zx-lan
ZXNet echo conference «zxnet.soft»
Hello, research
regarding the application level - maybe as a way out we can develop a super-simplified one
html option?
this would solve the memory problems. the site determines that the client is Speck, and
produces the corresponding document generated by the same php. and there is text with
minimum formatting and functionality.
What do you think, is it real?
Hello, moroz1999
Theoretically - real! But I have no practical realizations on my life’s path.
met...
The server actually has to look at the browser request, and based on the fields
request such as
"User-Agent,Accept,Accept-Language,Accept-Charset,Accept-Encoding", define,
what HTML format to send to the browser.
This is a wonderful method, unfortunately, it will be too narrowly applicable, since
most real servers simply won't pay much attention to these
parameters... Only servers specially adapted for Spectrum remain.
Here another “inconvenience” appears, connected with the fact that to implement
Such support will only be available on paid hosting services.
Hello jdigreze
only on paid hosting? What do you mean?
But the question is: where do you really need to go on the Internet from the Spectrum? first
Queue for Spectrum sites, right? I think there is support for this
It won’t be very difficult to implement the format on them.
Another option is to simplify the HTML on the go to an acceptable level, but it seems to me
unlikely.
Hello, moroz1999
mor> Well, the way I imagine it will have nothing in common with HTML
mor> a lot. to reduce space, you can replace tags with binary code, or
mor> how. Link the parameters purely to spec queries, taking into account
mor> familiarity and number of colors.
mor> pictures are in Speck format.
Then I advise you to take a closer look at the hypertext from Vitamin ;)
Again, I would like to be able to view any content from
Spectrum, but here only stripped-down rendering can help...
PS: Your idea is worthy of attention :)
Hello jdigreze
jdi> Then I advise you to take a closer look at the hypertext from Vitamin ;)
hmmm, thanks, I'll dig around..
jdi> Again, I would like to be able to view any content from
jdi> Spectrum, and here only stripped-down rendering can help...
well, then, in general, you don’t need to do anything - the PC is always at hand, and when not
at hand, a terminal connection to Linux will help.
in general, 90% of what we do has no real application for a long time, so
Personally, I’m more interested in the option when something is implemented on the Spectrum
exclusive.
jdi> PS: Your idea is worthy of attention :)
It will be worthy of attention if it finds application :)
In general, point by point, what is missing to access the Internet?
From the problems known to me:
1. amount of memory (the problem we are talking about above)
2. iron? (I don’t know, I haven’t followed the latest developments)
Hello, moroz1999
mor> 1. amount of memory (the problem we are talking about above)
mor> 2. iron? (not aware, did not follow the latest developments)
1. Amount of memory? I have KAY-1024... 1MB is quite a normal volume! Even
more than in PC 8086(88) ;)
2. Now there is an RS232 solution to support Hayes-modem by Caro, i.e. iron
for the implementation of PPP, SLIP, GPRS IS! It's a matter of software support...
Hello jdigreze
jdi> Theoretically - real! But for practical implementations I have my way
jdi> I haven’t met the path of life...
jdi> In fact, the server must look at the browser request, and based
jdi> request fields such as
jdi> "User-Agent,Accept,Accept-Language,Accept-Charset,Accept-Encoding",
jdi> determine which HTML format to send to the browser.
jdi> This is a wonderful method, unfortunately, it will be too narrowly applicable
jdi> since most real servers simply won’t pay
jdi> much attention to these parameters... Only the servers remain
jdi> specially adapted for Spectrum.
jdi> Here another “inconvenience” appears due to the fact that
jdi> such support will only be available on paid ones
jdi> hostings.
In fact, on the part of the spec, it is enough just to correctly form
GET request (I don’t see any problems with this) - and the requested CGI script
will throw out everything you need, in any language...
Hello, AL
Oh, I doubt that CGI will produce what is needed. A painfully familiar situation when
Most scripts ignore a lot. ;)
Here is an example of an experiment:
┌─- code ───
GET / HTTP/1.1
User-Agent: Spectrum
Host: 192.168.221.1
Accept: text/html, */*
Accept-Charset: isdos
HTTP/1.1 200 OK
Date: Tue, 11 Apr 2006 11:53:33 GMT
Server: Apache/2.0.50 (Win32)
Last-Modified: Wed, 08 Feb 2006 10:09:36 GMT
ETag: "c97a-10b-b1742881"
Accept-Ranges: bytes
Content-Length: 267
Content-Type: text/html; charset=ISO-8859-1
└── code ───
isdos encoding requested
ISO-8859-1 received
How do you want to fight this? ;)
Hello jdigreze
Hmm, why don’t you like the connection method via AY described by Alone Coder in
in my ALCOnews I just don’t remember the number...
And the game has already been written for him
Plus chat on two computers only
Hello jdigreze
By the way, enlighten me, why bother with all these headings? maybe
differentiate by file type or agree in advance on the type of index file?
Hello Firestarter
Well, by the way, my rs-232 is working great now, communication with the control center
stable ;)
However, the idea of “globalization” is in the air, or to be more precise,
there are those who want to introduce their iron friend to the global community
computers called the Internet. And, apparently, there is a reason for this desire.
