I agree that Comcast's IE 5.whatever only support is complete and total BS. Heck, their support is way way way better than RoadRunner. What a POS their website is. Navigating that steaming pile makes me cringe everytime I see it, not to mention that hey, are you a OH or TX resident? Different pages for you. How about ID? WA? Nope, all of you get different pages too. Oh, and we won't tell you they're different, so you get to wallow through their unhelp section looking for how to access that meager amount of webpage space you get (it'll be the last time I do that). And then the hoops you have to jump through (why the F*@& would I want my personal web pages published to the RR directory?) to get the damn thing up online. All in the sense of making it "easier" for me. Right. Give me the damn path to load files and the URL to hit, and I'm good.
So the short of it is, despite having better daytime speeds than when Comcast had the network, I'm probably still going to bail for the new Fiber that AT&T is hawking. T1's, even partial, are just too expensive at the moment although I'm considering it.
It's standard where MS is concerned. Try it sometime. It was one of the major points that people were pointing to as failures in NT 4.x onward (when the GDI two other modules were moved into kernel space and all semblance of security and user space/kernel space separation vanished). Yes, you can lock up your entire GUI, even today with the latest XP service packs (don't know about Vista - fortunately I don't have to use it). I'm not sure if you can completely lock out network access post SP2, but you certainly could prior to SP2 (VNC would drop, Remote Desktop would drop, and direct calls to ports 135-139 would time out).
BTW, PMMail (a former OS/2 app) proved that the single-threaded behavior of the entire MS Office suite was entirely due to bad coding. Why do I say this? Because you can test the Exchange server download with PMMail and still have a highly responsive app, not to mention OS. (PMMail uses multiple threads splitting even I/O into their own threads, one for in, one for out, separated from the main app thread) No - I'm not associated in any way with PMMail, it's merely an example that shows how badly MS apps are coded with regard to multi-threadedness.
It's irrelevant if Explorer is multi-threaded or not, when it's dependency, GDI in whatever today's incarnation is, is single-threaded and just about everything runs through GDI or one of the incredibly useful shared libraries that are also single-threaded.
Let's say you're downloading a rather large file from an Exchange server with Outlook over a slow connection. Most of your GUI functions just ceased to respond. That's a sign of a badly coded system with single-threaded dependencies.
Think of it as 3 multi-lane freeways intersecting and the resulting gridlock when everything has to go through single lanes.
Based on your inability to see that, I can't really find a reason to say anything else to you. Being stupid like you are doesn't mean you can ignore reality, and discussing things like this with zealots like you bores me. Care to support that? Here's a clue: the fact that it took a flaming vehicle with 2 men rolling out of it the next day in Glasgow before they knew who they were would be indicative of the camera failures (Piccadilly Circus is, after all, such a quiet out of the way place). The identities were confirmed by records and fingerprints regarding the vehicles in London. Not a single mention of video (cue Underdog theme) "come to save the day".
But, then again, I'm "stupid" and "ignore reality", nothing like your own worshipful self who's mere stated opinion carries the weight of worlds and is firmly supported by reality merely by its utterance.
I take it you assume that cameras can do 100% ID from a picture. Also the people checking the IDs couldn't see they were fakes, and you'd assume they were trained for that. You're also assuming that a heap-load of information is known about passengers, and is correct. Kind of like the 1 year old that was denied boarding on a plane because he showed up on the no-fly list last month.
Now let's add to this heavenly scenario that even supposedly secure military systems are getting hacked, and think about how much trouble would be caused if this networked system was hacked and say, for instance, your face was superimposed on some wanted persons record the day you were to board a flight? Sounds like fun to me.
It's really helping in London, isn't it? The most CCTV'd city in the world. Yep, certainly helped with the 4 suicide bombers. The 4 wanna bees a couple of weeks later probably couldn't even be convicted (in the US) based on the relatively bad quality of the video (makes identification based on the video alone rather uncertain). They didn't help with the 2 this past month either, in detecting them nor in preventing their intended acts.
