But given all the complaints I see hear on/. about how Word does such a crappy job of opening older files, how is MS going to claim that they are fully compliant either?
If it comes to a pissing contest, which is going to seem worse: MS can't read old MS files faithfully, or random vendor XYZ can't read old MS files faithfully.
Or is Word's ability to read old documents perhaps not as awful as is usually claimed here?
I suspect that Microsoft has near zero dev docs to work from, at least when it comes to emulating Word 95 or WP 5.1. In the early days MS was known as a very freewheeling culture, and internal docs were very rare.
Even more to the point, the guidance sections essentially say "the existing implementations are buggy and we can't actually describe the precise behavior for arbitrary input". In the case of previous versions of Word they probably just have copies of the old modules that they feed the data to. For WordPerfect emulation I bet someone just did a black box reverse engineering that *mostly* works. While it may be documented, the rules are going to seem completely arbitrary.
I am often faced with this kind of situation at work. The requirements calls for significant enhancement to a feature, and it seems clear that the best way to implement it is to write the code from scratch. But the feature still has to maintain backwards compatibility. So you start studying the code, and find that it has subtle bugs that you now have to recreate in a new implementation.
Digital watermarking is far more resilient than that. To destroy a Digimarc watermark in a photograph you have to do a surprising amount of manipulation of the image - it is by no means impossible to get rid of, but doing it without significantly altering your perception of the image is difficult (I am sure people have figured it out, but the amount of manipulation required made me wonder how the watermarking is done).
In some ways, watermarking video should be even easier. You could potentially watermark every frame, and use a number of different technologies throughout the video, making removal harder.
But having said all that, the problem that they are going to have is that the huge majority of material has been released into the wild with no watermarks whatsoever, so they are going to have to detect content without the assistance of watermarks.
I hear lots of people assert that most people run the classic theme in XP, but yet when I walk around the office I would guess that 95% of the computers I see run the standard XP theme. Can anyone point to actual stats on the usage of the various themes?
I can certainly believe that MS intentionally excludes downlevel operating systems to generate more revenue, but there is real cost to supporting multiple operating systems. If Win 2K is a supported platform, you can bet that the testing group is going to spend some serious time and money doing the testing on that platform.
I have often advocated dropping support for older operating system on that basis, even if I don't know of any reason why the product would not function. At some point you just decide that spending the testing effort to check that the product still works on a Win 95 box with 64MB of RAM isn't worth it.
Reveal codes is not a feature, it is a performance hack that allowed you to work around the fact that the screen couldn't give you true WYSIWYG. Modern word processors don't need reveal codes because you can see what the code does, and its presence becomes implicit.
For businesses I think there are two reasons to upgrade, neither of which you mentioned:
1. Improved support for remote administration. That is not an area I am at all familiar with, but I keep hearing that group policies are much more comprehensive in Vista.
2. Better security. I know, MS always claims the next release will be locked up tight, but to some extent they are correct - each version is tougher than the one before At this point there is no reason to believe that Vista will not be an improvement over the security in XP. Just having the default accounts run as non-admin, and the whole principle of LUA will serve to stop lots of attacks.
sorry, I meant that you had to document to Microsoft why you needed those privs. If the explanation is accepted then you can get the logo and still require admin.
Getting the Windows logo does require that your app run as non-admin, or you have to document what admin feature it needs and why. Wanting to store config data in Program files is not an acceptable reason for needing admin privs.
But I'm all for promoting the hell out of apps that don't require admin, and pushing much harder for LUA by default.
I believe the test has always been that the press cannot themselves break the law in acquiring classified information, but it is not a crime for them to publish information that they have innocently received - innocently meaning that some 3rd party brought to them.
I don't have a security clearance, so I should never see classified data. But if someone gives me classified data, I haven't done anything wrong, the person who gave it to me has done something wrong. That is the mechanism the press has used in the past
I agree that the warnings that come up seem reasonable. I'd like to see them consolidated into a single message box, but given the nature of communications between the app and the OS that seems unlikely.
