I was in San Francisco for WWDC this year, which was my first taste of the U.S.A., and I must say while I had a good time and was happy overall, I quickly developed a very real fear about venturing anywhere near a road - the road rules seem chaotic compared to Australia's, and most drivers seem to have only two settings: bat out of hell, or sunday.
Also, the freeways in L.A. are the worst made roads in the world. Sure, I haven't travelled on every single road in the world, but I can't easily imagine anything worse, even as a hypothetical.
For reference, the average number of road fatalities per year in Australia is roughly 9 per 100,000 people. In the U.S.A., it's closer to twice that, at around 15. See this blog. Up to date stats for Australia also available from the ABS (Australia Bureau of Statistics); can't find a U.S.A. equivalent.
The biggest problem with Australias roads, imho, is that idiots can get licenses, while good drivers can easily luck out and fail license tests. The number of young drivers I know who have *never in their entire life* parallel parked, or reversed into a parking spot, is just embarassing. And I know a number of people who've failed license tests because, for example, they braked suddenly to avoid an animal on the road (a dog, from memory). Well, that might be the right thing to do, they're told, but that's an automatic fail. Sheesh.
Then again, for whatever advantage we have in road safety, we make up for it with one of the worst public transport networks imaginable. Today, for example, I had to wait 45 minutes for a tram (which, ultimately was a bus chartered as a replacement), because the power was out over the last 7 or so km of the line. Fine, that happens. The annoying thing was that no one had any idea what was going on. There were at least half a dozen people from the tram company there, and none of them had any idea what was going on, excepting one had heard some mention of a bus. Or something. He only said anything at all after being prompted.
That's the third time this week I've had to wait an obscene amount of time for a tram. And it's not even Friday yet. But at least I get to rant about it.:P
Try as I might, I just can't take someone seriously - in any argument - when they can't even spell whatever they're arguing about.. MAC? Why do people persist with this errant capitalisation?
How much can someone possibly know about an entity if they can't even get the name right, I mean really...
The irony of course is that if we do let the U.S. do whatever the hell it wants on this, as usual, there won't be an Earth 500 years from now, so it really is a good solution.
Don't be so quick to wipe your hands clean because you felt "threatened". While I would probably defend myself to the death - as a basic animal instinct for survival - I wouldn't consider myself innocent, regardless.
On a higher level, if I had to resort to such measures I would always think of myself as guilty. I like to think that such things can always be avoided, and if I did not manage to do so, that is a failing in myself.
From a legal point of view, I believe in Australia (and parts of Europe) killing someone in "self defense" could lead to a manslaughter charge, regardless of how you felt at the time - although the court will evaluate the particular circumstances.
Too much tolerance for "reasonably certain" leads to prejudice and fear. This is a common theme in a lot of murder movies, particularly military ones - as a simple and extreme example, consider this: if someone shoots you and misses, do you really know they actually intended to hit you? There was a good episode of J.A.G. which covered this well... although imho the guy did really miss, and was a good enough shot second time round to get out of it.;)
Re:A very incomplete list off the top of my head
on
Longhorn Beta Begins
·
· Score: 3, Interesting
"One video showed two Notepads rotating around while still completely usable at the same time a video played in Media Player. Old apps will be compatible."
I've never heard of Rice Krispies, but Rice Bubbles are a very common breakfast cereal in Australia. They are exactly as the author describes them in the original story, so I suspect they're also available in the U.S.A.
No point beating around the bush - best to nip the whole problem in the bud.
FWIW, in a college I lived in for three years we had absolutely no security for as many as 1000 people, and we never had any significant network issues, despite the constant virii and other malware roaming around.
IMHO, the best solution is to just "shape" bad users down to the slowest speed possible - dialup if your switch supports QoS for it, otherwise just 10 m/bit or similar. One bad user getting disconnected and whining to someone above you could get you in a bit of trouble - but sapping their speed won't be a reprimandable offense, and will curtail a large part of the problem.
And I wouldn't worry too much about being speedy about removing the limits - just tell them the system is updated once a week, and the next update happens to be just under 7 days from whenever they demand it.:)
The author of this blog makes the assertion that it's too difficult to pre-compute hashes for a 42-character password. But of course the length of the password is [ultimately & eventually] irrelevant - what you want is anything that hashes to the same 20 bytes as the 'real' password. Granted, a 20-byte hash collision is still far harder to obtain than your typical 8-character passwords, but it does limit the usefulness of ever-increasingly-long passwords.... I daresay beyond 40 characters of unicode you're just adding fluff.
