I use Python on a regular basis, but I also found the BitTorrent code hard to read. What BitTorrent does it really cool, but the code really does seem messy.
However, I never was able to pin down exactly what it was about the style that I disliked. It makes me very uncomfortable to say "that code is bad" when I can't say why in a clear and concise manor.
I've never done much Python networking code. Perhaps the APIs lead to clunky code, especially when you are trying to do robust error handling.
Actually, how well does advertising work? Do they do anything at all other than make sure you know a given company is big and well established?
The rare commercial that I pay attention to is usally for some form of entertainment (like a movie or tv show). As someone else pointed out, these are more informative, partly because we don't already know what the new shows/movies will be, or what they'll be like.
Of course, I'm not a brand concious person, and most of the things I'm interested in aren't advertised. Or at least, not in enough detail to be at all informative.
I buy movies/books/music based on my own eccentric tastes. What's popular is not relevant. I buy (mostly) generic foods, and I try to have someone else (with taste) pick out my clothes. I go online and research before buying electronics.
Ads had no impact at all when I was shopping for a home mortgage, and little to no impact when I went car shopping.
I can only imagine two uses for ads.
1) If it's something totally new (which I'll hear about on slashdot if I'm going to be interested).
2) To prove that the company is well established. (Nike versus QuickFoot shoes).
Do they have any other impact on people? Do they really work the way most companies seem to think they do?
Are there really people out there who CARE about the new flavor of hand soap? Do they get excited about Palmolive commercials?
I recommend Smoothwall.
In my experience, it seems to be secure, it's definatly reliable and fast on slow hardware. I run this at home and at the office. The setup is simple, and it has all the features you are mentioning.
There is a free (as in GPL) release, and there are commercial versions with additional features (VPNs, etc).
I once had a security guard pull a gun because we were running around (well, sliding around in our socks) pointing fingers and saying bang-bang. The joys of graduate school....
I think it's obvious that a lot of the problems on the road come from the fact that driver's licenses are given out casually.
However (at least in most of the US), to not have a license is almost equivalent to being disabled. There are many things that you can't do, or which at least become much more difficult. Holding a job is a good example. There are ways around the problems, but (like a physical disability) it just makes doing XYZ that much harder.
If we had convenient/reliable public transportation, I think we would do as much to drop the number of idiot drivers as we would by making it harder to get a license. We'd solve a lot of other problems as well. Unfortunatly, a good solution to public transit is really hard to implement and make work.
They've had this for a while on the big trucks, where the system costs in the $100k range.
They really seem to work, because they've been shown to boost the safety records of the drivers/companies that use them.
I used to have an article about it laying around. I dug it up when a friend and I were discussing what it would take to really build a self-driving vehicle for mass production.
I remember his helping me get a cheap clone of an NE2000 card working on my 386sx.
At his prompting, I did some very minor driver tweaking. I was struting for weeks because I'd not only 'helped' a real linux developer find a problem with an obscure card, but had written linux kernel code!
Even if all I did was tweak some compile time parameters by following explicit instructions.
I also remember being stunned that that 386sx was so much faster (opening xterms, etc) than the room full of Sparcs in the nearby lab. It took me a while to realize that it had everything local, but that they were getting most of their filesystem through NFS.
I've done contract work for Tivo. They do NOT have a database of all this data tied to a specific person. Internally, they are very aware of the dangers, and very paranoid. This stuff is partially anonymized before it's uploaded, and is anonymized more fully as it's moved into their backend databases.
I should also point out that users have the option of opting in 'fully', or opting out 'fully' of data sharing. Most users stick with the default 'anonymous only' privacy option for their account. Probably because it's the default, and you have to ask customer service to change this one way or the other. Through maybe it's a question when you activate your account (I forget).
With fiber, however, who is prohibiting some other company from laying fiber just like Verizon?
