Conversationware

Not proper Unified Messaging. But I’ve promoted Unified Messaging for so long - 12 years and counting, and I have to say that the market did recently seem relatively cooler towards the subject until Microsoft made their Unified Commmunications announcements.

So, Microsoft’s approach is a good thing, but still, from being a hot subject a few years ago, there is more shrugging of shoulders than there used to be.

I detect that attitudes stem from just how much email is being received by key people - and it is those very people who normally most benefit from productivity and increased levels of service that Unified Messaging can deliver.

I can see the thought process…”I get way too many emails, and you want to stuff my mailbox full of voicemail?” It’s a false argument, as everyone has to handle voicemail anyway, so their real point is that currently, voicemail is instantly recognisable and can therefore be prioritised.

The worst result of bad unified messaging is a ‘lumping’ of important voice messages into a sea of email dross

Therefore - to mitigate this issue for potential users, make sure your new unified messaging system distinguishes voicemail from email properly. This can be done in a few ways;

• When looking at your Outlook/Notes inbox, the voicemail should show a phone icon instead of a straightforward email envelope
• There should be a “This is a Voicemail” type statement in the subject field, and the number of the person who left you the message.
• Voicemail should have notifications by phone, and especially SMS (if no mobile email available) as an overlay to their email box or desk phone, to include the sender’s info! (helping users prioritise)
• When users dial in by phone, the UM system should separate voice, fax and email into separate queues that the user can choose to access. Most people will choose the voicemail queue to action first.

Better still, why not explore how to get rid of email overload, and start reducing inappropriate traffic through email. This is how I got interested in new media solutions

I take this quote from Business Week Article from a Dresdner Kleinwort Web 2.0 technology pioneer,

Thill says using the wiki, along with blogs and instant messaging, has cut down his e-mail use by at least 75%, and his colleagues have reported similar results. Now, he only needs to go a single wiki page to view all the key metrics for 80 Web sites monitored by his department. Whereas sifting and sorting that data from 80 sources could otherwise take weeks, he says, through the wiki, each user only needs “about 30 seconds” to plug in his or her data and make it immediately available to the whole team, all of whom are promptly notified by instant message or e-mail.

All apart from the email notification bit - amazing how few people use newsreaders - this would hugely ease life for the usersphere. I’m not sure 75% is achievable for all of us, but I seriously think 50% of my emails are inappropriate to the medium…sent and received.

The Realtime Community blogged about Jaduka, which would seem to be an IVR with a web interface for callbacks.

One key point for me is that calling someone in this instance driven by a web interface to the call generator - which means you don’t need to know someone’s number to contact them. Looking forward, from a unified communications point of view some sort of availability before I clicked to call, plus the ability to converse with other media would be great too.

But, it strikes me that if the IVR were actually the PSTN switch, then this would be an enabler for web telephony. By that I mean not the transport of calls, but the call control. If you had a web interface for the actual call - you could introduce files, conference someone else in, go on a web tour….all generating minutes and service revenue for the Operator.

Shouldn’t this be an innovation already brought to market? Oh wait, the pstn network (or overblown IVR platform) as an intelligent and developery platform looks like it’s moving steps closer with Grand Central, as blogged on Oreilly Emerging Telephony blog.

Also, incidentally, for presence, who better to know the status of an telephony endpoint than the telephony network itself. For instance, if I click to call a colleague on a mobile phone, then the network has the status of that mobile phone that it could publish to previously trusted parties, in or outside of the enterprise.

A friend of mine pointed out that I was using the VOIP and IPT acronyms interchangeably and I shouldn’t have been.

At the time, I was putting the point of view that I couldn’t understand why many companies were installing VOIP/IPT.

He explained that there was nothing wrong with VOIP, as it reduces the cost of interconnecting telephony and applications (like my oft cited Unified Communications and Unified Messaging for example).

However, where IPT was concerned, the case for purchase is very much less clear.

On a new site, cabling could be reduced through having a new IPT using single cables instead of double the amount. Multiple sites might save on call costs, but the payback could be some time in coming.

But if you already have the site and cabling, then the impact and cost was potentially quite large, in terms of new LAN/WAN equipment, New IPT, New cabling infrastucture, and lots of new fancy handsets (Gartner says don’t buy them) for perhaps not a lot of incremental benefit.

I stand to be corrected by the way, as I’m not an IPT expert, but then neither are many other non-IPT companies. Which is one of the points, why install something that’s more complicated and that existing staff don’t yet understand.

Microsoft are saying a similar thing in this interview, although, it seems to say “you won’t need a telephone system” which would be an overly drastic step in my view.

Rather, I would say, like Gartner

With the money saved on IP phones, Mr. Hafner said companies should purchase UC applications. This option enables the user to improve productivity integrating communications applications with services such as instant messaging, unified messaging, presence, personal agent, conferencing and mobility services to create a converged desktop with the voice communications on the phone. These applications are far more productive than the screen on an IP phone and are about the same cost

I’m biased, but a lot of the customers I talk to can’t understand why they need to change phone systems either.

As I wrote about here, I’m now pretty much convinced that browser based applications are just better for users, and the help-desk, than using client installed software.

