Dynamic theme switching was disabled in bug 127784 before Mozilla 1.0. There's also a metabug keeping track of the bugs with dynamic theme switching (134260). Look at the bugs blocking 134260 and blocked by 127784 to get an idea of what kinds of problems Mozilla's dynamic theme switching had.
I can seamlessly switch between Japanese and Chinese input with windows.
I started using Japanese IME a few weeks ago. Alt+Shift for switching languages is convinient, even if it's not very responsive (the language bar doesn't update for about a second until after I press the shortcut). But is it possible to switch between Hiragana and Katakana with the keyboard? I haven't been able to figure out how. Also, do you know if it's possible to put the language bar on the taskbar without making it expanding the taskbar by a few pixels?
The main difference between that and normal Altavista is that it doesn't have an untargeted graphical banner ad. It has most of the same junk (refine your search, sponsored matches) taking up vertical space and pushing the first result past the bottom of my browser's viewport. In constrast, Google always shows at least one result without scrolling, and Alltheweb shows two or three with my custom Alltheweb style sheet. I browse in a maximized window at 800x600.
When I'm looking for information, I use many of the same methods to determine the credibility of the information as I would in real life:
Who wrote the information?
Would the author benefit in some way from misinforming me? (If I'm looking up the molecular weight of iron, I probably don't need to worry about intentional misinformation.)
Does the author use objective language where I think he should use subjective language?
When the author talks about things I already know about, is he correct and does he seem to be unbiased?
If it's a news article, do I trust the editors of the newspaper and am I aware of any bias that the newspaper has?
Is the information reasonable?
When I am considering purchasing from a site, I'm reluctant rely solely on its content and looks, because both can be stolen from other web sites. Instead, I do one or more of the following:
Check for a nice-looking domain name and URL.
Use the Google Toolbar to check the site's PageRank.
Search Google for the company's name to make sure the first result matches the URL I have.
PornzillaModifications - stealth profiles, image zoom, view (but don't download) all linked images, go to next/previous thumbnail gallery or image.
Leech - download all links from a page that have an extension in your list of extensions to download. The author didn't figure out how to send referrers with the requests, which is annoying because many porn sites require a correct referrer header, but there are several workarounds included with Leech.
Just type this URL into your address bar while viewing the disclaimer: javascript:for (i=0; t=document.getElementsByTagName ("textarea")[i]; ++i) void(t.readOnly=false)
Then you can delete all the text, change all the vowels to o's, or do whatever you do with annoying contracts in real life.
You can even keep this URL as a "bookmarklet" on your personal toolbar and click it whenever you need to edit a read-only textarea.
the AIM client will finally include a feature that lets you change how someone appears in your buddy list (e.g., "Jesse Ruderman" instead of "JesseRud")? I can't imagine AIM forcing users to deal with a buddy list full of 9-digit ICQ numbers. Other than automatic logging, this is the feature I'm hoping for the most in the official AIM client.
Unfortuneatly, it [the "linked images" bookmarklet] chokes on donkey balls on sites that check referrer headers before serving images.
Not anymore -- bbaetz, darin, and I fixed bug 123293 in August. If you find any specific sites or command sequences (such as "linked images" followed by View Image followed by Shift+Reload) that fail to send the referrer header in 1.2alpha or later, please file a bug and cc me.
Loading large tables and large quanity of images (thumbnails) are slower than IE. Download pre-buffering actually becomes a problem when you download large files, due to it downloading in your temp dir, then moving the file after its completed.
Is that bug 73757 or another dependency of 129923? It's not clear to me whether 73757 applies to all local files or only local files to be opened by helper apps.
1. Click hyperlinks. Can't beat the mouse for this, IMHO.
No, but you can get close. IE for Mac lets you type the first few characters of a link to focus the first matching link. Current Mozilla nightlies have a similar feature, but in Mozilla, you can type letters anywhere in the link text. Mozilla gives you more feedback as you use its version of the feature. Mozilla lets you use Ctrl+G to go to the next matching link. Mac IE's version currently works better for pages that use image links, as long as the images have alt text.
