Why shut it now? David and I kept it going as a live demo and as a conversation and discussion point. However Twitter is closing in on the day when using Twitter oAuth is mandatory. It wasn’t available when Phweet was designed and created. I wish it was. Still today in it’s current form it isn’t worth upgrading the coding.

A Few Observations:

Friction: Twitter is not a “friction” free environment and continues to have many points of failure. One form of friction relates to escalating conversations on and around Tweets. There’s a constant follow me follow you issue, send me your email, what’s your phone number? Every escalation that adds real value currently takes place off or outside Twitter. Phweet was one of the first services to try and solve this. Allowing easy escalation to voice calls without further exchange of personal information.

Second type of friction is in the flow. Twitter’s fail whale happens. That can mess up whether or not messages are even delivered by Twitter. More importantly while the flow is massive the medium still remains fairly asynchronous. There are no guarantees that the respondent will get the message or a notification in real-time. More advanced users may (eg using a Twitter App on mobile). Thus “signaling” with Twitter needs careful thought.

Twitter’s Future: Twitter’s future remains bright. Their approach to developers still encouraging. Twitter hasn’t killed the innovation engine around it. The first law of developing for/with Twitter remains. It’s simply “love Twitter”. I still do. David does too, or we wouldn’t spend time developing and learning from other Twitter Apps. We’d like to apply that expertise in “more”, “broader”, strategic and relationship driven opportunities. We’re busy with both social and communications related projects currently. Contact one of us.

Twitter @anywhere and geo-location will in time make Phweet like services even more relevant. The ready escalation on location to richer dialogues whether with a restaurant or a hot date will drive the need to have conversations that stretch quickly beyond 140 characters. In fact that’s often the case with Twitter. The answer to a tweet needs more than 140 characters.

Twitter Firehose: The firehose (total access to tweets as they happen) changes many opportunities for how developers harness Twitter. Many simple social media monitoring programs are moving here. Although that’s commodity turf IMHO. The firehose has many advantages for “communications” particularly in the corporate sense and in terms of  call centers. We’ve now implemented “watchlist” solutions.

I couldn’t write this post without looking at a few previous posts. In fact the last one on the blog sums up many learnings captured with Phweet. I see it also keeps quiet and I’m secretly pleased to see we’ve really progressed in the last year. That’s really what Phweet was about, making progress, taking a step forward, accelerating the future. No we didn’t knock off Skype.

If I have a frustration it is simply that “great” no “outstanding” work never found a sponsor. Strategically, Phweet was the most revolutionary approach to changing the way we make calls in 100 years. “context before call”, “no numbers / your choice of profileID / public records streamed / personal exchanges etc.

There’s no way to predict what solutions might have been. However, here’s a few predictions of how we could still make a difference.

1. Context before Call – Twitter is making more and more inroads with Customer Service. All have a problem escalating and handling the resolution. Adding the Twitter stream to the Call Center is going to be a winner. A talk request for customer service rather than the queue.

2. Identity Services. Location based services must separate and enable us to use the ID of our choice. Whether that’s a directory service like Facebook, or Twitter or Linked In. Or perhaps something else. We require more sophisticated CallerID’s. We want more transparency with our companies. The integration with social media and call centers is now an imperative.

3. Phone Numbers: This is old school. It’s almost tragic watching GoogleVoice and Ribbit and recently Line2 say yet another number usually on your mobile is the solution. The world is context first and escalate if required. Using multiple lines is a failure case where there is no other management or separation solution (eg personal vs family).

4. Voice is a Commodity - The argument is voice is just part of the web of exchanges. It’s cheap and anyone can add it to any webpage or enable it almost anywhere. We’re not quite there yet.

Some things are Phweet! Ah sweet! At least that’s the play on words I was looking for. I thought a world full of PhweetTalk might make Twitter a more interesting place. So this is my review.

10 things I have learned:

We haven’t yet changed the world. People still prefer to make their calls off or outside Twitter and without Phweet. While we garnered a lot of early attention the reasons why are not that hard to see.

