Just saying, that though this tab isolation deal brings benefits and is great as a last resort protection, preventing things from crashing should remain the top priority,
I'd argue the reverse. Get the tab isolation to be flawless. Then worry about making it not crash.
The one huge thing about OS X, as I understand it, was similar: memory protection. Earlier versions of Mac OS had none, and, in fact, this allowed for some cool hacks, like patching in virtual memory on an OS which didn't support it, or even compressing the RAM of a running program.
It also meant that one bad pointer in one app could bring down the whole system.
Bringing down one tab sucks. Bringing down all tabs is unacceptable.
even if a tab's crash didn't stop the other ones from working it could still ruin your day if you were doing important stuff on that tab.
True. I'm not suggesting they ignore that problem.
But then, suppose you are on Google's web app utopia, and you open up another tab to check out a random link someone sent you... that'll ruin your day, too. Which is more likely?
I'm assuming that the tab isolation takes very little effort to get working right, and very little effort to maintain, relative to overall stability. But if I was a manager, I'd definitely push for getting that perfected first, and then shifting focus to what goes on inside the tab.
And how do you know? It seems they're trying to provide a sandbox for plugins, so I'd think they could manage something similar for ActiveX.
If it is signed (and any stolen credit card can be used to "sign"/buy a cert - so where's that identity huh?), then any web page you go to can call an installed activex component.
Sorry, but I don't buy it -- even MS-signed activex controls ask me first before they activate.
Hence, the activex things that HP/Compaq and so on installed, that later turned out to have buffer overflows
Which would imply a security hole in them, and not in ActiveX.
oh yeah, even IE allows you to NOT let any activex run - why doesn't Chrome do that huh?
Don't ask me, ask the developers. Better yet, add it yourself.
So, I wouldn't write Chrome off just yet.
Re:An advantage of 64-bit Linux?
on
Chrome Vs. IE 8
·
· Score: 1
(would be better if simply clicking on "sad plugin" image (displayed in place of youtube video for example) would suffice...)
No, not really. Flash, in particular, integrates quite a lot with the page itself -- so that wouldn't be like forcing an image reload, it'd be more like trying to figure out how to restart a script that had thrown an exception.
we constantly monitor the performance of the system in regards to memory usage and we can use this data to infer when we need more memory and when other applications or the OS needs more memory.
So, they infer when something else needs RAM. But they don't actually know -- all they do is monitor available RAM. Not unique to Windows, and basically a hack for lack of OS support.
And another reason explicit OS support would help:
Exchange Store is not the only product behaving like this... SQL does something very similar, for example. That is one of reasons why we do not necessarily encourage putting SQL and Exchange on the same server, as they will be fighting over whatever RAM is in the server.
That seems nonsensical, when you think about it -- Linux filesystems don't "fight" over RAM. Some Linux servers have 10-15 partitions mounted (because some admins like to slice things up that much), but the same algorithm applies, if I understand it -- stupidly simple things like least-frequently-used algorithms determine what stays, and what goes.
So, if browsers did this, they'd end up fighting over RAM with anything else that did this.
Well, given a critical mass of developers who agree with me... hell, I could become the one publisher who gets it right. That's my master plan!
More seriously, there are publishers who get this right, and it's not always the publishers who have the brand recognition. And it's not always brand recognition, anyway.
I doubt I'll ever get through to the major publishers.
You seem to have confused 'could care less' with 'couldn't care less'.
You're right. I usually tend to avoid the term entirely, though -- after all, "couldn't care less" doesn't work, either. Take it in a different context, say: "I love you, Juliet! I couldn't possibly love you any less! I couldn't care less!"
Now, "could care less" is pretty pointless, but neither is what I want to say. What I should be saying is "really don't care" or "really couldn't care".
I have NEVER seen a trojan come from a pirated game. Ever.
I have. Mostly from cracks, but also from games.
I trust the pirates and crackers way more than i trust the game companys these days.
I'll agree with that -- partly because I can at least throw the pirated copy through antivirus/antispyware. But I will buy a legitimate game if I know ahead of time that there isn't going to be trouble.
