I have to say as a Java Programmer a lot of people end up with slow Java programs cause they are bad programmers not due to the language. Most people think oh Java I don't need to worry about pointers or memory usages it will take care of it for me, and they proceed to write absolute shit for code. The amount that I learned after 2 years of programming the language full time was amazing. You really need to understand what you are doing in the language when are generating garbage when you are doing stupid things like creating a vector with 10 elements and then trying to stick 10,000 objects into it, instead of presizing things intelligently to avoid all the memory copies. I also used to see a lot of stupid things like people copying arrays in for loops instead of using System.arrayCopy. Anyway I suggest people learn how to actually use the language before they blame the language for their own personal inadequiecies as programmers. Of course this is the slashdot crowd so most of you couldn't write more than 10 lines of code anyway. The only complaint that is valid I think is startup time. People should be running the server vm and there is a startup penalty you pay with it, but really if you start your application and it takes 2 seconds it isn't the end of the world considering you will get c++ performance. If you run the client vm you will get faster startup, but poor application speed. Also there are a ton of bad swing programmers out there, who either A don't understand multithreaded programming or B don't understand good gui design, or even worse both. I suggest people spend some time on JGoodies Website to learn something about good looking swing UI's.
Every Ph. D that I have worked with (just a couple) had Csci degrees. I found them to be inferior to getting someone with a B.S. or a M.S. The reason is they just sat back and thought about these great designs and very ivory tower like thinking, but when it came down to it, their designs weren't applicable for what we were doing. They took too much of an ideal way of implementing things instead of a pragmatic approach seen by people with lesser degrees. So personally if you interviewed with me, I count that as points against you if you have a Ph. D. I hope that we just had some crappy workers and all Ph. D's aren't like that though and I would enjoy being proven wrong, but for now, I don't think they add value.
Depending on whether they're using H.323 or SIP or another standard, the VOIP tends to use a mix of transport protocols. The call setup signalling does tend to use TCP, because that part needs to be reliable and it's ok if it's delayed a couple hundred milliseconds as long as it all arrives and in order. The voice content itself tends to be RTP (Real Time Protocol) over UDP over IP.
Actually most of the signaling is done in UDP as well. People just build some reliability into the protocol. SIP and MGCP will both run in UDP or TCP and I am seeing it used in UDP only due to scalability reasons. You can't accord to burn a dedicated socket for every phone. It just doesn't scale up to hundreds of thousands of phones. UDP on the other hand scales much better.
The main VoIP standard is H.323 - Check out OpenH323, an open-source implentation of this technology.
Obviously you don't really use VOIP in the real world as H.323 is not the main standard. SIP is the most common (unfortunately). Don't waste your time with H.323 no one wants it. Here is the link to OpenSIP
The most widely used VoIP protocol is H.323. H323 allows negotiation of a compression CoDec. The base (worst) codec which must be supported is G.711 (64kb/s - this is what goes down an ISDN line - this is regarded as lossless digital encoding).
H. 323 is not the most widely used protocol. Most gateways are running SIP these days. IAD's tend to prefer MGCP (as it is a much more flexable and powerful protocol than SIP). For phones probably the most common is SCCP (skinny) that Cisco is pushing, but their phones will also run SIP or MGCP. And there are many SIP Phones out there. H. 323 is a relic and is not being used much in the real world. It is a terrible protocol. None of our customers even want it. Everyone seems to want SIP that is where everything is headed.
The background white noise in an analog phone call is artificially inserted ("comfort" noise). Studies showed users were not comfortable with a "dead" line, where the static reassures them the connection is still active.
You are right about that. Any good voip system will insert comfort noise, just so the user know's they aren't disconnected. We always do that on our phones.
Using H.323 (the main VoIP standard) you can choose from a number of compression codecs, from 32kbits upwards, although there exists a couple of codecs which sound pretty good (coparable to a average analogue phone line) at 64kbit/s, to you can make the transition without effecting you capacity.
H.323 is not the main VOIP standard. Where have you been. Back in 1999 there was a big push for it, but people have gone away from it cause it sucks. The main voip standards are either SIP or MGCP. And of course Cisco SCCP(skinny). Also 90kb that someone said is hi, I have no problems speaking with my gf on our ip phones and she has a 64kbit uplink. We use the G. 723 standard for compression, but G. 729 would work as well, and it much more common in voip.
