Send Devl mailing list submissions to
devl at freenetproject.org
To subscribe or unsubscribe via the World Wide Web, visit
http://www.uprizer.com/mailman/listinfo/devl
or, via email, send a message with subject or body 'help' to
devl-request at freenetproject.org
You can reach the person managing the list at
devl-admin at freenetproject.org
When replying, please edit your Subject line so it is more specific
than "Re: Contents of Devl digest..."
Today's Topics:
1. Re: Problems for splitting (Timm Murray)
2. Re: Killing Freenet (Re: [freenet-devl] Aardvark) (Peter Todd)
3. Re: Killing Freenet (Re: [freenet-devl] Aardvark) (Peter Todd)
4. Re: Killing Freenet (Re: [freenet-devl] Aardvark) (Peter Todd)
5. Re: Aardvark (Chris Anderson)
6. Re: Killing Freenet (Re: [freenet-devl] Aardvark) (Tavin Cole)
7. Killing Freenet (Re: [freenet-devl] Aardvark) (Chris Anderson)
8. 0.3.7 problems (Tavin Cole)
9. Re: Killing Freenet (Re: [freenet-devl] Aardvark) (Peter Todd)
10. Re: Aardvark (Scott G. Miller)
11. Re: Killing Freenet (Re: [freenet-devl] Aardvark) (Mr.Bad)
12. Re: Killing Freenet (Re: [freenet-devl] Aardvark) (Mr.Bad)
13. Re: Aardvark (Mr.Bad)
14. Re: Killing Freenet (Re: [freenet-devl] Aardvark) (Mark J. Roberts)
15. Re: Killing Freenet (Re: [freenet-devl] Aardvark) (Mr.Bad)
16. Re: Killing Freenet (Re: [freenet-devl] Aardvark) (Peter Todd)
--__--__--
Message: 1
From: "Timm Murray" <ad...@madtimes.com>
To: <devl at freenetproject.org>
Subject: Re: [freenet-devl] Problems for splitting
Date: Sun, 4 Feb 2001 19:15:29 -0800
Reply-To: devl at freenetproject.org
----- Original Message -----
From: "Oskar Sandberg" <md98-...@nada.kth.se>
To: <devl at freenetproject.org>
Sent: Sunday, February 04, 2001 3:51 PM
Subject: Re: [freenet-devl] Problems for splitting
> On Sun, Feb 04, 2001 at 05:08:39PM -0800, Timm Murray wrote:
> > Or the splitting process I used wasn't as clean as we're all assuming.
It'd
> > be if someone who uses an RPM-based distro could test it (I don't,
anymore).
> >
> > NOTE: I did this using a Window's program, though I used the same
program
> > to split Slackware .tar.gz and .deb files without any problems. It was
only
> > the RPM that failed.
>
> So your program was fucked up and screwed up some newline or something.
Wouldn't that mess up the .tar.gz and .deb, too?
I don't know, maybe it was the program.
--__--__--
Message: 2
From: Peter Todd <ret...@home.com>
To: devl at freenetproject.org
Subject: Re: Killing Freenet (Re: [freenet-devl] Aardvark)
Date: Sun, 4 Feb 2001 20:10:37 -0500
Reply-To: devl at freenetproject.org
On Sun, 04 Feb 2001, you wrote:
>
> On Sun, Feb 04, 2001 at 05:10:30PM -0500, Tavin Cole wrote:
> > I was thinking of a totally in-Freenet system (where maintainers would
> > have an SSK space). Somehow the system has to integrate these different
> > SSK spaces.
> > Of course there's no harm in having an HTTP server gateway *also*.. that
> > would be cool.
>
> And make submissions rather easy....
And of course gives us legal issues...
Given that we can assume that linking is for all practical
purposes illegal (they'll find some way to make that true) the only
way that such a directory will survive will be if it's creators are
anonymous.
Not a bad idea though.
