Re: [freenet-support] idea

2011-03-03 Thread Matthew Toseland
On Friday 25 Feb 2011 19:53:49 folkert wrote:
> > 
> > We did think about an "is the lan trusted?" option some time ago. This 
> > would both announce and open FCP and Fproxy. Unfortunately defining "the 
> > lan" is hard, when big untrusted NATed networks (e.g. ISPs in russia etc) 
> > often use private address space, and autodetecting it *reliably* is also 
> > hard.
> 
> You could make it configurable. Default: do not trust.

That was the whole idea. Unfortunately even detecting the LAN reliably isn't 
that easy. It's possible, of course, but you end up with lots of convoluted 
possibilities, and asking users questions that they probably don't know the 
answer to ...
> 
> > You should however peer with the other computers on your LAN, if you know 
> > their operators, of course. Which hopefully you do if the LAN is trusted!
> 
> Troublesome if everybody's on dhcp.

Right, we would need to be able to bind to a range rather than an IP (pick any 
local interface in that range)...


signature.asc
Description: This is a digitally signed message part.
___
Support mailing list
Support@freenetproject.org
http://news.gmane.org/gmane.network.freenet.support
Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
Or mailto:support-requ...@freenetproject.org?subject=unsubscribe

Re: [freenet-support] idea

2011-02-25 Thread folkert
> > > What about that the freenet daemon periodically
> > > (configurable/disable-ble of course) announces itself on the lan(s) to
> > > which it is connected? That way freenet-nodes can interconnect and
> > > speed up distribution of data.
> 
> There is a plugin:
> MDNSDiscovery.
> This uses it to announce FCP, but you still need to open FCP to the LAN...

Or plugin to avahi.

> > Data distribution on Freenet doesn't work like that. Data segments are
> > actually spread all across Freenet, ideally with no particular peer
> > having a large portion of a large splitfile. I don't think having fast
> > random LAN connections would speed things up -- the bottleneck will
> > still be the LAN's connection to the Internet. (Not to mention the fact
> > that it would be at least somewhat less secure. (Better chance of
> > traffic analysis and such tricks against you.))
> 
> We did think about an "is the lan trusted?" option some time ago. This would 
> both announce and open FCP and Fproxy. Unfortunately defining "the lan" is 
> hard, when big untrusted NATed networks (e.g. ISPs in russia etc) often use 
> private address space, and autodetecting it *reliably* is also hard.

You could make it configurable. Default: do not trust.

> You should however peer with the other computers on your LAN, if you know 
> their operators, of course. Which hopefully you do if the LAN is trusted!

Troublesome if everybody's on dhcp.


Folkert van Heusden

-- 
MultiTail na wan makriki wrokosani fu tan luku den logfile nanga san
den commando spiti puru. Piki puru spesrutu sani, wroko nanga difrenti
kroru, tya kon makandra, nanga wan lo moro.
http://www.vanheusden.com/multitail/
--
Phone: +31-6-41278122, PGP-key: 1F28D8AE, www.vanheusden.com
___
Support mailing list
Support@freenetproject.org
http://news.gmane.org/gmane.network.freenet.support
Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
Or mailto:support-requ...@freenetproject.org?subject=unsubscribe


Re: [freenet-support] idea

2011-02-24 Thread Matthew Toseland
On Saturday 19 Feb 2011 19:24:30 Edzard Pasma wrote:
> 
> Op 19-feb-2011, om 18:21 heeft folkert het volgende geschreven:
> 
> >>> Ok, that was not your point :-) Ok currently maybe not too many  
> >>> nodes
> >>> in the net but maybe this changes when governments restrict  
> >>> access to
> >>> what you can browse. Here in Europe governments already start  
> >>> talking
> >>> about installing filters.
> >>
> >> This is why you /don't/ want any kind of broadcasting, or any other
> >> kind of leak of identifiable traffic. Just encrypted non-identifiable
> >> "noise".
> >
> > Then we definately need a solution around the seed nodes. I mean  
> > traffic
> > going to them is a big fat warning that someone is doing freenet :-)
> >
> > Folkert van Heusden
> 
> 
> The seednodes might then once be blocked by authorities in their  
> struggle against the evil. The solution seems to have as many  
> potential seednodes as their are users. Thus one automatically  
> becomes one after a while. The other way around, any foreigner is  
> then a potential seednode. Adddresses can dynamically be collected to  
> be used at a next start.

