Try working as a developer on a large corporate software product. An average build in around a gig a pop, and you either have people "backing" to them, or copying them wholesale. Add in the fact that these patterns are also really bursty (based on build publications) and you are talking serious overhead.
We don't use Samba as the primary fileserver, but the majority of the windows developers use a Samba mirror (or gateway) to the backing tree.
If copying a build goes from 20 minutes to 10 minutes, and then you multiply this across the number of users, you get a signifigant time savings. (Especially for things like build publications, because the temptation to waste time "while the build copies" is pretty high.:) )
I'll agree the analogy is not perfect, but the point is that there is are a lot of factors besides engine size or CPU frequency to consider.
Oh, and the Type R Civic displaces more (trivially) than the S2000, and is slower in both pickup and top speed. While they top out similarly, the.6 seconds faster from 0-60 means if both cars are stock, the Civic gets smoked.
Civic Type R
displacement: 1998 cc
top speed : 146 mph
0 - 60 : 6.8 seconds
S2000
displacement: 1997 cc
top speed : 150 mph
0 - 60 : 6.2 seconds
When I was growing up, I lived an a very friendly neighborhood in the Midwest. Everyone used to sit out on their front porches in the evenings both for the social contact and because air-conditioning was pretty rare at the time. (in our neighborhood at least)
The neighborhood kids and I had tons of fun each summer listening for a call "coming down the block". You could actually hear a sales call working its way from house to house, and amazingly they usually went by street address (accending).
So, when we heard a call we'd all take off "racing the call". The idea was to get to a each house right before the call got there. If you were successful (and the house was someone willing to play along) you picked up the phone and instead of saying "hello" or something, you'd say something along the lines of "we don't want any!" *click*. And then off to the next house we'd race.
It was great fun to listen to the telemarketer getting more and more confused as to what was going on. I have no idea if it cut down the calls, but it was great fun.
Yeah, it might be the same argument, but either way it is fairly pointless. This is why benchmarking is such a controversial subject. Do you measure pure operations per second? If so, which ops? Or do you measure the actual wallclock time it takes for real world programs to execute a set of common tasks? Again, which programs and which tasks? (Which doesn't even begin to get into the all-to-real problem of vendors adding hacks to screw with the benchmark ala the video card arms race....)
I don't know jack about Macs, but it wouldn't surprise me if some marketing drone had claimed that. They might have even had a few examples to back it up.
Ultimately, "speed" isn't really a singularly quantifiable entity. I can think of at least three different ways to measure speed:
1) Pure CPU operations per second. Good for hard math, but only part of the equation
2) Hardware speed across the entire system. If the CPU screams, but the memory subsystem drags ass, the usage speed is slower. I doubt VT's cluster would be very cool if it was using 9600 baud modems for interconnects.:)
3) Perceived speed. How fast does it feel like it runs? For example the Transmeta chips which "learn" and optimize will feel slower or faster depending on if the code has been run recently, but the hardware speed hasn't changed.
As a lot of others have pointed out, there is a lot more to "speed" that the frequency of the clockchip. I use a lot of IBM "P-series" (what used to be called RS/6000) machines at work. The clock speeds are generally low (sub 1Ghz) but in certain situtations these machines absolutely smoke. In other situtations, such are running Java, they drag ass. A lot has to do with how well the application using the processor takes advantage of the low-level capabilites.
Think of computers like cars. The Honda S2000 only has between 2 and 2.2 liters of displacement, but it will smoke most other cars which displace equal amounts because of how it is tuned, geared, built, etc...
More clock speed is in many respects a cop-out when it comes to performance. Faster chips give you faster code, but as the chips get faster, the code gets slower. Moore's law applies to hardware, not software.
Man, did I get off-topic or what? Anyway, in summary: yes, the G5 is faster clockwise, but there are a lot of other factors.
When I was in high school (at least 9 years ago) my dad's office got a copier which required a phone line. No one knew why till the techs started magically showing up to fix things.
Re:Yeah, but they should be making AIX boot faster
on
Booting Linux Faster
·
· Score: 1
I'm willing to bet this was a "J40" right? nice squat beige box, about 8 procs at 20Mhz a piece? I used to have one of those at work. The reason it's so slow (I'm told) is that it was one of the earlier SMP machines in that line.
