Yet around here people are glad of the "nice weather." This time of year there should be some serious snow on the ground around here, not partially green grass. Sooner or later, we're all going to pay for the "nice weather."
Every agency or tech company I worked with had management and techs available who'd do weekend interviews if it was the only time some one could get together. Far more often an evening or morning phone call was arranged.
Even if you're working 12 hours a day and commuting an hour each way, that's still only 14 hours. That leaves you 2-3 hours per workday with 7-8 hours of sleep. Your choice, but arranging an interview call might be something to consider doing with that time.
It may be that you can get a packet-switch for a couple hundred dollars nowadays, but have you ever tried to convince mom & pop to replace something that "works just fine" from their perspective? There are a lot of places using old-fashioned hubs, and that won't be changing until the equipment has to be replaced.
The most basic emotions/states -- fear, lust, hunger, etc. have their roots in basic survival drives. Those instinctual states drive more complex emotional states -- anger (fear opting for aggressive defense), love (lust opting for protective bonding), hatred (fear opting for defensive avoidance and/or aggressive defense), etc.
The military would want to keep those aggressive tendencies, whereas the general public would obviously prefer AIs that don't have a survival/self-defense initiative. Self-defense triggered by a fear of dissolution could have nasty consequences with AIs in control of armaments or heavy equipment.
How to avoid "emotional baggage" if the AI entity is to learn from it's experiences? Take regular backups and restore to a pre-bad-experience state? Restoration denies the AI the ability to learn the consequences of real world situations, but you do need some way to keep it from developing hatred or paranoid fear. Are AI psychoanalysts needed? Sanity evaluations?
Seems to me that modelling emotions with AIs is actually what buys most of the risks.
Personally I think the only solution is to never let an AI make a final decision -- always require that a human push that final button/pull the trigger, even if they're doing it remotely with an audio/video feed.
The problem isn't the download time, the problem is that you're saturating the network for the whole subnet/department/workgroup unless you've got a rather high-end switching infrastructure. If you're working for an SMB, they don't have such infrastructures -- just basic router-hubs and a hardware firewall at best.
But rather than keep the image on a USB drive, why not just a small wallet of bootable DVD installer burns? One DVD per machine type, one "optional" software DVD per business function. Update the images on a regular basis (I'd think monthly to be sufficient), and archive or discard the older versions according to your business needs.
If you're using corporate license keys instead of per-machine, you shouldn't even have to enter key strings -- just boot the installer, a couple clicks, and go do something else for a few minutes.
You know, if you then treat wiki edits as change requests to be reviewed and prioritized (i.e. moderated), you'd have a rather useful tool for communications between business requirements, implementation, deployment, and support teams.
I'd been thinking that a wiki might be more useful than JavaDoc for providing the business analysts with a view of an implementation model. They just need to know what attributes are available and where to find them, not the implementation details. Maybe provide links to the corresponding JavaDoc details.
If you had a way to share filtered content between wikis, you could create a few aspects and let a user set a preferred default aspect. Other aspects would be available as cross-links between the wikis. For example, you could have a wiki indexing the source directory and object hierarchy for developers. Another aspect wiki would focus on business attributes for the BAs. Realistically you could have a wiki aspect for any development/business role your infrastructure requires.
Wiki portals are obvious as well. Being able to pull together key info people need to do their jobs is the main purpose of intranet portals.
Search engines already crawl wiki content, so you don't need anything special to search across the wikis, but you might want a customized intranet search engine that lets you select aspect/subject wikis to be searched.
Instead of intranet based, you could do the same thing with/for a web-distributed development team.
A lot of content is better kept in forums, though. Wikis are tag-definition based searches, and forums let you group subject content without requiring a subject identifier.
I thought it might be a way of displaying "active" topics from subjects you hadn't subscribed to in your preferences. Certainly they all seemed to be articles worth skimming to decide if they're worth reading in detail.
The fundamental question that I see is to define "intelligence" and "thought". If an AI reacts within parameters that perfectly simulate an emotional response, will a human witness react to that apparent emotional state?