--
retep at penguinpowered.com http://retep.tripod.com
--__--__--
Message: 3
From: Peter Todd <ret...@home.com>
To: devl at freenetproject.org
Subject: Re: Killing Freenet (Re: [freenet-devl] Aardvark)
Date: Sun, 4 Feb 2001 20:21:51 -0500
Reply-To: devl at freenetproject.org
On Sun, 04 Feb 2001, you wrote:
> Requests to the new nodes bounce around randomly for a bit, but then reach
> the old nodes and get routed correctly.
>
> << old nodes >> -- newish nodes -- new nodes -- very new nodes
> (well-connected) (... poorly-connected ...)
>
> Requests to the new and very new nodes bounce around randomly forever and
> never make it to the old nodes.
It just occured to me that one way we could help this would be to run
our own programs that would automaticly request data and insert
data on new nodes that turn up on inform.php to get them established
in the network.
I've noticed this problem too. Most of the nodes on inform.php that
are working at all are very much useless.
--
retep at penguinpowered.com http://retep.tripod.com
--__--__--
Message: 4
From: Peter Todd <ret...@home.com>
To: devl at freenetproject.org
Subject: Re: Killing Freenet (Re: [freenet-devl] Aardvark)
Date: Sun, 4 Feb 2001 20:29:20 -0500
Reply-To: devl at freenetproject.org
On Sun, 04 Feb 2001, you wrote:
> We should really start to think hard about how we can encourage people
> onto Freenet. Aardvark and GJ's webpage are great, but neither has been
> updated much recently. We need a killer-app. Perhaps Espra will
> fulfill this role, but those guys have been so quiet, I am unsure of
> where their development is, and am still unclear as to what exactly they
> are trying to create.
For those of your who don't know... Aardvark is back up! 2001-02-04
works anyway.
Whatever Espra is I don't see anything happening with it. No code, no
nothing. We can assume it's dead.
> The obvious killer-app would be mp3s, but for this we need a method that
> updatable catalogs of mp3s can be stored on Freenet (preferably in
> hierarchical indexes rather than just flat files for scalability).
> FProxy could be used to browse and download these mp3s. There should be
> a very easy way to take someone else's catalog and incorporate it into
> your own (so that individual catalogs will tend to glob into larger and
> larger catalogs making them much more useful).
One problem I think we're going to see is initally people will insert
music that no-one else wants. If I inserted all of my death metal
Nine Inch Nails (NIN rulez! :)) stuff I get the feeling that no-one
would download it soon enough for it to stick around on the network.
Meanwhile the person who inserts some Spice Girls music is going to
see a cold day in hell before I bother downloading that crap. :)
My retrieval test is going very badly BTW... I'll post my day 2
results real soon now.
> The other obvious killer-app would be porn, but of course nobody here is
> interested in that ;-)
Yah, all the radiation from our monitors has done bad things to
cirtain places where the sun don't shine. :)
--
retep at penguinpowered.com http://retep.tripod.com
--__--__--
Message: 5
Date: Sun, 4 Feb 2001 21:17:31 -0500 (EST)
From: Chris Anderson <chris...@charm.net>
To: devl at freenetproject.org
Subject: Re: [freenet-devl] Aardvark
Reply-To: devl at freenetproject.org
On Sat, Feb 03, 2001, Scott G. Miller wrote:
> Theres no way its 200x.
Alright, it's probably not that bad.
At 133mhz, a 32 bit pci bus is supposed to be capable of more than 4
gbit/s of bandwidth, but the copy from kernel buffer space to user space
is limited to about 2 gbit/s. The most I can get out of a local freenet
file request is 18 mbit/s... 2 raindolls and a fishy plus some SHA crap
(gcc -O). That's about 110x. A Java node+client may be lucky to get 3
mbit/s. Now, if you start counting disk seek times or charge Netscape's
pathetic performance costs to the equation, it might turn out that
encryption costs aren't all that much of a burden. It will all depend on
what you do with the data.
