Breaking news: Apple will sue it over its Harmony technology that 'breaks' iTunes' FairPlay DRM to allow its music to play on the iPod. Apple points out that, as reported on a popular site called Slashdot, Real acknowledged in an SEC filing that a lawsuit from Apple would potentially be very damaging to the companies [sic] bottom line, as it accepts that a court might not agree that the reverse-engineering is legal.
Little Point Learning Perl If Not Already Learned
on
Learning Perl, 4th Ed.
·
· Score: 2, Interesting
Some may consider this a troll, but, knee-jerk reactions aside, from a professional standpoint, unless you have to maintain some legacy system that uses Perl, or an interest in historical programming languages, there is very little point in learning Perl. You can find all of its benefits in more modern non-compiled languages without the cruft of years of esoteric development. I will not mention other languages as that will just start a religious war over everyone's pet language, but, seriously, aspiring technologists take note and do your research.
If you already know Perl, that's great and good that you learned it when you learned it, but it doesn't change the above.
The authority on it is what constitutes the simplest solution to the problem (of builds in this instance). Introducing the markup overhead of XML for no practical benefit and essentially requiring programming in what is supposed to be a *configuration* file is certainly not that simplest solution. In this manner, XSLT, as well, falls down since much simpler and more powerful functional programming languages exist.
The W3C is not an authority on anything. They are more of an idea reviewing and specification factory and, to the extent that what they propose is adopted, those things that are adopted en masses gain a certain sense of 'authority' by common usage. However, this does not directly lead to the conclusion that those things are the best solutions, nor that their ideas are very good.
It has been a source of constant dismay to me that something has not arisen to replace Ant in for Java builds. The Ant project has created a terrific mess where people essentially *program in XML*. XML is a set of data structure markup semantics, not a programming language. No potential of repurposing ant build files exists, another knock against using XML. Therefore, from a software architecture perspective, Ant is one of the biggest misapplications of technology I have seen to date and despite that, staggeringly popular. I rest my case and shake my head.
I attended the Macromedia MAX Conference in New Orleans in early November. Please note that I do not work for Macromedia, I abhor Flash on Web sites except for very limited uses, and will have nothing to do with Cold Fusion, so to say I felt out of place at a Macromedia conference comes as sort of an understatement.:) My observations regarding Flash on mobile devices:
Flash for mobile devices has the moniker Flash Lite. Two versions exist: 1.0 and 1.1. As I understand it, Flash Lite came about when DoCoMo in japan approached Macromedia with an interest in coming up with an animation engine to spruce up the user interfaces of DoCoMo's phones. Macromedia cobbled something together by stripping down Flash 5 to a footprint suitable for small devices. Note that, as a result, Flash Lite uses ActionScript 1.0 instead of the current 2.0 in the latest PC Flash implementations, which ruffled the feathers of some of the Flash developers at the conference.
As of the conference, Macromedia had essentially zero penetration in the U.S. They recently got a little bit of penetration in Europe with T-Mobile, but Flash Lite at this stage exists almost solely in Japan with DoCoMo, though they mentioned they might have something going with KDDI, the, as I understand it, second largest carrier in Japan behind DoCoMo. Some of the DoCoMo phones in Japan actually use Flash Lite to render the user interface replete with 'cute' animations and such, some models using Flash Lite 1.0 and others 1.1.
The latest version of Flash MX Studio 2004 (right name?) has a profile for Flash Lite 1.1, so you can develop Flash Lite applications with it. However, Flash Lite Flash applications have extreme limitations - no bigger than 100K distributable and small runtime memory allowances. Ironically, they advised developers to use bitmaps rather than high-complexity vectors because the player on these limited phones cannot handle vectors very well.
The examples of applications and code I saw demonstrated a high level of 'hack' factor to get around these limitations and Flash Lite development in Flash MX Studio 2004 looks absolutely agonizing, though that may stem from my lack of experience with Flash development in general. Let me just say when you have to draw 'off stage' *visual* elements and click on them to input your 'script', which differs from frame to frame in a 'movie track', I want none of it. When you get layered inappropriate paradigms, you have trouble.
Macromedia did a good job of providing information about Flash Lite, but they face an uphill battle because they appeared to have an almost singular focus on pleasing carriers, not developers. This does not surprise me in the mobile world, which presents a generally toxic environment for independent developers, but suffice to say that they really want to make money off licensing the player to carriers in large volume. They need developers to create some compelling apps to encourage such licensing, but with no penetration in the U.S. and very rudimentary support for developers, this does not seem likely or wise for anyone except those targeting the Japanese market.