If the AI subsequently follows a reasonable state change/learning of responses to simulate the emotional components of a long-term relationship (including conflicts and resolution), will the human consider the machine a true "friend" from an emotional perspective?
If humanity as a whole generally treats such complex AIs as emotional intelligence, does that mean that they've been imbued with "soul"? Or does it mean that we've collectively decided to change the collective rules of acceptable reality?
Once that reality is granted by the majority, does that mean the minority is now wrong when they say "it's just a machine?"
Personally I agree emotional intelligence will be a long time coming. I'm sure there will some trumpeting machine "intelligence" long before anyone really tries to simulate emotional responses. Some people would even claim a complex inference engine makes "intelligent" decisions, though all it's really doing is weighing options within a defined set of rules.
Neural networks are trying to simulate meat-based building blocks. Why the fundamental assumption that an intelligence has to be constructed at all like a brain works? If the responses fit the pattern of intelligence, does that make the implementation intelligent?
In short, what will we do when we encounter a machine that passes the Turing Test?
DRM is nothing more than providing a secure time-dependant keyfile for accessing an encrypted container, the same as TrueCrypt.org does, or the commercial DriveCrypt product. Such encryption has valid uses without limiting the user/owner's access to the information.
It seems to me that by specifically forbidding the use of GPL software with DRM technologies that we're imposing a limit on the what the software can be used for. This is distinct from prior versions of the GPL, which specifically were designed to prevent such limitations.
I for one would stick to GPLv2. I agree that the *AA plans for DRM are draconian and should be illegal, but that's an abuse of a perfectly useful technology and we shouldn't ban technology just because it can be abused.
I did some subcontract work for them in Poughkeepsie and Austin, and found them to be very professional. The usual inter-office/dept/division politics of large organizations, typical annual budget shuffles, and the usual paperwork hassles getting hardware or software.
Their devtechs were the usual mix of average-good with the occasional star, but their systechs were top-notch across the board.
Projects certainly didn't "drag out". Sometimes releases had to be delayed to coordinate other systems or upgrades, but that's just normal integration issues. Every company has to deal with them in some form or fashion.
Their people are well paid. They get solid benefits. They get useful training and support budgets. Even little details like having decent food in the cafeteria is taken care of.
If a company's management is "greedy", the staff don't get those benefits or support.
Unless, of course, you're a SCOX lawyer. Then maybe IBM really is an evil tyrant from your viewpoint. It's called "competition", which is a game played by companies that sell "products". Something SCOX has no recent experience with.:p
The big three as well as reputable hotsite and support service contractors like SunGard do a great job of getting parts out according to the contracted SLA's.
But it sounds to me like the failure happened at a shop that was nickle and diming their way out of SLA support contracts. You know, the kind of place where the boss's nephew says he can build a server for half the price of the vendors, or where a bean counter "saves" the company a few thousand a year by cutting the SLAs that they "never use anyhow."
But they are talking about downloads and about DiVX/MPEG4 encoding, not streaming.
While they mention also making downloads for PSP etc. available, I see no reason to expect there won't be seperate downloads for different resolutions. Sure you wouldn't want to try to watch a PSP video on a projector or even a decent sized monitor, but properly encoded MP4?
Why slam a new service on it's video quality when they've selected a good codec capable of producing high quality video at a reasonable resolution? Comparing it to older streaming services simply isn't fair until we get a chance to see what kind of quality they're actually selling.
I have a number of DVD box sets that I bought at $35-60/season. Apparently there are consumers who'd consider $2/ep an acceptable price if you get to keep the episode and play it on any machine you want.
Nope, it just took a bit of digging to recall that that was a 10 megabyte HDD on the old 286, not gigs. OTOH I don't think I actually got a machine with a CDROM until the '90s, and the multi-gig drives were out by then. (Deskstar 3? 4? Can't remember what that first one was, but I remember thinking I had enough space for the rest of a lifetime!)
Cable providers get the majority of their feeds via digital satellite nowadays, not analogue. They run it through hardware that reduces the MPEG blocking artifacts and blast it down their analogue pipes. In some cases, the digital-analogue conversions are done rather close to your house with a digital main trunk.