I don't know of a list. In fact, I'd love to create one -- though it's possible someone will read this comment and beat me to it -- preferably with the DRM classified, not just a list of DRM-free versions.
Off the top of my head:
As others have mentioned, Stardock. There's also the non-Steam versions of Introversion games -- I can download the Linux versions a fixed number of times (3, I think), but once downloaded, there's absolutely no DRM I can find, not even a CD key. There's older games, like Duke Nukem 3D (pre-Atomic).
Native Linux versions of everything ID up to and including Doom 3 used no DRM beyond a CD key, and everything up to and including Quake 3 are open source, so even that can be removed. I believe Quake 4 phones home with its CD key, but if it can't phone home, it allows access -- so if id goes out of business, Quake 4 will still work.
In fact, most Linux ports won't include much DRM, mostly because it's hard to port that stuff.
There's other open source games, like Nexuiz, OpenArena, World of Padman, possibly Battle for Wesnoth -- take a look around the Ubuntu repositories to see what's available.
Penny Arcade's "Greenhouse", which seems to be targeting Steam (somewhat), uses DRM somewhat like Windows XP. That is, it phones home once at installation, and insists on phoning home whenever there's a hardware change, and requires an activation code. So, no hardware changes and no reinstalls mean no phoning home (more than once). And their activation server enforces absolutely nothing right now -- I assume they'll start blocking you if they see a few thousand installs, but there's actually nothing in place so far.
Valve's own Steam games use no DRM beyond what Steam does itself. I find Steam to be acceptable. I find SecuROM on top of Steam to be a deal-breaker.
No MMO is going to use DRM. They might use craptastic anti-cheating, but really, most MMOs will simply rely on the fact that you have to play on their servers. Sort of a constant, perfect phone-home, but one that's actually adding value to you, as a consumer.
And if an alcoholic saw a brewer putting kegs in high schools, he'd bloody well have some choice words for the breweries, not to mention the school administrators.
And my point is, breweries should be allowed to exist, at least for the adults. I'm not sure what you're actually proposing, but take a moment to remember how well it went when that wasn't the case -- that is, remember Prohibition?
And, hell, alcohol is addictive. Do you rant at Budweiser for profiting from addicts?
These guys aren't content with the level of pervasiveness of their games, so they're pushing it out to ever-more-intrusive platforms.
Again: How in the fuck is it "intrusive" for me to have the ability to play the game on the train? You said it yourself:
I'm ok with folks wanting to play games on the train or whatever.
No one's forcing you to get a mobile device that can play their game -- or to play their game at all.
If they were installing spyware, say, and injecting ads into every comment posted, that would be intrusive. I don't see how providing a new optional feature can be "intrusive".
if you hold to any theory of evolution, you now have to know that "punctuated equalibrium" cannot possibly work.
Why not?
Observation has never produced one evidence of it,
Grammar would help, too...
As a Creationist, you should be familiar with this argument: Absence of evidence is not evidence of absence.
major DNA shifts producing change from one phyla, genus, specie, or whatever to another have always proven detrimental to the life affected,
Citation needed. And again, "has always".
Bluntly, there is no real evidence that such things can happen.
Yet you provide not one shred of evidence that they can't.
Bluntly, evolutionary theorum has been classified as a religion in many courts,
Which is proof of nothing more than the courts' ignorance.
it takes a great deal more faith to believe in it than it does to belive in Scientific Creationism.
Well, let's see -- where is your evidence supporting creationism?
Assuming you're right, you've found some problems with evolutionary theory. Are you so narrow-minded that the only other possibility is for the Earth to be young, and created by a divine power?
#5 is procedural generation -- which suggests that, rather than drawing each individual texture, we'd write algorithms and let them generate themselves.
#7 is id's megatextures, which suggests that, rather than doing anything algorithmic, we'll just add more and more detail to a gigantic image.
These seem to be pretty much direct opposites of each other. Are they suggesting that each will be good for different areas? Or do they just not know what they're talking about?
If ActiveX is a security problem, having it in your browser makes your browser insecure.
Is it impossible to disable?
