Actually, on XP Pro at least, because of the 2 (or was it 1? Never tested...) user restriction, it seems to always resume the active session. As in, I log in, do stuff, and connect remotely - I get the same stuff back, it's not a new session. Go to a different machine and resume from there - it's still the same screen. (It just locks the local user out.) This was all using the same user account though.
Also, I'm not sure if VNC would be technically possible to allow more than one user (as in, account) to be logged in. VNC does have the advantage of being able to share the session though (without using RDP's remote assistance thing).
I wouldn't trust MS's security either - that's what cygwin + openssh is for;)
Yes, it's a known bug in Firefox, will be fixed post-1.0 (because the fix required is rather big can have significant regressions) since it's not in the 1.0 tree (aviary).
Bugzilla #217527 - don't comment please, especially "me too" / advocacy comments. See my comment history (cid 10414776) for stuff on this.
I thought that Safari (WebCore) ripped out the KPart stuff (since they're not using Qt anyway) and just kept the rendering logic?
(Please note: I don't keep track of KDE / Apple stuff, so may be wrong. I have not looked at embedding WebCore into anything at all - or KHTML for that matter.)
For Mozilla: about:config - find dom.event.contextmenu.enabled - set to false
No, I have no idea why the article summary says it blocks the context menu in Mozilla - it's even in Firefox's GUI*... ({Edit -> Prefrences | Tools -> Options} -> Web Features -> (Enable Javascript) Advanced... -> Disable or Replace Context Menus). Not that it does much good in this case, since it's actually a background image of a transparent picture.
Sorry, you have to use the Page Info dialog box (Tools -> Page Info), or the DOM Inpsector in this case.
* I'm bringing up Firefox because that's supposed to have the stripped-down UI.
Dictionary.com also has it; see the first entry, it's credited to Webster.
It's not the machines sense though; appearently something about boxwood.
Re:Still can't open message in a new window
on
Gmail Adds Features
·
· Score: 1
I know about it, but am waiting for the patch in bug 172962 as mentioned in gp. (Essentially, it does it integrated into the binary bits of the browser, becoming part of the internal features.) They're still working on it I guess. I should really get a nightly post-2004-10-04 to test what they have.
I hope it's not a sign of feature creep... What with live boomarks, this, style switcher, etc. (I like the last two and use Tbird instead for feeds, but still...)
Would it be possible to get signed installs? (For the patches anyway, if not releases)
Even in the case where the user must be prompted, making sure the patches are signed would be very helpful. Or are they already signed? (It didn't seem to be, but I can't really tell)
It looks like that standard disclaimer to make sure the Mozilla Foundation doesn't get sued by the government - I believe that IE also had the disclaimer (havn't checked in a while though). MoFo does have their servers in the States.
I assume a version without NSS (the HTTPS &c stuff) would be legal, and it's probably possible to obtain the code from intermediary countries anyway.
This is known and fixed in the trunk (so it's in Mozilla 1.8a4 for example); it will be fixed in Firefox post-1.0.
This exposed a problem with boxes though, and thus was not ported back to the Firefox (aviary) branch; if that gets fixed soon, both problems will likely be fixed for Firefox 1.0 (assuming the developers agree).
See bug 217527 on bugzilla.mozilla.org for some discussion; please avoid adding more comments to it unless it's actually useful (which is unlikely). Not linking on purpose because slashdotting a bug database isn't cool.
Dynamic theme switching was considered too buggy for 1.0.
(And now, the part that is not a dupe)
Set extensions.dss.enabled to true in about:config to use what they have so far. Some things might not work completely, but people seem to believe that it works mostly well enough anyway.
Don't change app.extensions.version as that might render some extensions obsolete (when you try installing them later). For reference, the XPI only makes these changes:
I'd be surprised if the exploit doesn't affect Linux - considering that the changes made in the patch are all in Javascript. (That's also the reason it can be patched like this, and that it was applied so quickly.)
But I can't actually tell, of course - I can't see the security-sensitive bug:) So I can't actually check if the actual Mozilla people believe it's cross-platform.
(Oh, and... Where do you get that V1.0? Considering it's not out yet... Nope, 1.0PR isn't 1.0 at all; it's 0.10...)
Sounds like what WinNY was supposed to be. See/. article where the creator was arrested (appearently the BBS that came with it wasn't private, to allow edits).