Of course the average consumer doesn't realize that, so they make arguments like yours, thinking it's similar to the old vinyl vs. CD argument. I remember vinyl audiophiles insisting their records sounded better than CDs even for groups that were using CD-rate digital mastering back in the '80s. They simply refused to accept that the "improvement" was signal smoothing that is now done in the digital domain by high-end audio players.
...recall than when CD-ROM arrived, it offered FAR more storage than hard disks of the day...
Err, not at all true. Multi-gig HDDs were definitely out before the 700-odd MB offered by CDs.
Personally my only use for BluRay or HD-DVD media is backups. DVD video at a reasonable viewing distance is just fine as is.
It's also worth noting that while they're talking up to 1080 (interlaced) lines of resolution for HD, that's really only about a 25% improvement in resolution over DVD. Sure it requires gobs more bandwidth, but it's not enough of an improvement to be visible in most cases. A good line doubler or the processing provided by hardware-accelerated processing displays (like ATI's) makes DVD virtually indistinguishable from HD visually.
Nope, I'm afraid the only real push for HD is DRM.
After all, once DRM is in place and your TiVO recognizes it, how long before the DRM flags disable fast forwards and such so you have to sit through the commercials? How long before the DRM flags auto-delete your video files in a week so you can't have a TiVO-equivalent record shows for you while you're on holiday, forcing you to buy the episodes?
The providers know the population won't buy into what follows the widespread rollout of DRM, that's why they're shutting down the analogue broadcasts in a few years. Of course even that will backfire on the politians who sold out to the media industry, as their constituents will roast them over a fire rather than be forced to buy a new TV.
How many people do you know that still watch a 10-15 year old 20-25" TV with shifty/smeared images rather than spend a paltry $150-200 on a larger new set? How many of them will pay the price of an HD-compatible set if they have any choice? For that matter, how many retirees could afford to go out and buy new TVs on a fixed budget?
That depends entirely on the technology being used. DiVX at 5-700MB/hour produces some damned fine video.
And you'd probably be rather annoyed to realize that the digital feeds your cable provider distributes are only a higher bit rate because most of the feeds are still using older MPEG formats instead of MPEG4.
I have an old car. It gets me around town. Why do I want to buy a new car just so I can pay Ford/GM/Chrysler/??? for a newer car that I don't need?
Except that your car is so old it doesn't really have keys or anti-theft protection, so criminals keep taking it and using it to try to run other people off the road, litter the streets with spam, or create traffic jams of old beaters tying up the interstate so no one else can use it (DOS attack). It's so old that it's just not safe any more, and should not be allowed on public roadways.
Besides, car prices have changed. That new car costs about half of what your old monitor did way back when...;)
Law is like programming. It requires nailing down all sorts of intricate little details, getting the syntax just right, and tweaking the niggly bits until everyone is satisfied.
As long as anything has two interpretations, you can bet that two sides on a dispute will argue about which interpretation is correct.
The main difference between programming and law is that programmers argue with a machine that can't change it's mind about the rules. There is no such predictable arbitrator for law.
Programmers remove bad code -- the law just keeps adding to the mudball without ever actually deleting the cruft. Imagine "debugging" a system when someone can bring up a (fixed) bug (case) from 10-15 years ago and actually have the courts/system accept the old bug as relevent to the current implementation...
Don't you mean "dead" as opposed to "tried and true"? From their products/tools page:
It is available for use on VAX computers, and both UNIX and MS-DOS based PCs. Each ITS contains a JOVIAL compiler (which can produce code executable on either a VAX, PC, or a MIL-STD-1750A computer), MIL-STD-1750A assembler, linker, and simulator/debugger. Staying current with today's technology, development is complete on a RISC version of the compiler. Other host platforms and target environments are available (e.g., AP-101, Z8002, and M680X0 processors).
Of course it's hard to get any details, as they don't even have any syntax or library help available unless you want to get the tools from them.
Besides, wasn't ADA created for exactly these types of detail-spec'd, high-reliability systems? (I haven't done mil-spec since '87-'88, but back then ADA was the big push.)