There is already a leading in Memory database that is extremely fast. Check out TimesTen. That is what we use. There is also another one called Polyhedra. But the redundancy on Polyhedra doesn't appear to be as good as TimesTen, and it doesn't support Unicode either.
I hate to point out the obvious here since the whole slashdot crowd seems reactionary, but the standards really mean nothing. They have standardized the language. Whoopie the syntax is standardized. They have not standardized the class libraries which are the important part. That is what everyone is going to program in. So if you implement that standard you can't do much with it, without class libraries. Imagine programing in C without the standard libraries as well you are limited in what you can do, without writing for a specific os, and in a language like C# (or even Java) that is compiled to an intermediate langange and runs in a virtual machine you can't do anything without class libraries. So go ahead and build your C# environment from the specs you still won't be able to run MS's code on it. The Java Bytecode is a spec that you could implement as well, you just don't have the right to use the Trademark. You have the same issue there where you need the class libraries. But I would personally rather use Java where there is community involvement in the development and evolution of the language, rather than have the standard dictated to me by MS. But what do I know I am just a software developer unlike most of the people who bitch that don't write and code (they just want all their code to be free).
And now that I've gone all through this, the sometimes-10, sometimes-7 digit dialing that IIRC is used in the Dallas Metroplex area vs always-10 digits still doesn't make a case for needing the 1 in front. In fact, without the 1, 7-digit dialing could still be assumed. So I'm still just as confused as you are.
The Dallas Metroplex has 10 digit dialing. There are 3 area codes (214, 972, 469). So you always have to dial 10 digits. I don't see what 11 digits buys new york over 10 though, cause if the first digit is always 1 then effectively you have gained nothing over just adding another area code.
One of my buddies used to work for them, before they laid people off in the downturn. They have some really cool stuff in their labs. 32 x 32 Fiber routers that do everything as optically and not electronicly. Probably one of the biggest innovations in fiber optics in a long time unfortunatly it came out when there is a big capacity glut in telecom. At least they got a supercomputer sale, so hopefully they will be able to survive on things like this until telco picks back up.
I actually Dual boot Windows and Linux at home, and run windows and Solaris Sparc at work. I do Java Development so the platform isn't as important. But at home I always keep Windows because of Money. Not to mention I have data going back to 1995 on every financial transaction I have made. I can't easily move it to something else, and there is nothing worth moving it to. GNU Cash just isn't there, and I refuse to use a double entry accounting system. I am not an accountant and don't want that, I just want to add and subtract to and from my accounts and have decent reports to go along with it. Not to mention downloading new transactions from my credit card accounts, and pretty good 401k and stock tracking tools. The freesoftware just isn't there yet. But when it comes to programming and network monitoring I feel linux, or Solaris provide much better environments to work in.
I wanted to support them, and thought about subscribing. But I've always had strong concerns about their financials, and was worried that after I forked over my 30 dollars that they'd go under. This is one of the reasons I'm reluctant to pony up money for any web site. There's no guarantee that even after I subscribe that the site will still be there for the length of my subscription. I know it's not much money, but still if I pay for a year, I want to know that the site will still be there at the end of that year.
That is just stupid. If you read the site every day you should have subscribed. I did the premium content is worth it, plus you can shut off all the stupid ads. If everyone out there that has your opinion of course it will fail. But if all of the daily readers subscribed (40,000 of us did), then maybe Salon wouldn't be having these problems. You are probably the type of person that would say oh I wanted to send money to the EFF, but I was afraid they would lose a court case. You will be one of the people pissing and moaning cause there are no good webmags left, but you refused to support them when they were here. I mean it is $30 not a large sum of money, so who cares if you only get 6 months out of it, I mean it is only a case and a half of good beer anyway.
Personally it seems to me that psuedo code on exams is easier than writing code in on a computer. They aren't looking at syntax, they are looking at your logic, and any decent programmer should be able to express the ideas on paper. Now would my exams have compiled probably not, cause I would have missed semicolons or mismatched parens, but that isn't the point of a paper test. They don't want you to waste your time getting it to compile and debugging it, they are concerned with your logic and your high level design abilities. They will find out if you can make your code work on the labs and the homework.