But I think the TFA probably did an upgrade install of Vista, which is why he was accustomed to having access to spots he shouldn't be messing with. I suspect the huge majority of Vista installs will be done by an OEM on a clean box, and that should deal with a lot of the problems right there.
I've been running as a limited user on my XP box following a recent rebuild, and aside from working around apps that do obnoxious things like store data in Program Files, I haven't had any problems to speak of. Keep your data in profile directories or in some tree you have access to and everything is fine.
My point is, a fresh install of Vista running apps made for Vista (again, like a store bought PC will have), and most users won't hit these security dialogs.
I don't have recent experiences with Dell, but last time I had one the only data you had for reinstallation was an image of the drive as it shipped to you in the first place. So the average consumer has no way to start with a clean install of Windows without the bloat, since the only image they have includes the bloat.
I've used the ZA firewall several times, and it can be a pain to get configured. When you first install it you get popups almost continuously until the core set of programs is added to the allow list.
With games, I've had problems where the firewall prompt pops up on the Windows desktop, with no indication in a full screen game that internet access is blocked. ZA also has some issues dealing with limited accounts, so I had to keep going back to my admin account.
And I have to agree with the people who say that most non-technical users will just allow everything by default. I'm a programmer and very PC literate, and sometimes I'm not sure what program is requesting access.
Finally a voice of reason. If I had points I would mod you up.
People, if you think it ain't worth $2.50, don't buy it. If enough people feel the same way the price will drop, either for this mod or for ones in the future. Micropayments like this have to be new to Bethesda, so it seems quite plausible that they set the price too high initially.
$2.50 may be the price for the early adopter fanboys. After a month it may drop to pull in a larger audience.
There is no set "worth" of something. By definition, if you buy it for $.250 it is worth $2.50 to you.
If DX10 will only be available on Vista, games will not start requiring DX10 until Vista has achieved significant market penetration. Anything else is suicide for the game vendors.
I did that in college ('81), hand assembling and keying into a pdp-8 via the front panel. That was after I already knew Basic and Pascal, but as an introduction to assembler, it was great. The pdp-8 had a very simple instruction set, and after that exercise you truly understood it.
It helps to be able to follow something all the way down to the bare metal.
I (like millions of others) just bought a copy of TurboTax. TurboTax Basic, TurboTax Premium, TurboTax Deluxe. Three versions, three prices. A big checklist on the back of the box listing what each one adds over the others.
There is some confusion of course, but it lets Intuit sell to customers willing to spend $25 and willing to spend $70. MS is doing the same thing here, with the difference that in many cases the OEM will be the one looking over the checklists and making the decision.
Actually, you may see some backlash because of the OEM tactics. Just like in the old days you would see vendors using slower RAM or graphics adapters using system RAM, now you will see OEMs including Home starter. Most consumers won't be savvy enough to spot the difference in a spec sheet, and end up with less than they thought they were getting.
From the positioning laid out in the article, I would expect 3 builds in retail - Home for the low priced, entry level white boxen. Home Premium for the high end PC lines, and Ultimate for gamer editions and geeks.
The business lines are likely to be sold directly to business accounts as license packs.
Segment fragmentation is always a risk, but this may be a workable strategy. The basic Home edition lets them do a defacto price drop on cheap boxes, thus helping to hold off the presence of Linux builds in retail.
At the other end they probably work in a price increase on the Ultimate edition.
Encryption by a major corp is just as safe as open-source.
If there is indeed a back door, the gov't will go to significant lengths to keep the backdoor secret. So they might be able to read your hard drive, but they won't use that evidence in a criminal trial, because doing so would divulge the existence of the backdoor.
Personally, I don't believe the backdoors exist. What is MS getting from the government that makes the risk of customer rejection because of the backdoor worthwhile? And if it was something big like dropping of the anti-trust lawsuits do you really believe that it could be kept secret? Hundreds of people would have to know about the deal, and somebody would talk.