No, Civ II & Civ III (and Alpha Centauri, as someone else said) all have this as well. It's a pity they haven't expanded on it a bit - I'd like to be able to select different attributes with which to colour code the map (e.g. military presence, to watch the build-up's and conflicts).
It's that simple really. OpenOffice/NeoOffice/whatever are not suitable replacements for MS Office on a Mac - and this "but Firefox is popular and it has a shit lowest-common-denominator interface" is not helping.
Well, each to their own. Different projects will have different requirements, in terms of creativity, for their code, documentation et al. In my experience writing it, documentation should be formulated, rigorous and consistant. Consequently, getting creative while writing it is rarely a good idea. Conversely, I've worked on projects coding in entirely the same way. But these are relatively rare exceptions.
I've had similar experiences and concerns. My conclusion is that you only get a few good hours of creative coding per day, if you're lucky. So if you spend that at work, you'll have none left for your own interests. While there's a few ways to solve this (not doing any real work at work is one;) ), I find the best is to alternate each day between menial and creative tasks. So set aside some days at work to do documentation, specing, testing or whatever, which will leave you with the motivation to do some actually coding when you get home. And then the converse, where you can still do useful things (e.g. documentation) at home, after a good day of coding at work.
Well, I suppose that's a fair critique of my comment; I didn't really present my point very clearly, I know.
What I meant was, if I sat down with the Linux kernel and essentially "paraphrased it", i.e. read each functional unit one by one, and then rewrote my own - without any direct copying - but functionally identical to the original, how would it sit with different groups for me to then release that source as my own (under my own license)?
It seems clear that in a traditional corporate view this is quite clearly immorale, if not illegal. There is of course some degrees of grey even in this example, but nonetheless they're all pretty dark anyway.
From the point of view of the free source proponent, this is probably a perfectly valid thing to do - you're reinventing your own wheel, with practical inspiration from an existing one.
I personally imagine the FSF as leaning more towards the corporate case, to be honest. The GPL to me is just a way of enforcing your own philosophy on other people, which of course fails if I go and do some trivial workaround like the example given. They'd say I've broken the GPL in concept, if not in law.
I suppose I didn't have a sharp point with my original example... more I was trying to highlight all the hypocracy coming from every angle. Ken Brown is pushing BSD-licenses, while accusing Linus of "paraphrasing" Unix, which is infact the kind of sharing the BSD philosophy encourages. The idea of Linux "paraphrasing" Unix is offensive to Linux, the FSF and most open-sourcies (mainly because they want to protect Linus at all costs, I suspect). So the irony here is that Ken Brown is accusing Linus of doing what he thinks is just fine anyway, and the Linux community is getting up in arms about it because *they're* the ones who's philosophy's been violated.
Since I'm a BSD-loving Mac user, I'm sort of once-removed from all this, so I can look and laugh. I don't mean to offend those involved with my indifference - I'd just love to see them embrace the entire problem, rather than hiding behind their stubborn walls of pre-scripted defenses and inflamed heads. While Mr. Brown's methods may leave a lot to be desired, it's no reason to discount his opinion universally. And vice versa, of course.
Sadly, I feel Mr. Brown is making more considerations for his opponents than they for him. Not much more, mind you, given his relative lack of response to the whole issue, but some.
It sounds to me like the non-too-subtle purpose of the controversial report is to promote BSD-style licenses over GPL (copyleft) ones. So if you can bear with me for a moment and ignore the issues of theft and whatever, the report is ultimately still pro-free-source. Just free as in freedom, true freedom, not the hypocritical so-called "free source" the FSF pushes. No wonder it's inciting such a response from that very community.
From my point of view, granted as a proponent of the BSD-style licenses, what I've read of the report seems reasonably valid. I'm prepared to take Linus' word that he wrote Linux, but it's undeniable that the early hacker ethic and rebellion against the establishment doesn't cast a favourable light over that generation. Imho, the kind of "borrowing", or more aptly put "inspiration", between different sources and histories is an essential part of modern free-source coding... it's naturally at ends with the capitalist model that prevailed through the corporate heads at that time, and that's probably the only reason why any of this is a real issue.