One important point that demonstrates this: Even if you came up with a fantastic app such that you could actually convince mobile users to download Flash Lite, there currently exists no way for them to do so or for you to bundle the Flash Lite player with your app because Macromedia wants license fees from carriers for the installation of the player on the mobile device and therefore does not provide free and ubiquitous downloads as it does with the Flash browser plugin.
If you want to start development and test on the phone, you need an advanced phone for which they have a beta client, such as a Sony P900 or recent Nokia Series 60. You also need to email them at a special email address to get added to their 'Flash Lite beta program' and may have to sign an NDA to get a version of the Flash Lite player to run on your phone, which I declined. I think, to test
The first images are not very good ones to start with. I suggest browsing down to the first set of images that do not have parts of the rover in them (a set of small hills on the horizon). Also, try resizing the browser so that only the two images you are trying to combine are in view and place the browser on a plain background such as a reasonably uncluttered desktop. Try both the cross-eyed and parallel set of images if you do not know your method - you'll know when you have it right because there will be a slight topographic roll to the surface nearby.
Once you get those, try keeping your eyes situated in the same position and scrolling the other images up or down into your field of view without looking up or down. This will allow you to view the more difficult images with parts of the rover in them, which have sharp depth transitions between the solar panels, airbags, and ground.
If we assume best case scenario for the transmission potential stated there and assume the direct-to-Earth rate averages the midpoint between the stated 12000bps and 3500bps, the total daily data for a single Martian day, direct-to-Earth and orbiter relay potential combined, is on the order of 17MB. The total data for the entire mission is on the order of 1,550MB.
Of course, this has to include protocol overhead, error, and operating instructions, but it provides one quantitative, if not qualitative, answer to how much data can be retrieved by the mission. In this case, a bit more than 2 CDs worth.
EverStoned and any others who like to listen to loud headphones, please read this immediately. Quote:
Hearing damage from headphones is probably more common than from loudspeakers, because many people exploit the acoustic isolation by listening at higher volumes. Moreover, the risk of hearing damage from headphones is higher than with loudspeakers, even at comparable volumes, due to the close coupling of the transducers to the ears. One of the benefits of headphone listening is the ability to detect musical details. Any hearing damage would have substantial impact on that experience. This article takes a look at the process of human hearing and offers guidelines for safe listening. (The information given here does not substitute for medical expertise. Readers should consult a physician for a diagnosis of hearing damage.)
If you are turning up the volume to drown out ambient noise, consider getting an set of earbuds like the Etymotic ER-4P.
EverStoned and any others who like to listen to loud headphones, please read this immediately. If you are turning up the volume to drown out ambient noise, consider getting an set of earbuds like the Etymotic ER-4P.
Have you tried the small dongle control where you can "dial down" the fan speed or are you generating too much heat for that? At 1500rpm, the fan is essentially silent. If you are at 1500rpm and it is noticable even with the case open you probably have a bad fan and should contact Zalman. Also, make sure it it attached securely and not vibrating (and transmitting those vibrations to the case as another poster mentioned).
The information you are interested in: I am running a 2000FP via DVI from a Radeon 8500 64MB (regular version, not LE) at 1600x1200x32bitx60Hz in both Windows 2000 Professional and Red Hat 8.0 (with both original release and recently updated kernels) dual booted. The only thing I had to install was the Radon 8500 Catalyst 2.5 drivers (version 6.10.13.6813 obtained from ATI Web site, just the drivers not the ATI control panel) for Windows. The default Red Hat 8.0 X11 works at this resolution straight out of the install. I am using the DVI cable that came with the monitor.
For more on exactly this topic, see Vmyths.com. It's over the top at times, but there is some interesting speculation, rumor, and commentary there. The delivery to China of a cache of viruses by the anti-virus industry is a particulary interesting and some other established media outlets are picking up the story as well. Not affiliated, yada, yada.
DRM can't keep you from reading the material, as long as you pay the price. Some say, Well, how can you take a paragraph and copy it anymore? That's what we normally consider to be fair use. But the fact is, you can still do that. You might not be able to cut and paste but as long as you can read it, you can type it... It's just not as easy as it could be but it's not any harder than it was 30 years ago.
