Transmission

There seems to be so many developers offering different WebUI improvements as replacements but very little improvement in the core.

The current core is somewhere between a mobile UI and a desktop UI.. Options are missing on mobile and it isn't a particularly powerful or efficient use of screen space on the desktop.

I am currently running https://code.google.com/p/transmission-control/ since it has an interesting take on doing both mobile and desktop very differently however there are parts of the other UIs that I wish where in the core normal UI.

Is there anything preventing these devs from contributing directly to the project instead of creating many many forks of the WebUI?

That's a good question. Seems like we all have our preferences and no consensus. I'm planning on an update to mine in a couple of weeks and then you'll see why. https://github.com/cfpp2p/transmission/ ... client-cfp

It would be really nice if even some basic fixes for how broken the webui is becoming under iOS where patched... Authentication seems to break both the desktop icon loading and resuming sessions.

lazybones wrote:There seems to be so many developers offering different WebUI improvements as replacements but very little improvement in the core.

The current core is somewhere between a mobile UI and a desktop UI.. Options are missing on mobile and it isn't a particularly powerful or efficient use of screen space on the desktop.

I am currently running https://code.google.com/p/transmission-control/ since it has an interesting take on doing both mobile and desktop very differently however there are parts of the other UIs that I wish where in the core normal UI.

Is there anything preventing these devs from contributing directly to the project instead of creating many many forks of the WebUI?

Please refer to the current web-UI as "Clutch" or just "the current web-UI". Core could be interpreted as the daemon or core bittorrent functionality.
I fear that there has been little development of Transmission as a whole. The core devs seem to want to touch the server-side web-UI as little as possible. The only upside is that the rpc-spec has been very stable. So this may give you a clue to why very little is happening server side and why it is only logical that lately more and more web-UI projects emerged. The interface (rpc-spec) is very straightforward and Clutch is old and ... sucks. Improving on Clutch is a quick win and so from my perspective I won quickly.

I created Shift for myself after a very long period of virtually no improvements to Clutch. Then I tried to address as much trac tickets as possible. Then I released it.
The amount of feedback was well ... underwhelming. Then I figured to scan this forum for problems or features that I addressed with Shift and post links to the Shift post with added explanations of what I actually fixed. I thought I was doing something good for the Transmission community. And then I got banned for "spamming the forum". My account was eventually unlocked but all the explanatory post were removed. Looking at all the junk being posted on this forum I guess the moderators have given up moderating. With the amount of posts being far from what it used to be, I guess the community has very little interest left as well.

To get back to your actual question: Yes there is. Little interest/lack of vision from the core dev team/person concerning the web-UI. Using subversion and have contributors deliver patches is also not a great way of doing things. The patches usually get stuck in trac with the core devs cherry-picking whatever they like. Moving the code to Github might spark new interest in collaboration.

Personally, I have not been invited to collaborate with the actual project and I do not have the time to maintain a complete fork. So I just contributed what I could: a far more functional web UI than what was there. (Even if you hated the green on black theme, lazybones!)

cfpp2p wrote:That's a good question. Seems like we all have our preferences and no consensus. I'm planning on an update to mine in a couple of weeks and then you'll see why. https://github.com/cfpp2p/transmission/ ... client-cfp

How is that Shift integration working out for you?

killemov wrote:How is that Shift integration working out for you?

I'm quite happy with it. I really like the way you've included everything RPC and I actually do like green on black theme and even tweaked it a little more to the oldest of the green on black theme, which I liked better yet.

killemov wrote:I fear that there has been little development of Transmission as a whole. The core devs seem to want to touch the server-side web-UI as little as possible. The only upside is that the rpc-spec has been very stable.

There is some stuff going on, but it seems uncoordinated. The web-UI is not getting dev attention since there is no-one established as the developer. Contributors who submit patches to the web-UI might do more if they'd wait it out and work their patches but devs don't have time to do it all this way. I think someone who'd be willing to develop for the web-UI is out there as long as devs were willing to put up with a few mistakes. With libtransmission, old bugs and crashes are not fixed before new ones even if there are indicators of a reasonable solution. I think the Devs are very busy and just don't have the quality time they need to invest properly. The rpc is very good and this allows someone to somewhat quickly add new features to libtransmission and integrate them easily to their preferred clients, such as just got done with my fork here: https://github.com/cfpp2p/transmission/ ... e8e05c0861 The rpc is in a good stable state, this is a good thing.

killemov wrote:The amount of feedback was well ... underwhelming.

I for one appreciate your work, although I haven't had time to do any development other than the theme tweaks. SHIFT has provided for me a very much needed full rpc via the server. Everything is quick, simple, clean, easy to understand and uncluttered of frivolous graphics, these things I like. When I get a chance I'd like to look more closely at your rpc implementation with SHIFT and put in the added features I've done at my fork.

killemov wrote:Moving the code to Github might spark new interest in collaboration.

There is an official unofficial fork but it seems disjointed and lagging behind the svn. https://github.com/wereHamster/transmis ... its/master I don't know why this is but if that were maintained properly then maybe things would move a little better. The existing community might be familiar with the svn but the GitHub could still be utilized, possibly.

killemov wrote: Personally, I have not been invited to collaborate with the actual project and I do not have the time to maintain a complete fork. So I just contributed what I could: a far more functional web UI than what was there. (Even if you hated the green on black theme, lazybones!)

As with most things in life it is better to ask than to WAIT to be invited.... I find it is rather rare you achieve anything waiting for someone else.

Given the fixes and enhanced RPC features I think I might give Shift a shot again however https://code.google.com/p/transmission-control/ I think is more true to transmissions look and feel... I am largely looking for a better / more powerful mobile experience... I like to add torrents on the run and under iOS at least that leaves few options that are missing features.

lazybones wrote:... Given the fixes and enhanced RPC features I think I might give Shift a shot again ...

I just committed a change that might just blow your pants off. Does "Dynamic column changes triggered by changing the status filter." sound good?