Automatically harvesting seednodes is a possibility. The problems with it are:
1. Many nodes have low uptime. This can be detected.
2. Many nodes have poor connectivity (NATed without port forwarding). This can 
be detected but is some additional work to be automated and reliable.
3. They could block *ALL* the seednodes. If we do what Tor did and have a 
server that sends you a small number of seeds out of the global collection, 
they can still harvest them using lots of gmail addresses, IP addresses etc. 
This is what the Chinese did with Tor. Note that a gmail address is just a 
CAPTCHA, and these can be solved in bulk cheaply.
3. Opennet is inherently harvestable: Even if we distribute the seednodes and 
take all other precautions, it is feasible to find all nodes in order to block 
them.
4. Opennet is grossly insecure. It may be possible to improve this a bit 
against an attacker who is not able to connect to all nodes, surround groups of 
nodes gradually and so on, but IMHO really good security on opennet is very 
unlikely.

Viva darknet!


signature.asc
Description: This is a digitally signed message part.
___
Support mailing list
Support@freenetproject.org
http://news.gmane.org/gmane.network.freenet.support
Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
Or mailto:support-requ...@freenetproject.org?subject=unsubscribe

Re: [freenet-support] idea

2011-02-24 Thread Matthew Toseland
On Friday 18 Feb 2011 19:40:59 Dennis Nezic wrote:
> On Fri, 18 Feb 2011 19:01:10 +0100, folkert wrote:
> > What about that the freenet daemon periodically
> > (configurable/disable-ble of course) announces itself on the lan(s) to
> > which it is connected? That way freenet-nodes can interconnect and
> > speed up distribution of data.
> 
> Data distribution on Freenet doesn't work like that. Data segments are
> actually spread all across Freenet, ideally with no particular peer
> having a large portion of a large splitfile. I don't think having fast
> random LAN connections would speed things up -- the bottleneck will
> still be the LAN's connection to the Internet. (Not to mention the fact
> that it would be at least somewhat less secure. (Better chance of
> traffic analysis and such tricks against you.))

If the requests are served from the other node's cache then the risk against a 
distant attacker is significantly reduced. The risk against that other node 
increases, of course - this is the tradeoff. Bottom line, more friends is 
better if it means you can turn off opennet, and if you do actually know them 
(even if you don't trust them absolutely, they're still better than opennet).


signature.asc
Description: This is a digitally signed message part.
___
Support mailing list
Support@freenetproject.org
http://news.gmane.org/gmane.network.freenet.support
Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
Or mailto:support-requ...@freenetproject.org?subject=unsubscribe

Re: [freenet-support] idea

2011-02-24 Thread Matthew Toseland
On Friday 18 Feb 2011 19:40:59 Dennis Nezic wrote:
> On Fri, 18 Feb 2011 19:01:10 +0100, folkert wrote:
> > What about that the freenet daemon periodically
> > (configurable/disable-ble of course) announces itself on the lan(s) to
> > which it is connected? That way freenet-nodes can interconnect and
> > speed up distribution of data.

There is a plugin:
MDNSDiscovery.
This uses it to announce FCP, but you still need to open FCP to the LAN...

I think this is also possible with UPnP. Not sure if we use it there.
> 
> Data distribution on Freenet doesn't work like that. Data segments are
> actually spread all across Freenet, ideally with no particular peer
> having a large portion of a large splitfile. I don't think having fast
> random LAN connections would speed things up -- the bottleneck will
> still be the LAN's connection to the Internet. (Not to mention the fact
> that it would be at least somewhat less secure. (Better chance of
> traffic analysis and such tricks against you.))