True, there is no way to plug the "analog hole." However, to revert back to it as the only mechanism for copying is to effectively undo 30 or more years of productivity enhancement through technology and features such as copy/paste. The point of technological advancement is to automate manual processes such as transcription.
Estimating the lack of productivity seems popularly acceptable for damages alleged by computer viruses. If we did a calculation for the lost productivity costs of DRM difficulties imposed on legitimate copying in business and academic work, it would likely be a large number, perhaps dwarfing the revenues protected by DRM. Therefore, Mr. Liebowitz's argument founders on a zero, or perhaps even negative, sum.
In the previous thread on this issue, I raised the question of legal liability of non-disclosure with the hypothetical case of a company hacked through this hole prior to the recent announcement, but after Microsoft learned of the hole, that suffered financial injury. Putting aside the specifics and focusing on the legal question, is this liability, if it exists, a business argument against non-disclosure?
Responses to the previous post indicate that the EULA is not a sufficient shield to hide behind and this is about willful non-disclosure of a known vulnerability. Does an implicit (in the legal sense) trust exist that vendors alert their customers to vulnerabilities as soon as they become known so that the customers can take immediate, non-patching action (e.g. disconnect from network, shut down affected services) to protect their systems until a patch is released?
Has that ever been tested in court, especially in light of an argument of gross negligence by the plaintiff? It seems posssible to enter into agreements and obtain a favorable decision notwithstanding such agreements in certain cases, especially those involving gross negligence or unreasonable clauses of those agreements.
What if a company was hacked during the last few weeks prior to full disclosure through this hole and suffered financial injury due to the intrusion? Is Microsoft perhaps negligent and legally liable for those injuries since it did not disclose the hole for five weeks after it knew of its existence?
"But if we could just move your molecules around, we could just take a "snapshot" of you when you were a healthy 18 year old, and then, years later, rearrange your decrepit 85 year old body, except the brain, to the original configuration."
Now that is a new definition of pain! Seriously though, great point about every kind of property changing into intellectual property in the presence of such technology.
I disagree as to the inadequacy of the sampling size. I have no knowledge of how many items actually are on Freenet. However, the margin of error on 1070 items of a population of n > 1070 to infinity is not so high as to make the stats worthless. You can make an argument against accuracy no matter what the size of the data set since the basis is probability, but 1070 in a data set is fairly trustworthy, though not overly so.
I think the real concerns regarding the sample here are methods and the possible subjectivity of a single data analyzer. For example, a particular search pattern could yield content of a certain type more regularly than others; what is porn to one person might be art to another.
Exhibit at Royal Observatory
on
Longitude
·
· Score: 2
There is an excellent exhibit of time and Harrison's work, including several of the clocks at different stages in evolution, at the Royal Observatory in Greenwhich, England. If you think his work is impressive from reading the book, you should see some of his contraptions in person!
"The "killer app" I'm waiting for are the solar power satellites: high capital cost, but the lowest kilowatt/hour to manhour ratio of any power source. Let's live off the Sun, oh yeah!"
Question of curiosity: How do you efficiently get tcf:àower from the satellites to the ground?
"The US has a lot of experience training for short-term mission. Now we need to learn what to do for long-term mission"
This is a very good point. I imagine it is one thing for astronauts to launch and work brutal hours in cramped, downright uncomfortable conditions with no clear "work day" knowing they are returning to relative vacation in a week or ten days; quite another to keep that tempo up for months on end. I agree some, if not most, of the problems are likely directly attributable to a short-term approach, by ground control, to long-term missions.
I cannot help but compare the problems you cite to the difficulties and tensions in large software projects. If the team on Alpha gets bigger, will the tensions relax (more hands to do the work) or intensify (more problems created, more interpersonal tension)? I ask because, with software projects, it can go both ways.
"So who was the wench in Paul's room when the Hunter Seeker attacked? Hi - not true to the book"
I noticed this as well, but I understand why: There is no context for it when it first appears if you not a Dune expert. Thus there needs to be another person in the room for there to be a dialogue in which the purpose and capabilities of this thing are explained. It would be awkward to try to explain it and the behaviour of Paul (standing stock still, grabbing it when Mapes comes in) before or after the fact without losing either the suprise or the suspense.
