I have a Samsung (YP-U1, usb stick), it's just a USB mass storage device. (I have not bothered to install their software and still have no clue what it looks like.) Maybe you do mean SanDisk. Or maybe it's only for some models, who knows.
Thanks:) And, yes, rofl. I think the best part is that it's in Simplified Chinese.
Fun, it's CSS writing-layout: tb-rl. Except that doesn't seem to exist in the newest CSS3 text draft (it used to exist in previous drafts). Sigh; would have been cool. (I don't blame MS, just the slow progression that is CSS3)
Does any actual sites with vertical Asian text exist? (BTW, it's top-down, right-to-left. Top-down left-right is weird.) Example, please! Mozilla does lack ruby support, but that's usually used more for Japanese.
No, that's a separate bug (dealing with how QuickTime decides to completely ignore the APIs, instead opting for looking up the executable and manually making up a command line). This one is QuickTime failing to handle a RTSP stream and instead crashing.
Unfortunately, for this particular exploit that would have no bearing. The whole point was that Firefox couldn't use plugins for links to unknown protocols (in this case, rtsp://) and therefore launches the system default protocol handler (though I do recall there was a warning with a "don't tell me again" checkbox)
As far as the Firefox side was concerned, there was no plugin. It's the standalone app that's being exploited.
It appears that they're clicking on an rtsp:// link. Since Firefox uses NPAPI, which as far as I can tell is too dumb to handle protocols (it's all content-type/file extension based), Firefox gives up and asks the OS to handle it. Which invokes separate processes.
Had the link been a application/x-rtsp (*.rtsp,*.rts) [1] file, it would be in a plugin (and therefore in the same process). Not that it would necessarily help... Firefox 2 and earlier are built with MSVC6, and it looks like IE was safe because it was VC8 (with more security checks). You'd want Firefox 3 betas for what little protection that might provide.
For an equivalent with the WMP plugin, mms:// vs *.asx.
[1] I just found some random thing in about:plugins that my copy of the quicktime plugin claims to support.
Firefox supports python, but only with a special build of Firefox (so as to not have a dependency on starting Firefox in normal builds, I think), as well as only in extensions/the app itself (not in any web pages). Python's security model doesn't quite jive with the old same-origin stuff web pages use, AIUI.
See: ActiveState Komodo.
As for pluggable languages... Are we talking about IE now?:) IE does do that, but you need local installers. JScript and VBScript are done to interface that way, and ActiveState (yes, them again) has perl bindings.
Actually, ES4 does do JSON (here, or whatever spec you can actually find). Without evaluating things, because eval() sucks. See also overview PDF, top of page 39.
(Yes, I find it horribly difficult to figure out what they actually agree on, which makes trying to figure things our really hard)
Mozilla currently already does (1.7 had new keywords like "let" and "yield"). If you leave the version off, it defaults to a mishmash where no new keywords are introduced (i.e. those things don't work, and you get compatibility), and only things that were previously syntax errors (such as destructing assignment - [a,b] = [1,2];) get upgraded.
So, yes, they're already doing this in practice - if the dialect isn't explicitly noted, it's "whatever won't break old scripts".
(note: JavaScript is Netscape/Mozilla's version - the standard is ECMAScript, and MS's implementation is JScript. And Adobe's version is ActionScript. Hence they can all do their own numbering...)
Okay, let's assume the almost-done is not caused by malicious junk torrents, and instead just that nobody has the last part. Presumably then, it's because everybody who did have it left.
So, if there was 1% overhead that went into parity... those people would just leave 1% sooner (since they can regenerate the files they need anyway). So everybody would be stuck at 98% and still unable to use the parity. That won't be helpful.
Parity is useful in newsgroups because your servers won't randomly run away from you when they're done, it's just lossy in a random fashion. For bittorrent, you're reliant on whoever's seeding to stick around.
On both Windows and Linux, it would probably make more sense to install your development Firefox into somewhere your user can write to (C:\users\foo //home/foo). It pays to have a stable version sitting around when the latest nightly doesn't start:)
You can run command prompt as admin (right click on it, it's an option); probably explorer as well (I've been doing this a lot in XP, so I assume it's the same). Just like you can su in Linux and keep a privileged shell open. No tabs though:p
How do you run, say, process explorer in non-admin? IIRC I get an "Allow" (and run as admin) and a "Cancel" (and quits the app) - even though all I want to do is mess with my own processes. This goes for some of the other apps too, actually, by procexp triggers this for me the most.