Have you given blanket permission to any company that wants to run cable through your land to do so? I've never given that permission to ANY company, but they do it anyway.
The local government used Imminent Domain to seize permssion for the cables to be run. Since the government is taking my land and giving it to a company, I feel it's perfectly reasonable for the government to tell that company what they can and can't do with it. Especially if what they are telling them is in my benifit.
In the early days of electricity and phone, the companies running the lines would have to come to an agreement with every person whose yard they used. One hold out could keep a neighborhood in the dark. I think the compromise made is reasonable, but I don't think it should be forgotten.
Actually, there DOES seem to be an emerging market for 'Art house' games. It's the direct download stuff that's being sold straight from websites.
They are usually lower budget, with lower production values, and even lower tech, but often with really good game play, and with original ideas.
An example of this is 'Uplink', or anything released by Ambrosia.
A couple of people can produce a game like this on nights and weekends over a couple of years. They can do this with almost nothing invested but time.
The model of direct only sales works pretty well for them. You can sell for $20, and see almost ALL of the the $20! And because the focus is on the game, not the technology, the games tend to sell for years, not for a few months.
However, console games can't be sold this way, since you can't release a game unless you get permission from Sony, MS, whoever. They take a big cut, and pretty much force you into the mainstream marketting model.
This means you have to compete based on the pretty pictures and mainstream appeal to get enough sales to be able to break even. A small game company is lucky to see $5 a copy per game sold, and probably has 3 million invested by the time to game goes to market.
Basically, CA works by guessing. That's why it's imperfect.
There are several different things you can try to key from..... the fades to black between TV and commercials, the louder audio in commercials, standardized commercial break intervals, the break in the video sync signal as commercials are patched into the real content, etc.
However, non of these things are perfect, and broadcasters aren't about to tag commercials in any convenient way. In fact, if DVRs were more popular, I would expect them to start trying to find ways to break CA, or related techniques.
I'm not sure exactly how Replay does it (being a TiVo owner), but it seems to me that for a reasonable chance at being reliable, the DVR would need to spend some time doing non-linear analysis of the recording before playing back with CA type behavior.
Still, without tagging by the broadcasters, CA type functionality will always be like spam filtering is. Pleasant when it works, but imperfect.
Do you remember when 'they' were publishing all known email addresses in a single paper book? I thought it was really cool when I saw it in a book store.
They had me listed five different times with addresses that pointed to different machines in the lab. 'dgarrett@alcor.engr.latech.edu', 'dgarrett@altar.engr.latech.edu', etc.
While I agree that timeshifting should be totally legal, there are other legetimate purposes.
For example, if I own the copyright to a DVD, then I should have the right to copy it. This isn't a problem for major productions, but what about my home movies?
What about content that has had the copyright released to the public? For example, I believe this happened to the movie 'Heavy Metal'. If copyright lifespans were closer to the original 14 years, this would be much more of an issue.
And yes, the DVD companies will replace a broken disc for the small fee of the original disk ($10-$30). They sometimes view your purchase as a license to view it and sometimes they treat it as a material object, depending on what is currently in their favor. The law is also mixed in this regard, but isn't always as slanted as the content owners would like.
If the purchase were treated as a license, then copying would be sanctioned, as long as you didn't distribute. And getting a fresh copy (from the company, or by coping from a friend) would be sanctioned as long as you had previously purchased a copy. Moving from one medium to another (tape versus CD) wouldn't matter as long as the content was the same.
If the purchase were treated as a material good, then there would be no restrictions on how you display it. Two examples come to mind.
1) You can't play a purchased CD or DVD in a public place, such as a bar, without special permission.
2) You can't build your own DVD player, or bypass things like region encoding.
I'd be happy with DRM as long as it worked in my favor as well. If 'the system' remembered everything I'd ever purchased, and would let me reaquire it in whatever format I want for the rest of my life, then I'd be more than happy to put up with reasonable restrictions. The big record companies don't appear be to capable of understanding why this could be considered fair.