We did think about an "is the lan trusted?" option some time ago. This would 
both announce and open FCP and Fproxy. Unfortunately defining "the lan" is 
hard, when big untrusted NATed networks (e.g. ISPs in russia etc) often use 
private address space, and autodetecting it *reliably* is also hard.

You should however peer with the other computers on your LAN, if you know their 
operators, of course. Which hopefully you do if the LAN is trusted!


signature.asc
Description: This is a digitally signed message part.
___
Support mailing list
Support@freenetproject.org
http://news.gmane.org/gmane.network.freenet.support
Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
Or mailto:support-requ...@freenetproject.org?subject=unsubscribe

Re: [freenet-support] idea

2011-02-19 Thread Dennis Nezic
On Sat, 19 Feb 2011 20:24:30 +0100, Edzard Pasma wrote:
> 
> Op 19-feb-2011, om 18:21 heeft folkert het volgende geschreven:
> 
> >>> Ok, that was not your point :-) Ok currently maybe not too many  
> >>> nodes
> >>> in the net but maybe this changes when governments restrict  
> >>> access to
> >>> what you can browse. Here in Europe governments already start  
> >>> talking
> >>> about installing filters.
> >>
> >> This is why you /don't/ want any kind of broadcasting, or any other
> >> kind of leak of identifiable traffic. Just encrypted
> >> non-identifiable "noise".
> >
> > Then we definately need a solution around the seed nodes. I mean  
> > traffic
> > going to them is a big fat warning that someone is doing freenet :-)
> >
> >
> > Folkert van Heusden
> 
> 
> The seednodes might then once be blocked by authorities in their  
> struggle against the evil. The solution seems to have as many  
> potential seednodes as their are users. Thus one automatically  
> becomes one after a while. The other way around, any foreigner is  
> then a potential seednode. Adddresses can dynamically be collected
> to be used at a next start.

Again, this is why darknet mode was implemented. There are no seednodes
in darknet, only friends you know and trust, and that cannot be
identified. (I believe Freenet 0.7 was intended to be a darknet, and
opennet (i.e. how Freenet 0.5 operates) was only later added to make
things easier for newbies.)
___
Support mailing list
Support@freenetproject.org
http://news.gmane.org/gmane.network.freenet.support
Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
Or mailto:support-requ...@freenetproject.org?subject=unsubscribe


Re: [freenet-support] idea

2011-02-19 Thread Edzard Pasma


Op 19-feb-2011, om 18:21 heeft folkert het volgende geschreven:

Ok, that was not your point :-) Ok currently maybe not too many  
nodes
in the net but maybe this changes when governments restrict  
access to
what you can browse. Here in Europe governments already start  
talking

about installing filters.


This is why you /don't/ want any kind of broadcasting, or any other
kind of leak of identifiable traffic. Just encrypted non-identifiable
"noise".


Then we definately need a solution around the seed nodes. I mean  
traffic

going to them is a big fat warning that someone is doing freenet :-)


Folkert van Heusden



The seednodes might then once be blocked by authorities in their  
struggle against the evil. The solution seems to have as many  
potential seednodes as their are users. Thus one automatically  
becomes one after a while. The other way around, any foreigner is  
then a potential seednode. Adddresses can dynamically be collected to  
be used at a next start.


Edzard Pasma.
___
Support mailing list
Support@freenetproject.org
http://news.gmane.org/gmane.network.freenet.support
Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
Or mailto:support-requ...@freenetproject.org?subject=unsubscribe


Re: [freenet-support] idea

2011-02-19 Thread folkert
> > Ok, that was not your point :-) Ok currently maybe not too many nodes
> > in the net but maybe this changes when governments restrict access to
> > what you can browse. Here in Europe governments already start talking
> > about installing filters.
> 
> This is why you /don't/ want any kind of broadcasting, or any other
> kind of leak of identifiable traffic. Just encrypted non-identifiable
> "noise".

