Entirely missing the parent's point. The military training is for teaching the people to handle guns responsibly. Marksmanship skills are a side benefit.
Even if it were fully open, there is still the issue that, as a standard, it sucks. For everything from dates to mathematical formula notation, it specifies it's own sub-standard, ignoring existing standards. And those sub-standards are in many cases substandard. For example, take the date format. It enforces the old MS bug of treating 1900 as a leap year (and to counter the inevitable "but, but, backwards compatibility", the import filter can handle it). It also states that any dates before the base-date (1900/1/1 or 1904/1/1, not specified which one you should use, so your software must be able to handle both cases) are "ill-formed". So, want to do date calculations with historical data? Sorry, no can do.
OOXML is a "standard" that's bloated by chronic NIH syndrome that MS is afflicted by. Before I'd even consider supporting it as an ISO standard, it would have to be redone to use existing international standards whenever that is possible, and extending an existing standard if one is appropriate, but lacking a few necessary features. Of course, once that's done, it would be essentially equal to ODF, possibly with some features that MS claims ODF doesn't have. But that's why ODF allows extensions in vendor-specific namespaces. (And, for that matter, why MS were invited to take part in ODF development process. So there wouldn't be those missing features that prevent them from using it. But MS refused. And then decided to make their own "standard")
No. "Blood alcohol level as guessed based on breath contents" should never be considered equal to "blood alcohol level as measured from blood sample". The first is always going to be based on assumptions that may or may not apply during a particular test. Like, for example, did droplet of the testee's spittle end up on the detector? Oh, whoops, detector gives a rather higher reading than it should.
The hardware patent was on one particular way of doing automatic harvesting. Not on the whole idea of doing automatic harvesting. On the other hand, the MP3 patent is on the very idea of encoding sound in a certain way. Not on a particular encoder that implements that idea.
Why would the competitors want to write drivers for your hardware? Care to explain how you conjecture publishing API specs would help the competitors? Your example in your original post isn't valid, since you'll be telling the whole world about such great new features in your own damn marketing materials.
Do you really not understand that the documentation required for writing a driver is not the full hardware spec? The only people you will be helping by releasing an API doc is driver writers. Any competitors will still have to reverse engineer the hardware if they want to find out how it works.
Because there is non-zero amount of work involved in reverse-engineering how the proprietary drivers talk with the hardware. It can be done that way, but it's a major waste of effort.
This is about whether or not to approve a standard. The people who do not want the standard approved have pointed out technical issues with it, severe enough that they should be fixed before the standard should even be considered for approval.
But you get to ignore all those problems, because obviously anything any of the opponents say can never be based on facts.
The whole bloody point of a standard is to make choice be among competing, but compatible vendors instead of among incompatible vendors. The problem, then, with OOXML is that it stacks the deck against anyone other than MS attempting to implement it, being essentially just an XMLised dump of MS Office's binary format with all the legacy cruft included. In fact, due to the legacy cruft, no-one besides MS can ever be fully compatible with the standard. All anyone else can do is claim compatibility with most of the standard, minus certain optional bits.
And those optional bits will be rather crucial, since they lock up any documents converted from legacy formats to OOXML from being correctly interpreted by anyone other than MS. Contrast this with ODF, where any vendor has available to them the information they need to implement the entire standard and the legacy documents will be converted to generic ODF markup that can be correctly interpreted by any vendor that does implement the entire standard.
And as for why I fear OOXML becoming an ISO standard is that it will then automatically become the dominant standard due to MS Office being the dominant product at the moment. No need for that pesky competition based on it's actual merits vs. ODF's merits.
Except we're talking about commodity hardware here, so the competitor can get their hands on the actual hardware. And assuming they are not totally incompetent, they should be able to find out how that hardware works through direct examination.
Pointing out that all of your customers probably haven't switched yet might work. So the default format should probably be something all of them (and, incidentally, you with OOo, too) can open. You don't want to piss customers off by sending them mail they can't read unless they make an unexpected investment, now do you?:D
The point of the nonsense spam is to mess with adaptive spam filters. The idea is to train the filters so that the relative weight of whether the message looks like gibberish or not is high compared to other factors. Then you make the spam messages you want to actually reach the recipient look like legitimate messages.