There was a setting somewhere in SMIT which whould reduce the boot time to a much more managable 2-3 minutes. The 20 minutes was the result of each of the procs being verified in series... AIX wasn't even starting.
You know, it's funny.... I've never done any "extreme" programming, and I too tend to get into the "do-not-talk-to-me-I'm-coding" mode. None the less, I've always been interested in trying XP or one of its variants. When I was a child, I spent countless hours programming in pairs with a succession of older, smarter kids. I really think that those early sessions played a huge part in my life. So, maybe the company doesn't win, but I wouldn't mind...:)
Peter Gutmann is a serious expert. I write security code for a living. (For IBM) Peter Gutmann has writen a few seminal papers such as "A Layman's Guide to ASN.1" which is required reading for anyone coming on the team.
Well, yes you usually know how much you get off the wire from the hardware, but it's the handling of the data afterwards which gets dicey. For example, the portion of SSH fixed had to do with a buffer which grew dynamically as additional information was added. Most buffer overflows are the result of arithimatic errors, not simply "reading more than they can". Think of it like doing a very long equation. Sure, you can mess up in the first steps, the but chances are much greater you're gonna forget to carry a remainder in step 213.
While your firewall machines may not be running Linux, the IPFilter function has the ability to limit accessibility based on frequency. For example, you can say:
"allow no more than X ping requests from Y in Z seconds"
Of course, you shouldn't have to resort to measure like this to deal with these specific people, but it does send a message.
So, we're talking 6 grand before you get into the incidentals. That's not exactly mass-market pricing. And I'm sure if you live in an apartment, your neighbors will love the rumbling bass and subsonics...
The problem is that the film industry has lost sight of what they "sell". At the simpliest level, they are selling: a story projected on a huge screen, sound better than you have at home, and a decent seating environment. No foreseeable technology is going to allow the general public to have the "big screen" experience at home in the near future. Therefore, the movie industry has something of unique value.
The problem is that as that by expanding into the home market, they gave up a lot of their uniqueness. When the cost of creating a copy of a movie for home use was high, they could make money because they could do it cheaply. However, as the cost to distribute lower quality formats falls, the "value" the studios offer to home users pluments.
Now, I'm certainly not denying the studio's invested a lot of capital and "own" the movie, but think of it like this: I go see a famous comedian in a club, and remember/write down all his jokes. I can go tell those jokes to my friends, or type them up and email them across the internet. Chances are, even though they are still funny, they are much better when you see the actual comedian perform them.
There is a reason film and music companies are called "media" companies. The idea is that they provide the "medium" which conveys content to end-users. Medium used to be expensive, now it's cheap. Their business model is broken, they spend tons on content and are trying to profit of the medium.
Actually, I've been playing with Thunderbird, and setting up encryption is pretty damn easy.
I work for a fairly "hip" company (IBM) and we have this nifty website you can go to, fill out a form, and they send you a nice little PKCS#12 file, signed by a real root cert and everything. To use this with Thunderbird, all you have to do is: "Manage Certificates" -> "Import" -> (password) -> done.
Of course, if you want to generate your own keys, that's a little harder, but nothing a simple script couldn't be created to do.
I'm sure I'll be lost at the bottom of the heap, but seeing as how the passenger section of commerical jets is basically a big metal tube, and we've already established a nice firm wall between the cabin and the cockpit, why not just go the extra mile and make it a Faraday cage? The sensors are outside the cabin, so they can still "sense", and inside the passengers can be as electronically noisy as they like. (Until they start interferring with each other) Of course you'd have to have something like screens over the windows....
This idea might even provide some shielding from some of the milder radation passengers are exposed to environmentally. (But likely not much, I suspect the soft stuff is already blocked by the aluminum skin of the plan, and the bad stuff tends to be more like gamma rays...)
I completely, totally, and utterly disagree with the above. I work at IBM now, and I can promise you there are very firm policies in place about not monitoring employee activities without a good cause and serious management oversight.
If you do something obviously stupid, and people see and complain, you will get looked at. But remember, if someone is doing something like looking a p0rn at work and the workplace doesn't take action, then the employer becomes liable for creating a "hostile" workplace.