Then we definately need a solution around the seed nodes. I mean traffic
going to them is a big fat warning that someone is doing freenet :-) 


Folkert van Heusden

-- 
Multi tail barnamaj mowahib li mora9abat attasjilat wa nataij awamir
al 7asoub. damj, talwin, mora9abat attarchi7 wa ila akhirih.
http://www.vanheusden.com/multitail/
--
Phone: +31-6-41278122, PGP-key: 1F28D8AE, www.vanheusden.com
___
Support mailing list
Support@freenetproject.org
http://news.gmane.org/gmane.network.freenet.support
Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
Or mailto:support-requ...@freenetproject.org?subject=unsubscribe


Re: [freenet-support] idea

2011-02-18 Thread Dennis Nezic
On Fri, 18 Feb 2011 21:35:00 +0100, folkert wrote:
> Well I was thinking maybe in the future we're all using mesh
> networking over wifi (or whatever wireless protocol we then have).
> http://en.wikipedia.org/wiki/Mesh_networking

Freenet-Darknet should work wonderfully over such a network :). (Well,
assuming you're not roaming. And that the meshes aren't too
isolated :p).

> Ok, that was not your point :-) Ok currently maybe not too many nodes
> in the net but maybe this changes when governments restrict access to
> what you can browse. Here in Europe governments already start talking
> about installing filters.

This is why you /don't/ want any kind of broadcasting, or any other
kind of leak of identifiable traffic. Just encrypted non-identifiable
"noise".

> Currently only for kiddy porn but I'm afraid that when such a filter
> is in place the step to block certain political views or so is much
> smaller.

Obviously. Their (Statist's) worst enemy is the free flow of
information.

> It is also a matter of convenience. If I visit some conference I don't
> want to be hassled with the need of configuring all kinds of software
> just to get "work done".

It's a tradeoff -- ease-of-use and anonymity.
___
Support mailing list
Support@freenetproject.org
http://news.gmane.org/gmane.network.freenet.support
Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
Or mailto:support-requ...@freenetproject.org?subject=unsubscribe


Re: [freenet-support] idea

2011-02-18 Thread folkert
> Normal people (your neighbour so to say) start to know these things like
> FreeNet/Tor. This week there was an item on national television in the
> Netherlands about how Tor helps oppressed people to get their
> opinions/news items/etc. out!

Oh it was also a rather positive news item.
http://nos.nl/artikel/219423-tor-digitale-solidariteit-met-middenoosten.html



Folkert van Heusden

-- 
Multitail es una herramienta flexible que permite visualizar los "log
file" y seguir la ejecución de comandos. Permite filtrar, añadir
colores, combinar archivos, la visualización de diferencias (diff-
view), etc.  http://www.vanheusden.com/multitail/
--
Phone: +31-6-41278122, PGP-key: 1F28D8AE, www.vanheusden.com
___
Support mailing list
Support@freenetproject.org
http://news.gmane.org/gmane.network.freenet.support
Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
Or mailto:support-requ...@freenetproject.org?subject=unsubscribe


Re: [freenet-support] idea

2011-02-18 Thread folkert
> > True, but it can be used to find a way to other nodes further down the
> > path. This way one doesn't need to connect to the central seednodes to
> > find ways to reach the global network/the rest of the freenet network.
> 
> Ah, for initial connection to Freenet that might be useful, although I
> don't think it'll be used too often. (I have trouble finding Freenet
> friends in my entire city -- let alone in my LAN :p.) You can add known

Well I was thinking maybe in the future we're all using mesh networking
over wifi (or whatever wireless protocol we then have).
http://en.wikipedia.org/wiki/Mesh_networking
Currently nobody is preventing you to anonymously buy a wifi card (with
a non-registered mac address - some cards even enable you to change it)
so then you could hook into the mesh without anyone knowing it is you.

Ok, that was not your point :-) Ok currently maybe not too many nodes in
the net but maybe this changes when governments restrict access to what
you can browse. Here in Europe governments already start talking about
installing filters. Currently only for kiddy porn but I'm afraid that
when such a filter is in place the step to block certain political views
or so is much smaller.
Normal people (your neighbour so to say) start to know these things like
FreeNet/Tor. This week there was an item on national television in the
Netherlands about how Tor helps oppressed people to get their
opinions/news items/etc. out!