At a somewhat higher level, I've always wanted a compiler that goes back and reoptimizes the executables on my machine based on usage patterns. At the very least, it sounds like a good idea.
One, I don't consider C++ to be a strongly typed language. Ada or Modula-2 yes, but not C/C++. The typing information is there, but it's easily bypassed, and often MUST be to do normal operations.
Two, language based compile time error checking mechanisms often introduce many new types or errors that are artificial. For example, have you ever worked on a project where some pieces (perhaps external libraries) use 'const' regularly, and some don't? You get a lot of compile errors and have to fight with them, despite the fact that the code and logic may be perfect!
Three, C/C++ are languages in which MANY kinds of behavior are undefined. This leads to what you (and I) call the worst kind of bug (random crashes, non-evident behavior). Truly high-level languages, have very little undefined behavior, and so don't have nearly so many of this kind of error. How much time have you spent tracing down a NULL pointer error, or an array overrun in C++? How much in Python?
I personally think that the best of all worlds is to create a language in which all behavior is defined, and the language is easy to work with.
Over time, we develop smart tools that do analysis of the code for possible run-time errors, and points them out to the programmer. When we learn how to build better error detection, we improve the tools, but don't have to change the language.
I think modern scripting languages are a step in this direction, and are an improvement over what we've had before. I think that the performance issues can and will be overcome if we decide that it's worth the effort. Psycho for Python is an example.
A very large portion of this expense was caused by congress dithering over the budget, and NASA doing a very poor job of handling congress.
When the project started, EVERY year, congress would budget it out and say "you get X small amount this year, you will get Y larger amount next year and following years". Then next year they would revise the Y amount down, and direct NASA to redesign so as to reduce the over all cost.
NASA spent BILLIONS on redesigns, and wasted BILLIONS because Y budget wasn't there to take advantage (or even maintain) things they built and/or started using the X budget.
Congress created a plan, then revised it every year through the entire project. NASA believed everything congress told them, and planned on congress sticking to it's promisses.
You can buy a tivo, and do anything you want to hardware wise. Mine is modified to add extra disk space. Tivo even goes to some (minimal) effort to keep that reasonably simple to do.
This is legal.
You can replace the software. They have made it difficult to do, but if you wipe everything (ROM and Drive) then start over, it works fine. And they don't really try to stop you. A reasonably standard flavor of linux runs on the box. Most of the hardware drivers (other than video in/out) are stock, or close enough.
This is legal.
If you try to modify the software they ship, they try to stop you. I consider that different. I'm not exactly sure if it's right, but it is a somewhat different.
I do happen to know that a significant portion of their motivation is this. They are convinced that if people find ways to record with a tivo, then extract a raw digital copy of the content then tivo (the company) will:
1) be sued out of existance (bogus reasons or not) 2) loose the right to make DirecTV receivers
They are probably right on both counts.
They are also trying to prevent people from getting the benefits of service without paying for it. I support them, when it comes to people taking advantage of the service tivo provides (copying guide data from one tivo to another, etc).
I don't particularly when it comes to people providing their own service information with no support from Tivo. As far as I'm concerned, falls under the category of "I paid for the box, it's mine".
You forget the DMCA. Modchips are a means for bypassing the security of the device. Therefore they are illegal, REGARDLESS of the legality of why you are trying to accomplish.
In effect, DMCA says that if a company adds security to a product (no matter how weak), they own that product, not you. Even if you're the one who paid for it. Sigh.
However, that only applies in the US. Or does it? Maybe because they were selling the modchips to US customers they might be able to pursue in a US court.
Slashdot Mirror
I use Python on a regular basis, but I also found the BitTorrent code hard to read. What BitTorrent does it really cool, but the code really does seem messy.
However, I never was able to pin down exactly what it was about the style that I disliked. It makes me very uncomfortable to say "that code is bad" when I can't say why in a clear and concise manor.