And it seems to me that ActiveX isn't itself a security problem -- it's how it's used, and it's people hitting 'OK' to everything. Do you actually read dialog boxes before reflexively clicking on them?
I don't dispute that, but again: Firefox takes longer to start than Konqueror. It may use less RAM overall over a browsing session, but it certainly seems to need more just to boot itself.
Not only is he more fit, but you can't even hire the other guy because he can't get licensed without jumping through the union hoops.
Is this actually true? That is:
Are there laws preventing me from ripping open my own wall and doing my own wiring?
If not, sounds like there are laws preventing me from hiring someone else to do so.
Either seems absurdly limiting. It's like the FDA -- I can see where you're not allowed to actually sell poison, and I can see where an herbal remedy can't actually be sold as "medicine" without a disclaimer somewhere.
I can see where it would be useful to have some sort of certification body, but not to actually have laws forcing you to hire someone who's certified. And even here, certification tends to mostly test how good you are at regurgitating pointless facts you'll never use again -- just like high school.
This is a young industry, and it's changing all the time. What you need to know changes all the time.
Given that, all the more reason for there to be some sort of standardization, if it could be done right -- because there's that much more opportunity for me to claim to know more than I do. Hell, I could probably put things like LISP and Haskell on my resume, and no one would bother to actually ask me anything about them.
(For the record, I do know a bit about LISP and Haskell, but I couldn't so much as write "Hello, World" in them anymore.)
Re:On tabs crashing
on
Chrome Vs. IE 8
·
· Score: 2, Insightful
Perhaps we should send a bug report to someone, or something. Let's just clean up, and start all over again though!
Yeah -- that's called "crashing". Or, in Chrome, that would be the "Sad Tab" -- it catches some exception, cleans up, and lets you hit "refresh" to start all over again.
Sadly, "starting all over again" generally means clobbering some state the user cared about. If my browser throws an exception as I type this, my comment is pretty much gone, and there's not much that can be done about that.
Webkit isn't written in Java. Nor is v8. And frankly, I'm glad.
Even if you assume all the glue is the JVM, it doesn't save you -- at some point, you're going to have to link to something written in C. And when Flash crashes, it's going to take down anything linked to it.
I'm not sure how they do it, but I understand Exchange Server does pretty much what you suggested the browser should do. It'll grab almost all the available memory to use as a cache so it can speed up user request processing. Then, as other applications request memory from the OS, Exchange Server will free the amount requested.
Given that Exchange Server would likely have a box dedicated to it, are you sure that it's actually doing this with other applications, outside the realm of Exchange? And not, say, something else that's actually part of the Exchange process?
And are you sure it's actually doing that caching itself, and not simply relying on the disk cache? Exchange is a mailserver -- relying on the disk cache would be the logical thing to do.
If there is a mechanism, and it's actually published, then that's a point for Windows (I know of no way to do that on Linux), and it's something browsers should take advantage of. I just doubt such a mechanism exists.
Since I believe ActiveX to have shitty security, why would I be using that or want to use that?
Since you wouldn't be using it, why do you care that it exists in the browser?
And yes, it should refuse to support it. After all, it's not going to support it on linux and osx. Why should it cripple itself on windows?
Again, this misguided assumption that by supporting something, it's magically crippled. I'd say Flash is far more crippling than ActiveX, for more reasons than one.
But if you feel that strongly, wait a bit, or go fix it yourself -- the source is available.
This issue doesn't touch me nearly as much -- I'm on Linux, and easily 99% of the software I use updates through a central package management system. Which means only one update program to run, or only one check made on boot -- far less duplication and bloat just for the updaters.
But then, Ubuntu didn't ask me to start checking for updates, it only asks me when it wants to install them.
Slashdot Mirror
Just saying, that though this tab isolation deal brings benefits and is great as a last resort protection, preventing things from crashing should remain the top priority,
I'd argue the reverse. Get the tab isolation to be flawless. Then worry about making it not crash.
The one huge thing about OS X, as I understand it, was similar: memory protection. Earlier versions of Mac OS had none, and, in fact, this allowed for some cool hacks, like patching in virtual memory on an OS which didn't support it, or even compressing the RAM of a running program.
