[Tuxman]

So you want a mod-driven BitTorrent-like Kad2 subnet?

[xSTHNSx]

Far as development goes in 3G-P2P in general our DHT based network was hindered from the start, I don't need to tell you that as you know this. Thea idea of KAD2 is client based and shouldn't be another network to bootstrap into more like additional support added to client which would benefit clients that supports it, since lacking in mainline.

Currently I can support 8gbps+ (1000MB/s). Hell if kMule was multi platform and could be deployed on LINUX as headless. I wouldn't mind being test for KAD2 SuperNode for kMule users. I mean way I see it is this the hardware is unused along with bandwidth which is wasted so why not use it. I have total of 8 nodes online now so far as i'm expending and add one every 2-3weeks. I mean we use smartphone now which has more power in single core than we did decade ago.

[DavidXanatos]

mmh... supernodes.... I wonder how one would approach implementing that. As others already pointed out the network needs a way to know how much resources the nodes have that brings you to an other issue how to integrate that properly into Kad routing. As "larger nodes" must not only index more files but ideally also "cover" more address/id space.

When requesting closest nodes the asked node must than not only use the distance to determine what nodes to give back but also possibly provide further away supernodes :/
That of cause complicates things somewhat. A tweaked the distance function is needed that would allow for nodes with a "further reach" The simplest fix would be for a node to provide an "acceptance range" a.k.a. the maximal distance the node accepts payloads from. Than the actual scoring for the "closest node" function would not be based on the distance but on how well the particular target ID fits within the nodes acceptance range.

[RejZoR]

Isn't Gnutella 2 (in Shareaza at least) using such approach? The leaf/hub system? Not sure how it works exactly, but it sounds similar to what KAD could be with nodes/supernodes...

[Sir_Boagalott]

I have a concept theory for link sites to use a Leaf Hub idea for Supernodes and Meganodes.

i.e. Have link sites act as Meganodes and publish your files to the link site. Link sites then send it to Supernodes (i.e. the faster clients they know on the site).

Something like this could potentially verify files too. i.e. w/e file list you send to the link site could be considered verified by you.

[pier4r]

Tuxman, on 22 February 2016 - 11:27 PM, said:

No, C# is not "easier to handle" nor "known by more people".
Technically, it even sucks more.

Hmm, maybe while you ahve a point regarding porting emule (the ones doing the porting should know c++ as well), but AFAIK in C# one can reach some complexity without caring about a lot of stuff, for example freeing memory and such. It was developed to be on the same level of Java for a reason and (again AFAIK) when i read around in programming discussion places ( reddit.com/r/programming reddit.com/r/learnprogramming, etc.) for windows based developers is quite common to use it.

If not, could you give me / us more info?

[xSTHNSx]

When it comes to programming you need to understand one thing. That there are more than one solution as with multiple approach to solving any given problem. The limitation is only based on the programmers ability to translate their ideas into expression based on the rudimentary knowledge at that given time. Now far as C# goes I once did a project back in 2004 which was targeted for Windows (NT 5.0), which started out great as the project Base 1.x grew so did the public userbase. So as there was the need for experience the first upgrade for Base 2.x was ditch the project and created separate core along with is own own internal kernel. As I grew as programmer over time based on what I learned over time from when I initially started the project along with multiple other projects I have done I decided to combine them all into Base 3.x single unified. For me to do it I had rewrite it in JAVA so all aspect of it can be deployed since now it was targeted for LINUX (Slackware 11). As some time passed the project later was upgraded to Base 4.x with ~elf for full multi platform and directly integrated into aspect of the OS with now its own shell, which was by then targeted and deployed under UNIX (FreeBSD 7.2). But when the project was done the final version over close to decade has passed and it was the most robust project in its field user who used claimed which was far superior than anything available to them even now. If I started the project out in JAVA than C# and didn't take the limitation of language for OS into consideration. Then I would have saved myself over 4months of actual coding over period of one year. Ever had carpal tunnel from writing codes looking at the screen for hours end while drinking redbull? and testing everything in detail with being very intrusive every single line by line how its handled and interpreted? as its integrated to OS taking live input from users connected to base deployed in shell thus subject to exploit and takeover. It is also good to write function based on your own given template so if something is changed within it you don't have to write all functions to accommodate to reflect it, as you can just amend to current. Now point of this story is that as time goes on so does the programmer as they learn and gain experience as they grow older and view things different than when they were younger and incorporate other things they have learned over the years. Now its clear to have objectives which even upon completion meets the goal for when it was initially developed. But it is imperative as time goes on those objectives can and will changes to reflect current timeline as it can be different now than from when it was developed, which is vital to its relevance for survival or it will fade into obscurity as you need to adapt for your new environment.