So, the only thing the London CCTV network has done to this point is do blanket surveillance on the populace, and maybe help nab a mugger or two (I won't discount this feature, but is this really worth the invasion of privacy?)
They did blast the footage of the terrorists in the news, we even saw it here in the US more than once across multiple days.
Wrong. Cameras could have seen them entering the airport, purchasing the ticket, boarding the plane. Real time CCTV law enforcement may be a ways off, but you are totally not giving this a chance. Reminds me of the missile defense thing, I kept hearing arguments like "This is physically impossible" which was complete rubbish.
So much for analytical thinking. Let's analyze that for a moment. At what exact point would you have known person 1 was a "terrorist" and not a normal passenger?
As for the missile defense system, that's a red herring. It's merely a question of creating a system physically able to target a moving object. That's leaps and bounds different from reading someone's mind or divining their intentions.
a) cameras don't stop terrorists. b) cameras won't even help after the fact, if they're a cell of suicide bombers. There's no one to track down.
Look at 9-11. They tracked down all 19 terrorists relatively quickly without invading other's privacy. In no way would 9-11 have been stopped with the surveillance system in place.
Camera footage does make for great fodder for the news though: "LOOK! Here they are, about to commit egregious violence on innocents" and then blast it 24/7 across the airwaves.
Such a system is a great way of spending great amounts of money and time and accomplishing little to nothing except terrorize your own populace and maybe throw a few innocents in jail to boot based on bad "evidence".
You're almost right - we actually paid billions for a fiber network we never received. Not only that, but it can be argued that the telco's don't actually own any of the wires/cables/fiber, as it was all subsidized by the people (taxes) one way or the other. The same can be said for cable companies.
While I'm not one to liek government interference in business, this really does appear to be a utility function that should be owned by the public, much the same as other municipality owned services.
I did read before I posted. Please don't make assumptions, as you have no knowledge of why I posted my statements.
From a high level, there's very little difference between Web Services and Web Interfaces on top of a web protocol when it comes to whether I can store something remotely. If you don't see that, then you're part of the problem that's generating those millions of obvious patents. From a user perspective, there's little difference between the two. From an implementation scenario, there's still little difference between the two: I make a request, then load a large amount of data, then retrieve it later from potentially a different machine. The specifics of getting it from A to B and back again are even relatively similar. They're certainly not patentable in my view. It's like saying "Ooo look, my tire uses Ti wheels instead of steel wheels - and I'm patenting that - no one else can use Ti wheels."
well, let's give you a couple of prior art examples:
Google's GMail - remember the GDrive plugin? MS's Outlook Web Access (yes, I know) with Public Folders - web interface, large data storage capabilities, and certainly scales.
There's just 2, and one dates back to around 1997.
So, not only is this not innovative, it's not even original. Slapping different access around the concept or doing it slightly better isn't innovative. Or are you going to argue that a titanium mousetrap is better because it's lighter and the trapping mechanism more rigid and more likely to snap the rodent's neck? That's not innovative, it's a refinement.
Exactly - Google isn't interested in hiring low-cost code-monkeys. It appears MS is, contrary to all protestations from MS. Having been involved with several firms that have abused^Wutilized the H1B visa program, I can honestly say that only 1 of them actually lived up to even half of what the program is supposed to be. The rest were utilized for cheap sweatshop labor doing tasks that most CS graduates could have accomplished after their sophomore year.
Then again, if I had to deal with MS's monolithic yet federated build scheme as described previously, I don't think I could make a better work case either. At some point, you just need to hire a bunch of shovelers to move the shit. Apparently MS hit that point at least 6 years ago.
Actually, I think it's more like Google is hogging all the talent in the western WA area. After all, you can go work as a small cog in the 80s/90s inertia machine that is MS for $90K/year, or go to quick and nimble and exciting tasks Google for $120K+. Heck, there are people who'd like to work at Google for less than 90K.
I doubt MS could hire/keep many competent programmers in the Redmond area of WA for under $200K that aren't already working for them. Heck, I'd probably go work at Expedia before MS, less corporate cruft.
