While all of these people say to just install over your old version, they would be wrong. Over time, old files start sifting through. One of the most common problems of not-uninstalling is your UA string stays on the old version, but your browser is definately the new version.
You'll want to take a look at bug 237727* to see that they are going to clear out some of the old files if you choose to reinstall over your old version. They have already done some good work on that bug for the next versions (FX 1.0PR and TB 0.8), but they just need to widdle down which files need to be wiped. I have faith that this'll be fixed by FX 1.0.
One solution was to automatically delete the entire install directory right before reinstall. This actually backfired when stupid users tried to install to "C:\Program Files\" and it wiped that folder clean. Oops.
What types of security bugs do you consider to be "critical"?
In general we consider critical security bugs to be those that allow execution of arbitrary code on users' systems or that otherwise allow access to users' confidential information. In the latter case we consider bugs to be critical only if they potentially expose high-value personal information (e.g., passwords, credit card numbers, and the like); in the context of the bug bounty program we do not consider bugs to be critical if they potentially expose only lower-value information (e.g., browsing history) or information that would be useful primarily for other exploits (e.g., the names of files or directories on the user's system).
Finally, in general we do not consider bugs that allow denial of service attacks to be critical in the sense described above.
I'm not sure how you know that SeaMonkey does this, but if it does, and Fx does not, file a bug. Take a look at the source. Bitching on slashdot'll get you nowhere.
I guess if you are going to make me fact-check, I might as well go all out. When I said IEHTML, I meant MSHTML -- thanks for not acronym nazi'ing me.
"According to Microsoft, IE 4 contains nine components that interoperate for full functionality (see Figure 2). These components include five DLLs--Wininet.dll, Urlmon.dll, Mshtml.dll, Shdocvw.dll, and Comctl32.dll--along with two executable files, Explorer.exe and Iexplore.exe, and a security and a Java virtual machine component. Both IE 3 and IE 4 require these DLLs to function and without them could not offer the displays we expect and the features we expect, nor access the Internet at all. Wininet.dll offers Internet services and capabilities, including HTTP and FTP access, modem dialing, and browser caching. Urlmon.dll lets developers treat URLs as if they were programmable objects and thus build them into their applications (Microsoft's own Office 97 does this). Mshtml.dll is the component that lets you view your folders as Web pages: Choose a disk drive from My Computer, for example, and what you see is Dynamic HTML, Microsoft-style, relying on this particular DLL. Shdocvw.dll (Shell Document View) lets developers build browser capabilities into software, while Comctl32.dll (Common Controls) provides low-level support for menus, toolbars, progress bars, and many other controls used across Windows applications." (emphasis mine) Source.
As you can see, at least 2 of the 5 DLLs IE needs are in use by windows. Sure, this is IE4, but it was the first in-depth analysis I found.
Firefox's chrome is very thin layer on top of gecko, so that doesn't take long. Loading up the static GRE is what takes it the longest time. MSHTML.DLL (as well as comctl32.dll) is already in use by the windows shell before anyone double-clicks on "The Internet".
IEHTML is loaded in the OS to display lots of things (folders, icons, your desktop, etc). It's part of the windows shell. The IE application just loads up the browser chrome, and uses the preloaded IEHTML to display websites.
Firefox (the quickest-launching of the Mozilla line) has to load up gecko, the rendering engine, each time a process starts. It's browser chrome is just some JavaScript, CSS and the data to be displayed (XUL), which is displayed using gecko. If your shell were to run on top of the GRE, and Firefox were allowed to share that GRE, it would load up almost instantly -- seconds before an IE that wasn't halfway loaded into memory.
Who would load up faster, Firefox or IE when both were forced to load everything from scratch? I don't know. It doesn't matter though. Fx loads fast enough for me now.
Firefox will have RSS reading built into it by 1.0. There's your Mac/Linux/Windows solution. =P
They call it livemarks and it autmatically picks up on sites that offer feeds. You can add a livemark by clicking on a button on an RSS/Atom Enabled website. It feels just like a folder of bookmarks (where each bookmark is an entry).