You mean you couldn't be bothered updating. The only place I've seen seen anything older than NT 3 in the past 10 years was on a dust-gathering basement clunker that we booted for giggles. As many places will sell used machines with a newer version of Windows installed for
Upgrade or get off the 'net. Your "beater" is a road hazard.
1. SuSE's implementation of the "Read/Copy/Update" algorithm
RCU was developed by Sequent as part of NUMA IIRC, with the express intent of it being used by multiple operating systems. I was working on a Sequent S81 years ago, and their sales reps were quite excited about the idea. IBM subsequently acquired the rights by purchasing Sequent. SCO will lose this one.
2. SuSE's implementation of NUMA Aware Locks
See above RE: RCU.
13. SuSE's implementation of load balancing
Nonsense. If SuSE used SCO/SVR load balancing, it would fall over at a pathetically small number of CPUs the way SCO's implementation does.
14. SuSE's implementation of PIDs
LMAO. Ah, now process id's are something new to SCO. Must be really good drugs for them to come up with this one...
15. SuSE's implementation of numerous kernel internals and APIs
Ah, the ever-popular boogeyman of "and other stuff". Without specifics, the courts will toss it.
16. SuSE's implementation of ELF
Lots and lots of really good drugs!
24. SuSE's implementation of numerous header files
Another "and other stuff" boogeyman.
Can't speak to the rest as I haven't a clue, but these points are largely the ones held against IBM, and I'm pretty confident that the best SCO can hope for on the points has already happened -- they weren't summarily dismissed as a frivolous lawsuit. But actually win any of them? Not likely, methinks.
Never used (most) of them, so I didn't comment on them.
But seriously, Python?!?!?! It's a spacing-dependent scripting language. Just because it's been used for bigger problems than it was originally intended for does not mean it's suddenly a general-purpose language.
I'm really curious as to why you lump in functional languages with your list of "type safe" languages. The very definition of functional languages is that they generalize implementations instead of defining type-specific implementations.
If removing type concepts and support are your idea of making something type-safe, you should love LISP. You're down to two types: atom and list. And technically the former is just a member of the latter, so you've only got one "complex" object to deal with.:p
Slashdot Mirror
Yet around here people are glad of the "nice weather." This time of year there should be some serious snow on the ground around here, not partially green grass. Sooner or later, we're all going to pay for the "nice weather."
Why is it surprising? This wasn't front-page news, so most people never even heard about it.
Certainly I don't remember reading any headlines about a US citizen having their citizenship revoked for forgetting their id.
Those who did hear about it were probably watching TV news and got a 30-60 second AV bite instead of any details.
Every agency or tech company I worked with had management and techs available who'd do weekend interviews if it was the only time some one could get together. Far more often an evening or morning phone call was arranged.
Even if you're working 12 hours a day and commuting an hour each way, that's still only 14 hours. That leaves you 2-3 hours per workday with 7-8 hours of sleep. Your choice, but arranging an interview call might be something to consider doing with that time.
It may be that you can get a packet-switch for a couple hundred dollars nowadays, but have you ever tried to convince mom & pop to replace something that "works just fine" from their perspective? There are a lot of places using old-fashioned hubs, and that won't be changing until the equipment has to be replaced.
The most basic emotions/states -- fear, lust, hunger, etc. have their roots in basic survival drives. Those instinctual states drive more complex emotional states -- anger (fear opting for aggressive defense), love (lust opting for protective bonding), hatred (fear opting for defensive avoidance and/or aggressive defense), etc.
The military would want to keep those aggressive tendencies, whereas the general public would obviously prefer AIs that don't have a survival/self-defense initiative. Self-defense triggered by a fear of dissolution could have nasty consequences with AIs in control of armaments or heavy equipment.
How to avoid "emotional baggage" if the AI entity is to learn from it's experiences? Take regular backups and restore to a pre-bad-experience state? Restoration denies the AI the ability to learn the consequences of real world situations, but you do need some way to keep it from developing hatred or paranoid fear. Are AI psychoanalysts needed? Sanity evaluations?
Seems to me that modelling emotions with AIs is actually what buys most of the risks.