That said, when I interview people I give them a problem. The point is to see how the person reacts under pressure, and to see the thought they put into the design while I am sitting there watching them. Does the code have to compile and work, no I am more concerned with orginization and problem solving. So that is my thought, they are testing something other than what you think. Then again some people just aren't good test taker's either, but that is too bad for them, cause most cs classes I took were weighted towards the exams and not much on the lab/homework.
Re:Existing system works - why change?
on
VoIP at $15 a Pop
·
· Score: 2, Interesting
The existing voice telephone system works and works very well. There is absolutly no business justification for moving voice from a stable, reliable, cost-effective voice network to an unstable, unreliable, increasing-cost data network. None whatsoever, except to pump up the earnings of Cisco.
The reason people are going to VOIP is that it costs less than circuit switched telephones. When you see the big phone companies roll this out, they won't be running it over the internet, they will be running it over private wans, and it will cost less than circuit switched calls, and the end user won't even know that this is happening. There are already long distance providers that are routing thier calls over IP.
I use VOIP all the time, it works great
on
VoIP at $15 a Pop
·
· Score: 2, Interesting
I work in telecom and we have a VOIP product. I have had an ethernet phone hooked up to my DSL for a long time. The only gateway that I have access to, is in Dallas, so I can only make local calls here, but I call a good friend of mine in Mexico everyday on this thing (We put a phone on her cable modem) and it works perfectly. Once in a while you will get some jitter or lost packets, but mostly it is just like talking on a real phone. When the quality starts to degrade it will be more like a cell phone and if the net is really messed up it will be unusuable. But 99% of the time we talk with no problems. I have also talked to someone from singapore on it, and that was crystal clear, they had no idea that I wasn't in the office and that I was at home. This is the future. Of course I don't think people will be running it over the standard internet, just because there is no QOS, but if all the backbone providers started prioritizing voice traffic people could do that. Also yeah it is easy to record a conversation like this, we have used a Radcom sniffer that we have and it will play back the audio stream. For encryption though I think it would be stupid to add it to the phone, people would be better off to setup an IPSec link between the people they want to call, just cause these phones are already expensive enough and they don't really have enough power to handle crypto onboard.
We have a heavily threaded C++ program that we develop at my company. It compiles under Sun Workshop 5.0, and Visual Studio. I would recommend getting workshop or the new Forte 6.0 C++ tools for Solaris on Sparc. They are great tools. The Front End for DBX makes debugging multiple threads very easy and when you dump a core, you get a call stack for all of your threads. That is much more easier in my opinion then when I was doing pthreads under xxgdb.
Another nice part about doing it on Solaris is that after you get all the bugs worked out of it, you can bring it back to a linux platform and run it there. Our Application compiles and runs under Solaris Sparc, Solaris x86, Linux x86, and Windows, so if you design it well you can get the portability with C++.
What do you think all the developers that Netscape pays to work on Mozilla is. It is a total payback. If you don't like the Netscape branded browser then run Mozilla, but it is kinda hypocritical to say that Netscape doesn't give back anything to Mozilla, when the vast majority of the Mozilla developers in fact work for Netscape.
Does anyone else ever get the feeling that Hemos doesn't actually read slashdot anymore he just posts stories. He has an amazing tendency to post stories that were just posted
yesterday! This isn't the first time either.
Don't forget Vampire Hunter D, and Ghost in the Shell. Both of those are excellent Anime. I just rented them a couple of weeks ago. Does anyone know of any place to rent Anime on DVD? Blockbuster's only has them on VHS.
I have an Fic SD-11 and a K7 running at 550 Mhz. I purchased it last November. It has run Redhat 6.1 and 6.2 with no problems. I have only run 2.2.X kernels on it (2.2.15 currently). It is pretty fast and runs excellent under linux.
Is there really any reason the changes would become a part of the main kernel if the rest of it is closed. I think part of the problem is that if it changes things in ways that slowdown a nonclustered kernel it won't make it, and if it doesn't should Linus incorporate the changes when the Userspace clustering software won't be open, so it won't be able to be used by the free software community as a whole?
This is consistant with other supreme court rulings that things like fighting words are not protected speech. This has nothing to do with free speech. The point of this is that they see no value in speech that is defimation. This is similar to why we have slander and libel laws, your speech isn't protected in those cases either. I beleve the original precident comes from chapell v. hardwick.
