I'm a karma ho but if you want to have a chat to the IE Program Manager, do it here.
Choice quotes include: "Not much defense on the PNG transparency issue other than GDI made it hard back in the day. We really want this fixed too and are working on it. No eta yet. I'll try to update folks as we work through this. Same with other decisions."
... and "Anyway, we've started the get the band back together as it were. So far, the new IE team really has been focused on security and taking care of key corporate customer issues. A big part of the security effort has been our push around XPSP2. This has been all-consuming to the point where we've essentially stopped our Longhorn work for now.".
They have also set up a Wiki for gathering ideas for bugfixes/improvements to IE. Not like they are lacking ideas already though...
No, TLS is short for Thread Local Storage, and in most UNIX-like operating systems the dynamic linker *is* a feature of the linker. The glibc symvers implementation is more advanced than any of the BSDs/Solaris last time I checked. Anyway, those were only some examples, I suggest you go read the glibc sources if you would like to find some more.
Yeah, I have to agree. There are a ton of people here dicking about with semantics and irrelevant details of the technology (but they BEAMED it to me!) and not considering the implications of their actions which is ultimately what the law is based on.
The whole reason they do signal encryption is because broadcasting is a technically efficient way to get video content to people BUT economically is hard to make work unless (like the BBC) society assumes everybody with a TV will want to watch it. Signal encryption is a good compromise between technical efficiency and economic reality. The people who feel they are owned something for nothing have to consider their actions in the following context: what would happen if everybody did this?
It then becomes pretty obvious that these people are not acting upon some superior moral reasoning, they are just hoping that the majority of people are honest and will pay for the service, as if they didn't there would be no satellite TV.
Interestingly, it appears that for now the DirecTV have beaten the pirates, check out this rather spiffy explanation of the technologies involved - basically the old "HU" cards had a weakness in that they couldn't monitor the external clock signal which let you disrupt its execution. The newer P4 cards check for things like this, and so they are currently "unhackable". How long they will remain this way I don't know, but it's obvious that DRM to some extents while always an arms race, can/does work to a great extent.
Screen resolution switching has been fixed for about a year and a half now. There is an applet in Gnome upstream you can use, and there is also a command line applet.
These sort of issues are being rapidly nailed, I suggest you try the latest "cutting edge" releases before complaining as what you want might already have been fixed.
Samba usage is likewise getting better, at least, you can view the available network shares in the file manager now.
Do you have any evidence for that, or are you just guessing?
I'll assume you're just guessing, because anecdotal evidence from my life suggests just the opposite.
The first thing you should understand is the that people I hang out with and work with in the digital realm are all fairly hard core geeks. These are not geeks as in "wrote a website that used Perl once" or "enjoys fiddling with kernel compile options", these people are hardcore in terms of "knows 6 dialects of assembly language", "programs satellites", "reverse engineers Windows for a living", "knows and uses 8 different languages" etc. They all use either Gnome or fluxbox/kahakai/whatever (sometimes both). I'm a geek. I get a kick out of writing programs like this. Yet I use Gnome, and I like it. What gives?
From what I read on Slashdot, I would believe that anybody who is even remotely geeklike would hate Gnome and run away from it. All I see is bitching about whatever it is the Gnome developers have done now, whether it be adopting a HIG, changing the button ordering, spatial nautilus or whatever. Yet all around me there are geeks using Gnome. In fact, only a few I know use KDE, and the ones that do tend not to be the serious coders as such but more the ones who enjoy fiddling with their computer, perhaps know a bit of scripting etc.
OK, so having countered some anecdotal assertions with even more anecdotal evidence, let me try and explain what I see.
The thing is, Nautilus prior to Gnome 2.6 was not very useful. At least, I never used it, and from talking to other people they seem to be pretty much the same. Why use the slow and cumbersome GUI when the command line was so much faster?
With Gnome 2.6, that changed. Once people got used to it, they found it was in some cases actually faster to use the spatial GUI than it was to use the command line. Not for everything! I'd never use spatial Nautilus (or, for that matter, any GUI file manager) to manage my source code trees, which are enormously deep. I do use it to manage my desktop and home folder, which is not that deep.
