On 2/2/06, Michael Cottenden <[EMAIL PROTECTED]> wrote: > As someone new to *, with limited Linux experience, trying to implement in a > corporate office, I'll throw in my two cents... > > I can pretty much manage the administration of the system via good old conf > editing. I've spent time with [EMAIL PROTECTED] and AMP and like AMP, but I > wanted to get > my hands dirty and understand exactly what the heck is going on in the > dialplan, etc. Having said that, my implementation is going much, much > slower than if I used AMP, but I'm learning so much along the way! :) > > But, to follow on Lee's point - * needs a simple interface for the user to > access all the features of the system they should be able to change. And the > receptionist needs an app to control incoming calls, transfer, etc. > > From my perspective, the user interface is more critical - if the * > community wants to see Asterisk adopted in the corporate PBX market. Having > a great interface for the administrator to configure the system is just not > as important. In my case, managing a couple hundred users - once the system > is in and working, the system changes are pretty trivial to manage. I don't > need a brilliant interface to do that. (But if there was one, I would use > it!)
When I started the Asterisk Documentation Project, my ORIGINAL plan was to develop a GUI for Asterisk. After simply doing some brainstorming, I decided documentation was a MUCH better idea because all a GUI does is confiscate the underlying power of the system. OK -- we've all heard that argument before, but I HEAR you. I know what you mean. HOWEVER -- I don't think you're going to see a GUI from the main people who develop Asterisk for a very long time (ever?) because those people know everything about Asterisk. They have absolutely no use for a GUI, and if you think they are going to build one for you (us), then you're probably dreaming :) The only GUI you are going to find are the ones created by the user community (not often), or if you build your own. At least for the time being. Maybe some time in the future Digium will see that it is advantageous to devote resources to writing a GUI for Asterisk, but I have a feeling that will be FAR into the future. Since I work with Asterisk every day directly, have even looked at some of the code, work on the bug tracker and just have generally been using it for a long time, I can see all the little things that need to be "fixed". Others can see these too. The problem with creating a general purpose GUI is that it does too much, and not enough of what you want. Also, AMP is not a good way (IMHO) to learn Asterisk because the dialplans and configurations it creates can look very messy. While it may take longer to get over the learning curve by actually learning Asterisk, you will be in better shape to manage a system and quickly fix things if they break; and if its not quickly, at least you'll have an idea where to start instead of learning Asterisk only after something breaks on you. GUIs are a very personal thing. While AMP is probably great for someone who simply wants to run a small PBX at home (you're going to be seeing home PBXs advertised in less than 12 months on TV on a regular basis I imagine), as soon as you start to run it in a company, the dynamic of the system totally changes and you need to be able to get to the guts a lot easier. After saying all that, I can actually envisage what a standard SOHO PBX GUI would look like and have been planning on developing something for people who come up to me and want to purchase a PBX from me, but who can't afford a full custom PBX, but want all the "basic" features, and would be willing to pay for "advanced" features. The system would obviously have to be modular, and you'd have to be able to go in and add your own custom dialplan logic without upsetting or confusing the GUI. I can already start to see the complexity in building such a system, and this is just for small business. While I'm not anti-GUI by any means (this is what I do -- build GUIs for Asterisk to allow me to manage a VSP) -- I'm a firm believer in building small, task specific systems. If you don't, you're going to end up trying to build the impossible. If a GUI is *really* important for a company, that's probably going to be the one thing they are either going to have to develop in house, or purchase. Asterisk is not free. Its a great tool kit, but it still costs you either in man hours, or in development costs -- just like all good open-source software should. It keeps people like me employed and fed :) (and you!). -- Leif Madsen. http://www.leifmadsen.com http://www.oreilly.com/catalog/asterisk