I use accesskeys (Alt+S, etc) and type-to-focus for links on my start page and type-to-focus for paths I follow frequently from my start page (classes -> bio -> schedule, mozillazine -> talkback, etc). I also use type-to-focus to search through a list of students and phone numbers where the names happen to be links. If I need to open a large number of links at once, I often use a bookmarklet : Linked Images for thumbnail links to images, Linked Pages for a list of links to non-image pages. I use the mouse for most other links.
2. Scroll. Since my hand is already on the mouse, the mouse wheel is perfect. Mouse wheels are pretty common these days.
I also use the mouse for scrolling often, even if my hand isn't on the mouse before I want to scroll. Why? It might have something to do with the way the down arrow key only goes one line in Mozilla and the pgdn key goes a little too far. Or maybe I miss "smooth scrolling" from IE. Or maybe it's because in a maximized Mozilla window, there is no space between the scrollbar and the edge of the screen, so it's easier to scroll using the mouse in Mozilla than in IE.
Slashdot disallows spidering of comments.pl and article.pl, but it allows spidering of static pages with URLs like http://slashdot.org/yro/00/08/30/1933205.shtml. These static pages have comments with score:1 or above in nested-but-not-indented format.
If you insert your page on your signature, or it appears on your user info every time you post a reply, does it count in favour of your PageRank(TM) in google?
Yes. My site started with a decent pagerank (4? 5?) because I set it as my Slashdot url as soon as I created it. My site now has a higher rank due to having "real links" in addition to links from each Slashdot comment I have ever posted under this username, but I think having my site high in Google rankings from the beginning helped me establish myself as an authority on my site's topic.
Slashdot makes it really easy to "cheat" with pagerank. Want to make it harder to cheat? Make sig and user urls go through a redirect from which robots like Googlebot are banned, or only list user urls on user info pages (like webmasterworld).
For starters, IE has been so much better than any other browser for so long that I've forgotten all the clicks
That's a joke, right? In IE, it takes five extra keypresses to follow a link using the keyboard: Ctrl+F, [some text in the link], Enter, Esc, Tab, Enter. In Mozilla, it used to take four (the Tab was not necessary), and currently takes one (Enter). IE has no shortcut for Save, Increase Text Size, View Source, or Go to www.google.com (Ctrl+Shift+F in Mozilla). As far as I know, IE does not have a mode in which the caret is visible in web pages (F7 in Mozilla).
I'm not sure that Ctrl+Shift+L is fantastically better than Ctrl+O for the open location menu.
Opening a given URL or filename. IE: Ctrl+O, Ctrl+L, or Alt+D. Mozilla: Ctrl+L, Ctrl+Shift+L, (Alt+D coming soon).
Opening a file using a file picker. IE: Ctrl+O, Alt+R. Mozilla: Ctrl+O.
IE does not have a huge advantage here. If you prefer entering URLs in a dialog rather than the address bar (why?), IE has a slight advantage. If you open local files often, Mozilla has a slight advantage.
last night i was trying to load up an old aim logfile (if you remember, aim actually had logging as a feature at one point in time)...So i loaded this aim html logfile (12 MEGS OW!!)
From your hard drive, right?
with mozilla and it liturally took for_ever to show it. Granted, it was a hefty logfile. So i fired up IE to view the logfile and it displayed it very quickly.
I think this is/was a paint starvation bug specific to the Windows platform. It's supposedly fixed (bug 129640) in Mozilla 1.2a. I haven't taken time to test the fix myself.
Networks without nick and channel registration tend to have inflated user counts due to people staying on all the time to guard their nicks and creating bots to guard channels. Which networks currently have nick and channel registration? Is it possible to get a count of users with idle time under 30 minutes, thus excluding most idle users and some bots?