> (and trustworthy) lan members to your list of darknet-friends, and
> connect that way to Freenet without using any seednodes. Or you can
> modify your seednodes.fref file by only including references to your
> LAN nodes.

It is also a matter of convenience. If I visit some conference I don't
want to be hassled with the need of configuring all kinds of software
just to get "work done". For that DHCP is too centralised (centralised
== bad).

> It's also not a good idea to be able to broadcast to anyone that you're
> using Freenet. (To prevent them from blacklisting you, et cetera.) (The
> whole point of Darknet mode was to make this impossible.)

If I'm at a conference, i don't have to register my mac address. So I
setup my wifi, connect to tor for regular internet traffic and freenet
for what it is for. Same thing for that mesh I wrote about above.


Folkert van Heusden

-- 
--
Phone: +31-6-41278122, PGP-key: 1F28D8AE, www.vanheusden.com
___
Support mailing list
Support@freenetproject.org
http://news.gmane.org/gmane.network.freenet.support
Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
Or mailto:support-requ...@freenetproject.org?subject=unsubscribe


Re: [freenet-support] idea

2011-02-18 Thread Dennis Nezic
On Fri, 18 Feb 2011 21:08:19 +0100, folkert wrote:
> > > What about that the freenet daemon periodically
> > > (configurable/disable-ble of course) announces itself on the lan
> > > (s) to which it is connected? That way freenet-nodes can
> > > interconnect and speed up distribution of data.
> > 
> > Data distribution on Freenet doesn't work like that. Data segments
> > are actually spread all across Freenet, ideally with no particular
> > peer having a large portion of a large splitfile. I don't think
> > having fast random LAN connections would speed things up -- the
> > bottleneck will still be the LAN's connection to the Internet. (Not
> > to mention the fact that it would be at least somewhat less secure.
> > (Better chance of traffic analysis and such tricks against you.))
> 
> True, but it can be used to find a way to other nodes further down the
> path. This way one doesn't need to connect to the central seednodes to
> find ways to reach the global network/the rest of the freenet network.

Ah, for initial connection to Freenet that might be useful, although I
don't think it'll be used too often. (I have trouble finding Freenet
friends in my entire city -- let alone in my LAN :p.) You can add known
(and trustworthy) lan members to your list of darknet-friends, and
connect that way to Freenet without using any seednodes. Or you can
modify your seednodes.fref file by only including references to your
LAN nodes.

It's also not a good idea to be able to broadcast to anyone that you're
using Freenet. (To prevent them from blacklisting you, et cetera.) (The
whole point of Darknet mode was to make this impossible.)
___
Support mailing list
Support@freenetproject.org
http://news.gmane.org/gmane.network.freenet.support
Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
Or mailto:support-requ...@freenetproject.org?subject=unsubscribe


Re: [freenet-support] idea

2011-02-18 Thread folkert
> > What about that the freenet daemon periodically
> > (configurable/disable-ble of course) announces itself on the lan(s) to
> > which it is connected? That way freenet-nodes can interconnect and
> > speed up distribution of data.
> 
> Data distribution on Freenet doesn't work like that. Data segments are
> actually spread all across Freenet, ideally with no particular peer
> having a large portion of a large splitfile. I don't think having fast
> random LAN connections would speed things up -- the bottleneck will
> still be the LAN's connection to the Internet. (Not to mention the fact
> that it would be at least somewhat less secure. (Better chance of
> traffic analysis and such tricks against you.))

True, but it can be used to find a way to other nodes further down the
path. This way one doesn't need to connect to the central seednodes to
find ways to reach the global network/the rest of the freenet network.


Folkert van Heusden

-- 
--
Phone: +31-6-41278122, PGP-key: 1F28D8AE, www.vanheusden.com
___
Support mailing list
Support@freenetproject.org
http://news.gmane.org/gmane.network.freenet.support
Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
Or mailto:support-requ...@freenetproject.org?subject=unsubscribe


Re: [freenet-support] idea

2011-02-18 Thread Dennis Nezic
On Fri, 18 Feb 2011 19:01:10 +0100, folkert wrote:
> What about that the freenet daemon periodically
> (configurable/disable-ble of course) announces itself on the lan(s) to
> which it is connected? That way freenet-nodes can interconnect and
> speed up distribution of data.

