To hydranode grow up need a Very good Gui and Full support to ed2k(we need kad and many features of official emule) , and bittorrent..

The gui is the principal part !
But we need full support the network, to not and like shareaza or mldonkey
 

Full unicode support is needed to please east asia users. As I know, there's a big potential group in China that would like to try Hydranode only if unicode support ready, for both ed2k and bt module.
 

i see two ways to improve hydranodes popularity, but i dont know, which is easier to achieve.

1. as you already mentioned some time ago, there is not only one group of user. so there has to be easy installation combined with easy usage, but with the option to fully customize as much as possible, and all this with most features/compatibility to existing clients. at time being i wouldnt use hn on win or linux, although i tried both. on windows hn simply misses some features i expect, as there are other application with them. (serverless proto in particular) on linux the situation isnt that better. imho its main competitor, mldonkey lacks the same features like hn, but with excessive, more complete ui-options (web, telnet, remotegui with support for all core-features, but the latest experimental), not to speak of the headache, hn gives for build tries. (for instance with a modern compiler like gcc4.1) i dont even tried to build it for x86_64 (i like native applications, even if the performance gain, or whatever is minimal)

2. a killerfeature, or should i say THE killerfeature for modern filesharing, a high performance anonymous filesharing protocol, which scales with the participating system, not against them. imho a mature distributed file storage protocol looks most promising (some interesting approaches you can find at www.infoanarchy.org/wiki/index.php/Distributed_file_storage).


so long,
 

The recent raids on German edonkey filesharers (who were logged via a donkey server) and the (questionable) Pirate Bay TorrentTracker crackdown should have made obvious once again the relative danger which is threatening large parts of p2p users. Many of them are now refraining or or moving on, just like so often before (Napster, FastTrack etc.; away from first and second generation networks).

ed2k has potentially become unsafe for use for a lot of people; especially without Kad (proof now that servers are logged with following raids). Adding Kad support to the ed2k module may delay the point Hydranode becomes unusable for many users - without, it is already.

I see adding another network like Gnutella is close to being useless because by the time Hydranode is the best client for it, noone will be willing to use Gnutella/not-anonymous networks anymore. Quite the same for perfecting BitTorrent (not that I couldn't and shouldn't become better in the meantime as it is already there).
There are also already some http/ftp download libs, also free ftp clients and binary downloaders. I don't want to be too pessimistic but for me this alone is not so likely to make a huge impact or attract enough people who are willing to donate for the development. Perhaps when Hydranode excels at it, but this will take time - who donates to another ftp client in infancy?

Encrypted, anonymous networks like AntsP2P, Mute, Waste - that's where newly created clients are going. (If not, they're one the many BT clones). (Leaving aside outside factors such as anonymous layer like i2p.net)
There people would also immediately see the value, even if a client/network it's not so feature-rich yet (because at the moment there are not so many usable networks out there yet).

A secure client which has so far made a significant impact is not existent. This could be Hydranode. If there would rise an actively developed encrypted+anonymous network in the near future, it will make an impact and wouldn't be overlooked like it would have been in the past; the need is there and it's growing. And as with P2P, the more people are already there, the more will come. My bet is that this is one of the few directions Hydranode could take and generate non-insignificant income. People would feel the need to have it developed, see it growing and would donate money to support the development.

For a e.g. BT client, I would see less need, also in fostering it with money. There are already lots of them, also many good ones. Also with the increasing legal outside pressure, user count is going down, not up. (Just imagine, BT module is finally done and it's perfect, but there's noone left willing to use BT). The same for being the ultimate downloader for every known weak/insecure protocol known to mankind (Gnutella and others). And the audience for FTP and Usenet is much smaller and fragmented. I'm not sure if you could raise enough funding this way.

The developer of Mute (mut-net.sf.net) tried to make a living of coding his anonymous (but not encrypted) client and network. IIRC His fund-raising goal where also more or less met, at least in the beginning, when the development was even in its infancy but still more active. Later on development got more rare and funding with it.

To me it looks like the most thriving anonymous network for p2p is antsp2p.sf.net, it's also in concept similar to ed2k (though the network has not emerged as a clear winner yet - seen from the protocol side. From the client side, Ants is written in Java; when given the choice users will decide against that). (Mute: won't have swarming, development slow; Waste: a small, private not mass-network; GNUNet: not made an impact in it's long history; Freenet - a datastore; I2P, Tor: layers/different scope, not quite ready). Good to know that the used protocols and concepts are known and every network which could make it has sources avaiable, so it wouldn't be Hydranode against the world, but HN could take a strong lead amongst users, especially with frontend design.

So far my view on the whole evolution of P2P. Would be great to get some information from you if you consider anonymous networks an option for the direction Hydranode might take.
 

I think we should invite a lot of developers for the HN develop. In this case you would be able to get a job.
 