Many (if not most) elections revolve around the platforms of the candidates, to the point where often the platforms are more important to the voters than the candidates themselves. I intend to avoid platforms altogether. In my opinion, political platforms are little more than pieces of driftwood politicians cling to when cast adrift in the Potomac, away from the voters. Through the internet I intend never to be away from the voters of our district long enough to need a platform. I can present issues that spring up to the voters directly instead of having to rely on a platform set in stone for two years.
How do you plan to deal with the volume of electronic input (including e-mail) you will recieve if elected, given that congressmen who do not specifically ask for electronic input already have trouble handling their e-mail? I understand that you plan to set up a Slashdot-like system for your Louisiana district with moderation, but moderation isn't perfect.
People can abuse moderation by only moderating up only posts whose conclusion they agree with, and I would expect this problem to be greater in an environment where important decisions may be made based on the comments.
Moderation is good for getting interesting ideas from multiple viewpoints (when it is not abused) and for finding the majority (when it is abused), but it is not ideal for reaching consensus, where the ideas backed by the best arguments and most trusted debaters win. How will you look at comments and decide how to vote on an issue?
In Congress, you may find yourself wanting to suggest compromises regarding bills and riders. How will you determine which issues are most important to your state and district?
Moderation does little to fix the simple problem that there are too many posts for everyone to read. Will you read all posts, or will you rely on moderation to filter up the best comments for you to read?
If it isn't a big enough security hole to warrant instant attention then it should not be hidden in bugzilla, so anyone can have a whack at fixing it.
The bug was public for two months before it was marked as security-sensitive. There isn't an army of coders who spend all of their time fixing known minor privacy bugs. The bug had the "privacy" keyword for almost two months before it was marked as security-sensitive, so it would not have been invisible to such an army.
I'm not saying it was a good idea to make it security-sensitive after it was open for a while. It wasn't a good idea in this case, because someone who had seen the bug while it was public decided to make it public again. I'm just saying that leaving it open probably would not have led someone to fix it immediately.
But why is it when its an IE bug, its a "Severe Security Exploit", and when its a Mozilla bug, its a "Privacy Leak"...
Umm, maybe because this bug isn't severe? It only lets a malicious site find out what URL you visit immediately after leaving the site. I'm much more concerned about IE's policy of allowing sites to read from and write to the clipboard than I am about this bug.
Slashdot Mirror
Dynamic theme switching was disabled in bug 127784 before Mozilla 1.0. There's also a metabug keeping track of the bugs with dynamic theme switching (134260). Look at the bugs blocking 134260 and blocked by 127784 to get an idea of what kinds of problems Mozilla's dynamic theme switching had.
Slashdot editors: in a few hours, please post an article entitled "University of Twente No Longer Slashdotted".
Thanks for the help.
I can seamlessly switch between Japanese and Chinese input with windows.
I started using Japanese IME a few weeks ago. Alt+Shift for switching languages is convinient, even if it's not very responsive (the language bar doesn't update for about a second until after I press the shortcut). But is it possible to switch between Hiragana and Katakana with the keyboard? I haven't been able to figure out how. Also, do you know if it's possible to put the language bar on the taskbar without making it expanding the taskbar by a few pixels?
No, you're not the only person who's tired of swoosh logos. The folks at 37 Signals who brought you eNormicom were tired of them back in Dec 2000.
The main difference between that and normal Altavista is that it doesn't have an untargeted graphical banner ad. It has most of the same junk (refine your search, sponsored matches) taking up vertical space and pushing the first result past the bottom of my browser's viewport. In constrast, Google always shows at least one result without scrolling, and Alltheweb shows two or three with my custom Alltheweb style sheet. I browse in a maximized window at 800x600.
- Who wrote the information?
- Would the author benefit in some way from misinforming me? (If I'm looking up the molecular weight of iron, I probably don't need to worry about intentional misinformation.)