[tront]

Having read all the discussion, I don't see why new development can't continue in a new branch. If it provides benefits everyone can appreciate, more and more users will migrate to it. Eventually it will become the new "baseline". Even new networks have a chance of succeeding from scratch, so why not a better client that's based around an existing network?

The draw to this network for me is the ability to find rare and old stuff, which I guess is the result of both self-publishing and being able to have a lot of active files simultaneously. I also like the decentralization of everything, including search. But most people don't care about these things, they're just after faster speeds and less waiting before transfer starts. The relative decline of eMule may be simply the result of new tools, like BT, being more suitable for certain jobs. The people who have stuck around are those who are after the strengths of this network.

This post has been edited by tront: 10 March 2016 - 12:47 PM

[Xeus32]

Dear all,

it's a long time that i think that this project are going to oblivion.
Speak about future it's too early, there are many small steps before start to implement new feature.
For example, I tried different time to compile eMule but the solution is for VS2008 and when i try to load the solution with VS2013 starts and error hell.

From my point of view, the first step is to have access to the repository and push the patches already available in the forum (ex: fox88), update the library and the compiler so a normal developer doesn't spend a lot of time to contribute.

Start to make some testing build and go straight to new "update release".

In a second fase, migrate to a more collaborative platform like git hub.
I'm working on a patch to support mkv container but it's need update different libraries and I don't want spend other time if the patch will not push in the main code because, it will unuseful.
Now I ask to Support if it's possibile help the team and give some help to push some small patch inside the main repository and make some testing build.


Best regard

[tHeWiZaRdOfDoS]

Meh, the "fast development setup" thing has been brought up several times... I think it's a bad approach because of the way it's set up right now, you cannot blindly hit "compile" and you're done but you have to learn about all the extra libraries that you are about to use which - IMHO - is a very good thing. As far as I am concerned, there's never come anything good from that "fast fast fast" mentality.

[Trax_dat]

tHeWiZaRdOfDoS, on 26 March 2016 - 09:21 AM, said:

As far as I am concerned, there's never come anything good from that "fast fast fast" mentality.

... say usually the slow people.

[tHeWiZaRdOfDoS]

Maybe... but I prefer thinking stuff through carefully and coming up with a solid solution over rushing stuff, regretting it afterwards and having to clean up and fixing stuff later on

[beleaguered]

It's sad because there hasn't been an update in 6 years & because of that (and that dumb priority system), emule has become obsolete. It's like winmx now. I tried using it again, just for old times, and it absolutely sucks. Whoever was/is in charge of emule and didn't allow other people to work on it, is selfish AF! I mean, I can use a torrent program with a VPN and get whatever I want. I really have no idea why anyone would choose to use Emule.

[xSTHNSx]

beleaguered, on 24 April 2016 - 01:08 AM, said:

It's sad because there hasn't been an update in 6 years & because of that (and that dumb priority system), emule has become obsolete. It's like winmx now. I tried using it again, just for old times, and it absolutely sucks. Whoever was/is in charge of emule and didn't allow other people to work on it, is selfish AF! I mean, I can use a torrent program with a VPN and get whatever I want. I really have no idea why anyone would choose to use Emule.