However, there has always been a certain disconnect with the web apps, through not being notified of new things (or worse, getting email notification overload).  I’m sure I’ve read about Sharepoint suffering from this. And from my own example, there was no way I could rely on myself to check web based email on a regular basis as days could fly by.

So therefore, browser based applications need a client!! Don’t they?

My initial inspiration for this came mainly because of how useful Google Mail and Talk were with desktop notifications, even though the email was browser accessible.

For me, notification has made a crucial difference to personal adoption of web based personal email, as I can forget to use the tools, but as long as someone else maintains the effort, the notification drags me along!

One other aspect of client based browser software has convinced me that this Enterprise 2.0 adoption discussion could benefit from something so simple as desktop notification, as follows:

My company installed and deployed a browser based unified communications application over the last few months, and we slowly came to the realisation that although management were actively encouraging adoption (which shouldn’t be underestimated), there was a problem with people not logging in.

The network effect of Metcalfe’s law means that the less people that are logging on, the less useful the system will be.

In our case, initially we heard comments like, ”I logged on, but wasn’t anyone else on there….”

So, three things happened to put this right

1. Management continued to encourage people into at least logging on

2. One of our very clever technical types wrote a installer with a wrapper which automatically started the system every time the user’s desktop machine switched on plus,  

Included in the software wrapper was a facility to prevent the user accidentally closing the window, asking “Are you sure you want to close this window?” which meant people were more likely to STAY attentive. If minimised, the app ran in the system tray.

- this wrapper was the major difference in my opinion!

3. We trained more heavily and over a more closely sustained period than before to encourage people to cross the ‘first click’ barrier.

The difference in Adoption is amazing, and as simple as that. These things helped us reach a critical mass, and the true value became apparent to all users. Fantastically it’s now creating user passion and some great feedback!

Of course, it helps that the application is of great benefit once the users get there, but at least the application has the chance to succeed or fail.

One more useful browser adoption feature tip is the multiple home pages facility within Internet Explorer 7. Now every time I physically open my browser, it means that my Google Reader and Google Docs open alongside, resulting in daily attention to other updates!

OK, time to stop now.

Browser based applications are better than client software. Particularly for Unified Communications.

That’s a feeling I’ve had that’s been growing, particularly after reading a lot about this new web version lately, is it Web 2.0, or should that be Enterprise 2.0?

Looking back, when experimenting with web based applications in the late 1990’s, there was a lot missing.

Desktop based software had a certain familiarity, and it didn’t cost as much as using the internet (I didn’t have broadband in the 90’s). And then, there was the instant reaction you got from the desktop as compared to the click and wait of the browser.

Times have changed

Browser and web technology have enabled a real ‘desktop feel’ to browser accessible software and Broadband means that there is practically no cost to the amount of time and data used. But, my feeling of companies’ attitudes is they still feel that browser based applications are not as good, and I have to conclude that this stance must be based on those early experiences.

I suspect anyone newer to computers will already be convinced that web is so much better than desktop, for a number of reasons, and perhaps its just that us old timers have learnt to live with the pain.

For example, whenever an old desktop application needs upgrading, there is a big upheaval, potentially hazardous to the health of helpdesk staff.

• What are the consequences if the upgrade stops half way through?
• What versions of operating systems are going to be supported on the new versions?
• What happens if people using different versions of the software have to interract?
• What other interoperability issues exist with completely unknown software from thousands of other software developers?
• What security implications are there for this desktop based software, is someone going to write a client based exploit?
• What backups are required for the desktop software held information?
• What about synchronisation?

Do all these questions place a huge demand on developers? Will it slow development down? You bet.

Client Server technology is great, but compare the upgrade process - with web based systems you can tell the users the system will be offline on Sunday night, upgrade the server. It’s done.

Browser based software benefits include

• Easier backups
• Easier failover and disaster recovery support
• Easier access from multiple machines
• Easier support
• Easier integration of the system facilities into other processes
• Easier collaboration
• Easier to outsource (hosted services anyone?)

A colleague at a friendly company mentioned recently that they had abandoned the project to rollout CTI telephony integrated to their Microsoft LCS client.

The problem was that all versions had to be in synch, or the rollout just didn’t work. The payback wasn’t considered worth the extra support calls to the helpdesk that the rollout was generating, so it was canned. And these guys supply UC technology! 

In other words, using several client based software solutions, integrated from multiple suppliers - as the word unified implies - was just asking for trouble.

I had a look around, and found this 2004 post which starts to crystalise the argument for me. I read “When wizards stay up late” in the past, and similar books which made the point that Microsoft empire was built on schmoozing developers. No more it seems.

Greg over at Cistera Networks blogged about the Merrill Lynch CTI rollout discussed at Voicecon recently and quoted Alok Kapoor who was in charge of the deployment.

It seems to back up what I was saying about VOIP not being required for Unified Communications.

“Deploying an IP-telephony infrastructure underneath the CTI application, he said, didn’t really offer much benefit in and of itself. “Your best-case scenario is par with what they had already,” i.e., the TDM PBX, Kapoor said. “There’s a limited opportunity for upside, and a great opportunity for downside.”

I guess you can prove anything with the internet if you try hard enough, but this was 25,000 advisors over 250 offices, which should be just about the sweetest spot for VOIP that you could find.

And yet, there is little upside….

Just to clarify, I’ve seen Unified Messaging on hundreds of TDM systems now and in the past, and there is absolutely no difference in it when deployed on IP. The connections are different, but that’s about it.

There seems to be a theme around the Unified Communications world - that UC is somehow married to Voice Over IP - summed up by this sort of website. They say that you can only get UC by buying the latest shiny VOIP telephone system. 

Rubbish. 

There is absolutely no reason why the transport of a voice call, whether traditional voice TDM or VOIP, has to have anything to do with the functionality or unification of communications types or the routing of inbound calls.

In fact, I hold the view that it is much easier to unify voice, IM, conferencing and desktop data sharing, if VOIP isn’t a pre-requisite.

For example, I am mobile for a lot of my working life and delivering incoming office based voice calls out through firewalls to a mobile device via IP is just not going to happen for quite some time.

What we users really want is for our contacts to be able to call one number and get hold of us…isn’t it?

Inbound routing is easier if my number called by contacts (my office number) is abstracted from the device I receive the call on. This almost demands that the office number is divorced from any particular device, especially VOIP devices.

So, the receiving device may be VOIP when I’m in the office (because the cabling was cheaper), but when I am at home, the call should route via GSM to my mobile handset device, as this is not only far cheaper than installing a VOIP phone at home and then maintaining Quality of Service with special broadband, but I then also don’t then have to think about differing mobile and a home location numbers, as inbound calls would always route to the mobile. I Still have desktop based call control, even if the call is routed via PSTN!

I should add that although I would happily pay for my contacts to get hold of me on the mobile, without having to know where I am, with the vodafone mobile plan we’re on it is actually free to send calls from the office to my mobile, thus removing the most common of objections. But the likelihood is that mobile and PSTN calls are going to get cheaper and cheaper for most companies anyway from now on.

Technology must enable the use of desk phone, mobile, hotel phone or home phone to receive calls as the need arises - and without asking the caller to know varying numbers - THERE MUST ONLY BE ONE!

So, a caller can ring my DDI, or speak my name at the speech directory, and they get through, regardless of my device and it’s own number.

There are several outstanding features of this method, as I register my location, my colleagues and callers don’t need to know my number, they just click to call, or they say my name and route to whichever number I’ve entered. For me, this is the only way to get rid of proliferation of numbers.

A very real additional benefit to this is that as I try to make outbound contact, my UC system calls me first at my configured location, and then dials through - my phone shows as busy, even if I’m on a device not normally owned or controlled by my own organisation telephone system.

Think about it!!!!  - am I ever going to get my mobile company to share presence and status information with my Instant Message client? No way. However, if I use my UC system to make contact via my entered number and then get it to call my contact, then it is simple, the fact is that if I am on a mobile call, I will still show as busy to my colleagues!

How do I enter my location?  - easily by web page, voice recognition, and via whichever tool I use for my call control (mute, hold, transfer, conference, record).

Moreover if my system can apply some logic - like if I’m not logged in anywhere, then try my mobile, so much the better.There are some subtle consequences of routing all your incoming and outbound calls through a single UC system at the office location, instead of through your mobile, amongst which are;

1. The ability to screen calls, or to route callers to to colleagues group or operator

2. The ablity to record the calls details in reporting (direct mobile to mobile calls are lost)

3. The ability to record calls content

4. The ability to show telephony status to colleagues 

A VOIP system can be deployed without proper UC, but then I am going to have a number of different addresses for my colleagues and contacts to call me on and communications will be separate and I still may not be able to do the above. I shouldn’t wonder that for many, this may throw into question why they are buying the shiny system in the first place. 

Why are things are better when communication is unified?

I like to explain that I don’t want to have to choose ahead of time whether I want to text, speak, share documents, present a point of view or share my desktop or video when having a conversation with you. 

If we have to re-convene to continue a part of our conversation, wherever it is taking us, then a lot of the context of the moment may be lost. If I have to book a web conferencing room and pay for it, lets face it, 90% of the time this potentially valuable conversation opportunity may never happen.

So, instead, let me instigate any mode of conversation with a single tool, and then let me add and extend with different modes during the session. 

So, the first requirement is for an interface that will allow us not only to dial, but crucially to then control a phone call - mute, hold, conference someone else in, record and so forth. This is pretty new for most of us, but it only the first step.

The interface must then also include the ability to start with, or later revert to, an Instant Message instead - and because IM includes presence, then it handily allows me to know if you’re there or not, or on the phone, before initiating. (This is a big subject of course).

Could I also launch a desktop share, a powerpoint presentation or review a word document? Could I do this even if the person doesn’t have the right software loaded, if they’re not a part of my own organisation?

Can I use this with whatever telephone I happen to be using, office phone, mobile, home phone.

Can I then just add people at will, instantly?

The answer is yes, and companies who buy VOIP systems will be working this out soon.

I love this application, it’s called ‘My Teamwork’

Disclosure, my company is starting to promote this software.