Don't get me wrong, I love Skype. It's saved me a fortune, and is way more convenient than the alternatives.

But sometimes it lets me down. Yesterday, I was expecting an important SkypeIn call at 4pm. Never came. I was online, for sure. Finished work after 5pm.

This morning when I log on at 3am (hello jetlag), I get the voicemail from that person -- timed at 4.15pm yesterday. So it never rang, and I didn't get notification of the voicemail. Annoying.

I've also had problems with conference call quality at times. SkypeOut isn't as good as BT's VoIP voice quality. SkypeIn is generally pretty good though.

What this is telling me is that the field is wide open for competition in the small-medium business space. And a telco brand could be just as good as an Internet one. I don't mind paying for business-class quality -- I just need something that works at a reasonable price. There are additional feature like web conferencing (synchronised Powerpoint, desktop sharing) that need to be in there too.

PS - Downloaded Sightspeed this morning. Looks like a nice product, but they make the users jump through far too many hoops to get going.

You can miss Martin just as often at the Telepocalypse.

The first new idea I've seen in a long time on the stale network neutrality debate came from following a comment to this post (whose conclusion doesn't match the quotation -- if big enterprises want to waste shareholder capital on me, please bring it on!). Anyhow, here's the deal:

The false dichotomy of net neutrality, and the Tariff Rebate Passthrough solution

...and the nutshell version from the previous comment:

The idea is Tariff Rebate Passthrough -- i.e., the ISP can charge by byte for QOS (but only by byte) and the information service provider (Google) can rebate the costs directly to the consumer (but only to the consumer). This works because it meets the need to pay for differentiated QOS, without letting the telecom companies' control over that payment become actual control over content. I.e., all the good parts of net neutrality are preserved, but there's no need to give something costly away for free.

Walking across the river this afternoon in central London. Thinking of home, missing the wife and kids. Have been on the road too much in the last month or two, and too much travel in prospect too.

So I want to call Dr Mrs G to pass on my heart's desires etc. Problem is, the younger daughter tends to sleep at various slightly random times. Too often I've called just as the little madam was falling asleep on mama's shoulder, and ruined the whole afternoon for my wife who then has a grumpy, sleepy baby who will whine all afternoon.

So what I want to be able to do is make the phone flash gently, or solicit an outbound call. No ringing!

You can't do this on the PSTN. Sure, you could have handsets that have custom rings per caller based on the caller ID. But I want control per call over the ring at the other end, and it doesn't support that feature. ... continue reading.....

Some more random thoughts on how our minds have been poisoned by 100+ years of Bell (or was it Meuccian?) telephony.

The signalling system in the analogue era was very simple. I want to talk, your phone rings, you pick up. We then enter a manual signalling exchange. "Hello, this is Mary." Confirms I got through to the right number and callee. "Hi Mary, this is Kevin calling. Is this a good time for a chat about next week's meeting?". Identity, availability.

Now imagine a system where we could press the green "call" button on our mobiles either once or twice. Pressing once would just request a call with the person. They would then have a queue of "people who want to talk to you", and those present/online would appear in that queue in time order. I could even, if calling from a PC or other rich UI, suggest times to call back. My phone would have a special ring for returned calls.

Alternatively, press the green button twice and make a normal interruptive "ring now!" call. ... continue reading.....

I could do a long critique of every softphone out there, and there's plenty to pick apart. I thought I'd just select one little detail to show why the portal IM clients and Skype remain top dog: they just deliver what the user wants, no hassles.

Every time I log in to Windows I get this:

Go away! Shoo! Don't irritate me with unnecessary login screens. Fade into the background. I don't want to think about you until you're needed. (If the wireless Internet connection comes up too slowly, it also tends to crash.)

I suppose I should also point out some of the other usability issues. As Amazon long-ago discovered, the way you present the login/new user screen makes a big difference. If it's confusing (high cognitive load) people bail out, probably (rightfully) assuming the rest of the experience inside will be equally bad.

Gizmo fluffs this with a strange radio button layout. In the user's mind, registering is a different process from logging in, even if the information requested is identical. The drop-down text entry box is the wrong cue for creating an account name, because it implies a selection of existing data. (Yahoo is superb at managing this process in a crowded namespace.) Gizmo operates from the perspective of the programmer, not the user. Contrast with Skype: ... continue reading.....

"WAP is Crap!"

Well, in fact it was quite good given the technology constraints it had to work within. As an implementation of the wired Web on mobile devices, it was well thought through, surprisingly effectively implemented, and funded to the gunnels.

