscape (Score:6, Linked)
by Skyshadow on Thursday August 29, @02:56PM
"Why should/would I use Netscape instead of Mozilla? Not getting enough pop-up windows in my life? Feel the need for a more closed solution?"
Outdoor digital photography, mostly in New Engl
Here's the google cache for the Sky Shadow page... oh wait. heh.
So close and yet so far from the world's perfect ID number
Actually it's not the same, it's more akin to JNI or the Java -> ActiveX bridge that Sun wrote a couple of years ago.
M$ wanted to change the language to make it incompatible, I just want Java objects exposed in XUL.
Arrogance is Confidence which lacks integrity. -- me
Microsoft's problem is that it tied IE to the underpinnings of Windows, which essentially means you have to keep IE around.
Really?
Then how was Microsoft able to release IE for Solaris and HP-UX?
I'm not sure if you're trolling or not... your argument is pretty bold yet lacking in supporting facts. But if you aren't trolling I would suggest you reverse the argument. I think it's more accurate to say that the more recent versions of Windows depend heavily on IE (consider it's integration into the shell).
Mozilla still doesn't seem to have the incremental layout capabilities of Netscape 0.9
.9 was nothing like it is now, probably at least an order of magnitude simpler. Compare the First early specs of HTML with HTML 4 and that doesn't even include CSS. HTML 2 (which your comparison browser couldn't even render because it was too complex) is a 77 page spec, HTML 4.0 (linked above) is close to 286 pages.
"incremental layout" depends a lot on the HTML complexity and the HTML author. you need to define the sizes of layout objects before you can lay out things past them. the IMG WIDTH and HEIGHT tags introduced by netscape helped this a lot, where you can say "hey, I'm blocked on getting this image, but I know what size it will be, so let me render the stuff after it and I'll worry about putting the image in later". Tables and CSS add to the complexity of determining sizes. You never really know the size of a table until after you read the trailing TABLE tag and you may even need to know the sizes of multiple elements inside the table until you load them, so you essentially have to grabthe whole table before you can show anything inside. The state of HTML at the time of Netscape
making as many connections as you wanted (later capped at 20)
It still does this, defaulted at 4. You can change this in user.js, it's just not a pref you can see in the UI anymore because folks abused it too much, and there definitely is a diminishing returns thing, and mostly - you just don't need to change it. HTTP 1.1 also lessens the need for this, drastically reducing the overhead for small objects, where socket start and teardown time is a much more significant part of the overall time.
These days the thing will freeze as it loads some plugin or other, maybe this is somehow harder than images
This is harder, and the memory requirements are huge. You're loadoing a bunch of new code, having to dynamically link stuff all over the palce, establish communication links, allocate memory, a nuch of stuff. The image library is already loaded, and showing an image takes a lot less memory than say, showing a 10 meg shockwave game.
It's hard to make comparisons now, since our browsers are required to do so much more. I tried to look at some old browsers just for the hell of it, and I couldn't even get NCSA Mosaic to run, just blew up on me.
- Does AOL even use their own Netscape product with their web subscriptions, or is that IE deal still in play? How retarded can you be?
It makes perfect sense unfortunately. But you need to know a little more about what really goes on and went on behind the scenes.When AOL first "integrated" Spry Browser into the AOL service, many many apps were written to build and serve the content they (AOL) use and show - like Rainman for one. If you dont know what it is, get a job for AOL or an AOL partner. What it means is (still) there is a lot of proprietary non-web ready content out there that needs to be changed over - including tons done by content providers who pay for the priviledge of serving their content via AOL - like WebMD used to and many other channels.
AOL cant switch over until that situation is dealt with - which means writing code for Mozilla/Netscape that allows Rainman generated content to be viewed, as well as many other proprietary formats AOL uses.
When MS decided they wanted a browser and failed miserably at the attempt of creating one, they "acquired" Spry and relabelled the browser IE... which is how AOL got stuck with it.
Hence, CompuServe (an AOL company) already has a Netscape version available while the AOL service does not.
-Rob
WebMaster:
BinFeeds
XXX Thumbnailed Image Newsgroups but
O'Reilly is publishing Creating Applications with Mozilla this month.
But I want real examples where people were fully unable to continue or cope because of an "inconsistent user interface".
Since when was "no users give up" the only criterion for evaluating a program's usability?
The shareholder is always right.