Then again, it's an interesting thought exercise - if I grabbed the latest copy of Linux (the kernel, the true "Linux"), and rewrote it in my own words, so to speak.. then posted that derived version as truly free (e.g. BSD licensed), who amongst the GPL and Linux crowd wouldn't scream blue murder?
...but nothing useful for your purposes, in this case. The local messaging uses a subset of the Jabber protocol (as opposed to the AIM protocol used for peer->server->peer messaging). I think the file transfer code is also based in part on Jabber, although I do know that some of the essential specifics are proprietary and undocumented (the original developer made the note a few weeks ago that he can't even remember how it works anymore).
So while it does borrow from Jabber, it doesn't "use" Jabber. You can't connect to a Jabber server, nor communicate directly between Jabber & iChat.
There are some chat clients out there with similar functionality on Windows (and I believe Linux), but they're somewhat hit-and-miss affairs, from my experience. I like the suggestion to just get an iBook - I already have one:) - but a more reasonable idea is probably to look at Proteus, as someone mentioned, and think about some sort of Windows/Linux/your-poison port.
I have a Panasonic NV-28 (no longer available, but the newer models are basically the same) and it does this just fine. There's significant latency though - about a quarter of a second, so pointing the camera at the screen [showing the video] doesn't quite work as well as it does on a proper setup. This may well be due to the fact that the machine doing this is a 400mhz G4, which can only *just* capture video at full speed, assuming I run in OS 9 (no pre-emption) and kill every other app (including the Finder). And the audio still loses sync with the video over time, although I think that's due to bugs in Final Cut Pro & Express.
Ironically, on a machine perfectly capable of handling this sort of video grunt - an 800mhz iBook - the hard drive is too slow... grrr..:)
I would have thought this is standard fare on all Firewire camera's... I certainly see no reason why they shouldn't be able to do it - after all, they can all output in realtime via component/separated/your-poison. In fact, I can output video from my computer to the camera [via Firewire], and output from the camera to a tv, all in real time (without the annoying lag for the reverse case).
Unfortunately I can't go the other way, so I can't quite build myself a psuedo-TiVo just yet.:)
10.3.3 is indeed much faster than previous version of Panther, but in my case this is because pre-binding now works. Previously, ever since upgrading to Panther from 10.2.8, any attempt to pre-bind anything would fail. I ended up with log files in the 20-30 meg range full of meaningless error messages. 10.3.3 has finally fixed that. As an example, iTunes now bounces just once (as it should), rather than about 15 times.
Pity it still takes about 30 seconds to load my music library after that, but I'll probably get an iPod in a few days, so that'll solve that problem.:)
[Sorry, just had to gloat about the iPod;)]
Unfortunately, startup is still much slower than earlier versions of Panther. It always sits on the "Loading login window" bit for about 30 seconds (i.e. more than half the total startup time), which is a 'bug' introduced in 10.3.2 I think. Previously this step took at most 3 seconds.
If I create a library that I wish to use in both open source and in proprietary products, GPLing it does not prevent me from putting it in my proprietary product - as copyright holder, I can do what I want. GPLing would prevent someone else from an "embrace and extend" attack, which BSD style licenses don't protect against.
But it prevents others putting it in their products, because it's offen not possible for commercial or even public institutions to release their entire source base. NDA's have been mentioned as one reason, another is the simple fact that you may want to use some source - with appropriate credit - but don't want to be forced into a relatively limited licensing scheme.
This is of course what instigated the BSD-vs-GNU religious war. You believe that embracing and extending source privately is an "attack" on... well, on what? On the other hand, I think the GPL is an attack on the developer's freedom. In my opinion it's an outright offense to force your views onto other people, which is what the GPL is doing. With a BSD-style license, you can have your own view on how source should be free, without forcing others to follow it. This is what's traditionally known as "freedom of thought". Of course, you're free to think otherwise.
The result of the GPL has been the evolution of a large but aggressive community devoted to it, while everyone else outside has no choice but to find alternatives. This is disappointing - many excellent developers are making their work inaccessible to the larger populace, for reasons that are often misguided, or by poor conclusions.
Ironically, proponents of the GPL claim it is designed to encourage benefits to the world at large. Yet in reality this is not true - your excellent source may, by virtue of being GPL'd, not be an option, and thus much of the world settles on what may well be an inferior - but free - solution. Thus, GPL'd source does not inherently improve our quality of life - it may indeed hinder it, if you take the view that GPL'd source is lost opportunity.