To me, the worst part of UAC was the blink as the screen switched. Hardly terrible.
(I did have a bunch of weird file permissions problems locally though - directories set as creator/owner not having full control, junctions from Application Data to AppData having explicit deny for some odd reason... But for all I know it could have been one of the apps I installed that messed it up)
This would be why CDIC exists (in Canada). I recall there being an US equivalent too; never dealt with other banks so I have no idea if that applies.
But basically - yes, banks don't have enough money for everybody to withdraw. If that happens, the bank basically has to fold. At which point the insurance kicks in. The thing is - the bank has to cease business when this happens (i.e. repercussions that make them want to avoid this happening). ISPs? They complain and charge more.
Current versions of Mozilla Thunderbird / Seamonkey still use mbox, yes. With *.msf index files (which can be blown away - you might end up losing some read state on newsgroup messages though).
The SSH handshake does not look like the SSL/TLS handshake (for one thing, SSH needs user/pass whereas most uses of SSL don't use client certs). So yes, protocol.
So is the same as <ß/>?
Yay for capitalization rules. Or are we going for the SQLite style "case insensitive for 7-bit ASCII but case sensitive elsewhere"?:)
There's other commercial uses of open source too - for example, Adobe uses Spidermonkey (the Mozilla JavaScript engine) in Acrobat Reader... How many machines have that?:)
Umm, the title is buggy; the summary says it passes the CSS3 selectors test, not all of CSS3. It's impossible to support all of CSS3, there's chunks not even started - e.g. scoping. TFA does mention text-shadow will work with Opera 9.5 though.
With Mozilla, you might want to look at good first bugs for things that might be easier for someone just starting. As the sibling says though, you want to build first (which can be a challenge in itself in the Mozilla case, sadly).
Mozilla-specific again - you might want to consider starting by writing an extension. Helps you learn the same things, but no build system hurdle and no need to wait for somebody for reviews before end-users can see it.
Slashdot Mirror
That is by design. Those two things are fixed and not scrollable.
I have a Samsung (YP-U1, usb stick), it's just a USB mass storage device. (I have not bothered to install their software and still have no clue what it looks like.) Maybe you do mean SanDisk. Or maybe it's only for some models, who knows.
Thanks :)
And, yes, rofl. I think the best part is that it's in Simplified Chinese.
Fun, it's CSS writing-layout: tb-rl. Except that doesn't seem to exist in the newest CSS3 text draft (it used to exist in previous drafts). Sigh; would have been cool. (I don't blame MS, just the slow progression that is CSS3)
Does any actual sites with vertical Asian text exist? (BTW, it's top-down, right-to-left. Top-down left-right is weird.) Example, please!
Mozilla does lack ruby support, but that's usually used more for Japanese.
No, that's a separate bug (dealing with how QuickTime decides to completely ignore the APIs, instead opting for looking up the executable and manually making up a command line). This one is QuickTime failing to handle a RTSP stream and instead crashing.
Really? it doesn't seem to be listed. Got a bugzilla bug #?
Unfortunately, for this particular exploit that would have no bearing. The whole point was that Firefox couldn't use plugins for links to unknown protocols (in this case, rtsp://) and therefore launches the system default protocol handler (though I do recall there was a warning with a "don't tell me again" checkbox)
As far as the Firefox side was concerned, there was no plugin. It's the standalone app that's being exploited.
It appears that they're clicking on an rtsp:// link. Since Firefox uses NPAPI, which as far as I can tell is too dumb to handle protocols (it's all content-type/file extension based), Firefox gives up and asks the OS to handle it. Which invokes separate processes.
Had the link been a application/x-rtsp (*.rtsp,*.rts) [1] file, it would be in a plugin (and therefore in the same process). Not that it would necessarily help... Firefox 2 and earlier are built with MSVC6, and it looks like IE was safe because it was VC8 (with more security checks). You'd want Firefox 3 betas for what little protection that might provide.
For an equivalent with the WMP plugin, mms:// vs *.asx.
[1] I just found some random thing in about:plugins that my copy of the quicktime plugin claims to support.
Firefox supports python, but only with a special build of Firefox (so as to not have a dependency on starting Firefox in normal builds, I think), as well as only in extensions/the app itself (not in any web pages). Python's security model doesn't quite jive with the old same-origin stuff web pages use, AIUI.
:) IE does do that, but you need local installers. JScript and VBScript are done to interface that way, and ActiveState (yes, them again) has perl bindings.
See: ActiveState Komodo.
As for pluggable languages... Are we talking about IE now?