Data distribution on Freenet doesn't work like that. Data segments are
actually spread all across Freenet, ideally with no particular peer
having a large portion of a large splitfile. I don't think having fast
random LAN connections would speed things up -- the bottleneck will
still be the LAN's connection to the Internet. (Not to mention the fact
that it would be at least somewhat less secure. (Better chance of
traffic analysis and such tricks against you.))
___
Support mailing list
Support@freenetproject.org
http://news.gmane.org/gmane.network.freenet.support
Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
Or mailto:support-requ...@freenetproject.org?subject=unsubscribe


Re: [freenet-support] idea

2011-02-18 Thread Roland Haeder
Hi,

what he maybe means is called Local Peer Discovery [1] and many
peer-to-peer software (like Azureus, Transmission) have this to discover
peers running on the local network. This makes sure that the peers are
not connecting over the public IP/interface (ppp0) but instead on the
private IP/interface (e.g. eth0).

E.g. in a NAT-ed environment with port forwards, a local host NATed by
the router cannot connect back to itself or any other (except services
on the router) peers over the public IP number. But of course, if no
firewall rule forbids this the host can connect to other "locally"
running hosts on the private IP number.

Regards,
Roland

[1]: http://en.wikipedia.org/wiki/Local_Peer_Discovery


signature.asc
Description: This is a digitally signed message part
___
Support mailing list
Support@freenetproject.org
http://news.gmane.org/gmane.network.freenet.support
Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
Or mailto:support-requ...@freenetproject.org?subject=unsubscribe

Re: [freenet-support] idea for wininstaller

2002-11-04 Thread murray
I tryed windinstaller but it gave me a warning that with explorer its
dangerous so I uninstall.

 THank You
- Original Message -
From: "Mathew Ryden" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Sunday, November 03, 2002 5:07 PM
Subject: Re: [freenet-support] idea for wininstaller


> - Original Message -
> From: "Vitenka - Zen" <[EMAIL PROTECTED]>
>
>
> > Damn - forgot my other suggestion.
> >
> > Why not register fcp:// or similar, so that we can give links into
freeweb
> from external websites
> > without having to rely upon guessing the client port the user has
chosen?
>
> How would fcp help in this at all?
>
> > Forgive me if that's alreasy been done somewhere.
>
> -Mathew
>
>
> ___
> support mailing list
> [EMAIL PROTECTED]
> http://hawk.freenetproject.org/cgi-bin/mailman/listinfo/support


___
support mailing list
[EMAIL PROTECTED]
http://hawk.freenetproject.org/cgi-bin/mailman/listinfo/support



Re: [freenet-support] idea for wininstaller

2002-11-04 Thread Mathew Ryden
- Original Message -
From: "Vitenka - Zen" <[EMAIL PROTECTED]>


> Damn - forgot my other suggestion.
>
> Why not register fcp:// or similar, so that we can give links into freeweb
from external websites
> without having to rely upon guessing the client port the user has chosen?

How would fcp help in this at all?

> Forgive me if that's alreasy been done somewhere.

-Mathew


___
support mailing list
[EMAIL PROTECTED]
http://hawk.freenetproject.org/cgi-bin/mailman/listinfo/support



Re: [freenet-support] idea for wininstaller

2002-11-03 Thread Vitenka - Zen
Damn - forgot my other suggestion.

Why not register fcp:// or similar, so that we can give links into freeweb from 
external websites 
without having to rely upon guessing the client port the user has chosen?

Forgive me if that's alreasy been done somewhere.