1. Most importantly Phweet works. It breaks new ground re callerID, context for a call, call in progress, and call records. The variations developed later including persistent Phweets have even more potential.

2. However, Twitter isn’t yet a mature or effective signaling system. There’s lots of confusion around @ messages, DM’s and the default settings for notifications and email. Many times we wished we had the money to enable reliable SMS and email notification services ourselves.

3. I still look at Phweet as a Clayton Christensen type innovation. It’s counter intuitive, disruptive and slightly broken in the beginning. But watch out!

4. We thought the “PhweetURL” in the tweet would be more viral than it was. At first it was. We had a dozen people in our “launch call”. However sharing and promoting URL’s on twitter isn’t as easy as it sounds or may look. Signing in to Twitter was also a barrier. Twitter really needs an oAuth identification service. Twitter should authorize and send Phweet the key. We also had too many public canceled sessions so the user didn’t get a reward.

5. Tweeters do talk. In fact Twebinars prove they also like conference calls. Yet our lack of an app integration made this hard to call in. We also didn’t put in a “billing system” in the Alpha which restricted the “convenience that can be made available. A few of us know. We ran PSTN accounts and the integrity and convenience of the system improved.

6. We failed to get one Twitter App developer on board. They could have integrated Phweet using the API so it was almost like having a Twitter buddy on speed dial although without the rude ringing. This was disappointing to us. I never wanted to be in the App business although (if I had the funds) we could develop two or three twitter related apps that would challenge the current leaders on the desktop and mobile. I believe we presented revenue models and options that should have been attractive. My guess is many different agendas.

7. We have built a really nifty personal call management system. You can take the call on any channel and Phweet is effectively channel agnostic. Still the feature was buried for first time users and without a payment plan we couldn’t fund the calls so the host always got a call back. That’s a problem we’ve now solved alas it awaits another iteration.

8. Many don’t see the potential for the anonymity / callerID that is and separation from the phone number. This is integral to effective location based services and managing your privacy.

9. Similarly, we’ve not yet executed an integrated package that puts the receiver in charge. Many still fail to see that traditional call escalation results in interruptions, voicemail and potential breaches of privacy. By contrast Phweet lets the receiver make a judgment based on context and their relationship whether or not to escalate to a call. We’re already doing this unofficially by texting first. Phweet must make that dumb SMS message smart.

10. The PhweetURL was always meant to become invisible. In time it will. It’s really an exchange contract that brokers and escalates access between two or more parties. It may be public or private. Without apps we could make the Phweetman the button we envisage or enable it to go to every web page. Again we know the next step.

There’s more of course. We’ve looked at statistics, we’ve looking as deeply as anyone into twitter user behavior. For now we didn’t get a million users quickly. That’s not to say I can’t see 10+ million in my future.

What I would do differently

I wouldn’t have launched a public alpha. There were huge benefits and yet I think we may have done better with building out business case privately. When that public alpha launched, the perception was we were too big really for angels and we needed to go for something bigger. Fact was we were perhaps somewhere in between. As a consequence we shot too high too fast when we had a million demands at once. So I would have spent the time sharing it with more of those I trust and know. No one told me this privately although a couple of influential bloggers who I’d have called my friends shunned mentions as if they had been slighted. It was never the objective. We did what we thought was best at the time and prepped it. I couldn’t see at the time how I could run an effective alpha. Part of our initial focus was to learn how a Phweet passed through the twittersphere. Many of my best friends weren’t even on Twitter despite my early advocacy. So this is really a catch22 reflection. Still it is a note for future iterations.

I spent hard earned money on help with the look and feel of the site. If we hadn’t gone to public alpha we could have saved this money and in the end it was more trouble than it was worth. It’s always an interesting trade-off. It’s also part of the set of skills that’s required to get a start-up like this off the ground. In reality I’m still much happier we spent some time on the “packaging” my real frustration is not having the resources to make improvements. Like anything you have to be prepared to throw it away. Still as they say. Nothing ventured nothing gained. Just two guys had real skin and labor developing Phweet.

