-->Aren't you a little curious about the PPC970 and what kind of performance to expect?
So far, nothing impressive. if you have any impressive thing about it, would you mind sharing it?
-->Why would you even write a three-paragraph post on a subject you have no clue about?
When we talk about things to come, it is easy to do mud-slinging since nothing can be proved. but that is not my style, so i will give you benefit of doubt. may be i am totally clueless.
-->Apple did it before anyone. Ever hear of the Newton?
I am old enough to have used GM refrigerator. That doesn't mean I would tell everyone GM makes refrigerator. Apple Newton is a thing of past. It survived when there wasn't any competition. Once competition sprang, it was dead. i am talking about today. this is one market segment, where style and differentiation still matters, for which apple is famous and they can make some inroad. Apple doesn't sell any PDA today, but I won't be surprised if they come out with one soon.
--> The PowerPC isn't a proprietary processor...
It is as proprietary as Sparc.
-->Apple will not destroy its PPC customer-...
Didn't they destroy Motorola 68k base?
...and provide Mac OS X for a tightly proscribed x86 configuration...
Why would apple make Mac OS for x86 now, if it is not for a fact that x86 hardware would be cheaper? Since MacOS would only work with Apple prescribed config, "installed base" could not be a valid cause.
...Microsoft as the dominant player in the x86 market, but he also knows that this is probably the perfect time to give them some competition...
This is what i meant by apple concentrating on software
-->Apple will continue to produce Mac OS X...
They would make it till it is profitable. Would that be their primary bread and butter in future? I guess no.
apple is totally out of enterprise and it cannot enter soon. it is mostly preferred by individuals and professionals but they too are getting tired of slow hardware. apple has been telling us that MHz doesn't matter. but day by day, intel is becoming more and more efficient (per MHz) and ultimately, MHz will matter. this means that apple must do something drastic and something soon. but what are the alternatives?
AMD will not be of much help. it needs lots of MS help in order to fight against Intel. any mistake, and MS will simply delay optimizing their OS and other software for AMD. AMD also needs to court Dell etc, which it can't do by having good relation with Apple. IBM power-pc will have same fate as motorola power-pc. IBM and Motorola, both have lots of things in life than just power-pc, so they won't give too much attention to it.
The truth is, apple as a proprietary processor company is dead. they must adopt mainstream x86, x86-64, itanium architecture if they want to succeed in PC market. they must focus more on software and less on hardware. that doesn't mean, they are limited to making el-cheapo x86 boxes only. they can make proprietary boxes which are MacOS optimized. they are also good at making good gadgets and they should utilize these expertize too. if MS can port windows to handhelds, why can't Apple do it?
no i didn't design one bit of the equipment. the equipement belongs one of the national labs and the only lab of its type in the world. the equipment was newly designed with lots of new hardware and real time software. have you heard of 350 GByes of uncompressed data on a single tape? well there were 11 of them in the experiment a decade ago! they obsoleted them more than a year ago! these data on each tapes are marked using time stamp (since the tape drives are located from east coast to hawaii). in order for the experiment to succeed, each block of data recorded at the same time must be combined. there is no independent way to say that the data tracks on two tapes are aligned other than the time stamp. you get meaningful data only if close to all the data are aligned perfectly (to within few microseconds). processing of these data is too expensive too. so trial and error is ruled out. basically, the committee felt that it wasn't worth salvaging the data and I got one more day to use this equipment.
one way to look at this experiment is like this. you have a very faint object that you are photographing. you also want 360 degree view and are using 10 cameras at different angles. Due to shaky-ness, you can't use long exposures. So you use multiple photos which you later combine in your computer. Assume that the object was moving randomly but you know the exact motion. Now if you forget to remember what time, each frame was taken, there is no way to do motion compensation and hence no way to superimpose the frames. now if your computer was too slow to superimpose the images, it may not be worth doing trial and error.
seriously, i did! during one of my scientific experiments (I believe it was in Jun-93), they added leap second in the middle of my experiment. The data taken from various places could not be combined together, since they didn't know at what time, leap second was adjusted at which place. So we had a 24 hours experiment on 300 million dollar equipment failed and 100's of manhours were lost in the process.
