The most common argument against including adblock by default is code bloat / UI bloat, but I have another reason. If enough people start blocking ad URLs based on regexps, sites will start mangling all of their URLs to make ads difficult for software to spot, and that type of ad-blocking will become useless.
Note that this is not an issue with pop-up blocking. If the browser blocks pop-ups correctly, there is no way a site can pop up a window except in response to certain events (like clicking a link).
Firefox on windows is still a little buggy, paticularly with autodissconnect. So Version 1.0 would be nice ASAP.
What is autodisconnect and how is Firefox buggy with it? Do you have reason to believe that the bug you care about will magically become fixed when and only when the version number has risen to 1.0?
Stephen Horlander and Kevin Gerich, the creators of Winstripe, blogged about the improvements to Winstripe in 0.9.1. I'm impressed with their openness.
(It might be that only one of them wrote that blog post. I'm confused because it's attributed to Horlander but is on Gerich's blog.)
An exploit has been discovered that can cause you to open a web site in Internet Explorer, which can then lead to a system compromise. Disabling JavaScript in Firefox does not make you immune. The full text of the exploit follows.
<b>Sorry, this site requires Microsoft Internet Explorer.</b>
The workaround is to e-mail the site admin, telling them that they are promoting the insecure practice of using Internet Explorer for sites that claim to require it. Unfortunately, some sites require Internet Explorer even to contact the site admin, leading to a catch-22.
I don't know what the actual exploit text looks like. It has to contain "script" because that's the name of the tag. It only has to contain "javascript" in the type attribute if it wants to be valid HTML.
IE is crappy coded (it's closed-source, so there's no 'second opinion' on the code).
The number of "second opinions" on code has more to do with code review process than it does with whether the program is open-source. mozilla.org requires most new code to be reviewed by 2 people. I think that does more for the quality of the code than the wide availability of source code.
Of the 50 or so security holes I've found in Mozilla (see my resume for a link to the list), I only found 2 of them by looking at the source code. To be fair, other people have reported buffer overflows, overflowable integers used to determine the amount of memory to allocate, and other security holes they did find by looking at the code. Also, I chose to look for security holes in Mozilla rather than IE or Opera because Mozilla is an open-source project and I want it to succeed.
It was never "open new tabs in the background". It was "open links in the background", which was misleading. People thought it meant loading links before you click on them.
Maybe "open new tabs in the background" would be better than "select new tabs opened from links". I dunno. I don't really like either wording -- I don't think either wording is clear.
The best I can come up with is "Keep the current tab selected when opening a link in a new tab". It's long, but I think the increased clarity is worth it. Being terse doesn't help if the reader has to read the sentence three times to understand it.
The discussion that resulted in the current wording was in bug 246274, so if you file a new bug suggesting a new wording, add a comment to bug 246274 linking to your bug.
Pipelining is the same as persistent connections except that the browser doesn't wait for the response to the second request before sending the third request. Pipelining saves the server and the user a little bit of bandwidth and make the page load faster, so I don't see why a server admin would not like pipelining. I don't know where you got the idea that pipelining was "hogging" anything.
The Firefox developers would not accept most of the options and features from TBE, and I agree mostly with them. One option that I think should be added to Firefox itself is making new-window links in web pages open in new tabs instead.
2. Create a testcase that shows the problem and is as simple as possible. (The Flash can probably be left alone, but the HTML/CSS/JS need to be simplified.) Attach the testcase to the bug.
Creating a testcase makes it much more likely that a Mozilla developer will look at the bug, and it might also give you an idea about how to work around the bug. Or you might even find out that it's a bug in your code, although it doesn't sound like it is.
You're right that "thrust" should be "acceleration", but you're wrong about g. G is the gravitational constant; g is the acceleration due to gravity at the Earth's surface.
* It helps me decide which sites to trust. If a site "should be" PR7 (given the industry and my guess as to how many people use the site) but is only PR4, I will suspect that it's fraudulent.
* I want to see the Pagerank of my own pages (for SEO purposes), competitors' pages, and friends' pages.
If they had the flexibility of a 4 to 6 hour delivery window, several different orders could be rounded up and put onto the same delivery van for economies of scale during the delivery process.
That's what Vons (aka Safeway) does, except that Vons has 2-4 hour delivery windows, not 4-6 hour windows. You pay $10 for delivery with a 2-hour window or $8 for a delivery with a 4-hour window.
If there was a way to encourage the customers to get their orders in at least 12 hours in advance
Vons has a simple way to do this: it requires customers to put in their orders about half a day in advance.
Programs like Opera prove that you can't stuff 1000 features into a program without creating a total mess.
No, programs like Opera suggest that it is difficult to stuff 1000 features into a program without creating a total mess. Opera could be cleaned up a lot without removing features. For example:
* the left panels could be "off" by default. * the left panels could be much easier to disable in the menus. * things that aren't preferences, such as the fixed built-in list of searches, could be moved from the "preferences" dialog to the help feature. * the wording of some preferences could be made more clear. * don't pop up zillions of dialogs the first time I use Opera, the first time I middle-click, the first time I move the mouse more than a pixel while right-clicking, etc.
There's a limit to how clean you can make an interface when you have too many features and options, but Opera doesn't come close to hitting that limit. My intuition is that a browser could have all of the features and options Opera has without being a total mess.
Slashdot Mirror
The most common argument against including adblock by default is code bloat / UI bloat, but I have another reason. If enough people start blocking ad URLs based on regexps, sites will start mangling all of their URLs to make ads difficult for software to spot, and that type of ad-blocking will become useless.
Note that this is not an issue with pop-up blocking. If the browser blocks pop-ups correctly, there is no way a site can pop up a window except in response to certain events (like clicking a link).