Personally I think the only solution is to never let an AI make a final decision -- always require that a human push that final button/pull the trigger, even if they're doing it remotely with an audio/video feed.
The problem isn't the download time, the problem is that you're saturating the network for the whole subnet/department/workgroup unless you've got a rather high-end switching infrastructure. If you're working for an SMB, they don't have such infrastructures -- just basic router-hubs and a hardware firewall at best.
But rather than keep the image on a USB drive, why not just a small wallet of bootable DVD installer burns? One DVD per machine type, one "optional" software DVD per business function. Update the images on a regular basis (I'd think monthly to be sufficient), and archive or discard the older versions according to your business needs.
If you're using corporate license keys instead of per-machine, you shouldn't even have to enter key strings -- just boot the installer, a couple clicks, and go do something else for a few minutes.
You know, if you then treat wiki edits as change requests to be reviewed and prioritized (i.e. moderated), you'd have a rather useful tool for communications between business requirements, implementation, deployment, and support teams.
I'd been thinking that a wiki might be more useful than JavaDoc for providing the business analysts with a view of an implementation model. They just need to know what attributes are available and where to find them, not the implementation details. Maybe provide links to the corresponding JavaDoc details.
If you had a way to share filtered content between wikis, you could create a few aspects and let a user set a preferred default aspect. Other aspects would be available as cross-links between the wikis. For example, you could have a wiki indexing the source directory and object hierarchy for developers. Another aspect wiki would focus on business attributes for the BAs. Realistically you could have a wiki aspect for any development/business role your infrastructure requires.
Wiki portals are obvious as well. Being able to pull together key info people need to do their jobs is the main purpose of intranet portals.
Search engines already crawl wiki content, so you don't need anything special to search across the wikis, but you might want a customized intranet search engine that lets you select aspect/subject wikis to be searched.
Instead of intranet based, you could do the same thing with/for a web-distributed development team.
A lot of content is better kept in forums, though. Wikis are tag-definition based searches, and forums let you group subject content without requiring a subject identifier.
I thought it might be a way of displaying "active" topics from subjects you hadn't subscribed to in your preferences. Certainly they all seemed to be articles worth skimming to decide if they're worth reading in detail.
The fundamental question that I see is to define "intelligence" and "thought". If an AI reacts within parameters that perfectly simulate an emotional response, will a human witness react to that apparent emotional state?
If the AI subsequently follows a reasonable state change/learning of responses to simulate the emotional components of a long-term relationship (including conflicts and resolution), will the human consider the machine a true "friend" from an emotional perspective?
If humanity as a whole generally treats such complex AIs as emotional intelligence, does that mean that they've been imbued with "soul"? Or does it mean that we've collectively decided to change the collective rules of acceptable reality?
Once that reality is granted by the majority, does that mean the minority is now wrong when they say "it's just a machine?"
Personally I agree emotional intelligence will be a long time coming. I'm sure there will some trumpeting machine "intelligence" long before anyone really tries to simulate emotional responses. Some people would even claim a complex inference engine makes "intelligent" decisions, though all it's really doing is weighing options within a defined set of rules.
Neural networks are trying to simulate meat-based building blocks. Why the fundamental assumption that an intelligence has to be constructed at all like a brain works? If the responses fit the pattern of intelligence, does that make the implementation intelligent?
In short, what will we do when we encounter a machine that passes the Turing Test?
DRM is nothing more than providing a secure time-dependant keyfile for accessing an encrypted container, the same as TrueCrypt.org does, or the commercial DriveCrypt product. Such encryption has valid uses without limiting the user/owner's access to the information.
It seems to me that by specifically forbidding the use of GPL software with DRM technologies that we're imposing a limit on the what the software can be used for. This is distinct from prior versions of the GPL, which specifically were designed to prevent such limitations.
I for one would stick to GPLv2. I agree that the *AA plans for DRM are draconian and should be illegal, but that's an abuse of a perfectly useful technology and we shouldn't ban technology just because it can be abused.
We talking the same big blue?