In both my reflections on 2008 and in the above I wish we had started with a few more resources. I felt close at the time it just didn’t quite gel that way. I’m sure David feels the same way. Both of us are capable of driving more than just ourselves. I still look at it with pride and think if this is what we could do just think what a team of 10 could do. It’s not a question of delegating everything; its more about ensuring everyone is doing what they are best at. I found myself caught in trying to do everything. The simple fact is few startups or web solutions start with so few resources and even less that provide VoIP solutions. Most have a small army.

If it is not clear in the above I would  approach funding differently. I’m a little horrified that I’m still stuck on really moving forward because of a complete lack of funds. Having sold big dollar consulting projects the costs to have driven our Phweet team forward could have already paid out many times over in some large organization with more resources. Still that’s not the prize that a startup VC or Angel looks for. I’m convinced that the ROI can be huge.

The future for Phweet

That will be a separate post. It’s dependent on and requires additional support from my friends. I require encouragement. So does David. We have a good understanding of services that we can harness in the next iteration. Our plans for Phweet won’t be beholden to “minutes” which is a death spiral for most VoIP companies.

What would thrill me the most and best kick start my new year would be your suggestions; public or private. So if it was yours… what would you do next? If you are as passionate as I am and want to press the innovation boundaries then join us, advise us, fund us etc. Certainly, now would be a good time to come forward.

• We want Multiple Identities: Our directory listings are increasingly fragmented, outdated and dispersed. Where once a white pages or yellow pages listing was enough we now have profiles on social networks, corporate profiles, personal websites and more. Each serves a different purpose, each a different share and a different way of revealing ourselves. No one wants one identity for all communications anymore!

• We use Multiple Channels: Communications channels are often not synchronized or updated. It’s a guessing game where you will find me. This results in lost opportunities and failures. The fact is you don’t know my number anymore anyways. You click on a name in your address book or an IM handle and we talk. The end point doesn’t matter. What matters is that the communication channel is a good connection and that it costs nothing or almost nothing. More importantly why should you have to judge or guess where to send the message or make the call.

• We want privacy without Interruptions: Sharing your mobile number may result in unwanted interruptions. Routing all calls to your mobile is the future. The problem is interruptions. If I give you my mobile number and you share it there is no way for me to ‘expire’ that connection and I may suffer further breaches of privacy. As a result, we are often highly protective of our mobile numbers. When it gets really bad people get a new mobile number. That’s not a solution.

• Desire control over access: Callers dictate when the calls happen without context and often without identification. It’s even worse as I wrote yesterday when it is an unknown number calling. We send those to voice mail and then may later have to listen to it or clear it. Let’s face it SMS is more and more popular for setting up calls and the “Available”, “Away” etc. that we have seen on IM accounts has become almost irrelevant.

• Broadcast and escalate conversations: It is hard to spontaneously escalate calls to conferences particularly if people are on different networks. This is one of the things that has always impressed me most about Skype and moving Skype multichats to conference calls with three to five people. We’ve enabled this with Phweet without number exchanges in multiple formats. You can even broadcast what you are talking about in real time. The future where you control the bridge and determine the services on the bridge is here. It’s a key differentiator. There’s more controls and power we will be offering to users.

• Communications on-the-move: Location based services are limited by access to numbers and concerns about sharing proximity info. The future is not going to be limited by the cost of the call or the need to know the communications channel. It should be about whether or not I want to allow access to me and under what circumstances and context I am willing to share. I feel I will share different profiles / directory listings and broadcast them to different destinations. The status update is key and so are my public profiles. At the end of the day it is my directory listing that will put you in touch with me. My listings on the move may be many and varied. They may or may not relate to each other. That’s fine by me.

I beleive Phweet addresses communications in a fragmenting world. We still have a way to go. Let me know which of these items are important to you. Which one’s you think really make a difference. If you were Phweet how would you use these attributes to move forward? Look forward to your comments.

Have you ever..

