Page 3 of 49

Re: Call for a web based lobby

PostPosted: 29 Jul 2013, 12:11
by Gyle
Raging_Squirrel wrote:no really, why should it be *WEB* based? do you have any solid argument about it?


I gotta echo this. I'm a complete techno noob so i'm probably not understanding the benefits. Can someone in the know please explain to me what a web based lobby will do for us that we don't already get???

Re: Call for a web based lobby

PostPosted: 29 Jul 2013, 12:16
by Ze_PilOt
dstojkov wrote:I find this game a niche for new knowledge and new challenge ... who know how far this can bring us !


I fail to see why a new lobby would so anything FAF is not already doing.

Re: Call for a web based lobby

PostPosted: 29 Jul 2013, 12:31
by Ze_PilOt
dstojkov wrote:Well ... kind of .. I broke up with one girl friend this week but no worries still have two left ...

My graphics card also gives up this week ... Now have to use my gaming laptop instead ...


If you are so eager to see the mod function, check the external contributor section of the forum, there is a lot of work to do, you can participate.

Re: Call for a web based lobby

PostPosted: 29 Jul 2013, 16:59
by dstojkov
Gyle wrote:
Raging_Squirrel wrote:no really, why should it be *WEB* based? do you have any solid argument about it?


I gotta echo this. I'm a complete techno noob so i'm probably not understanding the benefits. Can someone in the know please explain to me what a web based lobby will do for us that we don't already get???


The main Idea is for the continuity of the project.

The lobby right now is ok but think that this will may be last more that 10 years if some stuff has to be added to it and that one day Zep gives up on the project what will the community do ?
The more the project will fit the standard the better.

A web lobby is the more portable environment that we can get and if one day the source code of FA become open than we will have a full integrated system that with some rework will be able to run on all the know platform.

Re: Call for a web based lobby

PostPosted: 29 Jul 2013, 17:37
by Ze_PilOt
dstojkov wrote:The main Idea is for the continuity of the project.


By breaking the continuity with an existing one ?

dstojkov wrote:The lobby right now is ok but think that this will may be last more that 10 years if some stuff has to be added to it


Then contribute to add these things instead of re-inventing the wheel.

dstojkov wrote: and that one day Zep gives up on the project what will the community do ?


They will do the exact same thing that if you do own lobby and give up. Probably continue without me/you.
If you fear about the server, some other people already have the keys to it.
They can start a replicated server if I die under a car tomorrow.
And if I don't die, I will probably officially give the key to someone.

For the lobby itself, it's open-sourced, so no extra worries.

dstojkov wrote:The more the project will fit the standard the better..

I still fail to see how java is more standard than python.

dstojkov wrote:A web lobby is the more portable environment that we can get


And the more limited too (good luck launching FA as a process from a JRE in a browser and have control over it, for example).

I really fail to see what the advantage of a web lobby are. You are limited, and can't use web-based (php or whatever) stuff (because you run java in a browser, you will have to run a browser inside java inside a browser). Or communicate through JS... Hey, exactly what FAF is doing currently (see the map vault :) !

And java is a pain in the ass by itself. Asking for updates everyday day, I've uninstalled it from every computer I'm using.

Also, the portability of the lobby is a trivial issue, as FA itself is not portable. (and it seems that FAF run perfectly in linux without additional effort).
Unless you want a macOs support, the problem will be running FA on it.

dstojkov wrote:and if one day the source code of FA become open than we will have a full integrated system that with some rework will be able to run on all the know platform.


Sure, porting a directX application on macOs or linux is easy *irony*.

Seriously, even if we have some day the source code of FA (I really doubt it), porting it to mac is probably the last thing you should put efforts into.
A 64 bits version of it will be challenging enough.

Re: Call for a web based lobby

PostPosted: 29 Jul 2013, 18:48
by FireMessiah
Ze_PilOt wrote:They can start a replicated server if I die under a car tomorrow.
And if I don't die, I will probably officially give the key to someone.

your not allowed to die under a car tomorrow,
at least until GW is complete :)

FAF is already the best solution for FA, why on earth would we want a web based application?
Support FAF !

Re: Call for a web based lobby

PostPosted: 30 Jul 2013, 13:51
by dstojkov
... java is not beloved here nonetheless all decent company use java in production. I only saw one time script language in a startup ...

I think you should write a paper and explain to the world software industry how they are wrong ...

Re: Call for a web based lobby

PostPosted: 30 Jul 2013, 14:09
by Ze_PilOt
dstojkov wrote: all decent company use java in production.

Really ? I don't think so. It's kind of being highly deprecated since Oracle bought it.

Python in the pro world :
http://wiki.python.org/moin/Applications
Also used in about any professional VFX application and VFX house (my domain or work) :
- Every 3d app (Houdini, maya, XSI, Blender - Only max and lightwave don't use it)
- Nuke (2d compositing tool).
http://opensource.imageworks.com/ | http://code.google.com/p/alembic/ (Sony ImageWorks)
http://www.openexr.com/ (ILM)
http://www.disneyanimation.com/technology/partio.html (Disney/Pixar).

ie. you probably used one python app everyday without knowing : bittorent (now ĀµTorrent is c++, but it's closed sourced too).
Reddit is written in python.


Also, it's not because something is widely spread or hyped that it's the best solution for you case.
It highly depend of the usage you do and, mostly, the available time you have (coding FAF in java would probably have take me 400% of the time it took me with python).

Re: Call for a web based lobby

PostPosted: 30 Jul 2013, 14:18
by Ze_PilOt
I think you should write a paper and explain to the world software industry how they are wrong ...


I don't need to, they are many of that already :
http://www.guild.net/~schwern/papers/Wh ... r/why.html
http://www.screamingatmyscreen.com/2012 ... -use-java/
http://betanews.com/2012/09/03/you-dont-need-java/

I continue ? I can find you hundred of articles of why Java sucks. Actually, I can't think of any company saying "We will do it in Java" without being laugh at.
The last one was Google for Android, I don't think they will do it again with Oracle asking them 2.3 Millions for it.

Python (and pyQt) allow fast prototyping, so fast development, it's entirely modular so easy to add stuff on it (mandatory for the kind of work FAF is doing), and if you need performance, converting it to c++ code is kind of trivial. (look at the proxy Server code in my repository, there is a python version and the same in c++, the code is 90% the same, took me 15 min to convert).

Re: Call for a web based lobby

PostPosted: 30 Jul 2013, 14:38
by dstojkov
Well for writing prototype yes scripting is the best thing you could ever had. That is why I don't condemn script language. I saw all the benefit in the lua stuff in fa.

But for long running app they are not used. I will not try to argument on which techno is the best all have + and -
Still I will made it in java and without applet so that the client just need a web browser I mean you can say this is worthless or this is crappy well ..
Let's see when it will be in alpha, beta test if you will say the same but not now when it has not even begun !!!