First, gotta love the troll mod. You state facts, get modded troll.
setup? You mean the package installer, or the setup.exe? Or the msi script. Or the InstallAnywhere script? Or the drop the executable where ever install method? Or the blow up your system's registry installation script? Or any of the numerous other ways to screw up... err, install on Windows?
Wow, you're really out of touch there. Do you think it matters to an end user if they run setup.exe or setup.msi? Especially given that users either download and run the installer, or pop in a cd which runs it automatically. Most software for Windows is NOT the 'extract files somewhere' method. Let's see... out of the last 8 programs I installed on Windows... yep, not a single one had a setup.exe or setup.msi involved. Matter of fact, 3 had no install whatsoever. 2 were named exe's that performed whatever they needed to do. Then there were a couple of InstallAnywhere/Installshield installers. Oh, and let's not forget the MS named executable. That's just what I installed in the last 2 days. Oh, and this is all legitimate/legally installed software
I should probably also mentioned that wow, not a single CD among the bunch. And Autorun is the second thing I disable on any MS box I run, right after cleaning out the services.
Face it, there's no standard installer for any OS I'm aware of, not even Mac OSX, although they do have a pretty nice standard for application packaging, far better than MS's write everywhere installation. But that's just my opinion.
Um, yes there is. MSI IS the standard for Windows. Pretty much all software installations now use it, even if they look different. Again, if an application installs everywhere, that's the fault of the application developer. There are known best practices regarding what to put where. Interestingly enough, if more developers followed these standards, they'd not be triggering UAC in Windows Vista that some people have complained about. These standards haven't really changed much at all by the way since Windows 2000. Vista is simply now more forceful when an app does something it really shouldn't be.
By the way, if you target MSI, you can target Win2000 and higher. If you target deb, you get maybe a third of all linux distros. You still leave Redhat and Mandriva users in the cold. Sorry, your argument doesn't hold up. Actually, MSI is MS's stated standard, and they don't follow it. Could it be because it's insufficient to the task? I can't think of a single other reason they don't follow their own standard. Oh, and writing to MSI doesn't guarantee anything, as there are at least 3 versions I'm aware of out there, and maybe you don't want to install.NET 1.1 or 1.2?
But all that aside, it would be nice if there were a single packaging API for all linux distros. It would certainly make it simpler.
As for Vista being more forceful... no, it's more irritating in general. Yes, it's generally the developer's fault that all those UACs pop up. It's also MS's fault for encouraging those development practices in the first place. (Remember all those calls to "use the registry?)
And just so there's no mistake, I wasn't arguing about the situation, just pointing out that MS certainly isn't a shining ray of enlightment when it comes to installation, especially when you consider that it's a single vendor vs the multiple vendors in the Linux space.
Of course, what choice do they have if they want/need to run Windows? If enough of the system monitors your usage and activity, not using those services pretty much makes your computer a brick.
if the OS can function without an internet connection, it damn well better be able to function on a firewall that blocks access to MS servers. Don't run the software that "requires" those services. One of the first tasks of starting any new MS OS should be to remove about 60-80% of the "services" that start on startup. My XP installation runs between 4 and 9 services, and the 9 is a non-VM machine that I've not bothered to further shrink down yet.
Now, just simply block *all* access to those IPs. Of course... there goes your automatic updates as well. And you'd want automatic updates why? Much better to push updates out on your own schedule.
My experience was that installing software on Linux, especially when there wasn't even a packaged version of the software to install is nothing like Windows. With Windows, I run the setup program and I'm done. Most Linux software was not so friendly. setup? You mean the package installer, or the setup.exe? Or the msi script. Or the InstallAnywhere script? Or the drop the executable where ever install method? Or the blow up your system's registry installation script? Or any of the numerous other ways to screw up... err, install on Windows?
Face it, there's no standard installer for any OS I'm aware of, not even Mac OSX, although they do have a pretty nice standard for application packaging, far better than MS's write everywhere installation. But that's just my opinion.