However, unlike users (and most programmers) of the notorious PC,
We don’t have anything ready... Not even a stack... So we’re thinking about how it’s all going to be
"cram" into the limited amount of Spectrum resources. :)
Hello, moroz1999
Yes, in general, no one bothers with headlines yet. There is only one problem so far - no
mechanisms for multiple online connection of Spectrums with each other...
The "easiest" way is to use the Internet as a carrier of packages,
but as soon as it becomes possible to forward packets, there will immediately be a need for
access to ready-made standard services such as http, ftp and others.
By the way, can we return to the original direction? And then they moved on to the "cutting"
skins of an unkilled bear"... ;)
I downloaded RFCs in my spare time to get acquainted with the basics. In general,
the initial task for “punching through the wall” is PPP (Point-to-Point Protocol), because
is already a standard, and is not going to give up its position yet, although the implementation of SLIP is at
a couple of orders of magnitude simpler, but outdated.
Anyone have any thoughts on this?
Hello jdigreze
I started doing PPP, but I will return to it after the “rudiment” is ready
ZX-OS, for which you can write clients :D
Hello jdigreze
jdi> Oh, I doubt that CGI will produce what you need. Painfully familiar
jdi> is a situation where most scripts ignore a lot. ;)
jdi> Here is an example of an experiment:
jdi> ┌─- code ───
jdi>
jdi> GET / HTTP/1.1
jdi> User-Agent; Spectrum;
jdi> Host; 192.168.221.1;
jdi> Accept; text/html,; */*
jdi> Accept-Charset; isdos;
jdi>
jdi> HTTP/1.1 200 OK
jdi> Date; Tue; 11 Apr 2006 11:53:33 GMT
jdi> Server; Apache/2.0.50; (Win32)
jdi> Last-Modified; Wed; 08 Feb 2006 10:09:36 GMT
jdi> ETag; "c97a-10b-b1742881";
jdi> Accept-Ranges; bytes;
jdi> Content-Length; 267;
jdi> Content-Type; text/html;; charset=ISO-8859-1
jdi>
jdi> └── code ───
jdi> Isdos encoding requested
jdi> Received iso-8859-1
jdi> How do you want to fight this? ;)
but in no way - we were talking about specialized sites for spec
Hello jdigreze
jdi> In this case, there is only one small problem, very
jdi> many pages work correctly only with JavaScript...
jdi> But its implementation will be somewhat more complex than
jdi> simplified html.
WAP?
Hello jdigreze
jdi> Well, by the way, my rs-232 is working great now
jdi> connection with the PC is stable
jdi> However, the idea of “globalization” is in the air, and what if
jdi> more precisely, that is, those who want to introduce their iron friend to
jdi> to the worldwide computer community called the Internet. And apparently
jdi> there is a reason for this desire.
jdi> However, unlike users (and most programmers)
jdi> of the notorious PC, we don’t have anything ready... Not even a stack... Here we go
jdi> we are thinking about how to “cram” all this into a limited amount of resources
jdi> Spectrum.
the idea is this - just like for WAP, introduce a server that will prepare on the go
pages for spektrumists - for example, enter zx.tr.ru/www.ya.ru
so we will communicate using an already prepared protocol (simpler) with
Yandex and other wisdom.
Hello Eltaron
We know, we read it. ;)
My problem is different - there is no modem, since there is no phone... I mean, from
Telecommunications... And mobile without GPRS... Although this can be solved, by the summer I will take mobile
with a modem.
Hello,GriV
PPP is also a very simple protocol, in fact. A banal "daughter" from HDLC.
The only exception there is the initialization state machine, but it
Implementation in theory should not be very difficult.
Hello jdigreze
jdi> I downloaded RFCs in my spare time to get acquainted with the basics. B
jdi> in general, the initial task for “punching through the wall” is PPP
jdi> (Point-to-Point Protocol), because it is already a standard, and positions are still to be handed over
jdi> is not built, although the implementation of SLIP is a couple of orders of magnitude simpler, but it has become obsolete
jdi> yourself.
jdi> Who has any thoughts on this?
SLIP and PPP are still supported on systems - *nix and win - so
There is no point in talking about old age.
SLIP is very simple, so it seems to me that we should focus on it.
Hello Alex/AT
please summarize, otherwise I, as a person little familiar with practical
implementation of network protocols, it’s difficult to follow the train of thought :)
Hello,GriV
Gri> SLIP is very simple, so it seems to me that this is what I need
Gri> navigate.
The simplicity of the protocol does not matter.
The implementation of PPP is not much more complicated than SLIP.
When working via a modem connection, SLIP is not suitable.
For those interested, you can look at the source code for the reception and transmission procedures
packets via the PPP protocol, which is part of PING - what's so complicated about it :).
File: PPP.zip http://zx.pk.ru/attachment.php?attachmentid=3162
Hello, moroz1999
They answered - it’s simple, just take it and do it ((((-;
Hello,GriV
thank you, it helped a lot.
Hello, moroz1999
TCP/IP v Zx Spectrum
http://zxspectrum.hal.varese.it/tcpip.zip
:v2_cheer: