I think the first Star Trek I ever saw was a few repeats of the animated series. I'm pretty sure I saw that before the first movie. This may be why I don't self-identify as a trekker.
Let me be clear. It was the first time I got multiple troll moderations on the same post. (I think. Certainly first in a LONG time.)
If I intend to troll, I do so anonymously like a good citizen. Slashdot also used to have the feature that you could voluntarily relinquish your karma bonus on a post-by-post basis. I very quickly realised that if I used that judiciously, it effectively limited how much I'd get modded down on any individual post. Damn, I miss that.
Oh, a quick meta-comment. Of the 500+ comments I've made here in the last 13 odd years, this one is, to my knowledge, the first to get more than one "-1, Troll". For some reason, I find this hilarious.
I thought the first season of TNG was actually not that bad.
I don't want to get caught up defending throw-away remarks, but I'd like to point out two things:
1. It was far better than most things on US TV in 1987. But compared to later seasons, it sucked like a gravitational anomaly. 2. The problem was almost entirely budgetary.
With the deepest respect, I suspect you had trouble pushing through Voyager because you were a "committed Star Trek junkie". The question is about how to get non-Trekkers interested in Trek. I still think that Voyager is probably going to do it better than any other series.
My (admittedly non-scientific) experience is that non-Trekkers tend to like Voyager in a way that Trekkers don't.
The Tiffany Aching books (A Hat Full of Sky, The Wee Free Men and I Shall Wear Midnight) to which you prefer are just a little too advanced for an eight year old. Even a reasonably bright eight year old. About 10-12 is probably more realistic. Oh, and the same goes for Nation.
I totally agree with you about the Bromeliad trilogy (Truckers, Diggers and Wings), though. A bright eight-year-old would eat that up, though admittedly not get some of the jokes.
Start with Voyager. Seriously. Apart from the odd Q or Barclay episode, it doesn't require nearly as much pre-existing knowledge as any of the other next-gen series apart from the first season of TNG. And the first season of TNG mostly sucked.
Alternatively, you could always start with J.J. Abrams.
It's a personal opinion my my part. Kay's notion of OO was inspired by a biological system of cells, which communicate via (chemical) messages. Erlang's actors seem to fit this pattern more closely than, say, Smalltalk.
I didn't forget about Smalltalk. When Alan Kay coined the term "object oriented", he didn't intend that everything should be an object. Erlang is closer to Kay's vision than Smalltalk is.
Erlang's actor model is also just another tool in our toolbox.
Threads with shared memory, manual locking, mutexes, condition variables, restartable transactions and so on are hard to use. But that can also be okay. Sometimes the problem you're solving is so simple that it doesn't really matter, and sometimes it's so hard that there's no better way.
Having said that, of course, Erlang is closer to what Alan Kay meant when he coined the term "object oriented" than pretty much any other language has yet realised.
I don't propse to start another to-GC-or-not-to-GC thread here. I'll just say my piece and that's it.
GC is sometimes a bad idea, and sometimes a good idea. It depends entirely on precisely what the programmer is doing.
Programs written in languages without some facility to enable automated or semi-automated memory management (be it full GC, reference-counted smart pointers or what have you) tend to be inherently non-modular. Somebody has to know when nobody else wants some object any more. There are essentially two solutions: either everyone is responsible for co-operative book-keeping (introducing boilerplate everywhere), or the program is structured around the lifetimes of data (potentially limiting what you can do algorithmically or making high-level reorganisation unnecessarily painful).
Sometimes none of that matters, or other concerns dominate. That's okay. If you spend all of your career working in niches where memory management must be manual or else, and you enjoy it, then I'm glad you've found your calling.
But for the majority of is, GC is another tool in our toolbox. It is part of my job to decide which tools are suitable for a given task and which ones are not. Hell, most of the time, semi-automated reference counting is more than enough. But sometimes, only GC will do the job.
Actually, it's worse than that. Australia was still officially a British colony at the time (technically, it remained so until 1986), and was under intense pressure from the British government not to pursue computer research.
CSIRAC did have a few innovations, though. It was the first computer musical instrument, and ran one of the first high-level languages. Well, high-level by the standards of the day.
That isn't going to happen, because a typical new BBC show is absurdly short by US standards. No US network would commission a series that's only six episodes long. That would be a mini-series.
You'd think that this would be win-win for US networks. You don't have to invest too much in a show that won't last beyond the first season, and as an added bonus, you have more audience feedback sooner. But no, that would be too sensible.
It MAY LOOK this way, but "cloud" computing is nothing more than the resurection of time share systems of the past on mainframe and minis.
Only if you define "cloud computing" as "running jobs on shared infrastructure with quotas". Mainframes and minis of the past didn't have replicated filesystems, didn't scale to multiple sites and certainly didn't let you run potentially untrusted code in a sandbox.