Of course, I could be wrong about that, since that's just conjecture based on what I remember from my information theory courses.
The difference lies in whether you can, without having to resort to reverse engineering, understand all the details of what the document says. OOXML has several unspecified (technically specified, but the specification is just "Behaves as proprietary product X") behaviours, especially related to formatting.
MS isn't obligated to maintain ie32, afaik. And seeing as there are other browsers for 32-bit Windows, I doubt there would be any cause for legal action.
So according to your hypothetical people running companies, using the correct tool for the job is disorganized, and just using the hammer everywhere is organized?
It is you who should be kicked out of the theatre. You are interfering with everyone else watching & listening to the movie. Just because you don't like the movie doesn't mean you have the right to ruin it for everyone else.
Could be that you really have done your research before asking questions. Too bad some 10% - 20% of the other people asking questions haven't. RTFM is a valid response when it is clear the person asking the question hasn't even bothered taking a look at the manual. That they have not read the manual can be seen from, for example, the fact that the answer to the question of "How do I do X?" is in it's very own chapter of the manual, complete with screenshots of every step along the way. Of course, you can (and usually should) be polite when telling the losers to RTFM. But there does come a point where just telling them to RTFM is valid, because you know that the only reason they can be asking the question is that they couldn't be bothered to do any research at all before asking you (up to and including not reading the sticky topic titled "READ FIRST if you have questions about X" in the very same forum they are posting their question to).
Of course, I'm not sure I understand how they count the atoms. I'm also not sure how different isotopes of silicon are accounted for. (do they separating out one isotope and use it to make the sphere? Or are they assuming a certain mixture of specific isotopes?) But it seems like the big idea is to give a definition of the kilogram that references a more universal constant that could be replicated. To count the atoms: 1) Measure the size of the sphere. 2) Measure how far apart the individual silicon atoms are in the crystal lattice. Since the crystal is approximately perfect, that distance will be constant over the whole sphere. Through 1 & 2, the number of Si atoms in the sphere is calculable, with nice and low error margins. Isotope ratios can be calculated from another piece of Si from the same crystal the sphere is made of using mass spectrometry.
No. Spam used to reach you. Then you switched email providers to one that's wasting more effort on preventing spam reaching you than the one before. The point is, just because you aren't dealing with all the spam coming at you anymore doesn't mean someone else isn't.
Apps can be installed under ~. That should work, unless the app is retarded or execution of files under ~ is prevented for security reasons (aka "We don't trust you enough to let you do anything not explicitly allowed").
All I can say about that video is "root abuse, the pain, oh the pain".
Unless that was just a random stunt, I'd really recommend setting up a system where each student has their own account. Have it be mounted over network from a central fileserver, so they keep their data & configuration no matter which machine they happen to be using at the moment. You could also have that same data be mounted as a network drive Windows-side so the students can access their data regardless of which OS they are in.
Did you notice the parts about the whole specification not being available? Seeing as several parts of the specification just refer to other MS products and say "works as $PRODUCT does". That disqualifies OOXML from being a fully open format.
Most probably would. Remember that a part of military training is intended to turn normal people into sociopaths that can actually kill other humans in cold blood. Part of that is the dehumanisation of whomever is supposed to be killed today. Somewhat relatedly, your government has gotten itself the right to use troops in states other than where they are from (something called National Guard, IIRC).
Yes, they are. Certainly in the same scale. You could assign infinite as the value of a single human life, but then you end up being unable to be objective in several situations (for example, this, or evaluating risks). IMO, the value should be very high, because I'd like individual humans to be valued in those calculations. However, causing enough minor damage that fixing it in aggregate takes up several human lifetimes should be considered equivalent to taking up one human lifetime through the premature termination of a single life.
See, the hard thing is not making things run parallel. It's the synchronisation. Let's say there's some variable C that your procedure1 (=p1) and procedure2 (=p2) need to read and then alter. Assume that read & alter cannot be done as an atomic operation. How do you ensure that the procedures do clobber each other's modifications? That is, how do you prevent a sequence of events:
p1 reads C
p2 reads C
p1 writes to C
p2 writes to C - p1's change just got clobbered
What needs to happen is that first one procedure gets to perform it's read & alter on C, and only then can the other be allowed to access C. You need to temporarily prevent the parallel execution. The hard part is limiting the synchronized execution to only the minimum necessary to maintain correct execution.