Slashdot Mirror
Breaking news: Apple will sue it over its Harmony technology that 'breaks' iTunes' FairPlay DRM to allow its music to play on the iPod. Apple points out that, as reported on a popular site called Slashdot, Real acknowledged in an SEC filing that a lawsuit from Apple would potentially be very damaging to the companies [sic] bottom line, as it accepts that a court might not agree that the reverse-engineering is legal.
Some may consider this a troll, but, knee-jerk reactions aside, from a professional standpoint, unless you have to maintain some legacy system that uses Perl, or an interest in historical programming languages, there is very little point in learning Perl. You can find all of its benefits in more modern non-compiled languages without the cruft of years of esoteric development. I will not mention other languages as that will just start a religious war over everyone's pet language, but, seriously, aspiring technologists take note and do your research.
If you already know Perl, that's great and good that you learned it when you learned it, but it doesn't change the above.
The authority on it is what constitutes the simplest solution to the problem (of builds in this instance). Introducing the markup overhead of XML for no practical benefit and essentially requiring programming in what is supposed to be a *configuration* file is certainly not that simplest solution. In this manner, XSLT, as well, falls down since much simpler and more powerful functional programming languages exist.
The W3C is not an authority on anything. They are more of an idea reviewing and specification factory and, to the extent that what they propose is adopted, those things that are adopted en masses gain a certain sense of 'authority' by common usage. However, this does not directly lead to the conclusion that those things are the best solutions, nor that their ideas are very good.
It has been a source of constant dismay to me that something has not arisen to replace Ant in for Java builds. The Ant project has created a terrific mess where people essentially *program in XML*. XML is a set of data structure markup semantics, not a programming language. No potential of repurposing ant build files exists, another knock against using XML. Therefore, from a software architecture perspective, Ant is one of the biggest misapplications of technology I have seen to date and despite that, staggeringly popular. I rest my case and shake my head.
I attended the Macromedia MAX Conference in New Orleans in early November. Please note that I do not work for Macromedia, I abhor Flash on Web sites except for very limited uses, and will have nothing to do with Cold Fusion, so to say I felt out of place at a Macromedia conference comes as sort of an understatement. :) My observations regarding Flash on mobile devices:
Flash for mobile devices has the moniker Flash Lite. Two versions exist: 1.0 and 1.1. As I understand it, Flash Lite came about when DoCoMo in japan approached Macromedia with an interest in coming up with an animation engine to spruce up the user interfaces of DoCoMo's phones. Macromedia cobbled something together by stripping down Flash 5 to a footprint suitable for small devices. Note that, as a result, Flash Lite uses ActionScript 1.0 instead of the current 2.0 in the latest PC Flash implementations, which ruffled the feathers of some of the Flash developers at the conference.
As of the conference, Macromedia had essentially zero penetration in the U.S. They recently got a little bit of penetration in Europe with T-Mobile, but Flash Lite at this stage exists almost solely in Japan with DoCoMo, though they mentioned they might have something going with KDDI, the, as I understand it, second largest carrier in Japan behind DoCoMo. Some of the DoCoMo phones in Japan actually use Flash Lite to render the user interface replete with 'cute' animations and such, some models using Flash Lite 1.0 and others 1.1.
The latest version of Flash MX Studio 2004 (right name?) has a profile for Flash Lite 1.1, so you can develop Flash Lite applications with it. However, Flash Lite Flash applications have extreme limitations - no bigger than 100K distributable and small runtime memory allowances. Ironically, they advised developers to use bitmaps rather than high-complexity vectors because the player on these limited phones cannot handle vectors very well.
The examples of applications and code I saw demonstrated a high level of 'hack' factor to get around these limitations and Flash Lite development in Flash MX Studio 2004 looks absolutely agonizing, though that may stem from my lack of experience with Flash development in general. Let me just say when you have to draw 'off stage' *visual* elements and click on them to input your 'script', which differs from frame to frame in a 'movie track', I want none of it. When you get layered inappropriate paradigms, you have trouble.
Macromedia did a good job of providing information about Flash Lite, but they face an uphill battle because they appeared to have an almost singular focus on pleasing carriers, not developers. This does not surprise me in the mobile world, which presents a generally toxic environment for independent developers, but suffice to say that they really want to make money off licensing the player to carriers in large volume. They need developers to create some compelling apps to encourage such licensing, but with no penetration in the U.S. and very rudimentary support for developers, this does not seem likely or wise for anyone except those targeting the Japanese market.