So, for me and it seems many others that I know, spatial Nautilus is a win. Even for those who don't like it, it's not a big deal because almost universally when questioned they did not use Nautilus before.
Now, all this would be academic if spatial file management did not solve a real usability problem. Does it? I don't know 100%, it's too early to tell, but I do know one thing: I've met many, many Non-Geeks who don't really grok directory/folder hierarchies.
My mother is a classic example of this. She uses computers as part of her office job, but she does not grok file management. She knows how to go through the motions, but if anything changes, she is stuck. She doesn't really use directories, at least, not in a meaningful way. I've explained it to her of course, but she does not grok it (by "grok", I mean to have a zen-like understanding of something) in the same way we do.
Does spatial Nautilus solve her problem? Yes, I think so. I've seen a lot of evidence both from HCI texts I've read and real world experience watching friends and relatives use computers that many people don't connect with tree structures. Presenting a tree structure is a bad plan, they won't really understand it, and it's all too easy to end up with people saving files in the "wrong places" because they don't have any concept of where places are relative to each other.
So spatial Nautilus is about trying to help these people. It might well piss off some other people, but I've found that very few of these people really used GUI file management before - they were almost always shell users, so it's no big loss. And it's fairly easy to revert back to the old way, for those who did.
Free software isn't just for geeks anymore. There are some people who are trying to write software for whom computers are not a natural thing. Don't flame them just because it's not what you would want! Instead, understand their goals, and think critically about whether they are good or bad. I wish I saw more of that here.
To be fair I think double-check locking is only broken on non-x86 architectures where you need to erect memory barriers, or something similar. It's very subtle and wouldn't surprise me at all that your co-workers didn't realise it was broken.
Of course, I may have got totally the wrong idea, in which case please explain it to me too:)
ps. i am assuming you are talking about
if (condition) {// first check synchronized(whatever) { if (condition) {// second check } } }
glibc also does a lot more than the BSDs libcs, which are generally rather poor in terms of features, portability and so on. Besides on my system the size of glibc is negligable, only about 2mb (libc+pthreads+libm etc). Considering that it's shared between every app that's not something we need worry about.
Far more likely is that you were running more services in the background than you were on NetBSD.
Worst of all, IIRC Apple had known about the problem for over 6 months. The security problems Safari and OS X have make these IE problems look patsy in comparison - from reading the description you'd have to have a seriously good understanding of the internals of Internet Explorer to find these vulnerabilities, whereas the OS X/Help Viewer/DMG mounting vulnerabilities were very easy to understand and exploit.
Fortunately for Apple their market share is low enough that these exploits are mostly confined to theory - scummy spyware companies don't target OS X because the cost:benefit ratio isn't good enough. Same for Linux (which is equally not immune to URL handler/scripting vulnerabilities).
That's pretty anecdotal. I know a few people still using 500/600mhz PCs just fine, it's not like they *need* to upgrade. I'm sure your Powermac 9500 is just fine, but that doesn't mean PCs (even "typical windows" ones) have a short shelf life.
Like many things, you can get as much out of it as you are willing to.
Of course they let you do that - you can do whatever the hell you like, it's your system and all the source is under open licenses. What you mean is, Red Hat will not support you or give you binary upgrade RPMs for patches unless you are subscribed.
Note that the patches are still available, feel free to apply them yourselves, but if you want Red Hat to be the fall guy when things go wrong, they expect you to pay for it. Peoples time and expertise is not free in this case.
Personally, I'm just wondering exactly what ia32 chips will Linux and OpenBSD use NX on.
New ones. People will buy them because they want "Enhanced Virus Protection".
In the case of older chips: well, that is what exec-shield is for.
Re:why do you troll every Linux story?
on
The GNOME Roadmap
·
· Score: 1
Why do you think he's trolling? Yes, such a post would inspire lots of aggravted replies which is the hallmark of a good troll but that's not his fault. The guy makes a good point. An OS where you can see some software written for it, that you want to run, but that you can't easily install is a broken OS.