Slashdot Mirror
I agree that Comcast's IE 5.whatever only support is complete and total BS. Heck, their support is way way way better than RoadRunner. What a POS their website is. Navigating that steaming pile makes me cringe everytime I see it, not to mention that hey, are you a OH or TX resident? Different pages for you. How about ID? WA? Nope, all of you get different pages too. Oh, and we won't tell you they're different, so you get to wallow through their unhelp section looking for how to access that meager amount of webpage space you get (it'll be the last time I do that). And then the hoops you have to jump through (why the F*@& would I want my personal web pages published to the RR directory?) to get the damn thing up online. All in the sense of making it "easier" for me. Right. Give me the damn path to load files and the URL to hit, and I'm good.
So the short of it is, despite having better daytime speeds than when Comcast had the network, I'm probably still going to bail for the new Fiber that AT&T is hawking. T1's, even partial, are just too expensive at the moment although I'm considering it.
Pidgin?
It's standard where MS is concerned. Try it sometime. It was one of the major points that people were pointing to as failures in NT 4.x onward (when the GDI two other modules were moved into kernel space and all semblance of security and user space/kernel space separation vanished). Yes, you can lock up your entire GUI, even today with the latest XP service packs (don't know about Vista - fortunately I don't have to use it). I'm not sure if you can completely lock out network access post SP2, but you certainly could prior to SP2 (VNC would drop, Remote Desktop would drop, and direct calls to ports 135-139 would time out).
BTW, PMMail (a former OS/2 app) proved that the single-threaded behavior of the entire MS Office suite was entirely due to bad coding. Why do I say this? Because you can test the Exchange server download with PMMail and still have a highly responsive app, not to mention OS. (PMMail uses multiple threads splitting even I/O into their own threads, one for in, one for out, separated from the main app thread) No - I'm not associated in any way with PMMail, it's merely an example that shows how badly MS apps are coded with regard to multi-threadedness.
Actually, I said I was under the impression they would do one or the other, and that's given their previous history.
I have no idea what posting/blog you're referring to. Perhaps you would do well not to jump to conclusions and think someone's "misinformed"?
It's irrelevant if Explorer is multi-threaded or not, when it's dependency, GDI in whatever today's incarnation is, is single-threaded and just about everything runs through GDI or one of the incredibly useful shared libraries that are also single-threaded.
Let's say you're downloading a rather large file from an Exchange server with Outlook over a slow connection. Most of your GUI functions just ceased to respond. That's a sign of a badly coded system with single-threaded dependencies.
Think of it as 3 multi-lane freeways intersecting and the resulting gridlock when everything has to go through single lanes.
I was under the impression they were looking at doing precisely that, or at least the hardware specs.
Yes, it seems to be.
Based on your inability to see that, I can't really find a reason to say anything else to you. Being stupid like you are doesn't mean you can ignore reality, and discussing things like this with zealots like you bores me. Care to support that? Here's a clue: the fact that it took a flaming vehicle with 2 men rolling out of it the next day in Glasgow before they knew who they were would be indicative of the camera failures (Piccadilly Circus is, after all, such a quiet out of the way place). The identities were confirmed by records and fingerprints regarding the vehicles in London. Not a single mention of video (cue Underdog theme) "come to save the day".
But, then again, I'm "stupid" and "ignore reality", nothing like your own worshipful self who's mere stated opinion carries the weight of worlds and is firmly supported by reality merely by its utterance.
WebSphere might be salvageable by stripping out some of those proprietary add-ons. Lotus Notes however would have 0 developers.
I'd mod you up if I had points and I hadn't already posted... That's funny!
Hey, if it smells like 3 day old fish, it's fish.
I take it you assume that cameras can do 100% ID from a picture. Also the people checking the IDs couldn't see they were fakes, and you'd assume they were trained for that. You're also assuming that a heap-load of information is known about passengers, and is correct. Kind of like the 1 year old that was denied boarding on a plane because he showed up on the no-fly list last month.