One important point that demonstrates this: Even if you came up with a fantastic app such that you could actually convince mobile users to download Flash Lite, there currently exists no way for them to do so or for you to bundle the Flash Lite player with your app because Macromedia wants license fees from carriers for the installation of the player on the mobile device and therefore does not provide free and ubiquitous downloads as it does with the Flash browser plugin.
If you want to start development and test on the phone, you need an advanced phone for which they have a beta client, such as a Sony P900 or recent Nokia Series 60. You also need to email them at a special email address to get added to their 'Flash Lite beta program' and may have to sign an NDA to get a version of the Flash Lite player to run on your phone, which I declined. I think, to test
Combine with Second Life and seriously.
Not affiliated, not even a member, blah blah blah.
...rumors are that Dave Winer threw the first turd.
(Google cache, scroll down to highlighted words.)
The first images are not very good ones to start with. I suggest browsing down to the first set of images that do not have parts of the rover in them (a set of small hills on the horizon). Also, try resizing the browser so that only the two images you are trying to combine are in view and place the browser on a plain background such as a reasonably uncluttered desktop. Try both the cross-eyed and parallel set of images if you do not know your method - you'll know when you have it right because there will be a slight topographic roll to the surface nearby.
Once you get those, try keeping your eyes situated in the same position and scrolling the other images up or down into your field of view without looking up or down. This will allow you to view the more difficult images with parts of the rover in them, which have sharp depth transitions between the solar panels, airbags, and ground.
There is an interesting and informative entry on the NASA site regarding how much data can be transmitted back and forth between Earth and the rover:
http://marsrovers.nasa.gov/mission/comm_data.html
If we assume best case scenario for the transmission potential stated there and assume the direct-to-Earth rate averages the midpoint between the stated 12000bps and 3500bps, the total daily data for a single Martian day, direct-to-Earth and orbiter relay potential combined, is on the order of 17MB. The total data for the entire mission is on the order of 1,550MB.
Of course, this has to include protocol overhead, error, and operating instructions, but it provides one quantitative, if not qualitative, answer to how much data can be retrieved by the mission. In this case, a bit more than 2 CDs worth.
EverStoned and any others who like to listen to loud headphones, please read this immediately. Quote:
Hearing damage from headphones is probably more common than from loudspeakers, because many people exploit the acoustic isolation by listening at higher volumes. Moreover, the risk of hearing damage from headphones is higher than with loudspeakers, even at comparable volumes, due to the close coupling of the transducers to the ears. One of the benefits of headphone listening is the ability to detect musical details. Any hearing damage would have substantial impact on that experience. This article takes a look at the process of human hearing and offers guidelines for safe listening. (The information given here does not substitute for medical expertise. Readers should consult a physician for a diagnosis of hearing damage.)
If you are turning up the volume to drown out ambient noise, consider getting an set of earbuds like the Etymotic ER-4P.
EverStoned and any others who like to listen to loud headphones, please read this immediately. If you are turning up the volume to drown out ambient noise, consider getting an set of earbuds like the Etymotic ER-4P.
Have you tried the small dongle control where you can "dial down" the fan speed or are you generating too much heat for that? At 1500rpm, the fan is essentially silent. If you are at 1500rpm and it is noticable even with the case open you probably have a bad fan and should contact Zalman. Also, make sure it it attached securely and not vibrating (and transmitting those vibrations to the case as another poster mentioned).
wik
The information you are interested in: I am running a 2000FP via DVI from a Radeon 8500 64MB (regular version, not LE) at 1600x1200x32bitx60Hz in both Windows 2000 Professional and Red Hat 8.0 (with both original release and recently updated kernels) dual booted. The only thing I had to install was the Radon 8500 Catalyst 2.5 drivers (version 6.10.13.6813 obtained from ATI Web site, just the drivers not the ATI control panel) for Windows. The default Red Hat 8.0 X11 works at this resolution straight out of the install. I am using the DVI cable that came with the monitor.
For more on exactly this topic, see Vmyths.com. It's over the top at times, but there is some interesting speculation, rumor, and commentary there. The delivery to China of a cache of viruses by the anti-virus industry is a particulary interesting and some other established media outlets are picking up the story as well. Not affiliated, yada, yada.
DRM can't keep you from reading the material, as long as you pay the price. Some say, Well, how can you take a paragraph and copy it anymore? That's what we normally consider to be fair use. But the fact is, you can still do that. You might not be able to cut and paste but as long as you can read it, you can type it... It's just not as easy as it could be but it's not any harder than it was 30 years ago.