From Blake Ross's blog (a mozilla developer who is only 19) says "We are less concerned with actual download numbers than with trends--how many more or fewer people are using Firefox this week than last--and we're very happy with the trends we see now. Our goal, then, is to concern ourselves more with maintaining these trends than with the big picture: keep Firefox growing weekly and the rest will follow."Source.
Given what's been going on recently, however, it seems like only a matter of time before somebody justifies using it on Americans on the grounds of terrorism prevention.
I'm born, raised, and currently live in Missouri. I loathe that man.
A search for 'search' returns a list of websites that all contain the word search in their domain name, none of them are any of the good ones - Google, AV, Yahoo, and (even!) MSN.
Yeah, that gets me too. The slashdot bug is bug 217527 and as seen in comment 81 was backed out of the aviary (Firefox 0.9 and 1.0, and Thunderbird 0.7 and above) branch because it caused a regression (bug 246382). If you get a trunk build, the bug will be fixed.
* You'll have to copy/paste those links into your Address Bar, because bugzilla blocks links from slashdot.
Th reason your are still working towards creating proper semantic documents and separate content from design is because you have been doing HTML professionally for years. I know, because I have too. We had to hack out those tables and 1px spacer.gif files to get things to work cross-browser. The standards movement is just catching on.
It's like learning a language wrong the first time. You are relearning. She'll be learning HTML properly for the very first time. As long as she reads books and tutorials and doesn't look at random web-page source, she'll learn it the proper way.
What's new? Apparently a pop-up blocker, and extensions. (That sounds familiar.) Also they locked down the "default zone" so that if (when) a security breach occurs, a virus won't be running in a privileged security mode.
This all comes at a cost. Some old plug-ins don't work. At least that'll back the people off from bitching when Firefox 0.9 comes out and everybody has to fix their extensions.
The algorithm for the adaptive junk mail controls has been heavily redesigned to learn faster and catch more spam.
To get the best possible experience from the new junk mail controls, we highly recommend that you re-train the filters from scratch. Tools > Junk Mail Controls > Adaptive Filters > Reset Training Data. Be sure to train an equal number of good and junk messages. We recommend several hundred messages of each.
The enable/disable option for adaptive junk mail detection appears to apply to all accounts (Tools > Junk Mail Controls > Adaptive Filters). It is, however, a per account option. To set the option for a specific account, choose the account in the 'Account:' dropdown on the 'Settings' panel, then switch to the 'Adaptive Filters' panel and set the option. Repeat per account as needed.
There are a couple bugs on context menu hijacking (backend, firefox work, seamonkey work, etc) but if interested, you should start out at bug 117532. Links from slashdot are forbidden, so you'll have to copy and paste:
I've read your stuff, know of the visual identity team, and thank you for your work (UI is important to me), but are you sure you have the right authority to say these things?
I was under the assumption (from the 03/24/04 minutes) that the --enable-official-branding flag could only be set if you had foundation approval. It seems silly to have the flag that anyone can (and should) use on their builds instead of having it just built in automatically.
Obviously nobodys going to get into a fuss about personal builds. We're talking builds that get distributed in Mozilla's name. Debian can't just use the flag and compile, because they have to be approved to use that artwork.
First off, not being part of the Free Software Foundation does not make them not free. To say that without the triple negatives, Mozilla and Mozilla Firefox (the code) are free software. Here's a clip from the MPL tri-license for c files:
Alternatively, the contents of this file may be used under the terms of either the GNU General Public License Version 2 or later (the "GPL"), or the GNU Lesser General Public License Version 2.1 or later (the "LGPL"), in which case the provisions of the GPL or the LGPL are applicable instead of those above. If you wish to allow use of your version of this file only under the terms of either the GPL or the LGPL, and not to allow others to use your version of this file under the terms of the MPL, indicate your decision by deleting the provisions above and replace them with the notice and other provisions required by the GPL or the LGPL. If you do not delete the provisions above, a recipient may use your version of this file under the terms of any one of the MPL, the GPL or the LGPL.