Slashdot Mirror
I have to say as a Java Programmer a lot of people end up with slow Java programs cause they are bad programmers not due to the language. Most people think oh Java I don't need to worry about pointers or memory usages it will take care of it for me, and they proceed to write absolute shit for code. The amount that I learned after 2 years of programming the language full time was amazing. You really need to understand what you are doing in the language when are generating garbage when you are doing stupid things like creating a vector with 10 elements and then trying to stick 10,000 objects into it, instead of presizing things intelligently to avoid all the memory copies. I also used to see a lot of stupid things like people copying arrays in for loops instead of using System.arrayCopy. Anyway I suggest people learn how to actually use the language before they blame the language for their own personal inadequiecies as programmers. Of course this is the slashdot crowd so most of you couldn't write more than 10 lines of code anyway. The only complaint that is valid I think is startup time. People should be running the server vm and there is a startup penalty you pay with it, but really if you start your application and it takes 2 seconds it isn't the end of the world considering you will get c++ performance. If you run the client vm you will get faster startup, but poor application speed. Also there are a ton of bad swing programmers out there, who either A don't understand multithreaded programming or B don't understand good gui design, or even worse both. I suggest people spend some time on JGoodies Website to learn something about good looking swing UI's.
Every Ph. D that I have worked with (just a couple) had Csci degrees. I found them to be inferior to getting someone with a B.S. or a M.S. The reason is they just sat back and thought about these great designs and very ivory tower like thinking, but when it came down to it, their designs weren't applicable for what we were doing. They took too much of an ideal way of implementing things instead of a pragmatic approach seen by people with lesser degrees. So personally if you interviewed with me, I count that as points against you if you have a Ph. D. I hope that we just had some crappy workers and all Ph. D's aren't like that though and I would enjoy being proven wrong, but for now, I don't think they add value.
Depending on whether they're using H.323 or SIP or another standard, the VOIP tends to use a mix of transport protocols. The call setup signalling does tend to use TCP, because that part needs to be reliable and it's ok if it's delayed a couple hundred milliseconds as long as it all arrives and in order. The voice content itself tends to be RTP (Real Time Protocol) over UDP over IP.
Actually most of the signaling is done in UDP as well. People just build some reliability into the protocol. SIP and MGCP will both run in UDP or TCP and I am seeing it used in UDP only due to scalability reasons. You can't accord to burn a dedicated socket for every phone. It just doesn't scale up to hundreds of thousands of phones. UDP on the other hand scales much better.
The main VoIP standard is H.323 - Check out OpenH323, an open-source implentation of this technology.
Obviously you don't really use VOIP in the real world as H.323 is not the main standard. SIP is the most common (unfortunately). Don't waste your time with H.323 no one wants it. Here is the link to OpenSIP
The most widely used VoIP protocol is H.323. H323 allows negotiation of a compression CoDec. The base (worst) codec which must be supported is G.711 (64kb/s - this is what goes down an ISDN line - this is regarded as lossless digital encoding).
H. 323 is not the most widely used protocol. Most gateways are running SIP these days. IAD's tend to prefer MGCP (as it is a much more flexable and powerful protocol than SIP). For phones probably the most common is SCCP (skinny) that Cisco is pushing, but their phones will also run SIP or MGCP. And there are many SIP Phones out there. H. 323 is a relic and is not being used much in the real world. It is a terrible protocol. None of our customers even want it. Everyone seems to want SIP that is where everything is headed.
The background white noise in an analog phone call is artificially inserted ("comfort" noise). Studies showed users were not comfortable with a "dead" line, where the static reassures them the connection is still active.
You are right about that. Any good voip system will insert comfort noise, just so the user know's they aren't disconnected. We always do that on our phones.
Using H.323 (the main VoIP standard) you can choose from a number of compression codecs, from 32kbits upwards, although there exists a couple of codecs which sound pretty good (coparable to a average analogue phone line) at 64kbit/s, to you can make the transition without effecting you capacity. H.323 is not the main VOIP standard. Where have you been. Back in 1999 there was a big push for it, but people have gone away from it cause it sucks. The main voip standards are either SIP or MGCP. And of course Cisco SCCP(skinny). Also 90kb that someone said is hi, I have no problems speaking with my gf on our ip phones and she has a 64kbit uplink. We use the G. 723 standard for compression, but G. 729 would work as well, and it much more common in voip.