Cloud computing is to car pooling as a mainframe is to mass transit. It's certainly cheaper and more efficient to send 1000 people from A to B by bus or train than to use the equivalent number of cars, but it only works if a critical mass of people all want to go the same way.
As far as I know, there is nowhere left in the world that is as free as the United States [...]
I think I'm far more free in Australia than you are in the United States, and Norway has us both beaten. But perhaps we both have an unrealistic idea of what "free" means.
I'm always curious as to how people like you would answer the question. At what point do you stop opposing the violent solution.
My country, as we Australians are fond of saying, was not created by a war. The number of attempted uprisings can be counted on one hand. Needless to say, we don't have the same fixation with the violent solution.
But to answer your question: I'd consider it if I wasn't allowed to emigrate.
Disclaimer: I oppose the violent solution, especially given that institutional recourse has not yet been exhausted.
Having said that, there's a reason why Washington was nicknamed "Cincinnatus". Strong leaders who would voluntarily give up power when it's no longer needed have always been hard to find. These days, they're nonexistent.
Z39.50 and WAIS were implementing a client/server protocol that wasn't tied to any particular database storage backend.
That's certainly true. I only know of one Z39.50 database engine that actually speaks Z39.50 natively.
Nonetheless, Z39.50 was designed with SGML in mind. It implements a very flexible documents-with-nested-and-repeating-fields schema, and did so in 1988.
Some of them are. You need to send a script kiddie to catch a script kiddie.
Some of the engineers working at anti-virus companies are the most brilliant and talented hackers you've (clearly) never met. It's one of the few white-hat jobs left for diehard assembler programmers. You know how those old-skool mainframe skills, like binary patching executables? These people can still do it.
The problem is that, to use the biological analogy, Stuxnet isn't a virus. Viruses have small payloads. Stuxnet, on the other hand, is indistinguishable from a serious application. Asking an anti-virus company to track down Stuxnet is like sending a microbiologist to catch a serial killer.
For the purpose of this discussion, the distinction is irrelevant. In most jurisdictions, incidentally, using a gun to defend yourself from a fist is an unlawful escalation.
You say "thought control" like it's a bad thing, or like it's a one-size-fits-all thing. I don't know if you've ever had a close friend who was brought up in an abusive family, or a fundamentalist religion, or been through some kind of severe trauma when a child, but it really does scar someone for life. Exposure to a significant amount of modern pornography is that kind of harm.
Slashdot Mirror
I think the first Star Trek I ever saw was a few repeats of the animated series. I'm pretty sure I saw that before the first movie. This may be why I don't self-identify as a trekker.
Let me be clear. It was the first time I got multiple troll moderations on the same post. (I think. Certainly first in a LONG time.)
If I intend to troll, I do so anonymously like a good citizen. Slashdot also used to have the feature that you could voluntarily relinquish your karma bonus on a post-by-post basis. I very quickly realised that if I used that judiciously, it effectively limited how much I'd get modded down on any individual post. Damn, I miss that.
Enterprise is better than Nemesis. But I understand if you blocked that movie out.
Oh, a quick meta-comment. Of the 500+ comments I've made here in the last 13 odd years, this one is, to my knowledge, the first to get more than one "-1, Troll". For some reason, I find this hilarious.
I don't want to get caught up defending throw-away remarks, but I'd like to point out two things:
1. It was far better than most things on US TV in 1987. But compared to later seasons, it sucked like a gravitational anomaly.
2. The problem was almost entirely budgetary.
With the deepest respect, I suspect you had trouble pushing through Voyager because you were a "committed Star Trek junkie". The question is about how to get non-Trekkers interested in Trek. I still think that Voyager is probably going to do it better than any other series.
My (admittedly non-scientific) experience is that non-Trekkers tend to like Voyager in a way that Trekkers don't.
The Tiffany Aching books (A Hat Full of Sky, The Wee Free Men and I Shall Wear Midnight) to which you prefer are just a little too advanced for an eight year old. Even a reasonably bright eight year old. About 10-12 is probably more realistic. Oh, and the same goes for Nation.
I totally agree with you about the Bromeliad trilogy (Truckers, Diggers and Wings), though. A bright eight-year-old would eat that up, though admittedly not get some of the jokes.
Start with Voyager. Seriously. Apart from the odd Q or Barclay episode, it doesn't require nearly as much pre-existing knowledge as any of the other next-gen series apart from the first season of TNG. And the first season of TNG mostly sucked.
Alternatively, you could always start with J.J. Abrams.
It's a personal opinion my my part. Kay's notion of OO was inspired by a biological system of cells, which communicate via (chemical) messages. Erlang's actors seem to fit this pattern more closely than, say, Smalltalk.
If it helps overcome your disbelief, consider that Haskell is closer to Peter Landin's vision of functional programming than ISWIM.
I didn't forget about Smalltalk. When Alan Kay coined the term "object oriented", he didn't intend that everything should be an object. Erlang is closer to Kay's vision than Smalltalk is.
Erlang's actor model is also just another tool in our toolbox.