Outside of locked down corp desktops, I have never seen a machine where the user did not install something, even if it's only WinAmp.
Re:What applications are there
on
Mono Beta 2 Released
·
· Score: 5, Interesting
OK, so the fundamental issue here is one of toolkit quality.
You said that if you don't use GTK# your app will run on Windows. This is wrong. GTK# apps can work on Windows, or MacOS X just fine.
Understand this: the portability of an application is not defined by the type of machine (virtual or not) it's running on, it is by and large defined by the portability of the compilers and frameworks/libraries it relies upon.
GTK+ is a portable widget toolkit, it works pretty well on Windows and MacOS. The Win32 widget toolkit is not very portable, mostly because the only open source implementation is the Wine implementation and Wine by policy only concentrates on application compatibility, not on having nice pretty widgets.
So, if you are writing a.NET application you are best advised to use GTK# - this is true even if you are writing a program meant for Windows as in future if you wish to port things to another platform it will be a lot easier. There are a few other things to consider as well, such as the nicer API GTK has.
As to the DotGNU approach vs the Mono approach, basically I think you'd have to be insane to want to reimplement what Wine has done. Nobody is going to use System.Windows.Forms on Linux because it blows goats, everybody will use GTK# or (maybe when it is mature) Qt# - therefore a SWF implementation is useful only for application compatibility.
As to mapping S.W.F to Gnome/GTK, forget it. Back in the day (waaaaay back) Wine attempted to map the Win32 widget toolkit to Tk which was one of the better toolkits available back then. Didn't work. Widget toolkits differ too much to succesfully map between, and in particular the differential between a modern toolkit like GTK+ and Win32 is enormous - why do you think Microsoft are so keen to scrap it and start over with Avalon?
Your mother would not know about the secondary/middle-click clipboard unless she was told about it. She equally would not be using emacs, which is one of the few remaining broken apps (and you can unbreak it with a setting).
So, basically, the clipboard would work OK for your mother. It's power users who try and combine both mechanisms at once then get confused who it doesn't work for, but there's no way around that.
The real problems with the X clipboard have more to do with handling large quantities of data, and standardised data formats than middle click vs ctrl-c/v.
Have you actually used X much? I don't really see what the fuss is about - if you don't like the way the PRIMARY (middle click) clipboard works, don't use it. The standard Windows style clipboard is still there, still works fine, and is independent of the middle-click clipboard. If that is ever not true, you are using a buggy app that should be fixed.
Basically, this guy is complaining that things are working in the way they should, but not how he wants them. I'm not sure how to make it convenient for him without breaking other stuff. Look at it this way, he could just use the same way that works on Windows/MacOS, but he's not, then complains that it's broken. What's up with that?
No, you can't (easily) build this with WineLib, as it depends on various proprietary things like the MS platform SDK tools (mktypelib/midl), MFC etc. Wine has replacements for some of these things, though not MFC.
There isn't much point to porting things to WineLib though in a case like this. There are only limited places where you may want native UI - for instance the settings window you might want to port to GTK - but you can do that using a DLL meant to be compiled on Linux using Wine then using the MSVC++ produced EXE binaries.
The rest of the GUI appears to be almost totally themed, and I must admit it's very pretty.
BTW Shareaza already works ~ 95% under Wine, there are one or two bugs that would not be hard to fix or work around from inside the app. If anybody wanted to begin making it native they could do. Just ask on wine-devel for details (you should have a good understanding of windows+linux)
There are already nightly builds of Inkscape CVS available for autopackage 0.5.1 here. To those who haven't used autopackage before, download and run that file.
Note: there are known issues with certain (rare) setups which have a non standard umask and X security settings. If you are on a stock Red Hat/Fedora install all should go smoothly (let us know if it does not). If you have tweaked your umask or have X security too restrictive (programs run as root must be able to connect) things will break.
If you want to test inkscape quickly and you are on x86 Linux, this is an easy way to do it. Just be careful. autopackage is in beta. If it breaks you get to keep the pieces.