I did some subcontract work for them in Poughkeepsie and Austin, and found them to be very professional. The usual inter-office/dept/division politics of large organizations, typical annual budget shuffles, and the usual paperwork hassles getting hardware or software.
Their devtechs were the usual mix of average-good with the occasional star, but their systechs were top-notch across the board.
Projects certainly didn't "drag out". Sometimes releases had to be delayed to coordinate other systems or upgrades, but that's just normal integration issues. Every company has to deal with them in some form or fashion.
Their people are well paid. They get solid benefits. They get useful training and support budgets. Even little details like having decent food in the cafeteria is taken care of.
If a company's management is "greedy", the staff don't get those benefits or support.
Unless, of course, you're a SCOX lawyer. Then maybe IBM really is an evil tyrant from your viewpoint. It's called "competition", which is a game played by companies that sell "products". Something SCOX has no recent experience with. :p
The big three as well as reputable hotsite and support service contractors like SunGard do a great job of getting parts out according to the contracted SLA's.
But it sounds to me like the failure happened at a shop that was nickle and diming their way out of SLA support contracts. You know, the kind of place where the boss's nephew says he can build a server for half the price of the vendors, or where a bean counter "saves" the company a few thousand a year by cutting the SLAs that they "never use anyhow."
But they are talking about downloads and about DiVX/MPEG4 encoding, not streaming.
While they mention also making downloads for PSP etc. available, I see no reason to expect there won't be seperate downloads for different resolutions. Sure you wouldn't want to try to watch a PSP video on a projector or even a decent sized monitor, but properly encoded MP4?
Why slam a new service on it's video quality when they've selected a good codec capable of producing high quality video at a reasonable resolution? Comparing it to older streaming services simply isn't fair until we get a chance to see what kind of quality they're actually selling.
Hmm. 22 episodes at $2 each works out to $44.
I have a number of DVD box sets that I bought at $35-60/season. Apparently there are consumers who'd consider $2/ep an acceptable price if you get to keep the episode and play it on any machine you want.
"Too young to remember" -- I wish!
Nope, it just took a bit of digging to recall that that was a 10 megabyte HDD on the old 286, not gigs. OTOH I don't think I actually got a machine with a CDROM until the '90s, and the multi-gig drives were out by then. (Deskstar 3? 4? Can't remember what that first one was, but I remember thinking I had enough space for the rest of a lifetime!)
Err, you miss the point.
Cable providers get the majority of their feeds via digital satellite nowadays, not analogue. They run it through hardware that reduces the MPEG blocking artifacts and blast it down their analogue pipes. In some cases, the digital-analogue conversions are done rather close to your house with a digital main trunk.
Of course the average consumer doesn't realize that, so they make arguments like yours, thinking it's similar to the old vinyl vs. CD argument. I remember vinyl audiophiles insisting their records sounded better than CDs even for groups that were using CD-rate digital mastering back in the '80s. They simply refused to accept that the "improvement" was signal smoothing that is now done in the digital domain by high-end audio players.
Err, not at all true. Multi-gig HDDs were definitely out before the 700-odd MB offered by CDs.
Personally my only use for BluRay or HD-DVD media is backups. DVD video at a reasonable viewing distance is just fine as is.
It's also worth noting that while they're talking up to 1080 (interlaced) lines of resolution for HD, that's really only about a 25% improvement in resolution over DVD. Sure it requires gobs more bandwidth, but it's not enough of an improvement to be visible in most cases. A good line doubler or the processing provided by hardware-accelerated processing displays (like ATI's) makes DVD virtually indistinguishable from HD visually.
Nope, I'm afraid the only real push for HD is DRM.
After all, once DRM is in place and your TiVO recognizes it, how long before the DRM flags disable fast forwards and such so you have to sit through the commercials? How long before the DRM flags auto-delete your video files in a week so you can't have a TiVO-equivalent record shows for you while you're on holiday, forcing you to buy the episodes?
The providers know the population won't buy into what follows the widespread rollout of DRM, that's why they're shutting down the analogue broadcasts in a few years. Of course even that will backfire on the politians who sold out to the media industry, as their constituents will roast them over a fire rather than be forced to buy a new TV.