But, let's not deteriorate into a flame war [any further, at least]. The original aim was to make the source available to others in the industry. In a highly vertical industry, it is very unlikely that the GPL would allow any other company to use the source. As I originally said, there may not be a need for a BSD-style license; a compromise can be made. There are many such "creative commons" licenses available.
You might find that people in similar industries will be interested in integrating some particular level of your vertical solution into their own vertical solutions, which will mean they'll componetise them, and consequently make them even more widely applicable and attractive.
Just remember though that if you want to actually see your code used [legally] by commercial interests, the standard GPL probably isn't suitable. While a BSD-style license is perhaps going too far the other way, there are various "creative commons"-type licenses available which retain some degree of responsibility amongst users of your code, without ruling it for them.
I hate cheaters and those people who ask for help, then expect you to do the problem for them, while they "watch on and learn". Yeah, right.
However, as much as I wish they'd all fall off the edge of the earth, it takes a lot of true positives to make up for just one false positive. I know several people who've been wrongfully accussed of cheating. It's not pleasant, for the obvious reasons, and apart from alienating the accussed from the accussors - often their lecturers & tutors - it can have dramatic long term effects. I've seen one person who was very intelligent, but did the usual thing - didn't turn up to lectures, didn't bother doing homework if they were busy doing their own thing, etc. When they got near perfect marks on the exam, they "had to be cheating". They couldn't defend themselves through prior work, since they hadn't submitted any, so even though they were never formally punished - no hard evidence - they became completely switched off to all school work. Didn't bother turning up to exams, in the end, and that was the end of that.
Granted, it's not often that people are accussed of cheating simply because they do well, but it does happen. I don't want to be dramatic and say it "ruins peoples lives", but a certain someone would be doing a heck of a lot better if it weren't for that one wrong accusation.
...but what about an Xserve RAID? The 720 gig one is $5999, the 1260 gig one $7499, and the 2520 gig one $10999. The middle one sounds like what you're looking for, and still within your price range, going by what you said. Plus, you could add another 1260 gig to it at any time later - probably even more given the inevitable increase in drive capacity. That should keep you going for another two years at least.
And since it includes redundant power supplies and [I believe] some surge protection, it should keep your data much safer than some generic Lintel box.
Slashdot Mirror
Back in my day, we used to play this game called "sex" with our girlfriends. 'course, things may have changed a lot with this new online generation.
Sooner or later someone will have to introduce Bonk Bonk Revolution, just to maintain the population.
If he were American, he wouldn't be using an ad-blocker because it'd be illegal under some obscure clause of the DMCA.
Also, the freeways in L.A. are the worst made roads in the world. Sure, I haven't travelled on every single road in the world, but I can't easily imagine anything worse, even as a hypothetical.
For reference, the average number of road fatalities per year in Australia is roughly 9 per 100,000 people. In the U.S.A., it's closer to twice that, at around 15. See this blog. Up to date stats for Australia also available from the ABS (Australia Bureau of Statistics); can't find a U.S.A. equivalent.
The biggest problem with Australias roads, imho, is that idiots can get licenses, while good drivers can easily luck out and fail license tests. The number of young drivers I know who have *never in their entire life* parallel parked, or reversed into a parking spot, is just embarassing. And I know a number of people who've failed license tests because, for example, they braked suddenly to avoid an animal on the road (a dog, from memory). Well, that might be the right thing to do, they're told, but that's an automatic fail. Sheesh.
Then again, for whatever advantage we have in road safety, we make up for it with one of the worst public transport networks imaginable. Today, for example, I had to wait 45 minutes for a tram (which, ultimately was a bus chartered as a replacement), because the power was out over the last 7 or so km of the line. Fine, that happens. The annoying thing was that no one had any idea what was going on. There were at least half a dozen people from the tram company there, and none of them had any idea what was going on, excepting one had heard some mention of a bus. Or something. He only said anything at all after being prompted.
That's the third time this week I've had to wait an obscene amount of time for a tram. And it's not even Friday yet. But at least I get to rant about it. :P
Try as I might, I just can't take someone seriously - in any argument - when they can't even spell whatever they're arguing about.. MAC? Why do people persist with this errant capitalisation?
How much can someone possibly know about an entity if they can't even get the name right, I mean really...