You CAN relicense Mozilla code as GPL which is (as everyone knows) free software. That goes even without talking about MPL code and whether or not it is a license for "free software".
All of this has a big however.
However, the artwork is not licensed under the MPL. The artwork is their property and is not part of the "free software" code. It's their decision to make. Check out that link for the "why" (by ben goodger, lead programmer for firefox).
I think they are still being altruistic, but being smart about it. Like their name (which they had to fight the godzilla guy for), they retain a trademark on the graphics so that they can brand official mozilla-approved builds so that their user base knows they can trust their branded build.
If the Google search engine were free software, you wouldn't want to run into a really crappy google spin-off. "That was an incredibly crappy search." The Google name would be tarnished.
I do understand what you are saying, and agree with it on a limited basis: Mozilla was originally open-sourced so that they could benefit from the OSS developers that wanted to write a kick-ass browser. It wasn't so all man could be free in their web browsing (though that was a side-effect). I believe Mozilla to be pure in their intentions, and while there are sticky points (Debian), I think they've got smart enough guys to figure something out.
Fedora Core 1 forum posting with people who have already done it. It seems pretty easy from the looks of it. I'm going to do it just as soon as I get some free time...
And for the love of god, please read the whole thread. Don't ever install a kernel with rpm -Uvh. Leave yourself a backup (rpm -ivh).
As much as I am a mozilla supporter (I really love the element-animal products -- I've installed them on my parent's system), I don't think the idea of a google branded thunderbird would work well, at least until Ben finishes his smart update (due for firefox 1.0) and Scott encorporates that into Thunderbird. If google will be giving it out, it will need to be dummy proof, and we can't have X people playing around with Google Thunderbird 0.5 for the next few years.
It's not done yet, so you'll need to read some of the actual forum posts to figure out what you need to do, but they have an experemental version that I love.
Slashdot Mirror
and the newly acquired
http://www.firefox.com/
and
http://firefox.com/ (redirect to http://spreadfirefox.com/)
While all of these people say to just install over your old version, they would be wrong. Over time, old files start sifting through. One of the most common problems of not-uninstalling is your UA string stays on the old version, but your browser is definately the new version.
You'll want to take a look at bug 237727* to see that they are going to clear out some of the old files if you choose to reinstall over your old version. They have already done some good work on that bug for the next versions (FX 1.0PR and TB 0.8), but they just need to widdle down which files need to be wiped. I have faith that this'll be fixed by FX 1.0.
One solution was to automatically delete the entire install directory right before reinstall. This actually backfired when stupid users tried to install to "C:\Program Files\" and it wiped that folder clean. Oops.
* Bugzilla hates slashdot; copy and paste.
What types of security bugs do you consider to be "critical"?
Solution:
Disable Active Scripting.
Use another product.
I'm not sure how you know that SeaMonkey does this, but if it does, and Fx does not, file a bug. Take a look at the source. Bitching on slashdot'll get you nowhere.
I guess if you are going to make me fact-check, I might as well go all out. When I said IEHTML, I meant MSHTML -- thanks for not acronym nazi'ing me.
"According to Microsoft, IE 4 contains nine components that interoperate for full functionality (see Figure 2). These components include five DLLs--Wininet.dll, Urlmon.dll, Mshtml.dll, Shdocvw.dll, and Comctl32.dll--along with two executable files, Explorer.exe and Iexplore.exe, and a security and a Java virtual machine component. Both IE 3 and IE 4 require these DLLs to function and without them could not offer the displays we expect and the features we expect, nor access the Internet at all. Wininet.dll offers Internet services and capabilities, including HTTP and FTP access, modem dialing, and browser caching. Urlmon.dll lets developers treat URLs as if they were programmable objects and thus build them into their applications (Microsoft's own Office 97 does this). Mshtml.dll is the component that lets you view your folders as Web pages: Choose a disk drive from My Computer, for example, and what you see is Dynamic HTML, Microsoft-style, relying on this particular DLL. Shdocvw.dll (Shell Document View) lets developers build browser capabilities into software, while Comctl32.dll (Common Controls) provides low-level support for menus, toolbars, progress bars, and many other controls used across Windows applications." (emphasis mine)
Source.