I have always wanted a simple remote control which can just download a code-file for a given model. Here is how it should work
To program your remote to Yamaha-5550, you will download a file for this on your computer. Then use whatever port you like and connect your remote to computer and download the file to your remote. Viola! you got Yamaha-5550 in your remote. To do this, some industry consortium needs to come up with standards and then all new devices will come up with such a downloadable file.
I know programmable LCD remote controls exist, which can do just like this. But here are the shortcomings:
1) LCD Based: Needs too much battery power.
2) LCD Based: Hard to operate without backlight.
Even then, you must look at it. I operate almost all my remote by touch feeling without lookint at them. No good non-LCD remote available which can download files from internet.
3) No industry standard. Not many manufacturers make such directly downloadable files. So no guarantee that your equipment is listed.
4) No industry standard. Each remote has its own version of code-file.
5) LCD Based: Too expensive.
at $100 a piece, it is still pricey. last time, i bought my 512 MB CF card for $68. So CF cards, which come in different sizes, fits in PCMCIA slot with just a passive connector, requires no driver,.... is only 35 % more expensive. I guess, I will stick with CF. The prices of CF cards are falling faster than any microdrive, iomegra click drive, etc may even go below $100/GB before the time, this drive comes in market. I have PDA, Digital Camera which take CF card natively. Also I have a laptop and printer which read CF card with just a passive adapter.
grow up. we count clicks/second nowadays or is it ticks/second? whatever it is, 3 billion is the answer. WPM? I thought, we obsoleted those ugly English units in favor of French SI units. No, wait... I guess, again we are boycotting French units and adopting English units.
and what the hell is this "$" thing? CIA world factbook shows USD as being USA currency.
Many countries use dollar as their currency; so for offical purpose, US currency is designated as USD. Similarly, many countries use "Rupee" as currency, so Indian currency is abbreviated as INR (INdian Rupee).
The plan being expensive or not should be based on regional pricing. 825 rupees a month for "always on" connection is cheaper by existing Indian pricing (if you are a moderate user).
Yes, India IS poor, so almost everything is expensive compared to their income (isn't it the definition of being poor?). Instead of saying it too expensive, it should be looked at it this way. The average telephone bill in India is about Rs. 1500/month. So most people who have telephone, fall in the affordable range at this internet pricing.
how is selling a CD any different than selling a book? people are ready to pay for a book and primary reason for this is that there is no easy and cheap way of copy a book to its exact replica. OTOH, it is just too easy to copy CD. music industry's real dilemma is that they want to capitalize on existing system (CD players, DVD players, CD-ROMs etc) which were created in an era when copying them was not economical. if they are sincere about anti-piracy, they should become little bold and come out with
a totally new medium which is not possible to
duplicate. here is my suggestion:
Create a new media. Patent technology and license only with the terms that the device will only output analog audio.
Allow stand alone copiers. The copier will mark on the copy that it is a copy. It will also mark on original that a copy has been made. This way, people cannot sell copies in the market. At the same time, user can make as many copies as desired (for car, backup, portable etc). If a copy has been made, then original cannot be sold. The media will have a digital signature and write once media (small portion, where digital signature and copy information is written. The rest of the medium could be rewriteable. After few rewrites, you may run out of write-once area. At that point, the media can only be used to copy analog contents. Allow easy detection that this is a copied content and this media has been copied.
Hardware manufacturer should be given royalty for each album sold for upto some period of time, provided their hardware is not hacked. This will reduce the cost of hardware, and will encourage them to create more fool-proof system. Also, this would discourage illegitimate players, since they will become more expensive.
Allow unlimited analog copies.
The reason, why DVD copy protection failed is because of fatal mistake that hardware decoding will be too expensive and hence computer DVD-ROM should be allowed to have software decoding. Today we know, hardware decoder doesn't cost much. Ofcourse the second reason was how to display analong content on monitor. I am sure some solution could have been found for this too. The above scheme can also be applied to DVD-ROM.
Just because Landmark would be early adopter, doesn't guarantee success. Read this older story about Inacom at Microsoft site.
Funny thing is, the date on the article says May-2000 which is really really close to the date Inacom filed Chapter 7 and closed down completely.
there are some other stuffs that i don't like or it is just a rehash of what others have pointed out earlier. all of these issues are from programmers point of view and not for reading or writing xml.
* attributes and elements and mixed contents complicates programming. now you have too many different types of fields tied together in an ugly manner which introduces all sorts of programming complexity. i would rather have only elements.