Anonymous with his long post above managed to write my thoughts down. Full ack there.
chemical
 

Yeah... all that is written above. Plus, the idea of doing other programs with the engine is a good one.
And welcome back, Madcat!
 

trying to revolutionize those areas where others have been years before is a very risky approach if you want to base your future income on it.

re-inventing the ftp-client or it's user interface should be quite a challange with hundreds of existing clients.

imo the best approach of getting popularity would be to deliver an state of the art ed2k/torrent client that supports all the advanced features (e.g. emule/amule functionality plus what some of the emule mods might offer additionally, e.g. special seeding modes), has a nice (skinnable?) gui and maybe even offers features not supported by other clients (anonymity/cryptography should be the key words in these time).

if you manage to differentieate your client from the rest you'll get your reputation (and page impressions) in no time. but hopping from hydranode to another (more or less meaningless) project leaves a smack of unfinished business. i'm not saying hydranode's missing essential features here, but compared to other clients and guis, do you really have the superior client that rules them all at your hand already?

if you really want to implement something ftp-like why not include a ftp-server in hydranode and enable people who are marked as special friends the download of chunks/files directly via ftp to speed things up? that should also help with providers blocking p2p traffic if you could alternatively transfer via ftp protocol.

might be a stupid idea, but that's what my hungover mind just told me to write :)
 

I had already spend same money (moneybrooker) for the HN development.
I would spend even more money for an GNUNet module.
Promised.
 

You can create a new type of file that permit to start a download in all the networks that hydranode support, at the moment, edonkey and bittorrent.
I think that someone will add the hydranode links to his elinks/bittorrent page (or will create a new page) and this will incrase hydranode's popularity a lot.
An easy solution :)

I'm spanish so my english isn't very good... sorry
 

This post has been removed by the author.
 

>I would spend even more money for >an GNUNet module.
I would spend some money, too.
Some anonymous network support
would be a killer feature.
 

Well, in short time there are few things that I thing need improvement:

1) Support for proxy
2) Port testing in case you are behind a firewall or router, like emule morphxt does.
3) Local/remote authentication support for user interfaces/shells. (Planned).
4) Web interface for both use/administration. Administration could ask admin password for example. This depends on 3, you could admin remotely by hnsh or via web (any operating system has a web browser).
5) Better comand line interface, to be able to do all or at least most things you can do with the interface.
6) Some other network supported could help to consolidate hydranode as a real multinetwork client. 2 networks aren´t that much.
7) IP filtering?. By now this could be a little enhacement.

Some people talks about creating a new anonymous/encrypted network. I thing this is a very ambitious idea and is not easy to be achieved. How could one manage to conquer such BIG project? Just my thoughs.
 

There is a very good reason why emule is THE one and only ed2k client: It does what it's meant to, and it does it perfectly! Sad to say, but Hydranode is far behind emule at the moment.

It doesn't matter that hydranode can download from different networks unless it does it as good as the best clients out there.

Example:
emule user tries hydranode and thinks "nah emule is better" and doesn't even give a chance to hydranode.

Instead, he should think "Whoo this program has every single feature that makes emule so good! And omg I can download torrents too And it also has every single feature that makes Azureus my favorite!!"

So my point is:

Hydranode now downloads from 2 very popular network, ed2k and bittorrent. You have to polish ONLY these features until they are perfect before even planning anything else!! I can't stress this enough: WHAT YOU DO, DO IT PROPERLY!

And I'm not saying hydranode is bad, not at all!! But before even planning anything else like ftp clients, media players etc you have to make hydranode PERFECT on what it does. Why hurry with supporting more networks?
 

"Instead, he should think "Whoo this program has every single feature that makes emule so good!"

Please define what you're currently missing at Hydranode.
Just saying "eMule is better" doesn't really help to improve Hydranode.
 

"Please define what you're currently missing at Hydranode.
Just saying "eMule is better" doesn't really help to improve Hydranode."

That's very true! Well I try, here are few things...

GENERAL

- Hydranode should show on the ”home” tab which networks it is connected. (or like in emule, this information is visible all the time in bottom panel.)

- Better ed2k server management is needed. Right now all user can do with gui is connect to server or delete server from list. With emule you can also add servers or update server list from url.

-more ed2k settings needed!! (for example does hydranode update server list automatically, I personally don't like this feature in emule because it adds fake servers to list. I prefer adding my favorite servers by hand and use always them.


SEARCH

- In hydranode I can set file min and max sizes for search, that’s all. In emule I can also set file availability, file extension and complete sources + some not so important

- In emule I can do search by type, like media, program, video.. etc. very handy. HN could also do even better than emule by allowing user to edit default types. (for example if user wants to search audio files but don’t like .wma format, he could delete .wma from “audio” category)

- In emule you can choose search method (server, global server search, kad…)

- Emule finds MUCH more files with same search
 

[i]And I'm not saying hydranode is bad, not at all!! But before even planning anything else like ftp clients, media players etc you have to make hydranode PERFECT on what it does. Why hurry with supporting more networks?[/i]

I agree! I think he must finish first at 100% those two networks before the other ones. Maybe starting now Gnutella but in very low priority (and adapting code stuff if necessary in the process).