WARNING about FreePBX 2.8

[Lost Trunk]

I know that PBX in a Flash is only up to FreePBX 2.6, and perhaps that's where they should stay indefinitely, if this exchange on a FreePBX forum is indicative of the direction that FreePBX is taking:

freepbx.org/forum/freepbx/beta-program-issues/2-8-dial-pattern-and-rules

If I am understanding it correctly, this is a major change in the FreePBX interface that has the potential to seriously inconvenience many experienced FreePBX administrators. What also comes across in that thread, in my opinion anyway, is the arrogance of the FreePBX lead developer.

I can't understand why some companies have to keep "improving" a good thing until they get it horribly wrong. First Apple and their new iPhone (with the lousy reception if you hold it the way any normal person might) and now FreePBX???

 

[dswartz]

to be fair, there is nothing wrong (AFAIK) with 2.7, nor is it greatly difference behavior or appearance wise from 2.6.

 

[YoungOrtho]

That is really disappointing. One of my routes is at least as long as that posters, if not longer, and there is no way I would have entered them one by one, as I apparently would now have to with 2.8.

 

[blanchae]

. What also comes across in that thread, in my opinion anyway, is the arrogance of the FreePBX lead developer.

I don't see any arrogance there. I do see someone trying to balance the needs of experienced and new users. And he's made it quite clear that both have been taken into consideration. As he states, you can use any version of FreePBX you want or none at all. This is open source free software - you get what you pay for and you are getting an incredible resource for nothing. Unless you are paying for support then you have nothing to gripe about.

The issue is having two software methods of defining routes for the same page: text area vs. individual routes. It makes perfect sense that only one software method should be used. It would nice if he could provide a "3rd party" module that could be used by advanced users that would follow the new software method that has been chosen. It comes down to nice to have vs. need to have.

I applaud his efforts for the development of FreePBX - it was a monumental undertaking and one of the most significant steps in the adoption and evolution of asterisk.

I have attempted to assist user's in solving their problems in the FreePBX forums and it is extremely difficult to do. We are lucky with PiaF that we have a structured and stable platform to work from that is strict in its environment - that is thanks to our development team.

Unfortunately, in the FreePBX forum, there are no standards for platform installation. You have users installing it under Ubuntu, Redhat, Slackware, and any other number of scratch installations. Here I can just turn on my test server and compare my configs, the directories, premissions, etc... In the FreePBX forums, they don't have that luxury.

 

[Lost Trunk]

I don't see any arrogance there. I do see someone trying to balance the needs of experienced and new users. And he's made it quite clear that both have been taken into consideration. As he states, you can use any version of FreePBX you want or none at all. This is open source free software - you get what you pay for and you are getting an incredible resource for nothing. Unless you are paying for support then you have nothing to gripe about.

Excuse me? Let me break this down so that perhaps you understand it. The thing I think you are missing here is that when they make a change like this that adversely impacts their experienced users, it basically means that we either have to use their dumbed-down interface, or forgo any improvements that may be made in future versions. Sure, presumably we can stay at 2.6 or 2.7 forever, but consider that once 2.7 reaches "end of life", whenever that may be, there will be no more improvements (not even security fixes). Also, it would probably mean that you'd be locked into Asterisk 1.6, since if we go by past history, any major version upgrade of Asterisk will have changes that FreePBX will need to account for.

Now I grant that if you put the security issues aside (would you really want to do that?), you could probably run a system with FreePBX 2.7 and Asterisk 1.4 or 1.6 indefinitely (but if your hardware ever breaks down and you ever have to set up a new system, you'd better hope it's compatible with the old software).

What I really disagree with are two things you said in you post. First, you said, "I do see someone trying to balance the needs of experienced and new users. And he's made it quite clear that both have been taken into consideration." Please explain exactly how experienced users have been taken into consideration? Sounds like he's pretty much saying that if they don't want the dumbed-down version, they can just stay with older versions. That's not taking the experienced users into consideration, that's a slap in the face!

And second, I strongly disagree with your assertion that "Unless you are paying for support then you have nothing to gripe about." What if you are experienced enough that you don't need paid support? And anyway, FreePBX has always been FREE - it's in their name, for crying out loud. Thankfully, I've seen no indication on the FreePBX site that they only give consideration to users that pay them (well, except for that whole Custom Contexts thing), but I still think it's pretty arrogant to hand-wave away your experienced users the way the lead developer did in that thread.