* although xml is verbose, it has done some corner cutting, like empty element can be written as <element/>, or can be written as <element></element>. Different parsers treat these two things identically or differently, this again introduces some bugs. there should be exactly one syntax.
* white spaces: no easy way to handle this. either will look ugly in text editor or will require extra programming effort.
* comments: Comments are extremely hard to read too. also from programmers point of view, comments are ambiguous. should they be preserved or discarded? no clear guidelines.
* No private elements: when i write an xml documents, i would like to put additional private information. the only way to write this now is using comments, but that is not structured. there should be a way to write structured comments too. this is not there in c/java/c++ etc too. java uses/**...*/ notation to tell it is a javadoc comments. this is a hack and some central body has to enforce it. there is no built-in way. also sun uses special javadoc tags to indicate fields within javadoc elements. again this is a hack. why doesn't xml have a built-in way of structuring comments so that all parsers, viewers would honor it?
after reading most of the replies it seems that their primary focus is on OD or on single deal. true, OD is not a big player in software business and a single deal like this won't hurt anyone. but think this way: today OD, tomorrow, other retailers. most businesses buy their stuff from this retailers only, so most software vendors will have to start getting their apps MS certified.
today it is OD, tomorrow, it is OD + OM + Stampes + Amazon +.... Large companies tend to buy their software from distributers which allow automated electronic billing, EDI, blah... which cost millions of dollars. That means if any vendor wants to sell their apps through these outlets to companies, then they will have to get their apps MS certified.
i m sure OD is not doing this as s a generosity for MS. it must be getting extra discount or something for doing this. if that is the case, could it be legal? can coca-cola decide, which potato chips Safeway can sell in order to get price break on coke? OD doesn't do customer support for any of the apps they sell, so why they bother if the app is certified or not? it seems that if MS can get few large retailers to follow this strategy, then commercial windows developers won't have choice but to get their apps certified. in this case, MS could create a license for certification which is harmful to their competitors. Oh well, it seems OD has two choices: either sell MS products at competitive rate or sell third party apps from MS competitors who don't want their apps certified by MS. they chose the 1st.
the problem of mass standard is not that it was made 100 years ago. the problem is that there is no way to describe it so that it can be reproduced independently. as one of the famous scientist said,
"we can communicate our definition of length and time to aliens 1000 light years away (if they are listening us), but we can't tell them what we mean by 1 kilo".
Secondly, it doesn't matter either what exactly is 1 kilogram. what matters is some reference atomic mass and then pick up Avogadro number (based on existing 1 kilo mass) and then get rid of the existing standard. this would allow independent reproduction (e.g. 1 kg is equivalent of 6.02...... x 10^23 atoms of Oxygen 16 in certain energy state. this scheme too has problem. there is no practical way of verifying that you have met the standard definition. so, two scientists can argue that each is possesing exact 1 kg and this cannot be arbitrated.
difference would be even large if not...
on
LCD Overtaking CRT
·
· Score: 1
LCD would take over CRT by even bigger margin if it were not for bundling of CRT monitor with many PC systems. 9 out 10 bundled monitors are CRT. OTOH, most people purchase CRT separately. Thank god, we don't have MS selling hardware, else we would be paying for CRT monitor whether we buy or not!
I am the original AC who wrote this message, so i will try to reply.
First, I am not mis-informed. I have taken graduate level courses on plasma fusion in a top university in 90's. I am also in personal touch with several people who work on plasma research. Second, there were Joe Sixpacks who caused Chernobyl accident and also there was some news about potential chain reaction in Japan nuclear material storage facility which too was caused by Joe Sixpacks. Further, if you read, I said that the technology inherently is too complex for today's scientists and engineers to understand. Joe only comes into picture when we talk about cost (for commercial exploitation). In future, you will need lots of them and failure to find them would result into same as what happened to fission power (very limited use).
You are claiming that the only way to go is upward, but then when wasn't that the case? Has anyone demonstrated either theoretically or through scaled down prototype that that would be commercially viable? Absolutely not. I never said to give up plasma research. I simply said that we are still at experimental stage only and nowhere close to commercial exploitation. That is atleast 30-50 (or more) years away. Saying otherwise would be to raise false hope which many scientists do to get funding.