As you can see, at least 2 of the 5 DLLs IE needs are in use by windows. Sure, this is IE4, but it was the first in-depth analysis I found.
Firefox's chrome is very thin layer on top of gecko, so that doesn't take long. Loading up the static GRE is what takes it the longest time. MSHTML.DLL (as well as comctl32.dll) is already in use by the windows shell before anyone double-clicks on "The Internet".
IEHTML is loaded in the OS to display lots of things (folders, icons, your desktop, etc). It's part of the windows shell. The IE application just loads up the browser chrome, and uses the preloaded IEHTML to display websites.
Firefox (the quickest-launching of the Mozilla line) has to load up gecko, the rendering engine, each time a process starts. It's browser chrome is just some JavaScript, CSS and the data to be displayed (XUL), which is displayed using gecko. If your shell were to run on top of the GRE, and Firefox were allowed to share that GRE, it would load up almost instantly -- seconds before an IE that wasn't halfway loaded into memory.
Who would load up faster, Firefox or IE when both were forced to load everything from scratch? I don't know. It doesn't matter though. Fx loads fast enough for me now.
Firefox will have RSS reading built into it by 1.0. There's your Mac/Linux/Windows solution. =P
They call it livemarks and it autmatically picks up on sites that offer feeds. You can add a livemark by clicking on a button on an RSS/Atom Enabled website. It feels just like a folder of bookmarks (where each bookmark is an entry).
I don't know if he's a "programming god," but I seriously doubt he's "some highschool kid with all summer to screw around."
Funny thing is, he just turned 19. =)
I don't know if he's a "programming god," but I seriously doubt he's "some highschool kid with all summer to screw around."
Funny thing is, he just turned 19. =)
From Blake Ross's blog (a mozilla developer who is only 19) says "We are less concerned with actual download numbers than with trends--how many more or fewer people are using Firefox this week than last--and we're very happy with the trends we see now. Our goal, then, is to concern ourselves more with maintaining these trends than with the big picture: keep Firefox growing weekly and the rest will follow." Source.
A search for 'search' returns a list of websites that all contain the word search in their domain name, none of them are any of the good ones - Google, AV, Yahoo, and (even!) MSN.
Hrm...
Yeah, that gets me too. The slashdot bug is bug 217527 and as seen in comment 81 was backed out of the aviary (Firefox 0.9 and 1.0, and Thunderbird 0.7 and above) branch because it caused a regression (bug 246382). If you get a trunk build, the bug will be fixed.
* You'll have to copy/paste those links into your Address Bar, because bugzilla blocks links from slashdot.
Th reason your are still working towards creating proper semantic documents and separate content from design is because you have been doing HTML professionally for years. I know, because I have too. We had to hack out those tables and 1px spacer.gif files to get things to work cross-browser. The standards movement is just catching on.
It's like learning a language wrong the first time. You are relearning. She'll be learning HTML properly for the very first time. As long as she reads books and tutorials and doesn't look at random web-page source, she'll learn it the proper way.
As much as I am a fan of everything mozilla, and can't stand IE, the next version of IE will come out in Windows XP SP2. Lets get our facts straight.
What's new? Apparently a pop-up blocker, and extensions. (That sounds familiar.) Also they locked down the "default zone" so that if (when) a security breach occurs, a virus won't be running in a privileged security mode.
This all comes at a cost. Some old plug-ins don't work. At least that'll back the people off from bitching when Firefox 0.9 comes out and everybody has to fix their extensions.
