It is my understanding that BATMAN has come out of the freifunk community, and wikipedia seems to confirm this:
http://en.wikipedia.org/wiki/B.A.T.M.A.N.
And further, it appears that there is some amount of criticism of OLSR from the BATMAN camp:
http://www.open-mesh.net/wiki/the-olsr-story
This all makes sense to me, but what I don't understand is, why do all (or nearly all ?) of the freifunk networks use OLSR, and continue to use OLSR on the freifunk firmware ?
Are any of the freifunk networks using BATMAN ? If so, what routing metric are they using with it ?
Thanks.
who cares?
2009/12/11 George Sanders gosand1982@yahoo.com:
It is my understanding that BATMAN has come out of the freifunk community, and wikipedia seems to confirm this:
http://en.wikipedia.org/wiki/B.A.T.M.A.N.
And further, it appears that there is some amount of criticism of OLSR from the BATMAN camp:
http://www.open-mesh.net/wiki/the-olsr-story
This all makes sense to me, but what I don't understand is, why do all (or nearly all ?) of the freifunk networks use OLSR, and continue to use OLSR on the freifunk firmware ?
Are any of the freifunk networks using BATMAN ? If so, what routing metric are they using with it ?
Thanks.
B.A.T.M.A.N mailing list B.A.T.M.A.N@lists.open-mesh.net https://lists.open-mesh.net/mm/listinfo/b.a.t.m.a.n
Georg, you just put your fingers into an old wound.
On Dec 11, 2009, at 8:32 PM, George Sanders wrote:
It is my understanding that BATMAN has come out of the freifunk community, and wikipedia seems to confirm this:
http://en.wikipedia.org/wiki/B.A.T.M.A.N.
And further, it appears that there is some amount of criticism of OLSR from the BATMAN camp:
No, not every one in the Freifunk communities believes these claims.
It is probably better not to go back to these senseless "protocol wars" .
This all makes sense to me, but what I don't understand is, why do all (or nearly all ?) of the freifunk networks use OLSR, and continue to use OLSR on the freifunk firmware ?
Are any of the freifunk networks using BATMAN ? If so, what routing metric are they using with it ?
Thanks.
B.A.T.M.A.N mailing list B.A.T.M.A.N@lists.open-mesh.net https://lists.open-mesh.net/mm/listinfo/b.a.t.m.a.n
On Dec 11, 2009, at 8:32 PM, George Sanders wrote:
> >
It is my understanding that BATMAN has come out of the freifunk community, and
wikipedia seems to confirm this:
http://en.wikipedia.org/wiki/B.A.T.M.A.N.
And further, it appears that there is some amount of criticism of OLSR from
the BATMAN camp:
No, not every one in the Freifunk communities believes these claims.
It is probably better not to go back to these senseless "protocol wars" .
No - you misunderstand - I was just mentioning that as _the basis of_ my current understanding.
I have no interest in the previous politics, I was just making some assumptions, and these are the background for my assumptions.
This all makes sense to me, but what I don't understand is, why do all (or
nearly all ?) of the freifunk
networks use OLSR, and continue to use OLSR on the freifunk firmware ?
Are any of the freifunk networks using BATMAN ? If so, what routing metric
are they using with it ?
This is the meat of what I am asking - I am wondering what context is it proper to use BATMAN, and the reason I am confused about it is that I see most (all ?) of the freifunk networks using OLSR.
And as an aside, I am curious, if any freifunk networks are using BATMAN, what routing metric do they use ?
Again, no interest in past politics, just trying to answer the "where did you get that idea" question in advance :)
This all makes sense to me, but what I don't understand is, why do all (or
nearly all ?) of the freifunk
networks use OLSR, and continue to use OLSR on the freifunk firmware ?
One aspect was the OLSR-NG project, which I started. OLSR.org was at that time programmed very inefficiently. We (Hannes, Henning, Bernd, me) re-wrote large parts of OLSR and now it uses a < 5% CPU load on 200MHz linksys devices in a network of 400 nodes. The CPU improvements were dramatic! It makes sense to use trees instead of linked lists sometimes :) It makes sense, to use proper Dijkstra implementations instead of an O(n^2) implementation. So the main critique point against OLSR (high CPU usage) was actually proven to be wrong. It was simply the implementation which was broken (and got eventually fixed).
But for more reasons see below....
Are any of the freifunk networks using BATMAN ? If so, what routing metric
are they using with it ?
AFAIK the routing metric in BATMAN is "built-in" and not changeable. But please correct me if there have been changes to that recently. My know-how on this aspect might be outdated.
This is the meat of what I am asking - I am wondering what context is it proper to use BATMAN, and the reason I am confused about it is that I see most (all ?) of the freifunk networks using OLSR.