Firefox on windows is still a little buggy, paticularly with autodissconnect. So Version 1.0 would be nice ASAP.
What is autodisconnect and how is Firefox buggy with it? Do you have reason to believe that the bug you care about will magically become fixed when and only when the version number has risen to 1.0?
Firefox 0.9 includes fixes for at least two security holes. Please do not continue using 0.8.
There's also a version of the logo with the fox replaced by an anime fox-girl.
Stephen Horlander and Kevin Gerich, the creators of Winstripe, blogged about the improvements to Winstripe in 0.9.1. I'm impressed with their openness.
(It might be that only one of them wrote that blog post. I'm confused because it's attributed to Horlander but is on Gerich's blog.)
I don't know what the actual exploit text looks like. It has to contain "script" because that's the name of the tag. It only has to contain "javascript" in the type attribute if it wants to be valid HTML.
IE is crappy coded (it's closed-source, so there's no 'second opinion' on the code).
The number of "second opinions" on code has more to do with code review process than it does with whether the program is open-source. mozilla.org requires most new code to be reviewed by 2 people. I think that does more for the quality of the code than the wide availability of source code.
Of the 50 or so security holes I've found in Mozilla (see my resume for a link to the list), I only found 2 of them by looking at the source code. To be fair, other people have reported buffer overflows, overflowable integers used to determine the amount of memory to allocate, and other security holes they did find by looking at the code. Also, I chose to look for security holes in Mozilla rather than IE or Opera because Mozilla is an open-source project and I want it to succeed.
It's possible for a crash to be Firefox's fault even if it only happens with certain operating systems or with certain extensions installed.
Why are you searching for "javascript" rather than "script"?
Hopefully an IE patch will come out soon, although we all know how many users actually go to Windows Update...
I don't visit Windows Update every day either. I rely on update notification.
It's good that the site has such a short, easy-to-remember URL.
Warez group announces August 1, 2004 release date for cracked version of Doom III.
Because some popular servers, such as Apache 1.x, break when pipelining is used or under certain circumstances when pipelining is used.
It was never "open new tabs in the background". It was "open links in the background", which was misleading. People thought it meant loading links before you click on them.
Maybe "open new tabs in the background" would be better than "select new tabs opened from links". I dunno. I don't really like either wording -- I don't think either wording is clear.
The best I can come up with is "Keep the current tab selected when opening a link in a new tab". It's long, but I think the increased clarity is worth it. Being terse doesn't help if the reader has to read the sentence three times to understand it.
The discussion that resulted in the current wording was in bug 246274, so if you file a new bug suggesting a new wording, add a comment to bug 246274 linking to your bug.
Pipelining is the same as persistent connections except that the browser doesn't wait for the response to the second request before sending the third request. Pipelining saves the server and the user a little bit of bandwidth and make the page load faster, so I don't see why a server admin would not like pipelining. I don't know where you got the idea that pipelining was "hogging" anything.
The Firefox developers would not accept most of the options and features from TBE, and I agree mostly with them. One option that I think should be added to Firefox itself is making new-window links in web pages open in new tabs instead.
Two things you can do to help:
1. File a bug in bugzilla.mozilla.org.
2. Create a testcase that shows the problem and is as simple as possible. (The Flash can probably be left alone, but the HTML/CSS/JS need to be simplified.) Attach the testcase to the bug.
Creating a testcase makes it much more likely that a Mozilla developer will look at the bug, and it might also give you an idea about how to work around the bug. Or you might even find out that it's a bug in your code, although it doesn't sound like it is.
If a user can't figure out how to get the data from Firefox 0.8 into Firefox 0.9, then yes, Firefox 0.9's behavior qualifies as "destructive".
/me considers changing his sig to "This text is included to defeat the lameness filter, demonstrating that the lameness filter cannot pass itself."
You're right that "thrust" should be "acceleration", but you're wrong about g. G is the gravitational constant; g is the acceleration due to gravity at the Earth's surface.
IIRC, the "10% / 30 second rule" applies to samples (e.g. in an online store) but not to all possible uses.
By the way, what's 10% of a song that has repitition?
I want to see Pagerank for two reasons:
* It helps me decide which sites to trust. If a site "should be" PR7 (given the industry and my guess as to how many people use the site) but is only PR4, I will suspect that it's fraudulent.
* I want to see the Pagerank of my own pages (for SEO purposes), competitors' pages, and friends' pages.
If they had the flexibility of a 4 to 6 hour delivery window, several different orders could be rounded up and put onto the same delivery van for economies of scale during the delivery process.
That's what Vons (aka Safeway) does, except that Vons has 2-4 hour delivery windows, not 4-6 hour windows. You pay $10 for delivery with a 2-hour window or $8 for a delivery with a 4-hour window.
If there was a way to encourage the customers to get their orders in at least 12 hours in advance
Vons has a simple way to do this: it requires customers to put in their orders about half a day in advance.
I ordered from vons.com throughout college.
Programs like Opera prove that you can't stuff 1000 features into a program without creating a total mess.
No, programs like Opera suggest that it is difficult to stuff 1000 features into a program without creating a total mess. Opera could be cleaned up a lot without removing features. For example:
* the left panels could be "off" by default.
* the left panels could be much easier to disable in the menus.
* things that aren't preferences, such as the fixed built-in list of searches, could be moved from the "preferences" dialog to the help feature.
* the wording of some preferences could be made more clear.
* don't pop up zillions of dialogs the first time I use Opera, the first time I middle-click, the first time I move the mouse more than a pixel while right-clicking, etc.
There's a limit to how clean you can make an interface when you have too many features and options, but Opera doesn't come close to hitting that limit. My intuition is that a browser could have all of the features and options Opera has without being a total mess.