I've never done much Python networking code. Perhaps the APIs lead to clunky code, especially when you are trying to do robust error handling.
Actually, how well does advertising work? Do they do anything at all other than make sure you know a given company is big and well established?
The rare commercial that I pay attention to is usally for some form of entertainment (like a movie or tv show). As someone else pointed out, these are more informative, partly because we don't already know what the new shows/movies will be, or what they'll be like.
Of course, I'm not a brand concious person, and most of the things I'm interested in aren't advertised. Or at least, not in enough detail to be at all informative.
I buy movies/books/music based on my own eccentric tastes. What's popular is not relevant. I buy (mostly) generic foods, and I try to have someone else (with taste) pick out my clothes. I go online and research before buying electronics.
Ads had no impact at all when I was shopping for a home mortgage, and little to no impact when I went car shopping.
I can only imagine two uses for ads.
1) If it's something totally new (which I'll hear about on slashdot if I'm going to be interested).
2) To prove that the company is well established. (Nike versus QuickFoot shoes).
Do they have any other impact on people? Do they really work the way most companies seem to think they do?
Are there really people out there who CARE about the new flavor of hand soap? Do they get excited about Palmolive commercials?
I recommend Smoothwall. In my experience, it seems to be secure, it's definatly reliable and fast on slow hardware. I run this at home and at the office. The setup is simple, and it has all the features you are mentioning. There is a free (as in GPL) release, and there are commercial versions with additional features (VPNs, etc).
Then your wife would use it, not you. That way it could save her all the time and worry of planning your days for you!
And of course, when the phone gets things wrong.... it'll still be your fault.
I once had a security guard pull a gun because we were running around (well, sliding around in our socks) pointing fingers and saying bang-bang. The joys of graduate school....
I think it's obvious that a lot of the problems on the road come from the fact that driver's licenses are given out casually.
However (at least in most of the US), to not have a license is almost equivalent to being disabled. There are many things that you can't do, or which at least become much more difficult. Holding a job is a good example. There are ways around the problems, but (like a physical disability) it just makes doing XYZ that much harder.
If we had convenient/reliable public transportation, I think we would do as much to drop the number of idiot drivers as we would by making it harder to get a license. We'd solve a lot of other problems as well. Unfortunatly, a good solution to public transit is really hard to implement and make work.
They've had this for a while on the big trucks, where the system costs in the $100k range.
They really seem to work, because they've been shown to boost the safety records of the drivers/companies that use them.
I used to have an article about it laying around. I dug it up when a friend and I were discussing what it would take to really build a self-driving vehicle for mass production.
I remember his helping me get a cheap clone of an NE2000 card working on my 386sx.
At his prompting, I did some very minor driver tweaking. I was struting for weeks because I'd not only 'helped' a real linux developer find a problem with an obscure card, but had written linux kernel code!
Even if all I did was tweak some compile time parameters by following explicit instructions.
I also remember being stunned that that 386sx was so much faster (opening xterms, etc) than the room full of Sparcs in the nearby lab. It took me a while to realize that it had everything local, but that they were getting most of their filesystem through NFS.
What about the cumulative effect of a hundred phones at once? If you don't ban them, then almost all passengers will leave them on.
I'm not sure if this really matters or not.... I'm asking.
I've done contract work for Tivo. They do NOT have a database of all this data tied to a specific person. Internally, they are very aware of the dangers, and very paranoid. This stuff is partially anonymized before it's uploaded, and is anonymized more fully as it's moved into their backend databases.
I should also point out that users have the option of opting in 'fully', or opting out 'fully' of data sharing. Most users stick with the default 'anonymous only' privacy option for their account. Probably because it's the default, and you have to ask customer service to change this one way or the other. Through maybe it's a question when you activate your account (I forget).
Have you given blanket permission to any company that wants to run cable through your land to do so? I've never given that permission to ANY company, but they do it anyway.