--__--__--
Message: 6
Date: Sun, 4 Feb 2001 21:22:58 -0500
From: Tavin Cole <ta...@mailandnews.com>
To: devl at freenetproject.org
Subject: Re: Killing Freenet (Re: [freenet-devl] Aardvark)
Reply-To: devl at freenetproject.org
On Sun, Feb 04, 2001 at 08:10:37PM -0500, Peter Todd wrote:
> > > I was thinking of a totally in-Freenet system (where maintainers would
> > > have an SSK space). Somehow the system has to integrate these different
> > > SSK spaces.
> > > Of course there's no harm in having an HTTP server gateway *also*.. that
> > > would be cool.
> >
> > And make submissions rather easy....
>
> And of course gives us legal issues...
>
> Given that we can assume that linking is for all practical
> purposes illegal (they'll find some way to make that true) the only
> way that such a directory will survive will be if it's creators are
> anonymous.
If this is done properly, any website will be able to mirror the
in-Freenet directory and perhaps even make submissions to it. They can't
all be taken down and frankly I consider it a civic duty to link to
illegal information.
2600.net solved the linking problem by removing the anchor tags and
printing a plaintext list of URLs..
--
// Tavin Cole
--__--__--
Message: 7
Date: Sun, 4 Feb 2001 21:30:37 -0500 (EST)
From: Chris Anderson <chris...@charm.net>
To: devl at freenetproject.org
Subject: Killing Freenet (Re: [freenet-devl] Aardvark)
Reply-To: devl at freenetproject.org
On Sun, 4 Feb 2001, Peter Todd wrote:
> One problem I think we're going to see is initally people will insert
> music that no-one else wants. If I inserted all of my death metal Nine
> Inch Nails (NIN rulez! :)) stuff I get the feeling that no-one would
> download it soon enough for it to stick around on the network.
> Meanwhile the person who inserts some Spice Girls music is going to see
> a cold day in hell before I bother downloading that crap. :)
>
Mmmmm, Spice Girls... I think you could do a pirate radio type thing
though, upload a hour length date redirected MSK program list every day...
--__--__--
Message: 8
Date: Sun, 4 Feb 2001 22:02:32 -0500
From: Tavin Cole <ta...@mailandnews.com>
To: devl at freenetproject.org
Subject: [freenet-devl] 0.3.7 problems
Reply-To: devl at freenetproject.org
The freenet_server wrapper provided in the tarball ought to cd to
the Freenet directory before starting Freenet.. as things are, if
I don't do this manually, Fred doesn't work right and never launches
fproxy. It creates a .freenet dir in my home dir and a subdir under
there called '0' -- not a good port number ;).
--
// Tavin Cole
--__--__--
Message: 9
From: Peter Todd <ret...@home.com>
To: devl at freenetproject.org
Subject: Re: Killing Freenet (Re: [freenet-devl] Aardvark)
Date: Sun, 4 Feb 2001 22:04:11 -0500
Reply-To: devl at freenetproject.org
On Sun, 04 Feb 2001, you wrote:
> > Given that we can assume that linking is for all practical
> > purposes illegal (they'll find some way to make that true) the only
> > way that such a directory will survive will be if it's creators are
> > anonymous.
>
> If this is done properly, any website will be able to mirror the
> in-Freenet directory and perhaps even make submissions to it. They can't
> all be taken down and frankly I consider it a civic duty to link to
> illegal information.
True. And any part of a directory can't be forceably removed even if
it can be prevented from being updated by removing the updaters.
Other people can then easilly incorperate the directory into their
own.
Still I'm dubious that date-based updating will do a good job with
this even with data-based redirects and MSK's. (0.3.7 has date-based
redirects and not MSK's right?)