Slashdot Mirror
Entirely missing the parent's point. The military training is for teaching the people to handle guns responsibly. Marksmanship skills are a side benefit.
Even if it were fully open, there is still the issue that, as a standard, it sucks. For everything from dates to mathematical formula notation, it specifies it's own sub-standard, ignoring existing standards.
And those sub-standards are in many cases substandard. For example, take the date format. It enforces the old MS bug of treating 1900 as a leap year (and to counter the inevitable "but, but, backwards compatibility", the import filter can handle it). It also states that any dates before the base-date (1900/1/1 or 1904/1/1, not specified which one you should use, so your software must be able to handle both cases) are "ill-formed".
So, want to do date calculations with historical data? Sorry, no can do.
OOXML is a "standard" that's bloated by chronic NIH syndrome that MS is afflicted by. Before I'd even consider supporting it as an ISO standard, it would have to be redone to use existing international standards whenever that is possible, and extending an existing standard if one is appropriate, but lacking a few necessary features.
Of course, once that's done, it would be essentially equal to ODF, possibly with some features that MS claims ODF doesn't have. But that's why ODF allows extensions in vendor-specific namespaces. (And, for that matter, why MS were invited to take part in ODF development process. So there wouldn't be those missing features that prevent them from using it. But MS refused. And then decided to make their own "standard")
No. "Blood alcohol level as guessed based on breath contents" should never be considered equal to "blood alcohol level as measured from blood sample". The first is always going to be based on assumptions that may or may not apply during a particular test.
Like, for example, did droplet of the testee's spittle end up on the detector? Oh, whoops, detector gives a rather higher reading than it should.
The hardware patent was on one particular way of doing automatic harvesting. Not on the whole idea of doing automatic harvesting.
On the other hand, the MP3 patent is on the very idea of encoding sound in a certain way. Not on a particular encoder that implements that idea.
Why would the competitors want to write drivers for your hardware? Care to explain how you conjecture publishing API specs would help the competitors? Your example in your original post isn't valid, since you'll be telling the whole world about such great new features in your own damn marketing materials.
Do you really not understand that the documentation required for writing a driver is not the full hardware spec? The only people you will be helping by releasing an API doc is driver writers. Any competitors will still have to reverse engineer the hardware if they want to find out how it works.
Because there is non-zero amount of work involved in reverse-engineering how the proprietary drivers talk with the hardware. It can be done that way, but it's a major waste of effort.
Sorry about the solid block o' text. It was supposed to be paragraphed, but a freak accident with comment mode selection ate those :P
This is about whether or not to approve a standard. The people who do not want the standard approved have pointed out technical issues with it, severe enough that they should be fixed before the standard should even be considered for approval. But you get to ignore all those problems, because obviously anything any of the opponents say can never be based on facts. The whole bloody point of a standard is to make choice be among competing, but compatible vendors instead of among incompatible vendors. The problem, then, with OOXML is that it stacks the deck against anyone other than MS attempting to implement it, being essentially just an XMLised dump of MS Office's binary format with all the legacy cruft included. In fact, due to the legacy cruft, no-one besides MS can ever be fully compatible with the standard. All anyone else can do is claim compatibility with most of the standard, minus certain optional bits. And those optional bits will be rather crucial, since they lock up any documents converted from legacy formats to OOXML from being correctly interpreted by anyone other than MS. Contrast this with ODF, where any vendor has available to them the information they need to implement the entire standard and the legacy documents will be converted to generic ODF markup that can be correctly interpreted by any vendor that does implement the entire standard. And as for why I fear OOXML becoming an ISO standard is that it will then automatically become the dominant standard due to MS Office being the dominant product at the moment. No need for that pesky competition based on it's actual merits vs. ODF's merits.
Except we're talking about commodity hardware here, so the competitor can get their hands on the actual hardware. And assuming they are not totally incompetent, they should be able to find out how that hardware works through direct examination.
Pointing out that all of your customers probably haven't switched yet might work. So the default format should probably be something all of them (and, incidentally, you with OOo, too) can open. :D
You don't want to piss customers off by sending them mail they can't read unless they make an unexpected investment, now do you?