Now let's add to this heavenly scenario that even supposedly secure military systems are getting hacked, and think about how much trouble would be caused if this networked system was hacked and say, for instance, your face was superimposed on some wanted persons record the day you were to board a flight? Sounds like fun to me.
1) production is easier than you think
2) small and portable is hard, something most don't do
Unfortunately, it takes very little knowledge to be "successful" in such an endeavor.
It's really helping in London, isn't it? The most CCTV'd city in the world. Yep, certainly helped with the 4 suicide bombers. The 4 wanna bees a couple of weeks later probably couldn't even be convicted (in the US) based on the relatively bad quality of the video (makes identification based on the video alone rather uncertain). They didn't help with the 2 this past month either, in detecting them nor in preventing their intended acts.
So, the only thing the London CCTV network has done to this point is do blanket surveillance on the populace, and maybe help nab a mugger or two (I won't discount this feature, but is this really worth the invasion of privacy?)
They did blast the footage of the terrorists in the news, we even saw it here in the US more than once across multiple days.
Wrong. Cameras could have seen them entering the airport, purchasing the ticket, boarding the plane. Real time CCTV law enforcement may be a ways off, but you are totally not giving this a chance. Reminds me of the missile defense thing, I kept hearing arguments like "This is physically impossible" which was complete rubbish.
So much for analytical thinking. Let's analyze that for a moment. At what exact point would you have known person 1 was a "terrorist" and not a normal passenger?
As for the missile defense system, that's a red herring. It's merely a question of creating a system physically able to target a moving object. That's leaps and bounds different from reading someone's mind or divining their intentions.
2 things to add:
a) cameras don't stop terrorists.
b) cameras won't even help after the fact, if they're a cell of suicide bombers. There's no one to track down.
Look at 9-11. They tracked down all 19 terrorists relatively quickly without invading other's privacy. In no way would 9-11 have been stopped with the surveillance system in place.
Camera footage does make for great fodder for the news though: "LOOK! Here they are, about to commit egregious violence on innocents" and then blast it 24/7 across the airwaves.
Such a system is a great way of spending great amounts of money and time and accomplishing little to nothing except terrorize your own populace and maybe throw a few innocents in jail to boot based on bad "evidence".
You're almost right - we actually paid billions for a fiber network we never received. Not only that, but it can be argued that the telco's don't actually own any of the wires/cables/fiber, as it was all subsidized by the people (taxes) one way or the other. The same can be said for cable companies.
While I'm not one to liek government interference in business, this really does appear to be a utility function that should be owned by the public, much the same as other municipality owned services.
US sales should have a big jump this month. Fry's appears to be clearancing them as well this week. There was a shopping cart about half full of them.
I did read before I posted. Please don't make assumptions, as you have no knowledge of why I posted my statements.
From a high level, there's very little difference between Web Services and Web Interfaces on top of a web protocol when it comes to whether I can store something remotely. If you don't see that, then you're part of the problem that's generating those millions of obvious patents. From a user perspective, there's little difference between the two. From an implementation scenario, there's still little difference between the two: I make a request, then load a large amount of data, then retrieve it later from potentially a different machine. The specifics of getting it from A to B and back again are even relatively similar. They're certainly not patentable in my view. It's like saying "Ooo look, my tire uses Ti wheels instead of steel wheels - and I'm patenting that - no one else can use Ti wheels."
well, let's give you a couple of prior art examples:
Google's GMail - remember the GDrive plugin?
MS's Outlook Web Access (yes, I know) with Public Folders - web interface, large data storage capabilities, and certainly scales.
There's just 2, and one dates back to around 1997.
So, not only is this not innovative, it's not even original. Slapping different access around the concept or doing it slightly better isn't innovative. Or are you going to argue that a titanium mousetrap is better because it's lighter and the trapping mechanism more rigid and more likely to snap the rodent's neck? That's not innovative, it's a refinement.