It also meant that one bad pointer in one app could bring down the whole system.
Bringing down one tab sucks. Bringing down all tabs is unacceptable.
even if a tab's crash didn't stop the other ones from working it could still ruin your day if you were doing important stuff on that tab.
True. I'm not suggesting they ignore that problem.
But then, suppose you are on Google's web app utopia, and you open up another tab to check out a random link someone sent you... that'll ruin your day, too. Which is more likely?
I'm assuming that the tab isolation takes very little effort to get working right, and very little effort to maintain, relative to overall stability. But if I was a manager, I'd definitely push for getting that perfected first, and then shifting focus to what goes on inside the tab.
There's no sandbox.
Not a problem, if it never gets run.
And how do you know? It seems they're trying to provide a sandbox for plugins, so I'd think they could manage something similar for ActiveX.
If it is signed (and any stolen credit card can be used to "sign"/buy a cert - so where's that identity huh?), then any web page you go to can call an installed activex component.
Sorry, but I don't buy it -- even MS-signed activex controls ask me first before they activate.
Hence, the activex things that HP/Compaq and so on installed, that later turned out to have buffer overflows
Which would imply a security hole in them, and not in ActiveX.
oh yeah, even IE allows you to NOT let any activex run - why doesn't Chrome do that huh?
Don't ask me, ask the developers. Better yet, add it yourself.
So, I wouldn't write Chrome off just yet.
(would be better if simply clicking on "sad plugin" image (displayed in place of youtube video for example) would suffice...)
No, not really. Flash, in particular, integrates quite a lot with the page itself -- so that wouldn't be like forcing an image reload, it'd be more like trying to figure out how to restart a script that had thrown an exception.
Not quite:
we constantly monitor the performance of the system in regards to memory usage and we can use this data to infer when we need more memory and when other applications or the OS needs more memory.
So, they infer when something else needs RAM. But they don't actually know -- all they do is monitor available RAM. Not unique to Windows, and basically a hack for lack of OS support.
And another reason explicit OS support would help:
Exchange Store is not the only product behaving like this... SQL does something very similar, for example. That is one of reasons why we do not necessarily encourage putting SQL and Exchange on the same server, as they will be fighting over whatever RAM is in the server.
That seems nonsensical, when you think about it -- Linux filesystems don't "fight" over RAM. Some Linux servers have 10-15 partitions mounted (because some admins like to slice things up that much), but the same algorithm applies, if I understand it -- stupidly simple things like least-frequently-used algorithms determine what stays, and what goes.
So, if browsers did this, they'd end up fighting over RAM with anything else that did this.
Well, given a critical mass of developers who agree with me... hell, I could become the one publisher who gets it right. That's my master plan!
More seriously, there are publishers who get this right, and it's not always the publishers who have the brand recognition. And it's not always brand recognition, anyway.
I doubt I'll ever get through to the major publishers.
I didn't. That's why I'm modded "Funny".
You seem to have confused 'could care less' with 'couldn't care less'.
You're right. I usually tend to avoid the term entirely, though -- after all, "couldn't care less" doesn't work, either. Take it in a different context, say: "I love you, Juliet! I couldn't possibly love you any less! I couldn't care less!"
Now, "could care less" is pretty pointless, but neither is what I want to say. What I should be saying is "really don't care" or "really couldn't care".
I have NEVER seen a trojan come from a pirated game. Ever.
I have. Mostly from cracks, but also from games.
I trust the pirates and crackers way more than i trust the game companys these days.
I'll agree with that -- partly because I can at least throw the pirated copy through antivirus/antispyware. But I will buy a legitimate game if I know ahead of time that there isn't going to be trouble.
I don't know of a list. In fact, I'd love to create one -- though it's possible someone will read this comment and beat me to it -- preferably with the DRM classified, not just a list of DRM-free versions.
Off the top of my head:
As others have mentioned, Stardock. There's also the non-Steam versions of Introversion games -- I can download the Linux versions a fixed number of times (3, I think), but once downloaded, there's absolutely no DRM I can find, not even a CD key. There's older games, like Duke Nukem 3D (pre-Atomic).