- Does the author use objective language where I think he should use subjective language?
- When the author talks about things I already know about, is he correct and does he seem to be unbiased?
- If it's a news article, do I trust the editors of the newspaper and am I aware of any bias that the newspaper has?
- Is the information reasonable?
When I am considering purchasing from a site, I'm reluctant rely solely on its content and looks, because both can be stolen from other web sites. Instead, I do one or more of the following:That should have been "view but don't save", rather than "view but don't download". Sorry for the confusion.
Pornzilla Modifications - stealth profiles, image zoom, view (but don't download) all linked images, go to next/previous thumbnail gallery or image.
Leech - download all links from a page that have an extension in your list of extensions to download. The author didn't figure out how to send referrers with the requests, which is annoying because many porn sites require a correct referrer header, but there are several workarounds included with Leech.
Just type this URL into your address bar while viewing the disclaimer:
javascript:for (i=0; t=document.getElementsByTagName ("textarea")[i]; ++i) void(t.readOnly=false)
Then you can delete all the text, change all the vowels to o's, or do whatever you do with annoying contracts in real life.
You can even keep this URL as a "bookmarklet" on your personal toolbar and click it whenever you need to edit a read-only textarea.
the AIM client will finally include a feature that lets you change how someone appears in your buddy list (e.g., "Jesse Ruderman" instead of "JesseRud")? I can't imagine AIM forcing users to deal with a buddy list full of 9-digit ICQ numbers. Other than automatic logging, this is the feature I'm hoping for the most in the official AIM client.
In my nightly build the typeahead stuff caused the backspace key to stop behaving as a "backbutton-key" even when I've done no typing at all.
I think that's fixed in 1.2b (bug 172956).
Unfortuneatly, it [the "linked images" bookmarklet] chokes on donkey balls on sites that check referrer headers before serving images.
Not anymore -- bbaetz, darin, and I fixed bug 123293 in August. If you find any specific sites or command sequences (such as "linked images" followed by View Image followed by Shift+Reload) that fail to send the referrer header in 1.2alpha or later, please file a bug and cc me.
Loading large tables and large quanity of images (thumbnails) are slower than IE. Download pre-buffering actually becomes a problem when you download large files, due to it downloading in your temp dir, then moving the file after its completed.
Is that bug 73757 or another dependency of 129923? It's not clear to me whether 73757 applies to all local files or only local files to be opened by helper apps.
1. Click hyperlinks. Can't beat the mouse for this, IMHO.
No, but you can get close. IE for Mac lets you type the first few characters of a link to focus the first matching link. Current Mozilla nightlies have a similar feature, but in Mozilla, you can type letters anywhere in the link text. Mozilla gives you more feedback as you use its version of the feature. Mozilla lets you use Ctrl+G to go to the next matching link. Mac IE's version currently works better for pages that use image links, as long as the images have alt text.
I use accesskeys (Alt+S, etc) and type-to-focus for links on my start page and type-to-focus for paths I follow frequently from my start page (classes -> bio -> schedule, mozillazine -> talkback, etc). I also use type-to-focus to search through a list of students and phone numbers where the names happen to be links. If I need to open a large number of links at once, I often use a bookmarklet : Linked Images for thumbnail links to images, Linked Pages for a list of links to non-image pages. I use the mouse for most other links.
2. Scroll. Since my hand is already on the mouse, the mouse wheel is perfect. Mouse wheels are pretty common these days.
I also use the mouse for scrolling often, even if my hand isn't on the mouse before I want to scroll. Why? It might have something to do with the way the down arrow key only goes one line in Mozilla and the pgdn key goes a little too far. Or maybe I miss "smooth scrolling" from IE. Or maybe it's because in a maximized Mozilla window, there is no space between the scrollbar and the edge of the screen, so it's easier to scroll using the mouse in Mozilla than in IE.