The clipper chip had a backdoor for US law enforcement agencies. It foundered because of wide domestic opposition to such a backdoor, and near universal unwillingness of any foreign government to buy devices that the US could crack.
From accounts I read (primarily Levin's book "Crypto"), the NSA was caught completely offguard by the vehemence of the domestic reaction to the notion of key escrow. They expected it to get debated in a congressional committee or two, covered deep in a paper or on CSPAN, and get implemented without the man on the street ever even hearing about it. Instead it became front page news and public relations fiasco. Turns out ordinary people care about privacy in principle, if not in practice.
In the abstract I agree with the language neutral philosophy, but it really isn't that simple. I have used C/C++ for many years and have lots of experience with the IDE, the debugger, and the standard libraries. Even more important, I have lots of practice debugging strange behaviors. All in all I am pretty efficient in those languages.
I have some history with C# and Python. I know something about the available libraries but often have to stop to look up a call signature or search for a method that I know exists. The first time I encountered a Python bug where I thought I had copied a list but really only copied a reference it took me many hours to solve. The first time I needed to embed a resource in a C# program I had to go reading for a few hours. The point is, there are still lots of things lurking in those languages that I haven't had run into yet, and at some point I am going to run into them. I can work in those languages, and am perfectly willing to learn others. I've had to maintain some MC68340 assembler of late, and while I'm no expert I can muddle through.
Ask me to maintain a Lisp program and I'm going to spend lots of time learning Lisp before I am even minimally effective. Ask me to maintain C++ code and I can hit the ground running.
There are definitely cases where one language is preferred over another, and if the time horizon is long enough to eat the learning curve then you should use the best fit language. But if someone needs a script today and it is 10 lines of Perl or 200 lines of Python, I'll write it in Python because I can get those 200 lines done today. Maybe I could get 10 lines of Perl done, but the Python is more likely to work and more likely to be maintainable.
Re:NOT A Selling Point-But a "must have" for secur
on
Buy Vista or Else
·
· Score: 1
That is the default in Vista, and they have done significant work so that fewer things require admin access.
And in a home environment having the machine boot up into a userland account with no password is probably not a huge hole, so long as escalating privs to admin level requires a password. That would get around the problem of requiring a login.
In the huge majority of the cases, Best Buy isn't getting the float, the manufacturer is. In fact, I think every time I have seen a store rebate at BB it has been of the instant variety - markdown taken at register, no float involved.
I used to work for a software company that sold at retail, and we routinely had $30 rebates. There were a few reasons for that:
1. Rebate submissions ran about 30% of total sales. If everyone submitted the rebate it wouldn't have been that big. 2. Rebates were a way to lower the retail price without lowering the distributor price. We controlled the rebate, but the big software resellers had too much pricing clout to let us raise and lower the distributor price at will. If we changed the distributor price there was no guarantee that the street price would change accordingly. 3. If you sell goods to the government you are subject to GSA pricing rules. Rebates let you drop the street price without affecting GSA price.
We used a 3rd party to process rebate fulfillment. There was no attempt to make the rebates hard to get, the sales group could predict the actual response rates pretty well and set the prices accordingly.
While that does seem to be the standard that constitutional law has evolved to, what troubles me more about the wartime assertion is that this war is so vague and open ended.
If we accept the premise that wartime is different than peacetime and different rules apply, then the first thing I want to know is how do we know when the war is over? We've gone 4 years since 9/11 with no additional attacks on US soil. Is the war over now? Will it be over if we catch bin Laden and Mullah Omar?
In a conventional war against a country or polity of some sort, it is pretty clear when hostilities have ended. A war against individuals in hiding is not nearly so cut and dried.
My biggest issue with the wartime argument is that there is no clear end to this war, so we had better assume that these "wartime powers" are in fact permanent. That thought scares the crap out of me.
Slashdot Mirror
How is that situation any different than losing a CD you bought at the mall? Or forgetting the combination to a lock you purchased?