"I hate Windows 99, it's so crappy."...
The irony of course is that if we do let the U.S. do whatever the hell it wants on this, as usual, there won't be an Earth 500 years from now, so it really is a good solution.
Don't be so quick to wipe your hands clean because you felt "threatened". While I would probably defend myself to the death - as a basic animal instinct for survival - I wouldn't consider myself innocent, regardless.
;)
On a higher level, if I had to resort to such measures I would always think of myself as guilty. I like to think that such things can always be avoided, and if I did not manage to do so, that is a failing in myself.
From a legal point of view, I believe in Australia (and parts of Europe) killing someone in "self defense" could lead to a manslaughter charge, regardless of how you felt at the time - although the court will evaluate the particular circumstances.
Too much tolerance for "reasonably certain" leads to prejudice and fear. This is a common theme in a lot of murder movies, particularly military ones - as a simple and extreme example, consider this: if someone shoots you and misses, do you really know they actually intended to hit you? There was a good episode of J.A.G. which covered this well... although imho the guy did really miss, and was a good enough shot second time round to get out of it.
"One video showed two Notepads rotating around while still completely usable at the same time a video played in Media Player. Old apps will be compatible."
a ted%20Windows.sitx
MacOS X 10.2, August 23, 2002.
http://www.atzenbeck.de/research/wildWindows/
http://homepage.cs.latrobe.edu.au/wjtregaskis/Rot
I've never heard of Rice Krispies, but Rice Bubbles are a very common breakfast cereal in Australia. They are exactly as the author describes them in the original story, so I suspect they're also available in the U.S.A.
No point beating around the bush - best to nip the whole problem in the bud.
:)
FWIW, in a college I lived in for three years we had absolutely no security for as many as 1000 people, and we never had any significant network issues, despite the constant virii and other malware roaming around.
IMHO, the best solution is to just "shape" bad users down to the slowest speed possible - dialup if your switch supports QoS for it, otherwise just 10 m/bit or similar. One bad user getting disconnected and whining to someone above you could get you in a bit of trouble - but sapping their speed won't be a reprimandable offense, and will curtail a large part of the problem.
And I wouldn't worry too much about being speedy about removing the limits - just tell them the system is updated once a week, and the next update happens to be just under 7 days from whenever they demand it.
The author of this blog makes the assertion that it's too difficult to pre-compute hashes for a 42-character password. But of course the length of the password is [ultimately & eventually] irrelevant - what you want is anything that hashes to the same 20 bytes as the 'real' password. Granted, a 20-byte hash collision is still far harder to obtain than your typical 8-character passwords, but it does limit the usefulness of ever-increasingly-long passwords.... I daresay beyond 40 characters of unicode you're just adding fluff.
No, Civ II & Civ III (and Alpha Centauri, as someone else said) all have this as well. It's a pity they haven't expanded on it a bit - I'd like to be able to select different attributes with which to colour code the map (e.g. military presence, to watch the build-up's and conflicts).
It's that simple really. OpenOffice/NeoOffice/whatever are not suitable replacements for MS Office on a Mac - and this "but Firefox is popular and it has a shit lowest-common-denominator interface" is not helping.
Of course, you wouldn't want to take repeated photographs of the sun with a digital camera, because you'll invariably burn the CCD out pretty quickly.
Well, each to their own. Different projects will have different requirements, in terms of creativity, for their code, documentation et al. In my experience writing it, documentation should be formulated, rigorous and consistant. Consequently, getting creative while writing it is rarely a good idea. Conversely, I've worked on projects coding in entirely the same way. But these are relatively rare exceptions.
I've had similar experiences and concerns. My conclusion is that you only get a few good hours of creative coding per day, if you're lucky. So if you spend that at work, you'll have none left for your own interests. While there's a few ways to solve this (not doing any real work at work is one ;) ), I find the best is to alternate each day between menial and creative tasks. So set aside some days at work to do documentation, specing, testing or whatever, which will leave you with the motivation to do some actually coding when you get home. And then the converse, where you can still do useful things (e.g. documentation) at home, after a good day of coding at work.
Well, I suppose that's a fair critique of my comment; I didn't really present my point very clearly, I know.
What I meant was, if I sat down with the Linux kernel and essentially "paraphrased it", i.e. read each functional unit one by one, and then rewrote my own - without any direct copying - but functionally identical to the original, how would it sit with different groups for me to then release that source as my own (under my own license)?