It's attitudes like the lead developer's, and yours I might add, that make me wish there was another viable alternative to FreePBX out there. Since the Asterisk people can be just as infuriating at times (breaking existing functionality for no apparent good reason at each major upgrade), I'd love to see a project based entirely on alternatives to Asterisk and FreePBX, particularly if it were something that PBX in a Flash could get behind (because this is the best online community out here for this type of software!).

 

[kenn10]

Don't Throw The Baby Out With the Bath Water

I think FreePBX 2.8 is pretty good for a broader base of users. Philippe has tried to make it less manual and more GUI-based so that more mainstream users can navigate through the PBX and do their own programming.

Based on the number of newbies in all the forums and the insanely diverse operating systems I see people using, I'd say this is happening. Unfortunately, FreePBX and Asterisk forums are getting clogged with people who want to run Asterisk/FreePBX without first investing any time into reading and learning about the product. PIAF keeps things sane by locking in the operating system and FreePBX version while giving us the ability to upgrade at our own pace.

As stated above, I'm sure someone can easily write an add-on module which will import a spreadsheet of routes much as the Bulk DID and Bulk Extensions module work now.

FreePBX 2.8 is clearly a necessary evolution to make the product less engine dependant. Since FreePBX 3.0 will support Asterisk and FreeSwitch, there must be some bridge between current FreePBX and the re-engineered V3.

I think we need to look at this from a "big picture" view and not purely from a selfish "just for me" view. There is no other viable GUI for Asterisk and FreePBX can now compete with proprietary systems out there from the big boys.

If you want to see something completely foreign to you, load up FreePBX v3 (which isn't nearly ready for prime time or even casual use, in my opinion.) You won't recognize much compared to the current versions of FreePBX.

So I think labeling a thread as a "Warning about FreePBX 2.8" is a little overly dramatic and myoptic. Change happens. Good open source projects must have a strong leader or they will flounder and flail around with no direction. I don't consider Philippe a traitor to the community or get the impression that they are tying to go to a closed source distribution.

FreePBX still allows the creation of third-party modules to do whatever you want. If someone doesn't like the functionality, add-on modules can be created. You may have to pay someone to do it, but you can get additional functionality. Try that with Avaya, Nortel or Cisco systems. The richness of add-on modules currently available is a major plus but those add-on modules have to be kept current with changes to FreePBX. (Thus the debacle I went through to get funding to fix the user-written Custom-Contexts module.)

I'm using FreePBX 2.8 on Asterisk 1.6.2.9 and I am quite happy with it. I see no cause to proclaim the sky is falling. If you don't like the functionality, write an add-on module.

And finally, why the "attack mode" attitude to other posters? This isn't the Trixbox forum so lets keep it polite. We may have to agree to disagree but insults are not required or appropriate in this forum.

These are just my thoughts as a user.

 

[tm1000]

And if you read the end of that whole conversation you'll see that Micke Carlsson is writing an add-on module that will allow users such as yourself to type in the dial plans the old way.

I am already on it, I have the module so far to erase route patterns and soon it will be able to add new patterns.

Stay tuned for Bulk Routes in the Contributed Modules repository.

There's no need to freak out because FreePBX can ALWAYS be changed by the community, even by it's own developers.

Also I applaud blanchae and Kenn10. Thanks for keeping it civil.

 

[kenn10]

Voila! Like magic. The world is no longer spinning out of control.

 

[blanchae]

It's attitudes like the lead developer's, and yours I might add, that make me wish there was another viable alternative to FreePBX out there.

There is an alternative that is in the development stage: Asterclick. If you don't like the direction that FreePBX is going then you can support Asterclick and put your two cents in the early development cycle.

 

[Lost Trunk]

Voila! Like magic. The world is no longer spinning out of control.

For now I'm going to hold my enthusiasm in check, since it looks like Micke Carlsson is already hedging a bit on what he'd written earlier:

Well, I might drop the whole thing and maybe do it another way, but time will tell, it will not be for a couple of weeks as I have a lot of other things to do at work, remember, I only do this on my (limited) spare time.
If something is released it will only work for 2.8.
​

None of that sounds like he's really committed to making that module happen, but it's that last sentence that has me most concerned. So there will be a module (maybe, if he gets around to it) but it will only work for 2.8? Sounds like at best it's just delaying the problem for one version. Of course, the optimistic side of me hopes that once that module is developed, they will realize that it's something people need and use, and will continue to update it.