Can I turn off the preview pane yet?
o gm espane
http://texturizer.net/thunderbird/extensions/#t
What's New?
Improved Junk Mail Controls
The algorithm for the adaptive junk mail controls has been heavily redesigned to learn faster and catch more spam.
To get the best possible experience from the new junk mail controls, we highly recommend that you re-train the filters from scratch. Tools > Junk Mail Controls > Adaptive Filters > Reset Training Data. Be sure to train an equal number of good and junk messages. We recommend several hundred messages of each.
The enable/disable option for adaptive junk mail detection appears to apply to all accounts (Tools > Junk Mail Controls > Adaptive Filters). It is, however, a per account option. To set the option for a specific account, choose the account in the 'Account:' dropdown on the 'Settings' panel, then switch to the 'Adaptive Filters' panel and set the option. Repeat per account as needed.
There are a couple bugs on context menu hijacking (backend, firefox work, seamonkey work, etc) but if interested, you should start out at bug 117532. Links from slashdot are forbidden, so you'll have to copy and paste:
2
http://bugzilla.mozilla.org/show_bug.cgi?id=11753
Firefox will get the speed improvements, but since Firefox is already smaller and uses less, it won't be as significant (I think it is 3%?).
They basically rewrote the string implementation and it is "better faster stronger" than before.
So yeah, Firefox 0.9 will get a speed improvement too. (You can also grab a nightly. They have the improvements -- and more bugs.)
P.S. Also new in Mozilla 1.6 is the ability to block websites from hijacking your context menu (right click menu) in the browser. Yay!
I've read your stuff, know of the visual identity team, and thank you for your work (UI is important to me), but are you sure you have the right authority to say these things?
I was under the assumption (from the 03/24/04 minutes) that the --enable-official-branding flag could only be set if you had foundation approval. It seems silly to have the flag that anyone can (and should) use on their builds instead of having it just built in automatically.
Obviously nobodys going to get into a fuss about personal builds. We're talking builds that get distributed in Mozilla's name. Debian can't just use the flag and compile, because they have to be approved to use that artwork.
You CAN relicense Mozilla code as GPL which is (as everyone knows) free software. That goes even without talking about MPL code and whether or not it is a license for "free software".
All of this has a big however.
However, the artwork is not licensed under the MPL. The artwork is their property and is not part of the "free software" code. It's their decision to make. Check out that link for the "why" (by ben goodger, lead programmer for firefox).
I think they are still being altruistic, but being smart about it. Like their name (which they had to fight the godzilla guy for), they retain a trademark on the graphics so that they can brand official mozilla-approved builds so that their user base knows they can trust their branded build.
If the Google search engine were free software, you wouldn't want to run into a really crappy google spin-off. "That was an incredibly crappy search." The Google name would be tarnished.
I do understand what you are saying, and agree with it on a limited basis: Mozilla was originally open-sourced so that they could benefit from the OSS developers that wanted to write a kick-ass browser. It wasn't so all man could be free in their web browsing (though that was a side-effect). I believe Mozilla to be pure in their intentions, and while there are sticky points (Debian), I think they've got smart enough guys to figure something out.
Fedora Core 1 forum posting with people who have already done it. It seems pretty easy from the looks of it. I'm going to do it just as soon as I get some free time...
And for the love of god, please read the whole thread. Don't ever install a kernel with rpm -Uvh. Leave yourself a backup (rpm -ivh).
As much as I am a mozilla supporter (I really love the element-animal products -- I've installed them on my parent's system), I don't think the idea of a google branded thunderbird would work well, at least until Ben finishes his smart update (due for firefox 1.0) and Scott encorporates that into Thunderbird. If google will be giving it out, it will need to be dummy proof, and we can't have X people playing around with Google Thunderbird 0.5 for the next few years.
It's not done yet, so you'll need to read some of the actual forum posts to figure out what you need to do, but they have an experemental version that I love.