The speeds are asymmetric because the ISPs have found that people just don't care. They advertise downstream all over the place, but the common person (i.e., one who does not get the service with the specific intent to share) wouldn't even notice the upload cap - for things like browsing, it's just not a problem.
This means, of course, that they get to charge much more for high uploads... They win either way. Ever notice how the DSL/Cable TOS always specify that you're not supposed to be hosting a server?
You don't need to actually drop HTML, you just need something much more restricted. Say, RTF. (Or, for that matter, Tex...)
I imagine this is what Mozilla's Simple HTML mode does (and I assume other sane clients would have something similar). If you restrict the HTML inside mail to a very limited subset that can't do anything (and that includes no plugins!), it should be fine. Until you hit one of them image buffer overflow vulnerabilities...:)
If the bitmaps are local anyway, wouldn't it be easier to just change the bitmaps? Make them, for example, a a unique (per card) coloured pixel at some predetermined position (since you know where all the cards end up anyway) and you don't even have to worry about comparing pictures - just look at what colour the pixel is and derive the card from that. Maybe even code the suit and the value (number? whatever you call it) separately into different bytes...
> Problem is alot of the making-mozilla-pretty people are concentrating on firefox/thunderbird instead.
May partly because skinning all of Mozilla would take slightly more work (since there are so many components to skin) and thus requires more initial effort.
That would be bug 231062. Currently stuck, and I assume help would be welcome. Looks like they want to use WiX for it but can't find people who would be able to write up the needed files.
(That link to mozilla isn't gonna work; drag it to the tab bar please.)
Slashdot Mirror
Actually, on XP Pro at least, because of the 2 (or was it 1? Never tested...) user restriction, it seems to always resume the active session. As in, I log in, do stuff, and connect remotely - I get the same stuff back, it's not a new session. Go to a different machine and resume from there - it's still the same screen. (It just locks the local user out.) This was all using the same user account though.
;)
Also, I'm not sure if VNC would be technically possible to allow more than one user (as in, account) to be logged in. VNC does have the advantage of being able to share the session though (without using RDP's remote assistance thing).
I wouldn't trust MS's security either - that's what cygwin + openssh is for
You mean 98/ME. (No, 2k doesn't work - tried it before. Which sucks, because it means it will never capture keys like alt+tab)
This would mean kudos for the guys that provided it, I guess? Encryption can indeed be done via SSH, but then you'd have SCP for file transfers too
(I'm currently on cygwin + openssh + RDP on XP, so I have no idea how well TightVNC is right now.)
Yes, it's a known bug in Firefox, will be fixed post-1.0 (because the fix required is rather big can have significant regressions) since it's not in the 1.0 tree (aviary).
Bugzilla #217527 - don't comment please, especially "me too" / advocacy comments. See my comment history (cid 10414776) for stuff on this.
Depending on his setup, .arpa could mean within the network. (IIRC, stuff like 192.168.* gets resolved as *.in-addr.arpa)
:)
The OS/2 stuff is still cool though
I thought that Safari (WebCore) ripped out the KPart stuff (since they're not using Qt anyway) and just kept the rendering logic?
(Please note: I don't keep track of KDE / Apple stuff, so may be wrong. I have not looked at embedding WebCore into anything at all - or KHTML for that matter.)
For Mozilla:
about:config - find dom.event.contextmenu.enabled - set to false
No, I have no idea why the article summary says it blocks the context menu in Mozilla - it's even in Firefox's GUI*... ({Edit -> Prefrences | Tools -> Options} -> Web Features -> (Enable Javascript) Advanced... -> Disable or Replace Context Menus). Not that it does much good in this case, since it's actually a background image of a transparent picture.
Sorry, you have to use the Page Info dialog box (Tools -> Page Info), or the DOM Inpsector in this case.
* I'm bringing up Firefox because that's supposed to have the stripped-down UI.
boxen
Dictionary.com also has it; see the first entry, it's credited to Webster.
It's not the machines sense though; appearently something about boxwood.
I know about it, but am waiting for the patch in bug 172962 as mentioned in gp. (Essentially, it does it integrated into the binary bits of the browser, becoming part of the internal features.) They're still working on it I guess. I should really get a nightly post-2004-10-04 to test what they have.
I hope it's not a sign of feature creep... What with live boomarks, this, style switcher, etc. (I like the last two and use Tbird instead for feeds, but still...)