Your friend purchased a product, and was careless with the credentials necessary to use that product. What did Apple do wrong in this case?
I would prefer no DRM of course, but iTunes has DRM-lite. Of all the systems out there, it is by far the most consumer friendly.
But given all the complaints I see hear on /. about how Word does such a crappy job of opening older files, how is MS going to claim that they are fully compliant either?
If it comes to a pissing contest, which is going to seem worse: MS can't read old MS files faithfully, or random vendor XYZ can't read old MS files faithfully.
Or is Word's ability to read old documents perhaps not as awful as is usually claimed here?
I suspect that Microsoft has near zero dev docs to work from, at least when it comes to emulating Word 95 or WP 5.1. In the early days MS was known as a very freewheeling culture, and internal docs were very rare.
Even more to the point, the guidance sections essentially say "the existing implementations are buggy and we can't actually describe the precise behavior for arbitrary input". In the case of previous versions of Word they probably just have copies of the old modules that they feed the data to. For WordPerfect emulation I bet someone just did a black box reverse engineering that *mostly* works. While it may be documented, the rules are going to seem completely arbitrary.
I am often faced with this kind of situation at work. The requirements calls for significant enhancement to a feature, and it seems clear that the best way to implement it is to write the code from scratch. But the feature still has to maintain backwards compatibility. So you start studying the code, and find that it has subtle bugs that you now have to recreate in a new implementation.
Digital watermarking is far more resilient than that. To destroy a Digimarc watermark in a photograph you have to do a surprising amount of manipulation of the image - it is by no means impossible to get rid of, but doing it without significantly altering your perception of the image is difficult (I am sure people have figured it out, but the amount of manipulation required made me wonder how the watermarking is done).
In some ways, watermarking video should be even easier. You could potentially watermark every frame, and use a number of different technologies throughout the video, making removal harder.
But having said all that, the problem that they are going to have is that the huge majority of material has been released into the wild with no watermarks whatsoever, so they are going to have to detect content without the assistance of watermarks.
I hear lots of people assert that most people run the classic theme in XP, but yet when I walk around the office I would guess that 95% of the computers I see run the standard XP theme. Can anyone point to actual stats on the usage of the various themes?
I can certainly believe that MS intentionally excludes downlevel operating systems to generate more revenue, but there is real cost to supporting multiple operating systems. If Win 2K is a supported platform, you can bet that the testing group is going to spend some serious time and money doing the testing on that platform.
I have often advocated dropping support for older operating system on that basis, even if I don't know of any reason why the product would not function. At some point you just decide that spending the testing effort to check that the product still works on a Win 95 box with 64MB of RAM isn't worth it.
Reveal codes is not a feature, it is a performance hack that allowed you to work around the fact that the screen couldn't give you true WYSIWYG. Modern word processors don't need reveal codes because you can see what the code does, and its presence becomes implicit.
For businesses I think there are two reasons to upgrade, neither of which you mentioned:
1. Improved support for remote administration. That is not an area I am at all familiar with, but I keep hearing that group policies are much more comprehensive in Vista.
2. Better security. I know, MS always claims the next release will be locked up tight, but to some extent they are correct - each version is tougher than the one before At this point there is no reason to believe that Vista will not be an improvement over the security in XP. Just having the default accounts run as non-admin, and the whole principle of LUA will serve to stop lots of attacks.
I have no problem with that. I was just trying to explain the current rules.
sorry, I meant that you had to document to Microsoft why you needed those privs. If the explanation is accepted then you can get the logo and still require admin.
Getting the Windows logo does require that your app run as non-admin, or you have to document what admin feature it needs and why. Wanting to store config data in Program files is not an acceptable reason for needing admin privs.
But I'm all for promoting the hell out of apps that don't require admin, and pushing much harder for LUA by default.
I believe the test has always been that the press cannot themselves break the law in acquiring classified information, but it is not a crime for them to publish information that they have innocently received - innocently meaning that some 3rd party brought to them.
