First, Avaya does run on common x86 servers as well as virtualized and has for many years. Linux and Xen if fact. Many implementations of freeswitch/asterisk fell apart as they simply don't scale to thousands of users. Yes, I know about Kamailio. Scale doesn't just mean have 50,000 endpoints register to the system and handle the busy hour call volume. It needs to be secure (Avaya encrypts H.323 and SIP very easily (Cisco actually doesn't do well here)). Which means you need to be able manage certs for each gateway, phone and application server. It's needs to be manageable with have full access control and it needs to provide the tools support the environment. A rich ecosystem of proven integrations help here too. Further, contact centre, when you have 1000 agents, or 25000, the cost of each of those agents doing nothing on a revenue contact centre because of a system crash is wildly expensive. Further, in the event of a major failure of a DC, many such customers also require triple redundancy with each system being able to failover with limited or no impact to the business. Federal governments require certifications that are very arduous and require substantial techical resources and capital. I could go on, but this mostly covers Enterprise and federal government. Now for small businesses it's a completely different issue. FreePBX is ridiculously difficult to configure. Very few people have the skills to understand how to develop and build a dial plan, how to integrate with SIP Trunks from a carrier, and even if they manage that the gateways that are from these other companies have a completely different interface and a completely different skillset. It's simply too hard especially when I can go and pay a small amount per month and either get some kind of a cloud service. Probably the best example for a commercial/open source offer is Digium. But they haven't gained substantial market share compared to the likes of Avaya IP office, Shoretel, Cisco, etc. However, if you do the math, these small systems are actually cheaper than cloud provider because you can run them with very low recurring expense. In my opinion, Avaya doesn't have a long road ahead. But it had nothing to do with their technology, nothing to do with their business practices, it was entirely the destruction of a business by a private equity firm who thought they could make a quick buck but completely screwed up.

This is true, but new competitors also haven't been around as long as Avaya. Before Asterisk and other software hit the market, there were four players - Avaya, Cisco, Mitel and Nortel. Now there are dozens of large players and hundreds of medium sized ones. These companies are gaining market share in the enterprise - just checkout the exhibitor list for enterprise connect 17.

In addition, there are random local dudes who can set up an office Asterisk system based on commodity hardware for like $2000, and whom you can pay another $200 every couple of years when something needs to be changed. An Avaya system can't get close to that.

I've torn out several Asterisk systems because that random local dude grew tired of dealing with Asterisk. Toshiba, which is a competitor of Avaya, does get close to that.

> Asterisk is completely scriptable.

I'm absolutely positive Asterisk is better than Toshiba in that regard. I've invented some absolutely hideous hacks to get Toshiba systems do complicated call flow. But for a minuscule office with less than 8 phones, say a chiropractor or hand made guitar manufacturer, Toshiba does the basics, does it well, and is much easier to install. The systems don't do much, but they do it well. The pbx can hang in a damp and dirty location for years and still work. Internet service here is too unreliable to have a hosting solution or sip trunks, unless you want to pay AT&T thousands a month. The wiring can be 50 year old 25 pair, and the phones work fine. Businesses don't want to switch to ip phones since they don't have the cabling infrastructure for it.

Asterisk is completely scriptable. Somehow I doubt that Toshiba system can do what we've needed done with respect to different stations ringing at different times of day, complicated dialing permissions, logging Charter Business outages, recording customer service interactions, etc. Each of those things required a small amount of labor, and no equipment purchases.

Of course our dude might leave the industry someday. Then we might have to hire a different dude.

It's true that open source can replace some PBX functionality.

There are though, operations that require scale or functionality that either doesn't exist, or isn't quite bulletproof in the open source world.

SS7 connectivity, geographically spread call centers, priority queuing based on multiple concurrent data sources, drag-drop complex IVR software, natural language parsing, natural language generation, contextual screen pops with hooks to caller information, and so forth.

Some of those are available in open source, yes. Not necessarily proven at scale, easy to integrate, etc.

Of course, Avaya filing for bankruptcy hints that there aren't many customers that need this sort of stuff anymore. Either the support function is shifting to chat/email, or they are using SaaS vs their own equipment.