True, there is no way to plug the "analog hole." However, to revert back to it as the only mechanism for copying is to effectively undo 30 or more years of productivity enhancement through technology and features such as copy/paste. The point of technological advancement is to automate manual processes such as transcription.
Estimating the lack of productivity seems popularly acceptable for damages alleged by computer viruses. If we did a calculation for the lost productivity costs of DRM difficulties imposed on legitimate copying in business and academic work, it would likely be a large number, perhaps dwarfing the revenues protected by DRM. Therefore, Mr. Liebowitz's argument founders on a zero, or perhaps even negative, sum.
In the previous thread on this issue, I raised the question of legal liability of non-disclosure with the hypothetical case of a company hacked through this hole prior to the recent announcement, but after Microsoft learned of the hole, that suffered financial injury. Putting aside the specifics and focusing on the legal question, is this liability, if it exists, a business argument against non-disclosure?
Responses to the previous post indicate that the EULA is not a sufficient shield to hide behind and this is about willful non-disclosure of a known vulnerability. Does an implicit (in the legal sense) trust exist that vendors alert their customers to vulnerabilities as soon as they become known so that the customers can take immediate, non-patching action (e.g. disconnect from network, shut down affected services) to protect their systems until a patch is released?
Has that ever been tested in court, especially in light of an argument of gross negligence by the plaintiff? It seems posssible to enter into agreements and obtain a favorable decision notwithstanding such agreements in certain cases, especially those involving gross negligence or unreasonable clauses of those agreements.
What if a company was hacked during the last few weeks prior to full disclosure through this hole and suffered financial injury due to the intrusion? Is Microsoft perhaps negligent and legally liable for those injuries since it did not disclose the hole for five weeks after it knew of its existence?
"But if we could just move your molecules around, we could just take a "snapshot" of you when you were a healthy 18 year old, and then, years later, rearrange your decrepit 85 year old body, except the brain, to the original configuration."
Now that is a new definition of pain! Seriously though, great point about every kind of property changing into intellectual property in the presence of such technology.I disagree as to the inadequacy of the sampling size. I have no knowledge of how many items actually are on Freenet. However, the margin of error on 1070 items of a population of n > 1070 to infinity is not so high as to make the stats worthless. You can make an argument against accuracy no matter what the size of the data set since the basis is probability, but 1070 in a data set is fairly trustworthy, though not overly so.
I think the real concerns regarding the sample here are methods and the possible subjectivity of a single data analyzer. For example, a particular search pattern could yield content of a certain type more regularly than others; what is porn to one person might be art to another.
There is an excellent exhibit of time and Harrison's work, including several of the clocks at different stages in evolution, at the Royal Observatory in Greenwhich, England. If you think his work is impressive from reading the book, you should see some of his contraptions in person!
"The "killer app" I'm waiting for are the solar power satellites: high capital cost, but the lowest kilowatt/hour to manhour ratio of any power source. Let's live off the Sun, oh yeah!"
Question of curiosity: How do you efficiently get tcf:àower from the satellites to the ground?"The US has a lot of experience training for short-term mission. Now we need to learn what to do for long-term mission"
This is a very good point. I imagine it is one thing for astronauts to launch and work brutal hours in cramped, downright uncomfortable conditions with no clear "work day" knowing they are returning to relative vacation in a week or ten days; quite another to keep that tempo up for months on end. I agree some, if not most, of the problems are likely directly attributable to a short-term approach, by ground control, to long-term missions.
I cannot help but compare the problems you cite to the difficulties and tensions in large software projects. If the team on Alpha gets bigger, will the tensions relax (more hands to do the work) or intensify (more problems created, more interpersonal tension)? I ask because, with software projects, it can go both ways.
"insanity to put our people at the mercy of equipment that we didn't design and cannot control"
Like most computer users ;)?
"So who was the wench in Paul's room when the Hunter Seeker attacked? Hi - not true to the book"
I noticed this as well, but I understand why: There is no context for it when it first appears if you not a Dune expert. Thus there needs to be another person in the room for there to be a dialogue in which the purpose and capabilities of this thing are explained. It would be awkward to try to explain it and the behaviour of Paul (standing stock still, grabbing it when Mapes comes in) before or after the fact without losing either the suprise or the suspense.