It seems clear that in a traditional corporate view this is quite clearly immorale, if not illegal. There is of course some degrees of grey even in this example, but nonetheless they're all pretty dark anyway.
From the point of view of the free source proponent, this is probably a perfectly valid thing to do - you're reinventing your own wheel, with practical inspiration from an existing one.
I personally imagine the FSF as leaning more towards the corporate case, to be honest. The GPL to me is just a way of enforcing your own philosophy on other people, which of course fails if I go and do some trivial workaround like the example given. They'd say I've broken the GPL in concept, if not in law.
I suppose I didn't have a sharp point with my original example... more I was trying to highlight all the hypocracy coming from every angle. Ken Brown is pushing BSD-licenses, while accusing Linus of "paraphrasing" Unix, which is infact the kind of sharing the BSD philosophy encourages. The idea of Linux "paraphrasing" Unix is offensive to Linux, the FSF and most open-sourcies (mainly because they want to protect Linus at all costs, I suspect). So the irony here is that Ken Brown is accusing Linus of doing what he thinks is just fine anyway, and the Linux community is getting up in arms about it because *they're* the ones who's philosophy's been violated.
Since I'm a BSD-loving Mac user, I'm sort of once-removed from all this, so I can look and laugh. I don't mean to offend those involved with my indifference - I'd just love to see them embrace the entire problem, rather than hiding behind their stubborn walls of pre-scripted defenses and inflamed heads. While Mr. Brown's methods may leave a lot to be desired, it's no reason to discount his opinion universally. And vice versa, of course.
Sadly, I feel Mr. Brown is making more considerations for his opponents than they for him. Not much more, mind you, given his relative lack of response to the whole issue, but some.
It sounds to me like the non-too-subtle purpose of the controversial report is to promote BSD-style licenses over GPL (copyleft) ones. So if you can bear with me for a moment and ignore the issues of theft and whatever, the report is ultimately still pro-free-source. Just free as in freedom, true freedom, not the hypocritical so-called "free source" the FSF pushes. No wonder it's inciting such a response from that very community.
From my point of view, granted as a proponent of the BSD-style licenses, what I've read of the report seems reasonably valid. I'm prepared to take Linus' word that he wrote Linux, but it's undeniable that the early hacker ethic and rebellion against the establishment doesn't cast a favourable light over that generation. Imho, the kind of "borrowing", or more aptly put "inspiration", between different sources and histories is an essential part of modern free-source coding... it's naturally at ends with the capitalist model that prevailed through the corporate heads at that time, and that's probably the only reason why any of this is a real issue.
Then again, it's an interesting thought exercise - if I grabbed the latest copy of Linux (the kernel, the true "Linux"), and rewrote it in my own words, so to speak.. then posted that derived version as truly free (e.g. BSD licensed), who amongst the GPL and Linux crowd wouldn't scream blue murder?
The only enemy of true freedom is hypocracy.
As a Mac user, I hope so. :)
...but nothing useful for your purposes, in this case. The local messaging uses a subset of the Jabber protocol (as opposed to the AIM protocol used for peer->server->peer messaging). I think the file transfer code is also based in part on Jabber, although I do know that some of the essential specifics are proprietary and undocumented (the original developer made the note a few weeks ago that he can't even remember how it works anymore).
:) - but a more reasonable idea is probably to look at Proteus, as someone mentioned, and think about some sort of Windows/Linux/your-poison port.
So while it does borrow from Jabber, it doesn't "use" Jabber. You can't connect to a Jabber server, nor communicate directly between Jabber & iChat.
There are some chat clients out there with similar functionality on Windows (and I believe Linux), but they're somewhat hit-and-miss affairs, from my experience. I like the suggestion to just get an iBook - I already have one
I have a Panasonic NV-28 (no longer available, but the newer models are basically the same) and it does this just fine. There's significant latency though - about a quarter of a second, so pointing the camera at the screen [showing the video] doesn't quite work as well as it does on a proper setup. This may well be due to the fact that the machine doing this is a 400mhz G4, which can only *just* capture video at full speed, assuming I run in OS 9 (no pre-emption) and kill every other app (including the Finder). And the audio still loses sync with the video over time, although I think that's due to bugs in Final Cut Pro & Express.