The local government used Imminent Domain to seize permssion for the cables to be run. Since the government is taking my land and giving it to a company, I feel it's perfectly reasonable for the government to tell that company what they can and can't do with it. Especially if what they are telling them is in my benifit.
In the early days of electricity and phone, the companies running the lines would have to come to an agreement with every person whose yard they used. One hold out could keep a neighborhood in the dark. I think the compromise made is reasonable, but I don't think it should be forgotten.
How much are you offering?
;>
The price of a small phone or cable company seems like it would be perfectly acceptable.
Actually, there DOES seem to be an emerging market for 'Art house' games. It's the direct download stuff that's being sold straight from websites.
They are usually lower budget, with lower production values, and even lower tech, but often with really good game play, and with original ideas.
An example of this is 'Uplink', or anything released by Ambrosia.
A couple of people can produce a game like this on nights and weekends over a couple of years. They can do this with almost nothing invested but time.
The model of direct only sales works pretty well for them. You can sell for $20, and see almost ALL of the the $20! And because the focus is on the game, not the technology, the games tend to sell for years, not for a few months.
However, console games can't be sold this way, since you can't release a game unless you get permission from Sony, MS, whoever. They take a big cut, and pretty much force you into the mainstream marketting model.
This means you have to compete based on the pretty pictures and mainstream appeal to get enough sales to be able to break even. A small game company is lucky to see $5 a copy per game sold, and probably has 3 million invested by the time to game goes to market.
Basically, CA works by guessing. That's why it's imperfect.
There are several different things you can try to key from..... the fades to black between TV and commercials, the louder audio in commercials, standardized commercial break intervals, the break in the video sync signal as commercials are patched into the real content, etc.
However, non of these things are perfect, and broadcasters aren't about to tag commercials in any convenient way. In fact, if DVRs were more popular, I would expect them to start trying to find ways to break CA, or related techniques.
I'm not sure exactly how Replay does it (being a TiVo owner), but it seems to me that for a reasonable chance at being reliable, the DVR would need to spend some time doing non-linear analysis of the recording before playing back with CA type behavior.
Still, without tagging by the broadcasters, CA type functionality will always be like spam filtering is. Pleasant when it works, but imperfect.
Do you remember when 'they' were publishing all known email addresses in a single paper book? I thought it was really cool when I saw it in a book store.
They had me listed five different times with addresses that pointed to different machines in the lab. 'dgarrett@alcor.engr.latech.edu', 'dgarrett@altar.engr.latech.edu', etc.
While I agree that timeshifting should be totally legal, there are other legetimate purposes.
For example, if I own the copyright to a DVD, then I should have the right to copy it. This isn't a problem for major productions, but what about my home movies?
What about content that has had the copyright released to the public? For example, I believe this happened to the movie 'Heavy Metal'. If copyright lifespans were closer to the original 14 years, this would be much more of an issue.
And yes, the DVD companies will replace a broken disc for the small fee of the original disk ($10-$30). They sometimes view your purchase as a license to view it and sometimes they treat it as a material object, depending on what is currently in their favor. The law is also mixed in this regard, but isn't always as slanted as the content owners would like.
If the purchase were treated as a license, then copying would be sanctioned, as long as you didn't distribute. And getting a fresh copy (from the company, or by coping from a friend) would be sanctioned as long as you had previously purchased a copy. Moving from one medium to another (tape versus CD) wouldn't matter as long as the content was the same.
If the purchase were treated as a material good, then there would be no restrictions on how you display it. Two examples come to mind.
1) You can't play a purchased CD or DVD in a public place, such as a bar, without special permission.
2) You can't build your own DVD player, or bypass things like region encoding.
I'd be happy with DRM as long as it worked in my favor as well. If 'the system' remembered everything I'd ever purchased, and would let me reaquire it in whatever format I want for the rest of my life, then I'd be more than happy to put up with reasonable restrictions. The big record companies don't appear be to capable of understanding why this could be considered fair.