Exactly - Google isn't interested in hiring low-cost code-monkeys. It appears MS is, contrary to all protestations from MS. Having been involved with several firms that have abused^Wutilized the H1B visa program, I can honestly say that only 1 of them actually lived up to even half of what the program is supposed to be. The rest were utilized for cheap sweatshop labor doing tasks that most CS graduates could have accomplished after their sophomore year.
Then again, if I had to deal with MS's monolithic yet federated build scheme as described previously, I don't think I could make a better work case either. At some point, you just need to hire a bunch of shovelers to move the shit. Apparently MS hit that point at least 6 years ago.
Actually, I think it's more like Google is hogging all the talent in the western WA area. After all, you can go work as a small cog in the 80s/90s inertia machine that is MS for $90K/year, or go to quick and nimble and exciting tasks Google for $120K+. Heck, there are people who'd like to work at Google for less than 90K.
I doubt MS could hire/keep many competent programmers in the Redmond area of WA for under $200K that aren't already working for them. Heck, I'd probably go work at Expedia before MS, less corporate cruft.
Wow, you're really out of touch there. Do you think it matters to an end user if they run setup.exe or setup.msi? Especially given that users either download and run the installer, or pop in a cd which runs it automatically. Most software for Windows is NOT the 'extract files somewhere' method. Let's see... out of the last 8 programs I installed on Windows... yep, not a single one had a setup.exe or setup.msi involved. Matter of fact, 3 had no install whatsoever. 2 were named exe's that performed whatever they needed to do. Then there were a couple of InstallAnywhere/Installshield installers. Oh, and let's not forget the MS named executable. That's just what I installed in the last 2 days. Oh, and this is all legitimate/legally installed software
I should probably also mentioned that wow, not a single CD among the bunch. And Autorun is the second thing I disable on any MS box I run, right after cleaning out the services. Face it, there's no standard installer for any OS I'm aware of, not even Mac OSX, although they do have a pretty nice standard for application packaging, far better than MS's write everywhere installation. But that's just my opinion.
Um, yes there is. MSI IS the standard for Windows. Pretty much all software installations now use it, even if they look different. Again, if an application installs everywhere, that's the fault of the application developer. There are known best practices regarding what to put where. Interestingly enough, if more developers followed these standards, they'd not be triggering UAC in Windows Vista that some people have complained about. These standards haven't really changed much at all by the way since Windows 2000. Vista is simply now more forceful when an app does something it really shouldn't be.
By the way, if you target MSI, you can target Win2000 and higher. If you target deb, you get maybe a third of all linux distros. You still leave Redhat and Mandriva users in the cold. Sorry, your argument doesn't hold up. Actually, MSI is MS's stated standard, and they don't follow it. Could it be because it's insufficient to the task? I can't think of a single other reason they don't follow their own standard. Oh, and writing to MSI doesn't guarantee anything, as there are at least 3 versions I'm aware of out there, and maybe you don't want to install
But all that aside, it would be nice if there were a single packaging API for all linux distros. It would certainly make it simpler.
As for Vista being more forceful... no, it's more irritating in general. Yes, it's generally the developer's fault that all those UACs pop up. It's also MS's fault for encouraging those development practices in the first place. (Remember all those calls to "use the registry?)
And just so there's no mistake, I wasn't arguing about the situation, just pointing out that MS certainly isn't a shining ray of enlightment when it comes to installation, especially when you consider that it's a single vendor vs the multiple vendors in the Linux space.
if the OS can function without an internet connection, it damn well better be able to function on a firewall that blocks access to MS servers. Don't run the software that "requires" those services. One of the first tasks of starting any new MS OS should be to remove about 60-80% of the "services" that start on startup. My XP installation runs between 4 and 9 services, and the 9 is a non-VM machine that I've not bothered to further shrink down yet.
Face it, there's no standard installer for any OS I'm aware of, not even Mac OSX, although they do have a pretty nice standard for application packaging, far better than MS's write everywhere installation. But that's just my opinion.
Dropping down on coffee/tea/soda/caffiene carriers myself. Life seems to be moving at about 3/2 speed at the moment.