There is already a leading in Memory database that is extremely fast. Check out TimesTen. That is what we use. There is also another one called Polyhedra. But the redundancy on Polyhedra doesn't appear to be as good as TimesTen, and it doesn't support Unicode either.
I hate to point out the obvious here since the whole slashdot crowd seems reactionary, but the standards really mean nothing. They have standardized the language. Whoopie the syntax is standardized. They have not standardized the class libraries which are the important part. That is what everyone is going to program in. So if you implement that standard you can't do much with it, without class libraries. Imagine programing in C without the standard libraries as well you are limited in what you can do, without writing for a specific os, and in a language like C# (or even Java) that is compiled to an intermediate langange and runs in a virtual machine you can't do anything without class libraries. So go ahead and build your C# environment from the specs you still won't be able to run MS's code on it. The Java Bytecode is a spec that you could implement as well, you just don't have the right to use the Trademark. You have the same issue there where you need the class libraries. But I would personally rather use Java where there is community involvement in the development and evolution of the language, rather than have the standard dictated to me by MS. But what do I know I am just a software developer unlike most of the people who bitch that don't write and code (they just want all their code to be free).
And now that I've gone all through this, the sometimes-10, sometimes-7 digit dialing that IIRC is used in the Dallas Metroplex area vs always-10 digits still doesn't make a case for needing the 1 in front. In fact, without the 1, 7-digit dialing could still be assumed. So I'm still just as confused as you are.
The Dallas Metroplex has 10 digit dialing. There are 3 area codes (214, 972, 469). So you always have to dial 10 digits. I don't see what 11 digits buys new york over 10 though, cause if the first digit is always 1 then effectively you have gained nothing over just adding another area code.
One of my buddies used to work for them, before they laid people off in the downturn. They have some really cool stuff in their labs. 32 x 32 Fiber routers that do everything as optically and not electronicly. Probably one of the biggest innovations in fiber optics in a long time unfortunatly it came out when there is a big capacity glut in telecom. At least they got a supercomputer sale, so hopefully they will be able to survive on things like this until telco picks back up.
I actually Dual boot Windows and Linux at home, and run windows and Solaris Sparc at work. I do Java Development so the platform isn't as important. But at home I always keep Windows because of Money. Not to mention I have data going back to 1995 on every financial transaction I have made. I can't easily move it to something else, and there is nothing worth moving it to. GNU Cash just isn't there, and I refuse to use a double entry accounting system. I am not an accountant and don't want that, I just want to add and subtract to and from my accounts and have decent reports to go along with it. Not to mention downloading new transactions from my credit card accounts, and pretty good 401k and stock tracking tools. The freesoftware just isn't there yet. But when it comes to programming and network monitoring I feel linux, or Solaris provide much better environments to work in.
I wanted to support them, and thought about subscribing. But I've always had strong concerns about their financials, and was worried that after I forked over my 30 dollars that they'd go under. This is one of the reasons I'm reluctant to pony up money for any web site. There's no guarantee that even after I subscribe that the site will still be there for the length of my subscription. I know it's not much money, but still if I pay for a year, I want to know that the site will still be there at the end of that year.
That is just stupid. If you read the site every day you should have subscribed. I did the premium content is worth it, plus you can shut off all the stupid ads. If everyone out there that has your opinion of course it will fail. But if all of the daily readers subscribed (40,000 of us did), then maybe Salon wouldn't be having these problems. You are probably the type of person that would say oh I wanted to send money to the EFF, but I was afraid they would lose a court case. You will be one of the people pissing and moaning cause there are no good webmags left, but you refused to support them when they were here. I mean it is $30 not a large sum of money, so who cares if you only get 6 months out of it, I mean it is only a case and a half of good beer anyway.
Personally it seems to me that psuedo code on exams is easier than writing code in on a computer. They aren't looking at syntax, they are looking at your logic, and any decent programmer should be able to express the ideas on paper. Now would my exams have compiled probably not, cause I would have missed semicolons or mismatched parens, but that isn't the point of a paper test. They don't want you to waste your time getting it to compile and debugging it, they are concerned with your logic and your high level design abilities. They will find out if you can make your code work on the labs and the homework.