At a somewhat higher level, I've always wanted a compiler that goes back and reoptimizes the executables on my machine based on usage patterns. At the very least, it sounds like a good idea.
One, I don't consider C++ to be a strongly typed language. Ada or Modula-2 yes, but not C/C++. The typing information is there, but it's easily bypassed, and often MUST be to do normal operations.
Two, language based compile time error checking mechanisms often introduce many new types or errors that are artificial. For example, have you ever worked on a project where some pieces (perhaps external libraries) use 'const' regularly, and some don't? You get a lot of compile errors and have to fight with them, despite the fact that the code and logic may be perfect!
Three, C/C++ are languages in which MANY kinds of behavior are undefined. This leads to what you (and I) call the worst kind of bug (random crashes, non-evident behavior). Truly high-level languages, have very little undefined behavior, and so don't have nearly so many of this kind of error. How much time have you spent tracing down a NULL pointer error, or an array overrun in C++? How much in Python?
I personally think that the best of all worlds is to create a language in which all behavior is defined, and the language is easy to work with.
Over time, we develop smart tools that do analysis of the code for possible run-time errors, and points them out to the programmer. When we learn how to build better error detection, we improve the tools, but don't have to change the language.
I think modern scripting languages are a step in this direction, and are an improvement over what we've had before. I think that the performance issues can and will be overcome if we decide that it's worth the effort. Psycho for Python is an example.
A very large portion of this expense was caused by congress dithering over the budget, and NASA doing a very poor job of handling congress.
When the project started, EVERY year, congress would budget it out and say "you get X small amount this year, you will get Y larger amount next year and following years". Then next year they would revise the Y amount down, and direct NASA to redesign so as to reduce the over all cost.
NASA spent BILLIONS on redesigns, and wasted BILLIONS because Y budget wasn't there to take advantage (or even maintain) things they built and/or started using the X budget.
Congress created a plan, then revised it every year through the entire project. NASA believed everything congress told them, and planned on congress sticking to it's promisses.
This did not work out well.
You can buy a tivo, and do anything you want to hardware wise. Mine is modified to add extra disk space. Tivo even goes to some (minimal) effort to keep that reasonably simple to do.
This is legal.
You can replace the software. They have made it difficult to do, but if you wipe everything (ROM and Drive) then start over, it works fine. And they don't really try to stop you. A reasonably standard flavor of linux runs on the box. Most of the hardware drivers (other than video in/out) are stock, or close enough.
This is legal.
If you try to modify the software they ship, they try to stop you. I consider that different. I'm not exactly sure if it's right, but it is a somewhat different.
I do happen to know that a significant portion of their motivation is this. They are convinced that if people find ways to record with a tivo, then extract a raw digital copy of the content then tivo (the company) will:
1) be sued out of existance (bogus reasons or not)
2) loose the right to make DirecTV receivers
They are probably right on both counts.
They are also trying to prevent people from getting the benefits of service without paying for it. I support them, when it comes to people taking advantage of the service tivo provides (copying guide data from one tivo to another, etc).
I don't particularly when it comes to people providing their own service information with no support from Tivo. As far as I'm concerned, falls under the category of "I paid for the box, it's mine".
If you have a DirecTV account, you CAN transfer the lifetime to other boxes, as long as they are DirecTV.
You can also support multiple boxes on a single lifetime agreement, as long as they are in the same house.
You forget the DMCA. Modchips are a means for bypassing the security of the device. Therefore they are illegal, REGARDLESS of the legality of why you are trying to accomplish.
In effect, DMCA says that if a company adds security to a product (no matter how weak), they own that product, not you. Even if you're the one who paid for it. Sigh.
However, that only applies in the US. Or does it? Maybe because they were selling the modchips to US customers they might be able to pursue in a US court.