Microsoft has the luxury of being able to hire the best people
I think that's pretty insulting to all the people who work (for pay) on free software such as Linux. Or what, do you think Alan Cox - who was approached by Microsoft and turned them down - is not one of the best?
Please. Implying that the best people only care about the size of their salary is doing a disservice to all those who have other factors in their choice of work.
Can you actually show that Win16 support acted to the detriment of OS/2, or is that just an assumption you've made - "other people on Slashdot keep saying it so it must be true" ?
Actually, I've seen some pretty compelling arguments from people who used OS/2 at the time that Windows support did not harm it, and in fact probably helped - they have claimed that the main reason OS/2 failed (and of course there were many) was that IBM didn't market it well: they weren't even selling machines with it on themselves at one point.
Regardless, whether it hindered or helped OS/2 is largely academic. Application support is one of the big things currently stopping a mass migration to desktop Linux, along with inertia/lack of experience and some general immaturities in the technology. Nat Friedman of Novell has said that app compat is the number one blocker for their sales team.
So Wine really is necessary, simply because it doesn't make sense to rewrite every desktop program in the world to use the Linux APIs. To be frank, humanity has better things to do.
Disclaimer: I'm a Wine developer so am somewhat biased. But on the flip side, I wouldn't be working on Wine if I didn't think it was important.
Well, they are both gcc, so only the backend and assembler is different - the overhead of compiling C is mostly parsing and the tree manipulations, iirc. Oh, and creating lots of processes/files.
Slashdot Mirror
Choice quotes include: "Not much defense on the PNG transparency issue other than GDI made it hard back in the day. We really want this fixed too and are working on it. No eta yet. I'll try to update folks as we work through this. Same with other decisions."
They have also set up a Wiki for gathering ideas for bugfixes/improvements to IE. Not like they are lacking ideas already though ...
er, that should have read "is a feature of the C library" of course :) egg -> face
No, TLS is short for Thread Local Storage, and in most UNIX-like operating systems the dynamic linker *is* a feature of the linker. The glibc symvers implementation is more advanced than any of the BSDs/Solaris last time I checked. Anyway, those were only some examples, I suggest you go read the glibc sources if you would like to find some more.
The whole reason they do signal encryption is because broadcasting is a technically efficient way to get video content to people BUT economically is hard to make work unless (like the BBC) society assumes everybody with a TV will want to watch it. Signal encryption is a good compromise between technical efficiency and economic reality. The people who feel they are owned something for nothing have to consider their actions in the following context: what would happen if everybody did this?
It then becomes pretty obvious that these people are not acting upon some superior moral reasoning, they are just hoping that the majority of people are honest and will pay for the service, as if they didn't there would be no satellite TV.
Interestingly, it appears that for now the DirecTV have beaten the pirates, check out this rather spiffy explanation of the technologies involved - basically the old "HU" cards had a weakness in that they couldn't monitor the external clock signal which let you disrupt its execution. The newer P4 cards check for things like this, and so they are currently "unhackable". How long they will remain this way I don't know, but it's obvious that DRM to some extents while always an arms race, can/does work to a great extent.
These sort of issues are being rapidly nailed, I suggest you try the latest "cutting edge" releases before complaining as what you want might already have been fixed.
Samba usage is likewise getting better, at least, you can view the available network shares in the file manager now.
Do you have any evidence for that, or are you just guessing?
I'll assume you're just guessing, because anecdotal evidence from my life suggests just the opposite.
The first thing you should understand is the that people I hang out with and work with in the digital realm are all fairly hard core geeks. These are not geeks as in "wrote a website that used Perl once" or "enjoys fiddling with kernel compile options", these people are hardcore in terms of "knows 6 dialects of assembly language", "programs satellites", "reverse engineers Windows for a living", "knows and uses 8 different languages" etc. They all use either Gnome or fluxbox/kahakai/whatever (sometimes both). I'm a geek. I get a kick out of writing programs like this. Yet I use Gnome, and I like it. What gives?