The point of the nonsense spam is to mess with adaptive spam filters. The idea is to train the filters so that the relative weight of whether the message looks like gibberish or not is high compared to other factors. Then you make the spam messages you want to actually reach the recipient look like legitimate messages. Of course, I could be wrong about that, since that's just conjecture based on what I remember from my information theory courses.
The difference lies in whether you can, without having to resort to reverse engineering, understand all the details of what the document says. OOXML has several unspecified (technically specified, but the specification is just "Behaves as proprietary product X") behaviours, especially related to formatting.
MS isn't obligated to maintain ie32, afaik. And seeing as there are other browsers for 32-bit Windows, I doubt there would be any cause for legal action.
So according to your hypothetical people running companies, using the correct tool for the job is disorganized, and just using the hammer everywhere is organized?
It is you who should be kicked out of the theatre. You are interfering with everyone else watching & listening to the movie. Just because you don't like the movie doesn't mean you have the right to ruin it for everyone else.
Could be that you really have done your research before asking questions. Too bad some 10% - 20% of the other people asking questions haven't.
RTFM is a valid response when it is clear the person asking the question hasn't even bothered taking a look at the manual. That they have not read the manual can be seen from, for example, the fact that the answer to the question of "How do I do X?" is in it's very own chapter of the manual, complete with screenshots of every step along the way.
Of course, you can (and usually should) be polite when telling the losers to RTFM. But there does come a point where just telling them to RTFM is valid, because you know that the only reason they can be asking the question is that they couldn't be bothered to do any research at all before asking you (up to and including not reading the sticky topic titled "READ FIRST if you have questions about X" in the very same forum they are posting their question to).
1) Measure the size of the sphere.
2) Measure how far apart the individual silicon atoms are in the crystal lattice. Since the crystal is approximately perfect, that distance will be constant over the whole sphere.
Through 1 & 2, the number of Si atoms in the sphere is calculable, with nice and low error margins.
Isotope ratios can be calculated from another piece of Si from the same crystal the sphere is made of using mass spectrometry.
No. Spam used to reach you. Then you switched email providers to one that's wasting more effort on preventing spam reaching you than the one before. The point is, just because you aren't dealing with all the spam coming at you anymore doesn't mean someone else isn't.
Apps can be installed under ~. That should work, unless the app is retarded or execution of files under ~ is prevented for security reasons (aka "We don't trust you enough to let you do anything not explicitly allowed").
All I can say about that video is "root abuse, the pain, oh the pain". Unless that was just a random stunt, I'd really recommend setting up a system where each student has their own account. Have it be mounted over network from a central fileserver, so they keep their data & configuration no matter which machine they happen to be using at the moment. You could also have that same data be mounted as a network drive Windows-side so the students can access their data regardless of which OS they are in.
Ctrl+Enter for line break instead of send in Kopete.
Did you notice the parts about the whole specification not being available? Seeing as several parts of the specification just refer to other MS products and say "works as $PRODUCT does". That disqualifies OOXML from being a fully open format.
Most probably would. Remember that a part of military training is intended to turn normal people into sociopaths that can actually kill other humans in cold blood. Part of that is the dehumanisation of whomever is supposed to be killed today. Somewhat relatedly, your government has gotten itself the right to use troops in states other than where they are from (something called National Guard, IIRC).
Yes, they are. Certainly in the same scale.
You could assign infinite as the value of a single human life, but then you end up being unable to be objective in several situations (for example, this, or evaluating risks). IMO, the value should be very high, because I'd like individual humans to be valued in those calculations.
However, causing enough minor damage that fixing it in aggregate takes up several human lifetimes should be considered equivalent to taking up one human lifetime through the premature termination of a single life.
See, the hard thing is not making things run parallel. It's the synchronisation.
Let's say there's some variable C that your procedure1 (=p1) and procedure2 (=p2) need to read and then alter. Assume that read & alter cannot be done as an atomic operation. How do you ensure that the procedures do clobber each other's modifications?
That is, how do you prevent a sequence of events:
p1 reads C
p2 reads C
p1 writes to C
p2 writes to C - p1's change just got clobbered
What needs to happen is that first one procedure gets to perform it's read & alter on C, and only then can the other be allowed to access C. You need to temporarily prevent the parallel execution. The hard part is limiting the synchronized execution to only the minimum necessary to maintain correct execution.