:)
:)
Ironically, on a machine perfectly capable of handling this sort of video grunt - an 800mhz iBook - the hard drive is too slow... grrr..
I would have thought this is standard fare on all Firewire camera's... I certainly see no reason why they shouldn't be able to do it - after all, they can all output in realtime via component/separated/your-poison. In fact, I can output video from my computer to the camera [via Firewire], and output from the camera to a tv, all in real time (without the annoying lag for the reverse case).
Unfortunately I can't go the other way, so I can't quite build myself a psuedo-TiVo just yet.
10.3.3 is indeed much faster than previous version of Panther, but in my case this is because pre-binding now works. Previously, ever since upgrading to Panther from 10.2.8, any attempt to pre-bind anything would fail. I ended up with log files in the 20-30 meg range full of meaningless error messages. 10.3.3 has finally fixed that. As an example, iTunes now bounces just once (as it should), rather than about 15 times. Pity it still takes about 30 seconds to load my music library after that, but I'll probably get an iPod in a few days, so that'll solve that problem. :)
[Sorry, just had to gloat about the iPod ;)]
Unfortunately, startup is still much slower than earlier versions of Panther. It always sits on the "Loading login window" bit for about 30 seconds (i.e. more than half the total startup time), which is a 'bug' introduced in 10.3.2 I think. Previously this step took at most 3 seconds.
But it prevents others putting it in their products, because it's offen not possible for commercial or even public institutions to release their entire source base. NDA's have been mentioned as one reason, another is the simple fact that you may want to use some source - with appropriate credit - but don't want to be forced into a relatively limited licensing scheme.
This is of course what instigated the BSD-vs-GNU religious war. You believe that embracing and extending source privately is an "attack" on... well, on what? On the other hand, I think the GPL is an attack on the developer's freedom. In my opinion it's an outright offense to force your views onto other people, which is what the GPL is doing. With a BSD-style license, you can have your own view on how source should be free, without forcing others to follow it. This is what's traditionally known as "freedom of thought". Of course, you're free to think otherwise.
The result of the GPL has been the evolution of a large but aggressive community devoted to it, while everyone else outside has no choice but to find alternatives. This is disappointing - many excellent developers are making their work inaccessible to the larger populace, for reasons that are often misguided, or by poor conclusions.
Ironically, proponents of the GPL claim it is designed to encourage benefits to the world at large. Yet in reality this is not true - your excellent source may, by virtue of being GPL'd, not be an option, and thus much of the world settles on what may well be an inferior - but free - solution. Thus, GPL'd source does not inherently improve our quality of life - it may indeed hinder it, if you take the view that GPL'd source is lost opportunity.
But, let's not deteriorate into a flame war [any further, at least]. The original aim was to make the source available to others in the industry. In a highly vertical industry, it is very unlikely that the GPL would allow any other company to use the source. As I originally said, there may not be a need for a BSD-style license; a compromise can be made. There are many such "creative commons" licenses available.
You might find that people in similar industries will be interested in integrating some particular level of your vertical solution into their own vertical solutions, which will mean they'll componetise them, and consequently make them even more widely applicable and attractive.
Just remember though that if you want to actually see your code used [legally] by commercial interests, the standard GPL probably isn't suitable. While a BSD-style license is perhaps going too far the other way, there are various "creative commons"-type licenses available which retain some degree of responsibility amongst users of your code, without ruling it for them.
However, as much as I wish they'd all fall off the edge of the earth, it takes a lot of true positives to make up for just one false positive. I know several people who've been wrongfully accussed of cheating. It's not pleasant, for the obvious reasons, and apart from alienating the accussed from the accussors - often their lecturers & tutors - it can have dramatic long term effects. I've seen one person who was very intelligent, but did the usual thing - didn't turn up to lectures, didn't bother doing homework if they were busy doing their own thing, etc. When they got near perfect marks on the exam, they "had to be cheating". They couldn't defend themselves through prior work, since they hadn't submitted any, so even though they were never formally punished - no hard evidence - they became completely switched off to all school work. Didn't bother turning up to exams, in the end, and that was the end of that.
Granted, it's not often that people are accussed of cheating simply because they do well, but it does happen. I don't want to be dramatic and say it "ruins peoples lives", but a certain someone would be doing a heck of a lot better if it weren't for that one wrong accusation.
And since it includes redundant power supplies and [I believe] some surge protection, it should keep your data much safer than some generic Lintel box.