World: r3wp

It's basically a one way websocket communication... server -> browser 
 . Not sure what the real advantage would be, perhaps less overhead?

Interesting...Do you know if current browsers supporting ws, implement 
the EventSource interface too?

I believe Safari 5 does

yeah

here's a nice little tool
http://html5test.com

The latest Chrome does as well.

FF needs to step up to the plate

Section 9 "Connectionless push and other features" in the W3 doc 
is interesting.

My Chrome 5.0.375.99 scores a "no" at Server-Sent Events :(

There's a Chrome 6

http://www.chromium.org/getting-involved/dev-channel#TOC-Windows

If you haven't tried Safari lately, check out Safari 5.. not bad.

Never!

Apple software installers try to get every apple product downloaded 
onto your pc ...

Don't they all?

Another advantage of web apps :) At least you can block the ads.

Scrap that whole Safari plug.. thing is buggy with websockets.

good site to test web sockets http://jwebsocket.org/index.htm

I wrote a very simple ws client protocol, http://rebolforum.com/index.cgi?f=printtopic&topicnumber=47

I write the pick function to return none if an error occurs. whole 
the function is inside an attempt
pick port 1 ;--> none
first port ;--> still return error

but "first port" actually same with "pick port 1", it calls "pick", 
how it returns error?

I can't read unformatted source code .. .but isn't your pick missing 
an integer parameter?s

yes when I paste it to the forum, all formatting goes out..
in http protocol scheme, pick takes just one argument, port.

Really??

Never noticed ... let me look

p: open ws://localhost/test
pick p 1 ;if no data returns none
first p ;if no data returns error (but it calls pick??)

oops, sorry, there is no "pick" in http.
it is in ftp.

heh .. I was looking and looking and trying to find new glasses ...!

in pop scheme also, document says: "Read the Nth message from the 
POP port"
but pick function takes just port as an argument.

I think that's because the docs are wrong

It only allows you to pick the first message ... so that's why it 
ignores the integer parameter

no, it works, and the pick function is (in pop)
pick: func [
	"Read the Nth message from the POP port" 
	port
][
	read-message port 1
]

oh yes. but it works in my ws scheme.

well, I don't have a pop account to test it now

but the source there says to read the first message ... not the nth

I put my ws protocol here http://www.moldibi.com/download/ws.txt

I confused because, first p simply calls pick, my pick function inside 
an attempt block, so it cannot return an error.

but "first port" returns an error, if no data. its not a big deal. 
but spent some hours on it, then I saw that "pick port 1" returns 
none instead of error (as expected)

So I don't know what is different between "first port" and "pick 
port 1"

I don't understand your question ...

anyway, not a big problem.

ws protocol works,
run the chat server (see the post on the forum)
p: open ws://localhost/chat
insert p "hello"
pick p 1
== "hello"
insert p "/nick endo"
pick p 1
== "server: you known as endo"

and pick p 1000000 is the same?

I guess so :)

some ports a pick is a bit meaningless

yes, you can use only pick port 1.

if no data, then it returns none, so you can simply poll it. it doesn't 
wait for the data.

ws protocol is event based in its specification, so made it as no-wait. 
then and event based protocol can be built on this.

syntax highlighted version http://www.moldibi.com/rebol/ws.html

net-utils/net-install ws self 80

Endo, for the moment, use 4 spaces for tabs, and you're formatting 
will stay intact.  I'll change the script to replace tabs with spaces.

Endo, I updated to script to maintain formatting in a few new ways 
:)

Thank you Nick.