___
support mailing list
[EMAIL PROTECTED]
http://hawk.freenetproject.org/cgi-bin/mailman/listinfo/support



Re: [freenet-support] idea for wininstaller

2002-11-01 Thread [EMAIL PROTECTED]
>> Also a text tip "freenet is loading, please wait..." would prevent 
>> ultra-newbies to immediately try and click on it.
>
>These are good suggestions.  But the problem is that the freenet.log
>never actually says WHEN the node is ready.  It says "Starting
>interfaces" but even then the interfaces aren't actually startED.
>They're just startING.  The node would have to say "All interfaces
>are now started.  Node is ready." or something similar in order for
>this to be really useful.  And I'm not even sure that's possible
>(you'd have to ask someone who speaks Java).
>
>In my experience, I'll often get "connection refused" on port 
>for several seconds after "Starting interfaces" appears in the log.
>It can be very frustrating.

you have already spoken out the solution: make bunnyapp try to open connections to 
: for every second or 500 ms, and if the 
connection could be established, drop it and do not ping anymore, because you know the 
interface is up and running (or keep on pinging to check if fproxy 
is still alive). if the connection failed, then the interface is not ready right now, 
so ping your way along until it does.



The information transmitted is intended only for the person or entity to which it is 
addressed and may contain confidential and/or privileged material. Any 
review, retransmission, dissemination or other use of, or taking of any action in 
reliance upon, this information by persons or entities other than the intended 
recipient is prohibited. If you received this in error, please contact the sender and 
delete the material from any computer.






___
support mailing list
[EMAIL PROTECTED]
http://hawk.freenetproject.org/cgi-bin/mailman/listinfo/support



Re: [freenet-support] idea for wininstaller

2002-11-01 Thread Dave Hooper
>> but the tray utility appears almost instantly, it would be neat not to
>>  display the red rabbit while the node is loading but the red rabbit
>> with  the grean arrow on it.
>>
>> Also a text tip "freenet is loading, please wait..." would prevent
>> ultra-newbies to immediately try and click on it.
>
> These are good suggestions.  But the problem is that the freenet.log
> never actually says WHEN the node is ready.  It says "Starting
> interfaces" but even then the interfaces aren't actually startED.
> They're just startING.  The node would have to say "All interfaces are
> now started.  Node is ready." or something similar in order for this to
> be really useful.  And I'm not even sure that's possible
> (you'd have to ask someone who speaks Java).

It's possible, because the rabbit icon doesn't actually read freenet.log
anyway.  It could try talking to the node and change icon when it's clear
that freenet or the web interface are up and running.  I'll work on this as
soon as I can because it's a really good idea (one I had myself about a
year ago but never got round to implementing!)

d



___
support mailing list
[EMAIL PROTECTED]
http://hawk.freenetproject.org/cgi-bin/mailman/listinfo/support



Re: [freenet-support] idea for wininstaller

2002-11-01 Thread Greg Wooledge
Zlatin Balevsky ([EMAIL PROTECTED]) wrote:

> since Fred still takes up to 20 or 30 seconds to load on slower machine, 

You have some really WEIRD notions of what constitutes a "slower
machine".  If your node only takes 20 seconds to start up, you've
got a pretty fast machine, or a very small data store.

> but the tray utility appears almost instantly, it would be neat not to 
> display the red rabbit while the node is loading but the red rabbit with 
> the grean arrow on it.  
> 
> Also a text tip "freenet is loading, please wait..." would prevent 
> ultra-newbies to immediately try and click on it.

These are good suggestions.  But the problem is that the freenet.log
never actually says WHEN the node is ready.  It says "Starting
interfaces" but even then the interfaces aren't actually startED.
They're just startING.  The node would have to say "All interfaces
are now started.  Node is ready." or something similar in order for
this to be really useful.  And I'm not even sure that's possible
(you'd have to ask someone who speaks Java).

In my experience, I'll often get "connection refused" on port 
for several seconds after "Starting interfaces" appears in the log.
It can be very frustrating.

-- 
Greg Wooledge  |   "Truth belongs to everybody."
[EMAIL PROTECTED]  |- The Red Hot Chili Peppers
http://wooledge.org/~greg/ |



msg01881/pgp0.pgp
Description: PGP signature