Your whole article tone is such that it implies you oppose everything what I said. However, you are just picking out couple lines out of context and replying by saying the same thing what I have said in the rest of the article.
product activation does write a small amount of data to an area of the first track of the hard disk that is not used by the Windows file system, as do a number of other programs and utilities...
This technology in no way harms your hard drive or computer.
This reply from TurboTax is self contradictory. First it says that activation writes to the same area as is done by other programs and utilities and then it says that this is not harmful. How? You are writing on my disk in an area that may be used by other utilities and programs (you acknowledge this in your reply) and that also without informing me or taking my consent. And still you have guts to say, this is harmless?
How about reformatting my non-windows hard drive partitions? These too are not used by Windows File System. Wouldn't that be harmless too?
i have used several PCs/Sparcs and in all of them, before discarding, I have upgraded memory to 2x to 4x times originally it came with. We don't need more than 4 GB now; but if you were to buy a computer now, you need to make sure that you would be able to upgrade ram to 4x. This means if your need is greater than 1 GB, then 32 bit system is not suitable for you. At present, all my home and work machines (new ones) are ordered with 1 GB. This is at the border of 4x memory expansion possible with 32 bit system. So in a year or so, I will definitely not buy another 32 bit machine.
Slashdot Mirror
So far, nothing impressive. if you have any impressive thing about it, would you mind sharing it? -->Why would you even write a three-paragraph post on a subject you have no clue about?
When we talk about things to come, it is easy to do mud-slinging since nothing can be proved. but that is not my style, so i will give you benefit of doubt. may be i am totally clueless.
-->Apple did it before anyone. Ever hear of the Newton?
I am old enough to have used GM refrigerator. That doesn't mean I would tell everyone GM makes refrigerator. Apple Newton is a thing of past. It survived when there wasn't any competition. Once competition sprang, it was dead. i am talking about today. this is one market segment, where style and differentiation still matters, for which apple is famous and they can make some inroad. Apple doesn't sell any PDA today, but I won't be surprised if they come out with one soon.
It is as proprietary as Sparc.
-->Apple will not destroy its PPC customer-...
Didn't they destroy Motorola 68k base?
Why would apple make Mac OS for x86 now, if it is not for a fact that x86 hardware would be cheaper? Since MacOS would only work with Apple prescribed config, "installed base" could not be a valid cause.
This is what i meant by apple concentrating on software
-->Apple will continue to produce Mac OS X... They would make it till it is profitable. Would that be their primary bread and butter in future? I guess no.
AMD will not be of much help. it needs lots of MS help in order to fight against Intel. any mistake, and MS will simply delay optimizing their OS and other software for AMD. AMD also needs to court Dell etc, which it can't do by having good relation with Apple. IBM power-pc will have same fate as motorola power-pc. IBM and Motorola, both have lots of things in life than just power-pc, so they won't give too much attention to it.
The truth is, apple as a proprietary processor company is dead. they must adopt mainstream x86, x86-64, itanium architecture if they want to succeed in PC market. they must focus more on software and less on hardware. that doesn't mean, they are limited to making el-cheapo x86 boxes only. they can make proprietary boxes which are MacOS optimized. they are also good at making good gadgets and they should utilize these expertize too. if MS can port windows to handhelds, why can't Apple do it?
why bother? just by some mpaa, riaa, bsa stocks. now, if i could only find their stock symbols....
one way to look at this experiment is like this. you have a very faint object that you are photographing. you also want 360 degree view and are using 10 cameras at different angles. Due to shaky-ness, you can't use long exposures. So you use multiple photos which you later combine in your computer. Assume that the object was moving randomly but you know the exact motion. Now if you forget to remember what time, each frame was taken, there is no way to do motion compensation and hence no way to superimpose the frames. now if your computer was too slow to superimpose the images, it may not be worth doing trial and error.
seriously, i did! during one of my scientific experiments (I believe it was in Jun-93), they added leap second in the middle of my experiment. The data taken from various places could not be combined together, since they didn't know at what time, leap second was adjusted at which place. So we had a 24 hours experiment on 300 million dollar equipment failed and 100's of manhours were lost in the process.
so much for their 64-bit OS.
To program your remote to Yamaha-5550, you will download a file for this on your computer. Then use whatever port you like and connect your remote to computer and download the file to your remote. Viola! you got Yamaha-5550 in your remote. To do this, some industry consortium needs to come up with standards and then all new devices will come up with such a downloadable file.
I know programmable LCD remote controls exist, which can do just like this. But here are the shortcomings:
1) LCD Based: Needs too much battery power.
2) LCD Based: Hard to operate without backlight. Even then, you must look at it. I operate almost all my remote by touch feeling without lookint at them. No good non-LCD remote available which can download files from internet.
3) No industry standard. Not many manufacturers make such directly downloadable files. So no guarantee that your equipment is listed.
4) No industry standard. Each remote has its own version of code-file.
5) LCD Based: Too expensive.
at $100 a piece, it is still pricey. last time, i bought my 512 MB CF card for $68. So CF cards, which come in different sizes, fits in PCMCIA slot with just a passive connector, requires no driver, .... is only 35 % more expensive. I guess, I will stick with CF. The prices of CF cards are falling faster than any microdrive, iomegra click drive, etc may even go below $100/GB before the time, this drive comes in market. I have PDA, Digital Camera which take CF card natively. Also I have a laptop and printer which read CF card with just a passive adapter.
grow up. we count clicks/second nowadays or is it ticks/second? whatever it is, 3 billion is the answer. WPM? I thought, we obsoleted those ugly English units in favor of French SI units. No, wait... I guess, again we are boycotting French units and adopting English units.
Many countries use dollar as their currency; so for offical purpose, US currency is designated as USD. Similarly, many countries use "Rupee" as currency, so Indian currency is abbreviated as INR (INdian Rupee).
The plan being expensive or not should be based on regional pricing. 825 rupees a month for "always on" connection is cheaper by existing Indian pricing (if you are a moderate user).
Yes, India IS poor, so almost everything is expensive compared to their income (isn't it the definition of being poor?). Instead of saying it too expensive, it should be looked at it this way. The average telephone bill in India is about Rs. 1500/month. So most people who have telephone, fall in the affordable range at this internet pricing.
Create a new media. Patent technology and license only with the terms that the device will only output analog audio.
Allow stand alone copiers. The copier will mark on the copy that it is a copy. It will also mark on original that a copy has been made. This way, people cannot sell copies in the market. At the same time, user can make as many copies as desired (for car, backup, portable etc). If a copy has been made, then original cannot be sold. The media will have a digital signature and write once media (small portion, where digital signature and copy information is written. The rest of the medium could be rewriteable. After few rewrites, you may run out of write-once area. At that point, the media can only be used to copy analog contents. Allow easy detection that this is a copied content and this media has been copied.
Hardware manufacturer should be given royalty for each album sold for upto some period of time, provided their hardware is not hacked. This will reduce the cost of hardware, and will encourage them to create more fool-proof system. Also, this would discourage illegitimate players, since they will become more expensive.
Allow unlimited analog copies.
The reason, why DVD copy protection failed is because of fatal mistake that hardware decoding will be too expensive and hence computer DVD-ROM should be allowed to have software decoding. Today we know, hardware decoder doesn't cost much. Ofcourse the second reason was how to display analong content on monitor. I am sure some solution could have been found for this too. The above scheme can also be applied to DVD-ROM.
Now if I had few billion dollars....
Just because Landmark would be early adopter, doesn't guarantee success. Read this older story about Inacom at Microsoft site. Funny thing is, the date on the article says May-2000 which is really really close to the date Inacom filed Chapter 7 and closed down completely.
* attributes and elements and mixed contents complicates programming. now you have too many different types of fields tied together in an ugly manner which introduces all sorts of programming complexity. i would rather have only elements.
* although xml is verbose, it has done some corner cutting, like empty element can be written as <element/>, or can be written as <element></element>. Different parsers treat these two things identically or differently, this again introduces some bugs. there should be exactly one syntax.
* white spaces: no easy way to handle this. either will look ugly in text editor or will require extra programming effort.
* comments: Comments are extremely hard to read too. also from programmers point of view, comments are ambiguous. should they be preserved or discarded? no clear guidelines.
* No private elements: when i write an xml documents, i would like to put additional private information. the only way to write this now is using comments, but that is not structured. there should be a way to write structured comments too. this is not there in c/java/c++ etc too. java uses /**...*/ notation to tell it is a javadoc comments. this is a hack and some central body has to enforce it. there is no built-in way. also sun uses special javadoc tags to indicate fields within javadoc elements. again this is a hack. why doesn't xml have a built-in way of structuring comments so that all parsers, viewers would honor it?
triode!
after reading most of the replies it seems that their primary focus is on OD or on single deal. true, OD is not a big player in software business and a single deal like this won't hurt anyone. but think this way: today OD, tomorrow, other retailers. most businesses buy their stuff from this retailers only, so most software vendors will have to start getting their apps MS certified.
today it is OD, tomorrow, it is OD + OM + Stampes + Amazon + .... Large companies tend to buy their software from distributers which allow automated electronic billing, EDI, blah... which cost millions of dollars. That means if any vendor wants to sell their apps through these outlets to companies, then they will have to get their apps MS certified.
i m sure OD is not doing this as s a generosity for MS. it must be getting extra discount or something for doing this. if that is the case, could it be legal? can coca-cola decide, which potato chips Safeway can sell in order to get price break on coke? OD doesn't do customer support for any of the apps they sell, so why they bother if the app is certified or not? it seems that if MS can get few large retailers to follow this strategy, then commercial windows developers won't have choice but to get their apps certified. in this case, MS could create a license for certification which is harmful to their competitors. Oh well, it seems OD has two choices: either sell MS products at competitive rate or sell third party apps from MS competitors who don't want their apps certified by MS. they chose the 1st.
Secondly, it doesn't matter either what exactly is 1 kilogram. what matters is some reference atomic mass and then pick up Avogadro number (based on existing 1 kilo mass) and then get rid of the existing standard. this would allow independent reproduction (e.g. 1 kg is equivalent of 6.02...... x 10^23 atoms of Oxygen 16 in certain energy state. this scheme too has problem. there is no practical way of verifying that you have met the standard definition. so, two scientists can argue that each is possesing exact 1 kg and this cannot be arbitrated.
LCD would take over CRT by even bigger margin if it were not for bundling of CRT monitor with many PC systems. 9 out 10 bundled monitors are CRT. OTOH, most people purchase CRT separately. Thank god, we don't have MS selling hardware, else we would be paying for CRT monitor whether we buy or not!
My one and only problem with gas based bicycle is the amount of noise it makes. If someone can come up with a good silencer, then that would be great.
First, I am not mis-informed. I have taken graduate level courses on plasma fusion in a top university in 90's. I am also in personal touch with several people who work on plasma research. Second, there were Joe Sixpacks who caused Chernobyl accident and also there was some news about potential chain reaction in Japan nuclear material storage facility which too was caused by Joe Sixpacks. Further, if you read, I said that the technology inherently is too complex for today's scientists and engineers to understand. Joe only comes into picture when we talk about cost (for commercial exploitation). In future, you will need lots of them and failure to find them would result into same as what happened to fission power (very limited use).
You are claiming that the only way to go is upward, but then when wasn't that the case? Has anyone demonstrated either theoretically or through scaled down prototype that that would be commercially viable? Absolutely not. I never said to give up plasma research. I simply said that we are still at experimental stage only and nowhere close to commercial exploitation. That is atleast 30-50 (or more) years away. Saying otherwise would be to raise false hope which many scientists do to get funding.
Your whole article tone is such that it implies you oppose everything what I said. However, you are just picking out couple lines out of context and replying by saying the same thing what I have said in the rest of the article.
This technology in no way harms your hard drive or computer.
This reply from TurboTax is self contradictory. First it says that activation writes to the same area as is done by other programs and utilities and then it says that this is not harmful. How? You are writing on my disk in an area that may be used by other utilities and programs (you acknowledge this in your reply) and that also without informing me or taking my consent. And still you have guts to say, this is harmless? How about reformatting my non-windows hard drive partitions? These too are not used by Windows File System. Wouldn't that be harmless too?
i have used several PCs/Sparcs and in all of them, before discarding, I have upgraded memory to 2x to 4x times originally it came with. We don't need more than 4 GB now; but if you were to buy a computer now, you need to make sure that you would be able to upgrade ram to 4x. This means if your need is greater than 1 GB, then 32 bit system is not suitable for you. At present, all my home and work machines (new ones) are ordered with 1 GB. This is at the border of 4x memory expansion possible with 32 bit system. So in a year or so, I will definitely not buy another 32 bit machine.
And get fired, when someone finds a hole in IRM and your boss finds out.