--
retep at penguinpowered.com http://retep.tripod.com
--__--__--
Message: 10
Date: Sun, 4 Feb 2001 22:09:31 -0500
To: devl at freenetproject.org
Subject: Re: [freenet-devl] Aardvark
From: "Scott G. Miller" <scgmi...@indiana.edu>
Reply-To: devl at freenetproject.org
--neYutvxvOLaeuPCA
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable
On Sun, Feb 04, 2001 at 09:17:31PM -0500, Chris Anderson wrote:
> On Sat, Feb 03, 2001, Scott G. Miller wrote:
>=20
> > Theres no way its 200x.
>=20
> Alright, it's probably not that bad.
>=20
> At 133mhz, a 32 bit pci bus is supposed to be capable of more than 4
> gbit/s of bandwidth, but the copy from kernel buffer space to user space
> is limited to about 2 gbit/s. The most I can get out of a local freenet
> file request is 18 mbit/s... 2 raindolls and a fishy plus some SHA crap
I promise you can't transfer 18mbits/sec out of Java. You're comparing
apples to oranges.
> (gcc -O). That's about 110x. A Java node+client may be lucky to get 3
> mbit/s. Now, if you start counting disk seek times or charge Netscape's
> pathetic performance costs to the equation, it might turn out that
> encryption costs aren't all that much of a burden. It will all depend on
> what you do with the data.
>=20
You won't be transfering more than 1mbit/sec out of your node anyway, so
whats the big deal?
Scott
--neYutvxvOLaeuPCA
Content-Type: application/pgp-signature
Content-Disposition: inline
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.4 (GNU/Linux)
Comment: For info see http://www.gnupg.org
iD8DBQE6fhlrr9IW4v3mHtQRAtIbAJ4vzE+K5Pw+VMam0hgOV5vSwotEpgCgjXdI
kieZb7zBl2/y9Otd8BT/l9o=
=FFaW
-----END PGP SIGNATURE-----
--neYutvxvOLaeuPCA--
--__--__--
Message: 11
To: devl at freenetproject.org
Subject: Re: Killing Freenet (Re: [freenet-devl] Aardvark)
From: Mr.Bad <mr....@pigdog.org>
Organization: Pigdog Journal
Date: 04 Feb 2001 19:10:39 -0800
Reply-To: devl at freenetproject.org
>>>>> "TC" == Tavin Cole <tavin at mailandnews.com> writes:
TC> [...] They can't all be taken down [...]
Why do we even HAVE Freenet, then?
Considering that:
1) A Freenet directory _can_ be done with all communication
done within Freenet, and
2) The end user needs Freenet to view the linked resources
ANYWAYS,
why bother with Web-based stuff at all?
~Mr. Bad
--
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
/\____/\ Mr. Bad <mr.bad at pigdog.org>
\ / Pigdog Journal | http://pigdog.org/ | *Stay*Real*Bad*
| (X \x)
( ((**) "If it's not bad, don't do it.
\ <vvv> If it's not crazy, don't say it." - Ben Franklin
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
--__--__--
Message: 12
To: devl at freenetproject.org
Subject: Re: Killing Freenet (Re: [freenet-devl] Aardvark)
From: Mr.Bad <mr....@pigdog.org>
Organization: Pigdog Journal
Date: 04 Feb 2001 19:11:37 -0800
Reply-To: devl at freenetproject.org
>>>>> "PT" == Peter Todd <retep2 at home.com> writes:
PT> Still I'm dubious that date-based updating will do a good job
PT> with this even with data-based redirects and MSK's. (0.3.7 has
PT> date-based redirects and not MSK's right?)
It has both, AND filtering. The only thing is that filtering is turned
off by default.
~Mr. Bad
--
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
/\____/\ Mr. Bad <mr.bad at pigdog.org>
\ / Pigdog Journal | http://pigdog.org/ | *Stay*Real*Bad*
| (X \x)
( ((**) "If it's not bad, don't do it.