Contrary to a lot of public percerption, IBM is very liberal. The phrases "open-door policy" and "an equal day's work for an equal day's pay" both were coined by Watson. We've recognized same-sex unions for years, had company anti-discrimation policies long before it was the expected thing to do. I know I sound like a raving fanboy, and I'll be the first to admit that IBM also has its share of large company bureaucratic BS, but the important things which make my job pleasant on a human level are always done well.
Twice a year? Heh... we do it a _lot_ more often than that. It's kinda good, and kinda bad. The sight of someone rattling drawers in a dark office at odd hours is not so uncommmon, and thus it surprises you less.
It could be worse, I've been to IBM sites in the UK where the IBM police will actually stop you in the halls and check your badge if they don't know you.
To folks that don't work here, it seems extreme, but I know the number to get my AC turned on too. 15 RS/6000 machines in an office makes it toasty after hours. And since they range from 5K to 50K a piece, I'm glad security is keeping an eye on them. If they didn't, it's like that Aussie case posted yesterday.:)
Oh, and the penny thing must have been a local manager in overdrive. I know people with dollar bills, currency from other countries, etc on their desk. No big deal. They only care about high-value stuff. PDA's, laptops, etc.... because if it walks off, the workplace gets hostile and suspicious in a hurry...
I'm a current IBM employee. And yes, the rules are pretty strict about leaving stuff out unattended. But you have to remember we have a zillion different customers from all over the world. Some of them, like major banks, governments, etc, tend to stipulate a certain level of general security "cleanliness" as part of their contract. When security is a habit, you don't make dumb mistakes. When you only do it when you remember you're on a sensitive projuect, you're more likely to slip.
It's a pain, but you adjust quickly. I have a locked drawer on my desk. End of the day, everything paper on my desk goes in my desk. The next day, stuff comes out as I need it. Every coupla months, the drawer gets full. And all that semi-sensitive stuff goes en-masse to the confidential recycling bin. Clean office, and no slip-up's from double stakcing papers, etc...
I have actually been spending the last few days looking at the VIA EPIA mobo's. For less than $200 you can get everything you need except RAM, HD, and PSU. The thing keeping me from pulling the trigger is a kinda yucky trend of so-so linux support. For example, the M10000 (Which is the 1Ghz VIA system) is pretty cool (even though it needs a fan, unlike the lower end models), but the hardware MPEG is not supported under linux... And it looks like VIA is willing to take compatability where they can find it. If you read a lot of the forums, it seems like a lot of hackers are using winXP instead of linux.
Okay... encrypting backups is a Good Thing(tm) but if the backups came from is reasonably high-end, then you still have a problem. Big-iron doesn't sit around at the local Fry's. The lead time required to buy a million dollar server is substantial. And this assumes you have the capital to do so... And what do you say in the mean time? "We're sorry, Customs is closed for 6-8 weeks. Our new server is in the mail".
I'm not bashing your point, the number of people burned by lacking backups is a LOT larger than the number burned by someone walking (rolling) out with a mainframe. Backups are good. Encrypting them is better. But losing hardware, especially hard-to-replace hardware, is pretty stupid.
I had to visit the data center for a major financial center in Jersey City, NJ shortly after WTC. (A lot of the big iron is across the river from Manhattan... for price reasons more than security) Because of the sudden lack of available downtown office space, every available empty space in Jersey City was suddenly rented out.
So... I walked into see my customer. I was surprised a the new security in place. I showed my company badge, signed in, and was lead to a desk under a sign marked "High Value Transactions". Plopped me right down in front of a terminal. I was really confused. The setup was totally different than what I was expecting from previous visits. So I started looking around for people I knew, etc... After about 10 minutes I realized I was in the data center for the WRONG company!
So I got up and left. I have no idea how long I could have stayed there, or what I could have done. I suspect that if I had gotten out a screwdriver, I could have likely started shopping for hardware.
Moral of the story: chaos breeds insecurity, and an "official" plastic badge with your picture on it is shockingly powerful.
And the "can't happen to me" and "just a little won't hurt" mentalities build up too. When I was in high school I had a job cleaning environmental air samplers. I don't want to think about how much hexane I inhaled in those 4 years. I finally jury-rigged a vent hood out of a cardboard box and a box fan.
No health problems yet (10 years later) but I'll always wonder...
Slashdot Mirror
Try working as a developer on a large corporate software product. An average build in around a gig a pop, and you either have people "backing" to them, or copying them wholesale. Add in the fact that these patterns are also really bursty (based on build publications) and you are talking serious overhead.