From what I read on Slashdot, I would believe that anybody who is even remotely geeklike would hate Gnome and run away from it. All I see is bitching about whatever it is the Gnome developers have done now, whether it be adopting a HIG, changing the button ordering, spatial nautilus or whatever. Yet all around me there are geeks using Gnome. In fact, only a few I know use KDE, and the ones that do tend not to be the serious coders as such but more the ones who enjoy fiddling with their computer, perhaps know a bit of scripting etc.
OK, so having countered some anecdotal assertions with even more anecdotal evidence, let me try and explain what I see.
The thing is, Nautilus prior to Gnome 2.6 was not very useful. At least, I never used it, and from talking to other people they seem to be pretty much the same. Why use the slow and cumbersome GUI when the command line was so much faster?
With Gnome 2.6, that changed. Once people got used to it, they found it was in some cases actually faster to use the spatial GUI than it was to use the command line. Not for everything! I'd never use spatial Nautilus (or, for that matter, any GUI file manager) to manage my source code trees, which are enormously deep. I do use it to manage my desktop and home folder, which is not that deep.
So, for me and it seems many others that I know, spatial Nautilus is a win. Even for those who don't like it, it's not a big deal because almost universally when questioned they did not use Nautilus before.
Now, all this would be academic if spatial file management did not solve a real usability problem. Does it? I don't know 100%, it's too early to tell, but I do know one thing: I've met many, many Non-Geeks who don't really grok directory/folder hierarchies.
My mother is a classic example of this. She uses computers as part of her office job, but she does not grok file management. She knows how to go through the motions, but if anything changes, she is stuck. She doesn't really use directories, at least, not in a meaningful way. I've explained it to her of course, but she does not grok it (by "grok", I mean to have a zen-like understanding of something) in the same way we do.
Does spatial Nautilus solve her problem? Yes, I think so. I've seen a lot of evidence both from HCI texts I've read and real world experience watching friends and relatives use computers that many people don't connect with tree structures. Presenting a tree structure is a bad plan, they won't really understand it, and it's all too easy to end up with people saving files in the "wrong places" because they don't have any concept of where places are relative to each other.
So spatial Nautilus is about trying to help these people. It might well piss off some other people, but I've found that very few of these people really used GUI file management before - they were almost always shell users, so it's no big loss. And it's fairly easy to revert back to the old way, for those who did.
Free software isn't just for geeks anymore. There are some people who are trying to write software for whom computers are not a natural thing. Don't flame them just because it's not what you would want! Instead, understand their goals, and think critically about whether they are good or bad. I wish I saw more of that here.
To be fair I think double-check locking is only broken on non-x86 architectures where you need to erect memory barriers, or something similar. It's very subtle and wouldn't surprise me at all that your co-workers didn't realise it was broken.
:)
// first check // second check
Of course, I may have got totally the wrong idea, in which case please explain it to me too
ps. i am assuming you are talking about
if (condition) {
synchronized(whatever) {
if (condition) {
}
}
}
symbol versioning, TLS, non-sucky threading support, RTLD_NEXT, dlinfo() etc etc
Far more likely is that you were running more services in the background than you were on NetBSD.
Fortunately for Apple their market share is low enough that these exploits are mostly confined to theory - scummy spyware companies don't target OS X because the cost:benefit ratio isn't good enough. Same for Linux (which is equally not immune to URL handler/scripting vulnerabilities).
Like many things, you can get as much out of it as you are willing to.
Note that the patches are still available, feel free to apply them yourselves, but if you want Red Hat to be the fall guy when things go wrong, they expect you to pay for it. Peoples time and expertise is not free in this case.
New ones. People will buy them because they want "Enhanced Virus Protection".
In the case of older chips: well, that is what exec-shield is for.
And no the answer is not "Just use Debian".
Outside of locked down corp desktops, I have never seen a machine where the user did not install something, even if it's only WinAmp.
You said that if you don't use GTK# your app will run on Windows. This is wrong. GTK# apps can work on Windows, or MacOS X just fine.
Understand this: the portability of an application is not defined by the type of machine (virtual or not) it's running on, it is by and large defined by the portability of the compilers and frameworks/libraries it relies upon.