My point really was that for experienced users, the current method isn't broke. And every time they try to dumb things down for new users, they wind up breaking functionality for people who know what they are doing (for example, when they try to do input validation they sometimes don't allow things that are perfectly permissible in Asterisk). I heard a rumor a while back that they were going to implement an "expert mode" that would disable all input validation, but don't know if that ever actually happened (if it didn't, it should have).

I do understand that newbies seem to stumble on the concept of routes and trunks (heck, I even recall seeing some of Ward's early posts, where he'd included trunk dial rules that didn't do a thing because they didn't add or remove any digits). But at the same time, I don't think it's right to deliberately impose burdens on experienced users that did not previously exist. The additional Bulk Routes module would be a very acceptable solution in this case; but I just wish that it were part of the main FreePBX distribution. Contributed modules sometimes get neglected - again, the whole Custom Contexts thing - but this would be a different situation in that Custom Contexts added functionality that never existed before, whereas Micke Carlsson's proposed module would restore functionality that's always been present up until version 2.8.

No one is "attacking" anyone here, at least that's not the intent. Unfortunately I think some users of FreePBX start to feel about that software the way some Apple fanboys feel about Apple products, in that even when Apple seriously f**ks up, they can't bring themselves to say anything negative about the company or its leaders.

In this case I think that the lead developer is making a huge mistake by not considering the impact of this change on experienced users, and when it's pointed out to him he acts like he just doesn't seem to care. But if Micke Carlsson does come out with his proposed module, and if that module is maintained and updated with future versions, this will be a non-issue, however nothing had been said about the possibility of an additional module when I first started this thread, so at that point I don't think I was being overly dramatic. I would certainly hate to do an upgrade to 2.8 and THEN discover that I had to enter every single route pattern manually - actually hate is probably much too mild a word for the emotion I'd feel at that point - you might wake up to read a headline like "Man smashes phone server with sledgehammer while invoking voodoo curses against unnamed software developers!" Well, maybe the emotion would not be quite THAT strong, but my mood would definitely not be mellow if that happened, and I just didn't want anyone else to go through that.

 

[Lost Trunk]

There is an alternative that is in the development stage: Asterclick. If you don't like the direction that FreePBX is going then you can support Asterclick and put your two cents in the early development cycle.

I read that entire thread and I am confused as to exactly what AsterClick is supposed to do. It didn't appear to me to be a full replacement for FreePBX, but then again there was never a real concise description of exactly what it is supposed to do, unless I missed it somehow.

I note that they are looking for volunteers but they don't say if the software can run alongside your existing software, or if you have to dedicate a server to the cause. And also, it looks like what they really are looking for at this point is people with coding skills, which leaves some of us out for now.

All that said, the thread was rather interesting (just wish I understood it better) and if that really is going to become a FreePBX replacement, I will be watching it with a lot of interest. I don't have a spare server to dedicate to testing purposes and I certainly don't want to mess up our existing PBX, and I'm not a programmer, otherwise I'd consider getting in on this at the early stage.

 

[blanchae]

Asterclick is quite interesting. Here's my simple interpretation:

There are two parts: Asterclick server which runs on the server. It is an interface to the Asterisk AMI and CLI. The purpose is to make it easy to access from a web-browser. It runs on the server and interprets the UI (user interface) pages.

The second part is the UI SDK (software development kit) which is a based on javascript and XML. This is what the user sees and what interacts with AsterClick.

The advantages of AsterClick is the way it processes data and keeps a persistent connection open for real-time updates.

Eventually, you could interface it to PiaF by adding another Kennosoft icon on the main menu page.

Unfortunately, it is in the early stages and has a way to go before be in the beta stage. There's still the database to be figured out, etc..

 

[dswartz]

I have to admit, that as much as I like philippe, this was a bad change, just from the point of view of (unless absolutely necessary) "do not break working code". this is a pretty major change to make with no warning. I must also say that if I were the OP, I would not have appreciated a "you are free to use older releases if you wish" (paraphrasing here.) Sorry, but that *does* strike me as a bit arrogant, albeit not as bad as Steve Jobs saying "just don't hold it that way then!" about the iphone4 reception issues.

 

[Baylink]

> As he states, you can use any version of FreePBX you want or none at all.

That's intellectual dishonesty of the highest order, I'm afraid.

No, you can't merely say "I'm never going to upgrade past 2.7" in the real world, and the reasons why should really be intuitively obvious.