I don't have a security clearance, so I should never see classified data. But if someone gives me classified data, I haven't done anything wrong, the person who gave it to me has done something wrong. That is the mechanism the press has used in the past
I agree that the warnings that come up seem reasonable. I'd like to see them consolidated into a single message box, but given the nature of communications between the app and the OS that seems unlikely.
But I think the TFA probably did an upgrade install of Vista, which is why he was accustomed to having access to spots he shouldn't be messing with. I suspect the huge majority of Vista installs will be done by an OEM on a clean box, and that should deal with a lot of the problems right there.
I've been running as a limited user on my XP box following a recent rebuild, and aside from working around apps that do obnoxious things like store data in Program Files, I haven't had any problems to speak of. Keep your data in profile directories or in some tree you have access to and everything is fine.
My point is, a fresh install of Vista running apps made for Vista (again, like a store bought PC will have), and most users won't hit these security dialogs.
I don't have recent experiences with Dell, but last time I had one the only data you had for reinstallation was an image of the drive as it shipped to you in the first place. So the average consumer has no way to start with a clean install of Windows without the bloat, since the only image they have includes the bloat.
I've used the ZA firewall several times, and it can be a pain to get configured. When you first install it you get popups almost continuously until the core set of programs is added to the allow list.
With games, I've had problems where the firewall prompt pops up on the Windows desktop, with no indication in a full screen game that internet access is blocked. ZA also has some issues dealing with limited accounts, so I had to keep going back to my admin account.
And I have to agree with the people who say that most non-technical users will just allow everything by default. I'm a programmer and very PC literate, and sometimes I'm not sure what program is requesting access.
Finally a voice of reason. If I had points I would mod you up.
People, if you think it ain't worth $2.50, don't buy it. If enough people feel the same way the price will drop, either for this mod or for ones in the future. Micropayments like this have to be new to Bethesda, so it seems quite plausible that they set the price too high initially.
$2.50 may be the price for the early adopter fanboys. After a month it may drop to pull in a larger audience.
There is no set "worth" of something. By definition, if you buy it for $.250 it is worth $2.50 to you.
If DX10 will only be available on Vista, games will not start requiring DX10 until Vista has achieved significant market penetration. Anything else is suicide for the game vendors.
I did that in college ('81), hand assembling and keying into a pdp-8 via the front panel. That was after I already knew Basic and Pascal, but as an introduction to assembler, it was great. The pdp-8 had a very simple instruction set, and after that exercise you truly understood it.
It helps to be able to follow something all the way down to the bare metal.
I (like millions of others) just bought a copy of TurboTax. TurboTax Basic, TurboTax Premium, TurboTax Deluxe. Three versions, three prices. A big checklist on the back of the box listing what each one adds over the others.
There is some confusion of course, but it lets Intuit sell to customers willing to spend $25 and willing to spend $70. MS is doing the same thing here, with the difference that in many cases the OEM will be the one looking over the checklists and making the decision.
Actually, you may see some backlash because of the OEM tactics. Just like in the old days you would see vendors using slower RAM or graphics adapters using system RAM, now you will see OEMs including Home starter. Most consumers won't be savvy enough to spot the difference in a spec sheet, and end up with less than they thought they were getting.
From the positioning laid out in the article, I would expect 3 builds in retail - Home for the low priced, entry level white boxen. Home Premium for the high end PC lines, and Ultimate for gamer editions and geeks.
The business lines are likely to be sold directly to business accounts as license packs.
Segment fragmentation is always a risk, but this may be a workable strategy. The basic Home edition lets them do a defacto price drop on cheap boxes, thus helping to hold off the presence of Linux builds in retail.
At the other end they probably work in a price increase on the Ultimate edition.
Encryption by a major corp is just as safe as open-source.
If there is indeed a back door, the gov't will go to significant lengths to keep the backdoor secret. So they might be able to read your hard drive, but they won't use that evidence in a criminal trial, because doing so would divulge the existence of the backdoor.