And as an aside, I am curious, if any freifunk networks are using BATMAN, what routing metric do they use ?
Again, no interest in past politics, just trying to answer the "where did you get that idea" question in advance :)
(note: I am biased in this aspect, but I will try hard to stay as objective as possible now).
For the general understanding: the freifunk (and funkfeuer.at) networks were already using OLSR at the point in time when the BATMAN idea came. It was probably nearly impossible to make everyone switch from OLSR to BATMAN (or any other protocol). The problem in the freifunk networks is that there was never a centrally planned upgrade mechanism. Every node is owned by someone else.
The other reason is that - objectively speaking - not everyone in the freifunk networks believes in either the matureness / stability / scalability of BATMAN or has some other favorite routing protocol (Babel comes to mind, others use lots of BGP). BATMAN is of course a very smart, unique and interesting idea... but OLSR has it's advantages in the eyes of others (stable and mature plugin architecture, well tested, RFC standard, around 50 implementations of OLSR worldwide, academic background, etc )
In addition, there are also plenty of other MANET protocols: http://en.wikipedia.org/wiki/List_of_ad-hoc_routing_protocols To my knowledge, finding "the best MANET protocol" is an unsolved research topic :) There are always trade-offs.
Personally I am also starting to firmly believe that the layer 2.5 (or layer 3) routing protocol does not matter so much anymore if the lower levels (Wi-Fi) are not stable enough. We can probably fight all the time about routing protocols but miss the point that we need a good layer 1 and 2 as a basis ;)))
Ok, I hope I could answer that as objectively as possible.
a.
One aspect was the OLSR-NG project, which I started. OLSR.org was at that time programmed very inefficiently. We (Hannes, Henning, Bernd, me) re-wrote large parts of OLSR and now it uses a < 5% CPU load on 200MHz linksys devices in a network of 400 nodes.
Sorry, forgot to add: Sven-Ola, Markus and quite a few others.
The CPU improvements were dramatic! It makes sense to use trees instead of linked lists sometimes :) It makes sense, to use proper Dijkstra implementations instead of an O(n^2) implementation. So the main critique point against OLSR (high CPU usage) was actually proven to be wrong. It was simply the implementation which was broken (and got eventually fixed).
a.
Hi,
This is the meat of what I am asking - I am wondering what context is it proper to use BATMAN, and the reason I am confused about it is that I see most (all ?) of the freifunk networks using OLSR.
first of all I have to say not all Freifunk networks use OLSR but I think it is fair to say many of them do and here is why:
a) These networks use the "standard" freifunk firmware that you can find here: http://downloads.berlin.freifunk.net/ - this firmware has OLSR built-in (the maintainer is/was an olsrd.org developer).
b) Many networks were using olsrd.org as protocol before batman became stable, hence they face upgrade issues (in a decentral environment).
c) These networks are community driven networks and somewhat slow to embrace change. A whole network requires a lot of maintenance - you don't want to fiddle with the protocol every week (although quite some experimented with batman - Weimar comes to mind).
d) None of us is actively trying to "lobby" batman into the Freifunk networks. If they have questions they can contact us (some do) but other than that ..
In general we see a strong adoption in new networks (no dependency issues there) and recently people are quite interested in layer 2 which gives you roaming/meshwide DHCP/etc.
So, I can understand you - Freifunk initiated an innovation that it does not use. Our biggest user groups are outside of Freifunk - funny. :)
And as an aside, I am curious, if any freifunk networks are using BATMAN, what routing metric do they use ?
The TQ based algorithm.
Regards, Marek
So, I can understand you - Freifunk initiated an innovation that it does not use. Our biggest user groups are outside of Freifunk - funny. :)
Ok, thank you. Yes, that was the thing that confused me and I now have a good grasp of it - many thanks to all.
And as an aside, I am curious, if any freifunk networks are using BATMAN, what routing metric do they use ?
The TQ based algorithm.
Is there another name for this ? I see talk of _calculating_ TQ, but I don't know what the proper name of the routing metric is ...
On Saturday 12 December 2009 12:22:39 George Sanders wrote:
The TQ based algorithm.
Is there another name for this ? I see talk of _calculating_ TQ, but I don't know what the proper name of the routing metric is ...
Not sure it helps you but TQ is an abbreviation for TransmitQuality. We often refer to B.A.T.M.A.N. IV (IV is more than TQ) because the whole algorithm is important not just the way how to calculate the number.
Hope it helps, Marek
b.a.t.m.a.n@lists.open-mesh.org
-
George Sanders -
L. Aaron Kaplan -
Marek Lindner -
predrag balorda