Native Linux versions of everything ID up to and including Doom 3 used no DRM beyond a CD key, and everything up to and including Quake 3 are open source, so even that can be removed. I believe Quake 4 phones home with its CD key, but if it can't phone home, it allows access -- so if id goes out of business, Quake 4 will still work.
In fact, most Linux ports won't include much DRM, mostly because it's hard to port that stuff.
There's other open source games, like Nexuiz, OpenArena, World of Padman, possibly Battle for Wesnoth -- take a look around the Ubuntu repositories to see what's available.
Penny Arcade's "Greenhouse", which seems to be targeting Steam (somewhat), uses DRM somewhat like Windows XP. That is, it phones home once at installation, and insists on phoning home whenever there's a hardware change, and requires an activation code. So, no hardware changes and no reinstalls mean no phoning home (more than once). And their activation server enforces absolutely nothing right now -- I assume they'll start blocking you if they see a few thousand installs, but there's actually nothing in place so far.
Valve's own Steam games use no DRM beyond what Steam does itself. I find Steam to be acceptable. I find SecuROM on top of Steam to be a deal-breaker.
No MMO is going to use DRM. They might use craptastic anti-cheating, but really, most MMOs will simply rely on the fact that you have to play on their servers. Sort of a constant, perfect phone-home, but one that's actually adding value to you, as a consumer.
And if an alcoholic saw a brewer putting kegs in high schools, he'd bloody well have some choice words for the breweries, not to mention the school administrators.
And my point is, breweries should be allowed to exist, at least for the adults. I'm not sure what you're actually proposing, but take a moment to remember how well it went when that wasn't the case -- that is, remember Prohibition?
And, hell, alcohol is addictive. Do you rant at Budweiser for profiting from addicts?
These guys aren't content with the level of pervasiveness of their games, so they're pushing it out to ever-more-intrusive platforms.
Again: How in the fuck is it "intrusive" for me to have the ability to play the game on the train? You said it yourself:
I'm ok with folks wanting to play games on the train or whatever.
No one's forcing you to get a mobile device that can play their game -- or to play their game at all.
If they were installing spyware, say, and injecting ads into every comment posted, that would be intrusive. I don't see how providing a new optional feature can be "intrusive".
First: Learn to use paragraphs.
Now...
if you hold to any theory of evolution, you now have to know that "punctuated equalibrium" cannot possibly work.
Why not?
Observation has never produced one evidence of it,
Grammar would help, too...
As a Creationist, you should be familiar with this argument: Absence of evidence is not evidence of absence.
major DNA shifts producing change from one phyla, genus, specie, or whatever to another have always proven detrimental to the life affected,
Citation needed. And again, "has always".
Bluntly, there is no real evidence that such things can happen.
Yet you provide not one shred of evidence that they can't.
Bluntly, evolutionary theorum has been classified as a religion in many courts,
Which is proof of nothing more than the courts' ignorance.
it takes a great deal more faith to believe in it than it does to belive in Scientific Creationism.
Well, let's see -- where is your evidence supporting creationism?
Assuming you're right, you've found some problems with evolutionary theory. Are you so narrow-minded that the only other possibility is for the Earth to be young, and created by a divine power?
Not at the level evolutionism tries to do, tens of orders of magnitude beyond the edge of the observed data.
Again: Big bang? Carbon dating?
#5 is procedural generation -- which suggests that, rather than drawing each individual texture, we'd write algorithms and let them generate themselves.
#7 is id's megatextures, which suggests that, rather than doing anything algorithmic, we'll just add more and more detail to a gigantic image.
These seem to be pretty much direct opposites of each other. Are they suggesting that each will be good for different areas? Or do they just not know what they're talking about?
Or maybe I don't know what I'm talking about.
...whereas the higher-end Minis come with XP.
If ActiveX is a security problem, having it in your browser makes your browser insecure.
Is it impossible to disable?
And it seems to me that ActiveX isn't itself a security problem -- it's how it's used, and it's people hitting 'OK' to everything. Do you actually read dialog boxes before reflexively clicking on them?
It's possible I'm missing something...