Threads with shared memory, manual locking, mutexes, condition variables, restartable transactions and so on are hard to use. But that can also be okay. Sometimes the problem you're solving is so simple that it doesn't really matter, and sometimes it's so hard that there's no better way.
Having said that, of course, Erlang is closer to what Alan Kay meant when he coined the term "object oriented" than pretty much any other language has yet realised.
I don't propse to start another to-GC-or-not-to-GC thread here. I'll just say my piece and that's it.
GC is sometimes a bad idea, and sometimes a good idea. It depends entirely on precisely what the programmer is doing.
Programs written in languages without some facility to enable automated or semi-automated memory management (be it full GC, reference-counted smart pointers or what have you) tend to be inherently non-modular. Somebody has to know when nobody else wants some object any more. There are essentially two solutions: either everyone is responsible for co-operative book-keeping (introducing boilerplate everywhere), or the program is structured around the lifetimes of data (potentially limiting what you can do algorithmically or making high-level reorganisation unnecessarily painful).
Sometimes none of that matters, or other concerns dominate. That's okay. If you spend all of your career working in niches where memory management must be manual or else, and you enjoy it, then I'm glad you've found your calling.
But for the majority of is, GC is another tool in our toolbox. It is part of my job to decide which tools are suitable for a given task and which ones are not. Hell, most of the time, semi-automated reference counting is more than enough. But sometimes, only GC will do the job.
Actually, it's worse than that. Australia was still officially a British colony at the time (technically, it remained so until 1986), and was under intense pressure from the British government not to pursue computer research.
CSIRAC did have a few innovations, though. It was the first computer musical instrument, and ran one of the first high-level languages. Well, high-level by the standards of the day.
That isn't going to happen, because a typical new BBC show is absurdly short by US standards. No US network would commission a series that's only six episodes long. That would be a mini-series.
You'd think that this would be win-win for US networks. You don't have to invest too much in a show that won't last beyond the first season, and as an added bonus, you have more audience feedback sooner. But no, that would be too sensible.
It MAY LOOK this way, but "cloud" computing is nothing more than the resurection of time share systems of the past on mainframe and minis.
Only if you define "cloud computing" as "running jobs on shared infrastructure with quotas". Mainframes and minis of the past didn't have replicated filesystems, didn't scale to multiple sites and certainly didn't let you run potentially untrusted code in a sandbox.
Cloud computing is to car pooling as a mainframe is to mass transit. It's certainly cheaper and more efficient to send 1000 people from A to B by bus or train than to use the equivalent number of cars, but it only works if a critical mass of people all want to go the same way.
I think I'm far more free in Australia than you are in the United States, and Norway has us both beaten. But perhaps we both have an unrealistic idea of what "free" means.
I'm always curious as to how people like you would answer the question. At what point do you stop opposing the violent solution.
My country, as we Australians are fond of saying, was not created by a war. The number of attempted uprisings can be counted on one hand. Needless to say, we don't have the same fixation with the violent solution.
But to answer your question: I'd consider it if I wasn't allowed to emigrate.
Disclaimer: I oppose the violent solution, especially given that institutional recourse has not yet been exhausted.
Having said that, there's a reason why Washington was nicknamed "Cincinnatus". Strong leaders who would voluntarily give up power when it's no longer needed have always been hard to find. These days, they're nonexistent.
Z39.50 and WAIS were implementing a client/server protocol that wasn't tied to any particular database storage backend.
That's certainly true. I only know of one Z39.50 database engine that actually speaks Z39.50 natively.
Nonetheless, Z39.50 was designed with SGML in mind. It implements a very flexible documents-with-nested-and-repeating-fields schema, and did so in 1988.
The anti-virus companies are just script kiddies.
Some of them are. You need to send a script kiddie to catch a script kiddie.
Some of the engineers working at anti-virus companies are the most brilliant and talented hackers you've (clearly) never met. It's one of the few white-hat jobs left for diehard assembler programmers. You know how those old-skool mainframe skills, like binary patching executables? These people can still do it.
The problem is that, to use the biological analogy, Stuxnet isn't a virus. Viruses have small payloads. Stuxnet, on the other hand, is indistinguishable from a serious application. Asking an anti-virus company to track down Stuxnet is like sending a microbiologist to catch a serial killer.
What I find especially cute is that nobody in this thread seems to have heard of Z39.50 or WAIS.
For the purpose of this discussion, the distinction is irrelevant. In most jurisdictions, incidentally, using a gun to defend yourself from a fist is an unlawful escalation.
This being slashdot, I figured most people here had experienced "those" kids at school.
Pornography is to sex as jocks are to witty banter.
You say "thought control" like it's a bad thing, or like it's a one-size-fits-all thing. I don't know if you've ever had a close friend who was brought up in an abusive family, or a fundamentalist religion, or been through some kind of severe trauma when a child, but it really does scar someone for life. Exposure to a significant amount of modern pornography is that kind of harm.