The difficulty was that it was in general a solution to a problem the users didn't have. The power of the wired Web is the hyperlink and browsing of information. Users spend a lot of time "transaction hunting", where you decide where to put your money and attention. The wired Web is about bubbling up of important, interesting and useful information. This doesn't match the use case of the wireless Web, which is about quick hits with sites where you already have a relationship.

All this is well documented. So it's rather sad that the industry is about to go through the same harrowing learning process all over again with mobile instant messaging.

Once more, there's a well-established and successful model from the wired Internet. "Presence" as it is usually constituted grew up from the always-off world of dial-up Internet. Online rendezvous was hard, presence solved that problem. For the first time, you could have multiple conversations on the go at once. Distance didn't matter, a novelty for those separated by countries and continents. Parents and partners were excluded from this private chat world.

Mobile IM is also the solution to a crisis the user doesn't have. The buddy list reflects a closed world that doesn't match the openness of the actual tools the users prefer, namely SMS and voice. We already have a universal identifier system, the phone number. Users already manage multi-threaded conversations using SMS. The idea of the "chat window" doesn't make sense on mobile. The interruption model doesn't match, either. A new IM whilst you're browsing the web means a flashing taskbar icon and minor context change from one app to another. Mobile interruptions mean suspending real life. That's why you ask the sender to stump up a few cents to demonstrate the value of the interruption.

Web 1.0: Lots of websites which offer personalised portals with domain names like "my.foo-inc.com".

Web 2.0: Shouldn't we see lots more sites with domains like "our.corporate-inc.com"? Their absence speaks volumes.

Exercise for the reader: is it possible to transition an institution from control to co-creation of value, or can you only build such edifices on greenfield sites? Or to be more blunt and specific, does the journey from Telco 1.0 to Telco 2.0 on average require the capital and goodwill to be split apart and re-cycled via the bankruptcy courts or distressed asset sales?

Martin thinks aloud at Telepocalypse.

Too busy getting ready for Telco 2.0 event to write anything too substantive, so here's a quick thought.

I've tried a whole lot of Skype phones. I've seen a whole load more. I've not liked a single one. As I don't like to slag off products from small companies, I simply write nothing.

I don't use any of them. Why is this?

They put the traditional keypad at the front of the experience. But I don't use Skype to make PSTN calls much. I use it for what it does best: contacting a small circle of friends and colleagues. That means putting the buddy list up front, a multi-modal UI for navigating voicemail, and enabling features llike easy set-up of conference calls by showing multiple buddies. Make PSTN calling the exception, not the rule.

Naturally, a big screen is a given; some way of navigating my long buddy list quickly (hint: not clickety-click up/down buttons); and the display of presence and mood message of each person. Probably also the ability to list just a few entries from a single group as the default too. Wireless, too. After all, we're trying to get away from the grounding of a PC headset and expand into the wider domestic/office context.

My kids need a device that has two buttons: London grandparents; Vilnius grandparents. They don't use telephony the same way, so need their own device built around their needs.

The only exceptions to the "they're all crap" statement are the conference call phones which are really just microphone/speaker extensions: do one thing well, and the user is happy. But bad imitations of PSTN phones aren't it.

Push Martin's buttons at Telepocalypse.net.

I was being interviewed for a podcast last night. As always, the purpose of the "stupid network" is to enable crazy new things, not connectivity arbitrage. The setup was that I'm in my hotel room using the woefully over-contended in-room Internet access. The caller could only record calls made using his landline phone, so he called me on my SkypeIn number.

The audio experience was OK, but about that of a typical cellular call. Not ideal for a podcast.

This does, however, provide great fodder for a "Voice 2.0"-ish story. Normally, VoIP uses the UDP protocol for media transmission. If the packet doesn't get there within 300ms, or whatever, forget it. No point in asking for reliability and re-transmission of lost data. The TCP protocol is used for signalling and other purposes where a reliable, in-sequence connection is required. ... continue reading.....

Consider this.

I'm a cheapskate, and I'm with Tesco Mobile's prepaid plan. I hardly use my mobile except as a camera and for brief voice notes. Under $10/month expenditure.

Tesco's MVNO only offer Web (ports 80/443 HTTP/HTTPS) access on their GPRS gateway. This is a means of the host operator (in this case, O2) to segment the market and avoid competition from the MVNO for its premium customers.

Now, if you have neutrality rules, you get two unwanted effects:

• Tesco may have to close down their GPRS service, because it discriminates against service providers who happen not to use HTTP as their only protocol. The customer loses if the only type of Internet access allowed is 100% unfiltered.
• Tesco can never expand the service to, for example, allow POP email access whilst disallowing VoIP by inducing jitter and using deep packet inspection. The customer loses again -- in this case the marginal one who may even be willing to pay a little more.