:) )
We don't use Samba as the primary fileserver, but the majority of the windows developers use a Samba mirror (or gateway) to the backing tree.
If copying a build goes from 20 minutes to 10 minutes, and then you multiply this across the number of users, you get a signifigant time savings. (Especially for things like build publications, because the temptation to waste time "while the build copies" is pretty high.
I'll agree the analogy is not perfect, but the point is that there is are a lot of factors besides engine size or CPU frequency to consider.
.6 seconds faster from 0-60 means if both cars are stock, the Civic gets smoked.
Oh, and the Type R Civic displaces more (trivially) than the S2000, and is slower in both pickup and top speed. While they top out similarly, the
Civic Type R
displacement: 1998 cc
top speed : 146 mph
0 - 60 : 6.8 seconds
S2000
displacement: 1997 cc
top speed : 150 mph
0 - 60 : 6.2 seconds
When I was growing up, I lived an a very friendly neighborhood in the Midwest. Everyone used to sit out on their front porches in the evenings both for the social contact and because air-conditioning was pretty rare at the time. (in our neighborhood at least)
The neighborhood kids and I had tons of fun each summer listening for a call "coming down the block". You could actually hear a sales call working its way from house to house, and amazingly they usually went by street address (accending).
So, when we heard a call we'd all take off "racing the call". The idea was to get to a each house right before the call got there. If you were successful (and the house was someone willing to play along) you picked up the phone and instead of saying "hello" or something, you'd say something along the lines of "we don't want any!" *click*. And then off to the next house we'd race.
It was great fun to listen to the telemarketer getting more and more confused as to what was going on. I have no idea if it cut down the calls, but it was great fun.
Yeah, it might be the same argument, but either way it is fairly pointless. This is why benchmarking is such a controversial subject. Do you measure pure operations per second? If so, which ops? Or do you measure the actual wallclock time it takes for real world programs to execute a set of common tasks? Again, which programs and which tasks? (Which doesn't even begin to get into the all-to-real problem of vendors adding hacks to screw with the benchmark ala the video card arms race....)
:)
I don't know jack about Macs, but it wouldn't surprise me if some marketing drone had claimed that. They might have even had a few examples to back it up.
Ultimately, "speed" isn't really a singularly quantifiable entity. I can think of at least three different ways to measure speed:
1) Pure CPU operations per second. Good for hard math, but only part of the equation
2) Hardware speed across the entire system. If the CPU screams, but the memory subsystem drags ass, the usage speed is slower. I doubt VT's cluster would be very cool if it was using 9600 baud modems for interconnects.
3) Perceived speed. How fast does it feel like it runs? For example the Transmeta chips which "learn" and optimize will feel slower or faster depending on if the code has been run recently, but the hardware speed hasn't changed.
As a lot of others have pointed out, there is a lot more to "speed" that the frequency of the clockchip. I use a lot of IBM "P-series" (what used to be called RS/6000) machines at work. The clock speeds are generally low (sub 1Ghz) but in certain situtations these machines absolutely smoke. In other situtations, such are running Java, they drag ass. A lot has to do with how well the application using the processor takes advantage of the low-level capabilites.
Think of computers like cars. The Honda S2000 only has between 2 and 2.2 liters of displacement, but it will smoke most other cars which displace equal amounts because of how it is tuned, geared, built, etc...
More clock speed is in many respects a cop-out when it comes to performance. Faster chips give you faster code, but as the chips get faster, the code gets slower. Moore's law applies to hardware, not software.
Man, did I get off-topic or what? Anyway, in summary: yes, the G5 is faster clockwise, but there are a lot of other factors.
When I was in high school (at least 9 years ago) my dad's office got a copier which required a phone line. No one knew why till the techs started magically showing up to fix things.
I'm willing to bet this was a "J40" right? nice squat beige box, about 8 procs at 20Mhz a piece? I used to have one of those at work. The reason it's so slow (I'm told) is that it was one of the earlier SMP machines in that line.
There was a setting somewhere in SMIT which whould reduce the boot time to a much more managable 2-3 minutes. The 20 minutes was the result of each of the procs being verified in series... AIX wasn't even starting.