If he's made the change I believe he's made, and has the opinion about people's reactions that he appears to, I predict a fork in fairly short order.

 

[Baylink]

Voila! Like magic. The world is no longer spinning out of control.

Well, no.

Philippe has *still* exhibited the attitude which he has, and it *still* is dismissive to power users.

So the fact that his back can be gone around on this one small point isn't really material to this discussion...

 

[tm1000]

Wow.

People make mistakes. It's a piece of software. The world is not going to burn. People just need to stop freaking out.

Yes I don't like the change either but I didn't come here stating that this is the worst thing ever when *IF* I would have *READ* the thread on FreePBX I would have *SEEN* that another FreePBX developer is going to create a bulk module...

Also there already is a Fork and it's not Asterclick....it's V3 which is not using any of the same code as v2. In fact the lead development team is completely different from v2 (Philippe is not that involved)

 

[tm1000]

......

I think even I could write this module for FreePBX....


http://pbxinaflash.com/community/threads/missing-bulk-dial-patterns-in-freepbx-2-8.7677/

 

[Baylink]

"People make mistakes".

Sure. But not pertinent here. PL is making an *architectural design decision* that a fairly large, vocal, and thoughtful collection of his users disagree with, for good reasons.

And he appears explicitly to not give a crap what they think or why.

That's a mistake alright, but not on the scale I believe you mean.

You can argue with me/us, if you like, on whether that's really what he thinks... but not on the results if in fact it *is*.

 

[drclue]

A little clarification on AsterClick

I read that entire thread and I am confused as to exactly what AsterClick is supposed to do. It didn't appear to me to be a full replacement for FreePBX, but then again there was never a real concise description of exactly what it is supposed to do, unless I missed it somehow.

There are two components to AsterClick.

One is the AsterClick server side component whose
core (but not only) purpose is to remap the
AMI interface into an XML interface that communicates
over HTML5 WebSockets. Being as the socket server
is written in PHP, it's pretty easy to add features to
and even a plug-in framework is being worked out.

The other AsterClick component is an easy to
use event driven client side Javascript class that
employs a WebSockets interface to send commands and
receive events without any need for polling. This library
allows web developers to easily create a wide variety
of Asterisk Applications.

I note that they are looking for volunteers but they don't say if the software can run alongside your existing software, or if you have to dedicate a server to the cause. And also, it looks like what they really are looking for at this point is people with coding skills, which leaves some of us out for now.

AsterClick can run along side any tool, and can even be used
within existing web based tools if one wants or even without
any web tools or without a web server if one desires.

AsterClick by itself is a tool for creating HTML5 web
applications that can manipulate the Asterisk AMI interface
and afford a defined XML communications protocol for
leveraging other functionality server side.

All that said, the thread was rather interesting (just wish I understood it better) and if that really is going to become a FreePBX replacement, I will be watching it with a lot of interest. I don't have a spare server to dedicate to testing purposes and I certainly don't want to mess up our existing PBX, and I'm not a programmer, otherwise I'd consider getting in on this at the early stage.

AsterClick itself is not a replacement for FreePBX,
but it could be used to make FreePBX better or
in FreePBX plugins , or yes to create a FreePBX type program
if someone was that ambitious. Replacing FreePBX is
not the primary goal for AsterClick.

The actual test case application for AsterClick is a HUD
replacement being made for one of the sponsors of the
open source AsterClick project.

Now while the AsterClick server and client side Javascript
class do work and are stable, they are still in development
and while copies are available to download, this is
a developers version, with Alpha, Beta and Production versions
some ways down the road.

If you want to tinker with creating AsterClick applications,
it should work great with any AMI command, but
one should not expose the interface to
any public accessible network as the security features
have yet to be implemented.

 

[drclue]

Actually

There's still the database to be figured out, etc..

Not so much figuring to do. The DBget and DBput functions for
asterisk's database system are already supported.

Additionally AsterClick's XML classes actually come straight
from my existing MySql tools and the AsterClick configuration
file already contains entries for integrating MySql among other
things with but a little bit of work to pull those items
into the AsterClick XML stream.

For now though , the focus is on those aspects required to
support the HUD application.

For the near term though , I think one will just have to wait
and see how the FreePBX folks resolve the duality of the
user base so as to serve both the geek and the novice.

It really is important to serve both , but at the same time
that can be a real challenge. For folks to get angry with the
developers is probably not going to help much, as
programming is as much art as science and both a matter of
evolution and iteration.