Re: [twsocket] OT: IPv4 address shortage reaching critical stage

2010-01-21 Thread A Drent
IMHO it's not very easy to keep backwards compatibility. And this is very important. -- Just curious, why? AFAIK Ip6 is not compatible to Ip4, so why not implement a new branch? Or is the idea that one component should be able to handle all? albert -- To unsubscribe or change your

Re: [twsocket] OT: IPv4 address shortage reaching critical stage

2010-01-21 Thread Fastream Technologies
On Thu, Jan 21, 2010 at 9:55 AM, A Drent albert.dr...@rug.nl wrote: IMHO it's not very easy to keep backwards compatibility. And this is very important. -- Just curious, why? AFAIK Ip6 is not compatible to Ip4, so why not implement a new branch? Or is the idea that one component

Re: [twsocket] OT: IPv4 address shortage reaching critical stage

2010-01-21 Thread Paul
Just curious, why? AFAIK Ip6 is not compatible to Ip4, so why not implement a new branch? Or is the idea that one component should be able to handle all? albert I think so--another branch will cause a lot of rework for future updates, wouldn't it? You can't make a version for IPv6 only.

[twsocket] http design problem

2010-01-21 Thread Paul
I'm testing an application that has to control a remote machine. The communication goes through an intermediate http server that connects 2 http clients. A--Server--B-machine computer A controls computer B wich controls the machine. When computer B is located in the US, some message aren't

Re: [twsocket] http design problem

2010-01-21 Thread Francois PIETTE
I'm testing an application that has to control a remote machine. The communication goes through an intermediate http server that connects 2 http clients. A--Server--B-machine computer A controls computer B wich controls the machine. When computer B is located in the US, some message aren't

Re: [twsocket] OT: IPv4 address shortage reaching critical stage

2010-01-21 Thread Francois PIETTE
Just curious, why? AFAIK Ip6 is not compatible to Ip4, so why not implement a new branch? Or is the idea that one component should be able to handle all? Yes, of course. I think so--another branch will cause a lot of rework for future updates, wouldn't it? Indeed. And some maintenance

Re: [twsocket] http design problem

2010-01-21 Thread Paul
Yes, using HTTP/1.1 you have perstant connexions. See Keep-Alive in the source code. That's what I'm using, but it's not the same. When I post a stream (10k as test) to the server which then sends it to the other computer, it's about 90 times faster then sending 1000 10byte-strings. of couse,

Re: [twsocket] OT: IPv4 address shortage reaching critical stage

2010-01-21 Thread Anton Sviridov
Arno Garrels wrote: IMHO it's not very easy to keep backwards compatibility. Maybe, but the question is what specifically should be done to make ICS support IPv6. On the component level all IP's are operated as strings - no problem here. Maybe you'll just need to implement a pair of new

Re: [twsocket] http design problem

2010-01-21 Thread Francois PIETTE
Yes, using HTTP/1.1 you have perstant connexions. See Keep-Alive in the source code. That's what I'm using, but it's not the same. When I post a stream (10k as test) to the server which then sends it to the other computer, it's about 90 times faster then sending 1000 10byte-strings. of couse,

Re: [twsocket] http design problem

2010-01-21 Thread Paul
I don't agree with you. HTTP protocol has all support for real keep alive connections. request can even be sent before the previous answer has been received. BUT most commercial webserver do break the connection very quickly. Probably to be able to support a large number of clients. If you

Re: [twsocket] http design problem

2010-01-21 Thread Francois PIETTE
I don't agree with you. HTTP protocol has all support for real keep alive connections. request can even be sent before the previous answer has been received. BUT most commercial webserver do break the connection very quickly. Probably to be able to support a large number of clients. If you

Re: [twsocket] http design problem

2010-01-21 Thread Paul
Well, proxies may also break the connections. Never had problems with that, but new requests are delayed somehow. Paul -- To unsubscribe or change your settings for TWSocket mailing list please goto http://lists.elists.org/cgi-bin/mailman/listinfo/twsocket Visit our website at

[twsocket] New HTTP/REST/JSON demo

2010-01-21 Thread Francois PIETTE
Hi ! I wrote a new demo showing how to use ICS HTTP client component to access a REST API (Google Web Search) with JSON. To parse JSON, I used SuperObject which is freeware and very nice. The demo is named OverbyteIcsRestJasonClientDemo.proj and is available from the SVN repository at