You know, it's funny.... I've never done any "extreme" programming, and I too tend to get into the "do-not-talk-to-me-I'm-coding" mode. None the less, I've always been interested in trying XP or one of its variants. When I was a child, I spent countless hours programming in pairs with a succession of older, smarter kids. I really think that those early sessions played a huge part in my life. So, maybe the company doesn't win, but I wouldn't mind... :)
Peter Gutmann is a serious expert. I write security code for a living. (For IBM) Peter Gutmann has writen a few seminal papers such as "A Layman's Guide to ASN.1" which is required reading for anyone coming on the team.
Well, yes you usually know how much you get off the wire from the hardware, but it's the handling of the data afterwards which gets dicey. For example, the portion of SSH fixed had to do with a buffer which grew dynamically as additional information was added. Most buffer overflows are the result of arithimatic errors, not simply "reading more than they can". Think of it like doing a very long equation. Sure, you can mess up in the first steps, the but chances are much greater you're gonna forget to carry a remainder in step 213.
While your firewall machines may not be running Linux, the IPFilter function has the ability to limit accessibility based on frequency. For example, you can say:
"allow no more than X ping requests from Y in Z seconds"
Of course, you shouldn't have to resort to measure like this to deal with these specific people, but it does send a message.
Okay... I know...feeding the trolls.
42" plasma =~ $4000
Decent Home Theater =~ $2000
So, we're talking 6 grand before you get into the incidentals. That's not exactly mass-market pricing. And I'm sure if you live in an apartment, your neighbors will love the rumbling bass and subsonics...
The problem is that the film industry has lost sight of what they "sell". At the simpliest level, they are selling: a story projected on a huge screen, sound better than you have at home, and a decent seating environment. No foreseeable technology is going to allow the general public to have the "big screen" experience at home in the near future. Therefore, the movie industry has something of unique value.
The problem is that as that by expanding into the home market, they gave up a lot of their uniqueness. When the cost of creating a copy of a movie for home use was high, they could make money because they could do it cheaply. However, as the cost to distribute lower quality formats falls, the "value" the studios offer to home users pluments.
Now, I'm certainly not denying the studio's invested a lot of capital and "own" the movie, but think of it like this: I go see a famous comedian in a club, and remember/write down all his jokes. I can go tell those jokes to my friends, or type them up and email them across the internet. Chances are, even though they are still funny, they are much better when you see the actual comedian perform them.
There is a reason film and music companies are called "media" companies. The idea is that they provide the "medium" which conveys content to end-users. Medium used to be expensive, now it's cheap. Their business model is broken, they spend tons on content and are trying to profit of the medium.
Actually, I've been playing with Thunderbird, and setting up encryption is pretty damn easy.
I work for a fairly "hip" company (IBM) and we have this nifty website you can go to, fill out a form, and they send you a nice little PKCS#12 file, signed by a real root cert and everything. To use this with Thunderbird, all you have to do is: "Manage Certificates" -> "Import" -> (password) -> done.
Of course, if you want to generate your own keys, that's a little harder, but nothing a simple script couldn't be created to do.
I'm sure I'll be lost at the bottom of the heap, but seeing as how the passenger section of commerical jets is basically a big metal tube, and we've already established a nice firm wall between the cabin and the cockpit, why not just go the extra mile and make it a Faraday cage? The sensors are outside the cabin, so they can still "sense", and inside the passengers can be as electronically noisy as they like. (Until they start interferring with each other) Of course you'd have to have something like screens over the windows....
This idea might even provide some shielding from some of the milder radation passengers are exposed to environmentally. (But likely not much, I suspect the soft stuff is already blocked by the aluminum skin of the plan, and the bad stuff tends to be more like gamma rays...)
I completely, totally, and utterly disagree with the above. I work at IBM now, and I can promise you there are very firm policies in place about not monitoring employee activities without a good cause and serious management oversight.
If you do something obviously stupid, and people see and complain, you will get looked at. But remember, if someone is doing something like looking a p0rn at work and the workplace doesn't take action, then the employer becomes liable for creating a "hostile" workplace.
Contrary to a lot of public percerption, IBM is very liberal. The phrases "open-door policy" and "an equal day's work for an equal day's pay" both were coined by Watson. We've recognized same-sex unions for years, had company anti-discrimation policies long before it was the expected thing to do. I know I sound like a raving fanboy, and I'll be the first to admit that IBM also has its share of large company bureaucratic BS, but the important things which make my job pleasant on a human level are always done well.