\ <vvv> If it's not crazy, don't say it." - Ben Franklin
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
--__--__--
Message: 13
To: devl at freenetproject.org
Subject: Re: [freenet-devl] Aardvark
From: Mr.Bad <mr....@pigdog.org>
Organization: Pigdog Journal
Date: 04 Feb 2001 19:12:24 -0800
Reply-To: devl at freenetproject.org
>>>>> "SGM" == Scott G Miller <scgmille at indiana.edu> writes:
SGM> I promise you can't transfer 18mbits/sec out of Java. You're
SGM> comparing apples to oranges.
Your comparison of his comparison to a comparison of apples and... Aw,
fergettit.
~Mr. Bad
--
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
/\____/\ Mr. Bad <mr.bad at pigdog.org>
\ / Pigdog Journal | http://pigdog.org/ | *Stay*Real*Bad*
| (X \x)
( ((**) "If it's not bad, don't do it.
\ <vvv> If it's not crazy, don't say it." - Ben Franklin
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
--__--__--
Message: 14
Date: Sun, 4 Feb 2001 21:16:59 -0500 (EST)
From: "Mark J. Roberts" <m...@statesmean.com>
To: <devl at freenetproject.org>
Subject: Re: Killing Freenet (Re: [freenet-devl] Aardvark)
Reply-To: devl at freenetproject.org
On Sun, 4 Feb 2001, Peter Todd wrote:
> Still I'm dubious that date-based updating will do a good job with
> this even with data-based redirects and MSK's. (0.3.7 has date-based
> redirects and not MSK's right?)
0.3.7 has MSKs. However, until Bad gets off his bad crazy lazy ass and
finishes freenetmirror, there's no way to insert them. (Also no way to
debug them, which is troubling...)
--
Mark Roberts
mjr at statesmean.com
--__--__--
Message: 15
To: devl at freenetproject.org
Subject: Re: Killing Freenet (Re: [freenet-devl] Aardvark)
From: Mr.Bad <mr....@pigdog.org>
Organization: Pigdog Journal
Date: 04 Feb 2001 19:17:13 -0800
Reply-To: devl at freenetproject.org
>>>>> "MJR" == Mark J Roberts <mjr at statesmean.com> writes:
MJR> 0.3.7 has MSKs. However, until Bad gets off his bad crazy
MJR> lazy ass and finishes freenetmirror, there's no way to insert
MJR> them. (Also no way to debug them, which is troubling...)
Actually, Gianni's PutFiles does MSKs. Check GJ's home page for
details.
I've got MSK's working in freenetmirror and hope to get it out
soon. Still have some little difficulties, but no biggy.
~Mr. Bad
--
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
/\____/\ Mr. Bad <mr.bad at pigdog.org>
\ / Pigdog Journal | http://pigdog.org/ | *Stay*Real*Bad*
| (X \x)
( ((**) "If it's not bad, don't do it.
\ <vvv> If it's not crazy, don't say it." - Ben Franklin
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
--__--__--
Message: 16
From: Peter Todd <ret...@home.com>
To: devl at freenetproject.org
Subject: Re: Killing Freenet (Re: [freenet-devl] Aardvark)
Date: Sun, 4 Feb 2001 22:12:28 -0500
Reply-To: devl at freenetproject.org
On Sun, 04 Feb 2001, you wrote:
> >>>>> "TC" == Tavin Cole <tavin at mailandnews.com> writes:
>
> TC> [...] They can't all be taken down [...]
>
> Why do we even HAVE Freenet, then?
>
> Considering that:
>
> 1) A Freenet directory _can_ be done with all communication
> done within Freenet, and
Or at the very least using plain old email. We don't need no fancy
web-based submission service.
My retrieval tests BTW are not going all that well. On my second day
it's taking 5 hops to get a 1 byte file and 7 for 256 and 1024 1meg
files took 20 hops on the first day and wasn't retrievable on the
second. I was retrieving using known good nodes (2 of them) that
worked for other stuff.
--
retep at penguinpowered.com http://retep.tripod.com
--__--__--
_______________________________________________
Devl mailing list
Devl at freenetproject.org
http://www.uprizer.com/mailman/listinfo/devl
End of Devl Digest