That said, when I interview people I give them a problem. The point is to see how the person reacts under pressure, and to see the thought they put into the design while I am sitting there watching them. Does the code have to compile and work, no I am more concerned with orginization and problem solving. So that is my thought, they are testing something other than what you think. Then again some people just aren't good test taker's either, but that is too bad for them, cause most cs classes I took were weighted towards the exams and not much on the lab/homework.
The existing voice telephone system works and works very well. There is absolutly no business justification for moving voice from a stable, reliable, cost-effective voice network to an unstable, unreliable, increasing-cost data network. None whatsoever, except to pump up the earnings of Cisco.
The reason people are going to VOIP is that it costs less than circuit switched telephones. When you see the big phone companies roll this out, they won't be running it over the internet, they will be running it over private wans, and it will cost less than circuit switched calls, and the end user won't even know that this is happening. There are already long distance providers that are routing thier calls over IP.
I work in telecom and we have a VOIP product. I have had an ethernet phone hooked up to my DSL for a long time. The only gateway that I have access to, is in Dallas, so I can only make local calls here, but I call a good friend of mine in Mexico everyday on this thing (We put a phone on her cable modem) and it works perfectly. Once in a while you will get some jitter or lost packets, but mostly it is just like talking on a real phone. When the quality starts to degrade it will be more like a cell phone and if the net is really messed up it will be unusuable. But 99% of the time we talk with no problems. I have also talked to someone from singapore on it, and that was crystal clear, they had no idea that I wasn't in the office and that I was at home. This is the future. Of course I don't think people will be running it over the standard internet, just because there is no QOS, but if all the backbone providers started prioritizing voice traffic people could do that. Also yeah it is easy to record a conversation like this, we have used a Radcom sniffer that we have and it will play back the audio stream. For encryption though I think it would be stupid to add it to the phone, people would be better off to setup an IPSec link between the people they want to call, just cause these phones are already expensive enough and they don't really have enough power to handle crypto onboard.
We have a heavily threaded C++ program that we develop at my company. It compiles under Sun Workshop 5.0, and Visual Studio. I would recommend getting workshop or the new Forte 6.0 C++ tools for Solaris on Sparc. They are great tools. The Front End for DBX makes debugging multiple threads very easy and when you dump a core, you get a call stack for all of your threads. That is much more easier in my opinion then when I was doing pthreads under xxgdb.
Another nice part about doing it on Solaris is that after you get all the bugs worked out of it, you can bring it back to a linux platform and run it there. Our Application compiles and runs under Solaris Sparc, Solaris x86, Linux x86, and Windows, so if you design it well you can get the portability with C++.
Actually Germany is approximately the same size as the state of Wisconsin, so it isn't even near Texas or Arizona.
As for Canada, they are running a huge deficit and having problems supporting their socialism (At least accoriding to a Canadian friend of mine.)
What do you think all the developers that Netscape pays to work on Mozilla is. It is a total payback. If you don't like the Netscape branded browser then run Mozilla, but it is kinda hypocritical to say that Netscape doesn't give back anything to Mozilla, when the vast majority of the Mozilla developers in fact work for Netscape.
Does anyone else ever get the feeling that Hemos doesn't actually read slashdot anymore he just posts stories. He has an amazing tendency to post stories that were just posted yesterday! This isn't the first time either.
Don't forget Vampire Hunter D, and Ghost in the Shell. Both of those are excellent Anime. I just rented them a couple of weeks ago. Does anyone know of any place to rent Anime on DVD? Blockbuster's only has them on VHS.
I have an Fic SD-11 and a K7 running at 550 Mhz. I purchased it last November. It has run Redhat 6.1 and 6.2 with no problems. I have only run 2.2.X kernels on it (2.2.15 currently). It is pretty fast and runs excellent under linux.
I think you can get a Swiss bank account with an initial deposit of $5000 if you want an oversea's account.
Is there really any reason the changes would become a part of the main kernel if the rest of it is closed. I think part of the problem is that if it changes things in ways that slowdown a nonclustered kernel it won't make it, and if it doesn't should Linus incorporate the changes when the Userspace clustering software won't be open, so it won't be able to be used by the free software community as a whole?
This is consistant with other supreme court rulings that things like fighting words are not protected speech. This has nothing to do with free speech. The point of this is that they see no value in speech that is defimation. This is similar to why we have slander and libel laws, your speech isn't protected in those cases either. I beleve the original precident comes from chapell v. hardwick.