You may want to check if you have a corrupted autocomplete database in Firefox.
~/.firefox//formhistory.dat I believe. Killing it should kill all of your form histories.
And now I just want to open them in new tabs :p
Will see if the patches they're doing on bug 172962 for Firefox would work around it. That would apply to every site though...
I guess this would be where Opera with its MDI view has an advantage.
Would it be possible to get signed installs? (For the patches anyway, if not releases)
Even in the case where the user must be prompted, making sure the patches are signed would be very helpful. Or are they already signed? (It didn't seem to be, but I can't really tell)
It looks like that standard disclaimer to make sure the Mozilla Foundation doesn't get sued by the government - I believe that IE also had the disclaimer (havn't checked in a while though). MoFo does have their servers in the States.
I assume a version without NSS (the HTTPS &c stuff) would be legal, and it's probably possible to obtain the code from intermediary countries anyway.
This is known and fixed in the trunk (so it's in Mozilla 1.8a4 for example); it will be fixed in Firefox post-1.0.
This exposed a problem with boxes though, and thus was not ported back to the Firefox (aviary) branch; if that gets fixed soon, both problems will likely be fixed for Firefox 1.0 (assuming the developers agree).
See bug 217527 on bugzilla.mozilla.org for some discussion; please avoid adding more comments to it unless it's actually useful (which is unlikely). Not linking on purpose because slashdotting a bug database isn't cool.
Dynamic theme switching was considered too buggy for 1.0.
(And now, the part that is not a dupe)
Set extensions.dss.enabled to true in about:config to use what they have so far. Some things might not work completely, but people seem to believe that it works mostly well enough anyway.
Don't change app.extensions.version as that might render some extensions obsolete (when you try installing them later).
For reference, the XPI only makes these changes:
pref("app.version", "0.10.1");
pref("general.useragent.vendorSub", "0.10.1");
I'd be surprised if the exploit doesn't affect Linux - considering that the changes made in the patch are all in Javascript. (That's also the reason it can be patched like this, and that it was applied so quickly.)
:) So I can't actually check if the actual Mozilla people believe it's cross-platform.
But I can't actually tell, of course - I can't see the security-sensitive bug
(Oh, and... Where do you get that V1.0? Considering it's not out yet... Nope, 1.0PR isn't 1.0 at all; it's 0.10...)
Looking on bugzilla.mozilla.org found bug 2920; from there got a link to this extension. Havn't tried it, but sounds like it should work.
Might be something you want to look at.
Sounds like what WinNY was supposed to be. See /. article where the creator was arrested (appearently the BBS that came with it wasn't private, to allow edits).
The speeds are asymmetric because the ISPs have found that people just don't care. They advertise downstream all over the place, but the common person (i.e., one who does not get the service with the specific intent to share) wouldn't even notice the upload cap - for things like browsing, it's just not a problem.
This means, of course, that they get to charge much more for high uploads... They win either way. Ever notice how the DSL/Cable TOS always specify that you're not supposed to be hosting a server?
You don't need to actually drop HTML, you just need something much more restricted. Say, RTF. (Or, for that matter, Tex...)
:)
I imagine this is what Mozilla's Simple HTML mode does (and I assume other sane clients would have something similar). If you restrict the HTML inside mail to a very limited subset that can't do anything (and that includes no plugins!), it should be fine. Until you hit one of them image buffer overflow vulnerabilities...
If the bitmaps are local anyway, wouldn't it be easier to just change the bitmaps? Make them, for example, a a unique (per card) coloured pixel at some predetermined position (since you know where all the cards end up anyway) and you don't even have to worry about comparing pictures - just look at what colour the pixel is and derive the card from that. Maybe even code the suit and the value (number? whatever you call it) separately into different bytes...
Works for me.
That did happen to me though when bugwramgler (an extension) was installed for some reason.
Mozilla/5.0 (Windows; U; Windows NT 5.1; rv:1.7.3) Gecko/20040915 Firefox/0.10
> Problem is alot of the making-mozilla-pretty people are concentrating on firefox/thunderbird instead.
May partly because skinning all of Mozilla would take slightly more work (since there are so many components to skin) and thus requires more initial effort.
That would be bug 231062. Currently stuck, and I assume help would be welcome. Looks like they want to use WiX for it but can't find people who would be able to write up the needed files.
(That link to mozilla isn't gonna work; drag it to the tab bar please.)