How many people do you know that still watch a 10-15 year old 20-25" TV with shifty/smeared images rather than spend a paltry $150-200 on a larger new set? How many of them will pay the price of an HD-compatible set if they have any choice? For that matter, how many retirees could afford to go out and buy new TVs on a fixed budget?
...poor video quality compared to my TV.
That depends entirely on the technology being used. DiVX at 5-700MB/hour produces some damned fine video.
And you'd probably be rather annoyed to realize that the digital feeds your cable provider distributes are only a higher bit rate because most of the feeds are still using older MPEG formats instead of MPEG4.
To paraphrase:
Except that your car is so old it doesn't really have keys or anti-theft protection, so criminals keep taking it and using it to try to run other people off the road, litter the streets with spam, or create traffic jams of old beaters tying up the interstate so no one else can use it (DOS attack). It's so old that it's just not safe any more, and should not be allowed on public roadways.
Besides, car prices have changed. That new car costs about half of what your old monitor did way back when... ;)
Law is like programming. It requires nailing down all sorts of intricate little details, getting the syntax just right, and tweaking the niggly bits until everyone is satisfied.
As long as anything has two interpretations, you can bet that two sides on a dispute will argue about which interpretation is correct.
The main difference between programming and law is that programmers argue with a machine that can't change it's mind about the rules. There is no such predictable arbitrator for law.
Programmers remove bad code -- the law just keeps adding to the mudball without ever actually deleting the cruft. Imagine "debugging" a system when someone can bring up a (fixed) bug (case) from 10-15 years ago and actually have the courts/system accept the old bug as relevent to the current implementation...
Don't you mean "dead" as opposed to "tried and true"? From their products/tools page:
Of course it's hard to get any details, as they don't even have any syntax or library help available unless you want to get the tools from them.
Besides, wasn't ADA created for exactly these types of detail-spec'd, high-reliability systems? (I haven't done mil-spec since '87-'88, but back then ADA was the big push.)
"Millions?"
You mean you couldn't be bothered updating. The only place I've seen seen anything older than NT 3 in the past 10 years was on a dust-gathering basement clunker that we booted for giggles. As many places will sell used machines with a newer version of Windows installed for Upgrade or get off the 'net. Your "beater" is a road hazard.
1. SuSE's implementation of the "Read/Copy/Update" algorithm
RCU was developed by Sequent as part of NUMA IIRC, with the express intent of it being used by multiple operating systems. I was working on a Sequent S81 years ago, and their sales reps were quite excited about the idea. IBM subsequently acquired the rights by purchasing Sequent. SCO will lose this one.
2. SuSE's implementation of NUMA Aware Locks
See above RE: RCU.
13. SuSE's implementation of load balancing
Nonsense. If SuSE used SCO/SVR load balancing, it would fall over at a pathetically small number of CPUs the way SCO's implementation does.
14. SuSE's implementation of PIDs
LMAO. Ah, now process id's are something new to SCO. Must be really good drugs for them to come up with this one...
15. SuSE's implementation of numerous kernel internals and APIs
Ah, the ever-popular boogeyman of "and other stuff". Without specifics, the courts will toss it.
16. SuSE's implementation of ELF
Lots and lots of really good drugs!
24. SuSE's implementation of numerous header files
Another "and other stuff" boogeyman.
Can't speak to the rest as I haven't a clue, but these points are largely the ones held against IBM, and I'm pretty confident that the best SCO can hope for on the points has already happened -- they weren't summarily dismissed as a frivolous lawsuit. But actually win any of them? Not likely, methinks.
Not likely at all...
Never used (most) of them, so I didn't comment on them.
But seriously, Python?!?!?! It's a spacing-dependent scripting language. Just because it's been used for bigger problems than it was originally intended for does not mean it's suddenly a general-purpose language.
I'm really curious as to why you lump in functional languages with your list of "type safe" languages. The very definition of functional languages is that they generalize implementations instead of defining type-specific implementations.
If removing type concepts and support are your idea of making something type-safe, you should love LISP. You're down to two types: atom and list. And technically the former is just a member of the latter, so you've only got one "complex" object to deal with. :p