Twice a year? Heh... we do it a _lot_ more often than that. It's kinda good, and kinda bad. The sight of someone rattling drawers in a dark office at odd hours is not so uncommmon, and thus it surprises you less.
:)
It could be worse, I've been to IBM sites in the UK where the IBM police will actually stop you in the halls and check your badge if they don't know you.
To folks that don't work here, it seems extreme, but I know the number to get my AC turned on too. 15 RS/6000 machines in an office makes it toasty after hours. And since they range from 5K to 50K a piece, I'm glad security is keeping an eye on them. If they didn't, it's like that Aussie case posted yesterday.
Oh, and the penny thing must have been a local manager in overdrive. I know people with dollar bills, currency from other countries, etc on their desk. No big deal. They only care about high-value stuff. PDA's, laptops, etc.... because if it walks off, the workplace gets hostile and suspicious in a hurry...
I'm a current IBM employee. And yes, the rules are pretty strict about leaving stuff out unattended. But you have to remember we have a zillion different customers from all over the world. Some of them, like major banks, governments, etc, tend to stipulate a certain level of general security "cleanliness" as part of their contract. When security is a habit, you don't make dumb mistakes. When you only do it when you remember you're on a sensitive projuect, you're more likely to slip.
It's a pain, but you adjust quickly. I have a locked drawer on my desk. End of the day, everything paper on my desk goes in my desk. The next day, stuff comes out as I need it. Every coupla months, the drawer gets full. And all that semi-sensitive stuff goes en-masse to the confidential recycling bin. Clean office, and no slip-up's from double stakcing papers, etc...
I have actually been spending the last few days looking at the VIA EPIA mobo's. For less than $200 you can get everything you need except RAM, HD, and PSU. The thing keeping me from pulling the trigger is a kinda yucky trend of so-so linux support. For example, the M10000 (Which is the 1Ghz VIA system) is pretty cool (even though it needs a fan, unlike the lower end models), but the hardware MPEG is not supported under linux... And it looks like VIA is willing to take compatability where they can find it. If you read a lot of the forums, it seems like a lot of hackers are using winXP instead of linux.
Okay... encrypting backups is a Good Thing(tm) but if the backups came from is reasonably high-end, then you still have a problem. Big-iron doesn't sit around at the local Fry's. The lead time required to buy a million dollar server is substantial. And this assumes you have the capital to do so... And what do you say in the mean time? "We're sorry, Customs is closed for 6-8 weeks. Our new server is in the mail".
I'm not bashing your point, the number of people burned by lacking backups is a LOT larger than the number burned by someone walking (rolling) out with a mainframe. Backups are good. Encrypting them is better. But losing hardware, especially hard-to-replace hardware, is pretty stupid.
I had to visit the data center for a major financial center in Jersey City, NJ shortly after WTC. (A lot of the big iron is across the river from Manhattan... for price reasons more than security) Because of the sudden lack of available downtown office space, every available empty space in Jersey City was suddenly rented out.
So... I walked into see my customer. I was surprised a the new security in place. I showed my company badge, signed in, and was lead to a desk under a sign marked "High Value Transactions". Plopped me right down in front of a terminal. I was really confused. The setup was totally different than what I was expecting from previous visits. So I started looking around for people I knew, etc... After about 10 minutes I realized I was in the data center for the WRONG company!
So I got up and left. I have no idea how long I could have stayed there, or what I could have done. I suspect that if I had gotten out a screwdriver, I could have likely started shopping for hardware.
Moral of the story: chaos breeds insecurity, and an "official" plastic badge with your picture on it is shockingly powerful.
And the "can't happen to me" and "just a little won't hurt" mentalities build up too. When I was in high school I had a job cleaning environmental air samplers. I don't want to think about how much hexane I inhaled in those 4 years. I finally jury-rigged a vent hood out of a cardboard box and a box fan.
No health problems yet (10 years later) but I'll always wonder...
I have to admit I agree.... but it's becoming more common. I work for IBM (so obviously we tend to use that feature more than some people....)
You can set Notes up to user X.509v3 certs which makes life easier. Of course establising the trust model becomes fun...
If your company uses an email system like Lotus Notes, which has built in encryption, signing, etc... then a far amount.
Of course this is MS so....