I don't know why people think deploying torrent over VPN is going to do anything or any other network for that matter. It doesn't make anything faster matter of fact it would make it slower due to com/decom of packets with encryption, VPN also does not make you anonymous. Now what you can do is give all your bandwidth to our ED2K network so others can download and if you need anything then find a community as we have STF/TvU which index. Abandoning the network doesn't help anyone and torrent is not the answer. Like many here we all share the same idea and we know because of the stupidity of few obstructionist and refusal for them to change things for better even now crippled the network for past decade. As users slowly fade away these idiots still in current timeline talking about backward compatible support and overheads.

[pier4r]

beleaguered, on 24 April 2016 - 03:08 AM, said:

It's sad because there hasn't been an update in 6 years & because of that (and that dumb priority system), emule has become obsolete. It's like winmx now. I tried using it again, just for old times, and it absolutely sucks. Whoever was/is in charge of emule and didn't allow other people to work on it, is selfish AF! I mean, I can use a torrent program with a VPN and get whatever I want. I really have no idea why anyone would choose to use Emule.

As discussed billions of times (you can make a search) the philosophy is different (i do not know about winmx).

Torrent: let's help a master releaser a bit, while we get what we want. (unless enforced ratios are in place)
Emule: let's share between us, otherwise we do not get what we want, with the speed that we have (normally small).

[hooligan3000]

xSTHNSx, on 24 April 2016 - 08:40 , said:

beleaguered, on 24 April 2016 - 01:08 AM, said:

It's sad because there hasn't been an update in 6 years & because of that (and that dumb priority system), emule has become obsolete. It's like winmx now. I tried using it again, just for old times, and it absolutely sucks. Whoever was/is in charge of emule and didn't allow other people to work on it, is selfish AF! I mean, I can use a torrent program with a VPN and get whatever I want. I really have no idea why anyone would choose to use Emule.

I don't know why people think deploying torrent over VPN is going to do anything or any other network for that matter. It doesn't make anything faster matter of fact it would make it slower due to com/decom of packets with encryption, VPN also does not make you anonymous. Now what you can do is give all your bandwidth to our ED2K network so others can download and if you need anything then find a community as we have STF/TvU which index. Abandoning the network doesn't help anyone and torrent is not the answer. Like many here we all share the same idea and we know because of the stupidity of few obstructionist and refusal for them to change things for better even now crippled the network for past decade. As users slowly fade away these idiots still in current timeline talking about backward compatible support and overheads.

a vpn is in ger. a must have when u share critical files ;-)

[xSTHNSx]

@hooligan3000

I just sent you PM check it out you might like them. There is no way you can beat that for that price.

[Fensergeist]

Hello. My name is Alberto Salvia Novella, and I had been member of the support staff of eMule for two years in the past. Also I have written part of the official eMule documentation myself.

Currently I coordinate Ubuntu OS papercuts quality assurance, and I'm the person who prioritises nearly all bugs in it. That is the person who warrants that half of the Internet works without bugs. I have been doing so since 2011.

I'm currently studying computer engineering, and I'm certified as professional lean manager and community manager. I bought my first software development kid when I was twelve, and I have used around 50 different operating systems.


And here's what I wanted to tell you: skip the words, see the facts. If the current opinion is bringing bad results, for sure is the wrong one!

This post has been edited by Fensergeist: 22 October 2016 - 01:48 PM

[Fensergeist]

Instead of being that careful, do this:

https://gettingreal....in_the_Wild.php
http://forum.amule.o...09703#msg109703

[DavidXanatos]

xSTHNSx, on 24 April 2016 - 08:40 AM, said:

[... ], VPN also does not make you anonymous.

Depends on the provider, there are many that will even protect the privacy of nazis and similar to the bitter end.

Quote

Now what you can do is give all your bandwidth to our ED2K network

And without a VPN be sued the same way as BitTorrent users are? P2P is P2P and if it is not intrinsically anonymized; protocol differences don't matter.

Quote

Like many here we all share the same idea and we know because of the stupidity of few obstructionist and refusal for them to change things for better even now crippled the network for past decade. As users slowly fade away these idiots still in current timeline talking about backward compatible support and overheads.

Amen!