Actually, ES4 does do JSON (here, or whatever spec you can actually find). Without evaluating things, because eval() sucks. See also overview PDF, top of page 39.
(Yes, I find it horribly difficult to figure out what they actually agree on, which makes trying to figure things our really hard)
Mozilla currently already does (1.7 had new keywords like "let" and "yield"). If you leave the version off, it defaults to a mishmash where no new keywords are introduced (i.e. those things don't work, and you get compatibility), and only things that were previously syntax errors (such as destructing assignment - [a,b] = [1,2];) get upgraded.
So, yes, they're already doing this in practice - if the dialect isn't explicitly noted, it's "whatever won't break old scripts".
(note: JavaScript is Netscape/Mozilla's version - the standard is ECMAScript, and MS's implementation is JScript. And Adobe's version is ActionScript. Hence they can all do their own numbering...)
Okay, let's assume the almost-done is not caused by malicious junk torrents, and instead just that nobody has the last part. Presumably then, it's because everybody who did have it left.
So, if there was 1% overhead that went into parity... those people would just leave 1% sooner (since they can regenerate the files they need anyway). So everybody would be stuck at 98% and still unable to use the parity. That won't be helpful.
Parity is useful in newsgroups because your servers won't randomly run away from you when they're done, it's just lossy in a random fashion. For bittorrent, you're reliant on whoever's seeding to stick around.
On both Windows and Linux, it would probably make more sense to install your development Firefox into somewhere your user can write to (C:\users\foo / /home/foo). It pays to have a stable version sitting around when the latest nightly doesn't start :)
:p
You can run command prompt as admin (right click on it, it's an option); probably explorer as well (I've been doing this a lot in XP, so I assume it's the same). Just like you can su in Linux and keep a privileged shell open. No tabs though
How do you run, say, process explorer in non-admin? IIRC I get an "Allow" (and run as admin) and a "Cancel" (and quits the app) - even though all I want to do is mess with my own processes. This goes for some of the other apps too, actually, by procexp triggers this for me the most.
To me, the worst part of UAC was the blink as the screen switched. Hardly terrible.
(I did have a bunch of weird file permissions problems locally though - directories set as creator/owner not having full control, junctions from Application Data to AppData having explicit deny for some odd reason... But for all I know it could have been one of the apps I installed that messed it up)
There's also the kdawson update to a blog post... dated 2006/10/05. It's only ten and a half months ago, it's still new, right? :)
This would be why CDIC exists (in Canada). I recall there being an US equivalent too; never dealt with other banks so I have no idea if that applies.
But basically - yes, banks don't have enough money for everybody to withdraw. If that happens, the bank basically has to fold. At which point the insurance kicks in. The thing is - the bank has to cease business when this happens (i.e. repercussions that make them want to avoid this happening). ISPs? They complain and charge more.
Current versions of Mozilla Thunderbird / Seamonkey still use mbox, yes. With *.msf index files (which can be blown away - you might end up losing some read state on newsgroup messages though).
The SSH handshake does not look like the SSL/TLS handshake (for one thing, SSH needs user/pass whereas most uses of SSL don't use client certs). So yes, protocol.
I believe you're looking for `openssl s_client`. SSH is a... rather different protocol.
So is the same as <ß/>? Yay for capitalization rules. Or are we going for the SQLite style "case insensitive for 7-bit ASCII but case sensitive elsewhere"? :)
I think you understood that, but I figured it was better to point it out.
There's other commercial uses of open source too - for example, Adobe uses Spidermonkey (the Mozilla JavaScript engine) in Acrobat Reader... How many machines have that? :)
Umm, the title is buggy; the summary says it passes the CSS3 selectors test, not all of CSS3. It's impossible to support all of CSS3, there's chunks not even started - e.g. scoping. TFA does mention text-shadow will work with Opera 9.5 though.
With Mozilla, you might want to look at good first bugs for things that might be easier for someone just starting. As the sibling says though, you want to build first (which can be a challenge in itself in the Mozilla case, sadly).
Mozilla-specific again - you might want to consider starting by writing an extension. Helps you learn the same things, but no build system hurdle and no need to wait for somebody for reviews before end-users can see it.
New Safari user here too. Yay win32.
The little book icon on the left end of the bookmarks toolbar, or Bookmarks -> Show All Bookmarks.
Firefox has no importer from Firefox. Firefox has importers for all of those things listed.
:)
Netscape 9 is heavily based on Firefox. Hence it inherits all the bookmarks importers...