VBR lame-encoded MP3s, with not a speck of DRM! Effective by design!!!
Yeah, that's a start.
I tend to seek out FLAC, these days. I'd certainly settle for 256-bit DRM-free AAC (far better quality than those MP3s), but Apple won't let me.
I don't dispute that, but again: Firefox takes longer to start than Konqueror. It may use less RAM overall over a browsing session, but it certainly seems to need more just to boot itself.
Yes, there is. Many pieces of software on windows do this.
Can you point to that?
The vast, vast, vast majority of web browser users are running on single user machines that do web browsing, email and maybe one or two others things.
In which case, there shouldn't be anything else eating up RAM and purging that cache.
Not only is he more fit, but you can't even hire the other guy because he can't get licensed without jumping through the union hoops.
Is this actually true? That is:
Are there laws preventing me from ripping open my own wall and doing my own wiring?
If not, sounds like there are laws preventing me from hiring someone else to do so.
Either seems absurdly limiting. It's like the FDA -- I can see where you're not allowed to actually sell poison, and I can see where an herbal remedy can't actually be sold as "medicine" without a disclaimer somewhere.
I can see where it would be useful to have some sort of certification body, but not to actually have laws forcing you to hire someone who's certified. And even here, certification tends to mostly test how good you are at regurgitating pointless facts you'll never use again -- just like high school.
This is a young industry, and it's changing all the time. What you need to know changes all the time.
Given that, all the more reason for there to be some sort of standardization, if it could be done right -- because there's that much more opportunity for me to claim to know more than I do. Hell, I could probably put things like LISP and Haskell on my resume, and no one would bother to actually ask me anything about them.
(For the record, I do know a bit about LISP and Haskell, but I couldn't so much as write "Hello, World" in them anymore.)
Perhaps we should send a bug report to someone, or something. Let's just clean up, and start all over again though!
Yeah -- that's called "crashing". Or, in Chrome, that would be the "Sad Tab" -- it catches some exception, cleans up, and lets you hit "refresh" to start all over again.
Sadly, "starting all over again" generally means clobbering some state the user cared about. If my browser throws an exception as I type this, my comment is pretty much gone, and there's not much that can be done about that.
Webkit isn't written in Java. Nor is v8. And frankly, I'm glad.
Even if you assume all the glue is the JVM, it doesn't save you -- at some point, you're going to have to link to something written in C. And when Flash crashes, it's going to take down anything linked to it.
And no, that doesn't catch programmer brainfarts.
Fine. s/Crysis/Doom3.
I'm basing my criticism of Crysis largely on this comic -- is the narrative actually better than that?
I'm not sure how they do it, but I understand Exchange Server does pretty much what you suggested the browser should do. It'll grab almost all the available memory to use as a cache so it can speed up user request processing. Then, as other applications request memory from the OS, Exchange Server will free the amount requested.
Given that Exchange Server would likely have a box dedicated to it, are you sure that it's actually doing this with other applications, outside the realm of Exchange? And not, say, something else that's actually part of the Exchange process?
And are you sure it's actually doing that caching itself, and not simply relying on the disk cache? Exchange is a mailserver -- relying on the disk cache would be the logical thing to do.
If there is a mechanism, and it's actually published, then that's a point for Windows (I know of no way to do that on Linux), and it's something browsers should take advantage of. I just doubt such a mechanism exists.
Since I believe ActiveX to have shitty security, why would I be using that or want to use that?
Since you wouldn't be using it, why do you care that it exists in the browser?
And yes, it should refuse to support it. After all, it's not going to support it on linux and osx. Why should it cripple itself on windows?
Again, this misguided assumption that by supporting something, it's magically crippled. I'd say Flash is far more crippling than ActiveX, for more reasons than one.
But if you feel that strongly, wait a bit, or go fix it yourself -- the source is available.
Valid points.
This issue doesn't touch me nearly as much -- I'm on Linux, and easily 99% of the software I use updates through a central package management system. Which means only one update program to run, or only one check made on boot -- far less duplication and bloat just for the updaters.
But then, Ubuntu didn't ask me to start checking for updates, it only asks me when it wants to install them.