GTK+ is a portable widget toolkit, it works pretty well on Windows and MacOS. The Win32 widget toolkit is not very portable, mostly because the only open source implementation is the Wine implementation and Wine by policy only concentrates on application compatibility, not on having nice pretty widgets.
So, if you are writing a .NET application you are best advised to use GTK# - this is true even if you are writing a program meant for Windows as in future if you wish to port things to another platform it will be a lot easier. There are a few other things to consider as well, such as the nicer API GTK has.
As to the DotGNU approach vs the Mono approach, basically I think you'd have to be insane to want to reimplement what Wine has done. Nobody is going to use System.Windows.Forms on Linux because it blows goats, everybody will use GTK# or (maybe when it is mature) Qt# - therefore a SWF implementation is useful only for application compatibility.
As to mapping S.W.F to Gnome/GTK, forget it. Back in the day (waaaaay back) Wine attempted to map the Win32 widget toolkit to Tk which was one of the better toolkits available back then. Didn't work. Widget toolkits differ too much to succesfully map between, and in particular the differential between a modern toolkit like GTK+ and Win32 is enormous - why do you think Microsoft are so keen to scrap it and start over with Avalon?
Check in with #autopackage on freenode, or autopackage-dev before you do that so we can give tips and so you can keep track of what's going on.
So, basically, the clipboard would work OK for your mother. It's power users who try and combine both mechanisms at once then get confused who it doesn't work for, but there's no way around that.
The real problems with the X clipboard have more to do with handling large quantities of data, and standardised data formats than middle click vs ctrl-c/v.
Basically, this guy is complaining that things are working in the way they should, but not how he wants them. I'm not sure how to make it convenient for him without breaking other stuff. Look at it this way, he could just use the same way that works on Windows/MacOS, but he's not, then complains that it's broken. What's up with that?
There isn't much point to porting things to WineLib though in a case like this. There are only limited places where you may want native UI - for instance the settings window you might want to port to GTK - but you can do that using a DLL meant to be compiled on Linux using Wine then using the MSVC++ produced EXE binaries.
The rest of the GUI appears to be almost totally themed, and I must admit it's very pretty.
BTW Shareaza already works ~ 95% under Wine, there are one or two bugs that would not be hard to fix or work around from inside the app. If anybody wanted to begin making it native they could do. Just ask on wine-devel for details (you should have a good understanding of windows+linux)
Note: there are known issues with certain (rare) setups which have a non standard umask and X security settings. If you are on a stock Red Hat/Fedora install all should go smoothly (let us know if it does not). If you have tweaked your umask or have X security too restrictive (programs run as root must be able to connect) things will break.
If you want to test inkscape quickly and you are on x86 Linux, this is an easy way to do it. Just be careful. autopackage is in beta. If it breaks you get to keep the pieces.
I think that's pretty insulting to all the people who work (for pay) on free software such as Linux. Or what, do you think Alan Cox - who was approached by Microsoft and turned them down - is not one of the best?
Please. Implying that the best people only care about the size of their salary is doing a disservice to all those who have other factors in their choice of work.
The administrator client apparently crashes if you try and use the remote console feature.
Actually, I've seen some pretty compelling arguments from people who used OS/2 at the time that Windows support did not harm it, and in fact probably helped - they have claimed that the main reason OS/2 failed (and of course there were many) was that IBM didn't market it well: they weren't even selling machines with it on themselves at one point.
Regardless, whether it hindered or helped OS/2 is largely academic. Application support is one of the big things currently stopping a mass migration to desktop Linux, along with inertia/lack of experience and some general immaturities in the technology. Nat Friedman of Novell has said that app compat is the number one blocker for their sales team.
So Wine really is necessary, simply because it doesn't make sense to rewrite every desktop program in the world to use the Linux APIs. To be frank, humanity has better things to do.
Disclaimer: I'm a Wine developer so am somewhat biased. But on the flip side, I wouldn't be working on Wine if I didn't think it was important.
Well, they are both gcc, so only the backend and assembler is different - the overhead of compiling C is mostly parsing and the tree manipulations, iirc. Oh, and creating lots of processes/files.