Personally, I don't believe the backdoors exist. What is MS getting from the government that makes the risk of customer rejection because of the backdoor worthwhile? And if it was something big like dropping of the anti-trust lawsuits do you really believe that it could be kept secret? Hundreds of people would have to know about the deal, and somebody would talk.
The clipper chip had a backdoor for US law enforcement agencies. It foundered because of wide domestic opposition to such a backdoor, and near universal unwillingness of any foreign government to buy devices that the US could crack. From accounts I read (primarily Levin's book "Crypto"), the NSA was caught completely offguard by the vehemence of the domestic reaction to the notion of key escrow. They expected it to get debated in a congressional committee or two, covered deep in a paper or on CSPAN, and get implemented without the man on the street ever even hearing about it. Instead it became front page news and public relations fiasco. Turns out ordinary people care about privacy in principle, if not in practice.
In the abstract I agree with the language neutral philosophy, but it really isn't that simple. I have used C/C++ for many years and have lots of experience with the IDE, the debugger, and the standard libraries. Even more important, I have lots of practice debugging strange behaviors. All in all I am pretty efficient in those languages.
I have some history with C# and Python. I know something about the available libraries but often have to stop to look up a call signature or search for a method that I know exists. The first time I encountered a Python bug where I thought I had copied a list but really only copied a reference it took me many hours to solve. The first time I needed to embed a resource in a C# program I had to go reading for a few hours. The point is, there are still lots of things lurking in those languages that I haven't had run into yet, and at some point I am going to run into them. I can work in those languages, and am perfectly willing to learn others. I've had to maintain some MC68340 assembler of late, and while I'm no expert I can muddle through.
Ask me to maintain a Lisp program and I'm going to spend lots of time learning Lisp before I am even minimally effective. Ask me to maintain C++ code and I can hit the ground running.
There are definitely cases where one language is preferred over another, and if the time horizon is long enough to eat the learning curve then you should use the best fit language. But if someone needs a script today and it is 10 lines of Perl or 200 lines of Python, I'll write it in Python because I can get those 200 lines done today. Maybe I could get 10 lines of Perl done, but the Python is more likely to work and more likely to be maintainable.
That is the default in Vista, and they have done significant work so that fewer things require admin access.
And in a home environment having the machine boot up into a userland account with no password is probably not a huge hole, so long as escalating privs to admin level requires a password. That would get around the problem of requiring a login.
In the huge majority of the cases, Best Buy isn't getting the float, the manufacturer is. In fact, I think every time I have seen a store rebate at BB it has been of the instant variety - markdown taken at register, no float involved.
I used to work for a software company that sold at retail, and we routinely had $30 rebates. There were a few reasons for that:
1. Rebate submissions ran about 30% of total sales. If everyone submitted the rebate it wouldn't have been that big.
2. Rebates were a way to lower the retail price without lowering the distributor price. We controlled the rebate, but the big software resellers had too much pricing clout to let us raise and lower the distributor price at will. If we changed the distributor price there was no guarantee that the street price would change accordingly.
3. If you sell goods to the government you are subject to GSA pricing rules. Rebates let you drop the street price without affecting GSA price.
We used a 3rd party to process rebate fulfillment. There was no attempt to make the rebates hard to get, the sales group could predict the actual response rates pretty well and set the prices accordingly.
While that does seem to be the standard that constitutional law has evolved to, what troubles me more about the wartime assertion is that this war is so vague and open ended.
If we accept the premise that wartime is different than peacetime and different rules apply, then the first thing I want to know is how do we know when the war is over? We've gone 4 years since 9/11 with no additional attacks on US soil. Is the war over now? Will it be over if we catch bin Laden and Mullah Omar?
In a conventional war against a country or polity of some sort, it is pretty clear when hostilities have ended. A war against individuals in hiding is not nearly so cut and dried.
My biggest issue with the wartime argument is that there is no clear end to this war, so we had better assume that these "wartime powers" are in fact permanent. That thought scares the crap out of me.