Except SS7 you can software based solutions for all the rest costing much less then hardware based solutions from Avaya, Cisco and similar mammoth companies.

Often people think that these hardware based products offers better quality but I often seen the opposite.

Yes, updated to denote many of the customers that need this are moving to SaaS. SaaS that's probably not 100% based on open source though.

Or use a hosted QueueMetrics cluster with Asterisk and a little bit of scripting to connect the dots. It costs a small fraction and gets you most of the functionality.

> are very good free alternatives which can be managed also by non-technical people with zero telecom experience.

Are you kidding me?

I run my own PBX (would count as a SMB from a user/call count perspective). It doesn't run on any open source software, because all FOSS PBXes I looked at where ridiculously hard to configure. FreePBX is a great example there.

Anyhow, I mainly looked into AskoziaPBX and Vodia after that.

Askozia terminated their "diet" version and costs 250+ € (only one variant is offered).

Vodia has a free version (5 extensions, which is exactly enough), and has a more granular licensing scheme. I didn't fully setup Askozia testing, but it looked quite nice. Vodia was also reasonably easy to configure (about 1-2 days initial setup and after a couple weeks I got all the kinks fixed).

They have a pretty good name in contact centre technology.

I think the bigger shift is that people don't need desk phones anymore. The move to mobile is killing the need for fancy PBX gear in business.

I used to be the lead developer on an IVR/PBX SaaS solution. It hasn't been necessary to have dedicated hardware for years (ever since out-of-band DTMF has been a thing). The only reason that companies buy it is because they're getting conned into it.

We had a server through which the conversation (SIP, converted from landlines) went, and enabled the agent to interact with the call through either a web page, or hitting keys on their phone. Transfers, blind or not, holds, recordings, assistance, we could handle all of it. To dumb phones. Literally things that weren't able to display the time of day.

Mobile is doing diddly squat, frankly. Same as softphones. The voice quality over wi-fi is terrible, and what's the point of using a mobile device if it needs to be plugged in anyway. Softphones never took off because call centres couldn't reuse their headphone equipment, and having a headphone's "pick up" button work on a softphone was impossible, at scale. It also meant that actually interacting with the call was tricky (change focus of the window to find the numpad, then either click on digits, or type on the keyboard).

One major game changer, however, seems to be WebRTC. A lot of companies are investing big on it (see Dixa and COLT for example).

I really think this is a generational problem. Callcentres of old want the highly integrated hardware. Train your people once, and then they can become robots. New, modern places want the ability to switch providers. They want their agents to be close to the customer, and even closer to the tracking/reporting software stack, without going through 10 layers of proprietary integration.

WebRTC also means that you can simply talk to the person on your website, while you see what they see, and can walk them through whatever they need.

Well the contact centers I know are not going mobile but going to soft phones on the pc.

I don't think many businesses are quite ready for this, but a CTO can dream...

Some are. The problem is that what is WebRTC depends on Google's whims of the day - so your browsers update and all of a sudden your agents can't work anymore.

We used SipX in prod twice and both times were positive experiences. Solid system, easy deployment...superb.

SipXecs is really great. Nortel actually bought PingTel and called it SCS (terrible name). For a moment it looked like there was some hope that the industry was going to see the light, but after the acquisition Avaya promptly killed it off.

Fortunately, because it was open source, we now have SipXecs. Last I worked with it (5 years ago), I'd be comfortable implementing it for a client with less than 1500 users and no complex contact centre. I imagine it is much more capable now. I still recommend it.

I was excited to find a VOIP service for windows -- but Mizu is not free for Commercial use. Their basic price is competitive with a few other solutions at that scale, though.

All free and open source solutions only serve companies up to a couple of hundred employees.

This is a big misconception.

I have experience with Asterisk and the Mizu VoIP softswitch only. Asterisk is quite limited on capacity, but that mizu server can handle 10000 simultaneous calls on a cheap Xeon server (full B2B calls with billing and RTP proxy and all the PBX stuff such as IVR, voicemail, conference).

This capacity is enough for a whole smaller country and these kind of software solutions can be easily scaled.

Of course, you can buy the same capacity 100x more expensive from a hardware vendor.

You can cluster together multiple Asterisk boxes and scale up as much as you need.