Slashdot disallows spidering of comments.pl and article.pl, but it allows spidering of static pages with URLs like http://slashdot.org/yro/00/08/30/1933205.shtml. These static pages have comments with score:1 or above in nested-but-not-indented format.
If you insert your page on your signature, or it appears on your user info every time you post a reply, does it count in favour of your PageRank(TM) in google?
Yes. My site started with a decent pagerank (4? 5?) because I set it as my Slashdot url as soon as I created it. My site now has a higher rank due to having "real links" in addition to links from each Slashdot comment I have ever posted under this username, but I think having my site high in Google rankings from the beginning helped me establish myself as an authority on my site's topic.
Slashdot makes it really easy to "cheat" with pagerank. Want to make it harder to cheat? Make sig and user urls go through a redirect from which robots like Googlebot are banned, or only list user urls on user info pages (like webmasterworld).
For starters, IE has been so much better than any other browser for so long that I've forgotten all the clicks
That's a joke, right? In IE, it takes five extra keypresses to follow a link using the keyboard: Ctrl+F, [some text in the link], Enter, Esc, Tab, Enter. In Mozilla, it used to take four (the Tab was not necessary), and currently takes one (Enter). IE has no shortcut for Save, Increase Text Size, View Source, or Go to www.google.com (Ctrl+Shift+F in Mozilla). As far as I know, IE does not have a mode in which the caret is visible in web pages (F7 in Mozilla).
I'm not sure that Ctrl+Shift+L is fantastically better than Ctrl+O for the open location menu.
Opening a given URL or filename.
IE: Ctrl+O, Ctrl+L, or Alt+D.
Mozilla: Ctrl+L, Ctrl+Shift+L, (Alt+D coming soon).
Opening a file using a file picker.
IE: Ctrl+O, Alt+R.
Mozilla: Ctrl+O.
IE does not have a huge advantage here. If you prefer entering URLs in a dialog rather than the address bar (why?), IE has a slight advantage. If you open local files often, Mozilla has a slight advantage.
Now, if you'll excuse me, I'm going over to Bugzilla to complain about the lack of Ctrl-Tab and Ctrl-Shift-Tab for switching between tabs.
Already fixed in nightlies.
last night i was trying to load up an old aim logfile (if you remember, aim actually had logging as a feature at one point in time)...So i loaded this aim html logfile (12 MEGS OW!!)
From your hard drive, right?
with mozilla and it liturally took for_ever to show it. Granted, it was a hefty logfile. So i fired up IE to view the logfile and it displayed it very quickly.
I think this is/was a paint starvation bug specific to the Windows platform. It's supposedly fixed (bug 129640) in Mozilla 1.2a. I haven't taken time to test the fix myself.
Networks without nick and channel registration tend to have inflated user counts due to people staying on all the time to guard their nicks and creating bots to guard channels. Which networks currently have nick and channel registration? Is it possible to get a count of users with idle time under 30 minutes, thus excluding most idle users and some bots?
is when you misspell "l33t" as "l77t".
If it isn't a big enough security hole to warrant instant attention then it should not be hidden in bugzilla, so anyone can have a whack at fixing it.
The bug was public for two months before it was marked as security-sensitive. There isn't an army of coders who spend all of their time fixing known minor privacy bugs. The bug had the "privacy" keyword for almost two months before it was marked as security-sensitive, so it would not have been invisible to such an army.
I'm not saying it was a good idea to make it security-sensitive after it was open for a while. It wasn't a good idea in this case, because someone who had seen the bug while it was public decided to make it public again. I'm just saying that leaving it open probably would not have led someone to fix it immediately.
But why is it when its an IE bug, its a "Severe Security Exploit", and when its a Mozilla bug, its a "Privacy Leak"...
Umm, maybe because this bug isn't severe? It only lets a malicious site find out what URL you visit immediately after leaving the site. I'm much more concerned about IE's policy of allowing sites to read from and write to the clipboard than I am about this bug.