• needed a talking place where people can come and go? A virtual watercooler or VoiceIRC.
• wanted to have an hangout or virtual cafe to visit for talking and chatting with friends or family?
• worked from home and just wished you could drop in to a conversation?
• been part of a group of experts or help desk that answer customer inquiries?
• wanted to run a open focus group around a discussion item
• had to manage disaster management and needed a quick point of contact?
• invited people to talk about a topic?
• run a 24hour talkathon?

Now you can set up your very own Phweet Coffee Talk Cafe in less than a minute. Then invite your friends and colleagues and watch the conversation grow. Each time someone joins “visits” you will get a direct notification from @phweet with the name of who joined and and you can jump into session with them. At first you will want to do this everytime. But later when others are jumping in and you are busy you may just want to skip it. From time to time you may want to promote your URL and bring new people into the conversation.

We need your feedback. Tell us how you want to use it. Help us make it better.

So what happens in practice?
Someone joins the URL Eg http://phweet.com/45bs and everyone that is subscribed get’s a direct message from @phweet that  @twittername has joined http://phweet.com with @names. If you follow the link you can join in or just skip it if you are busy. Right now everyone that joins is auto subscribed. You become a member of this group. Here’s how Sheryl describes this as hyperconnected.

What’s important? The host doesn’t have to be in on the session. Anyone can visit the URL at anytime. Hang out for awhile and if someone else turns up dial-in or take the call on your laptop. Anyone in the Phweet Cafe can invite new users. The host retains the control to “kick” anyone that is disruptive.

This means your friends can meet each other here. At minimum they have you as a common connection even if you aren’t there.

How Create your own Phweet Coffee Talk Cafe:

• Log in to Phweet
• Follow the Link “Start a Coffee Talk Cafe”
• Write your message which sums up what you are going to talk about. This is what I wrote for a http://phweet.com/45bs: “is a conversation cafe to talk about Phweet. Subscribe and visit when others do.”

For further instructions follow these page by page details here

Notes on creating a Phweet Cafe

1. Leave the box un-checked if you want total control over who can join
the Phweet session. (This means they cannot join until you as the
“host” approve or decline their participation. This is just like a
traditional 1 to 1, or 1 to many Phweet.) This may be more appropriate if you want to limit participation to an organization.

2 Tick the box RECOMMENDED and now anyone can join and instantly be
talking with anyone currently in the session. As the host if you really don’t like them you can “kick” them from the call. This makes it much easier for you and your friends to grow this Phweet Cafe as the “host” is no longer an approval bottleneck.

Then Twitter broadcasts the launch of your Phweet. This Phweet session is public. You can now promote this persistent URL for as long as you want. When people visit you will be notified by a direct message from @phweet. It’s best if you set your Twitter notifications (in Twitter Settings) to send you an SMS everytime you get a direct message. That way you can dial in quickly from where ever you are. (Currently US numbers only)

What I’ve learned so far:

Invite a number of people to join you. Remember that people may visit your Phweet Cafe at anytime of day. So potentially the more the merrier. That way you no longer have to jump in all the time.

Everyone is subscribed when they join for the first time. If the updates from the group become overbearing then all you have to do is unsubscribe. Unsubscribing will also remove this persistent Phweet from your list of pending Phweets. If you have it bookmarked you can always return later as long as the host has not ended it.

Note currently you can create as many persistent Phweets as you like. However, that’s likely to become a premium service.

How is this different from other conferencing systems.
You no longer have to remember any number to dial in. All anyone has to know is the PhweetURL. You can use it at both set times or spontaneously at anytime. People can come and go.

How might you use this?
The more visible and more you promote the URL the more likely you are to stimulate the conversation.

Undirected Phweets can be started in “auto-accept” mode where any visitor to the URL is “auto-approved” and immediately joins the Phweet. The host can still “kick” them out. Otherwise, normal request/approve controls are in place, where the host must approve (once) those users that want to join, the same way it works for normal Phweets. Hosts can also create the Phweet URL with normal host acceptance controls, and then “invite” specific users to join, sort of like a “broadcast”, which essentially pre-approves those specific users, but still requires anyone else to ask to join and to be approved by the host.

Perhaps we should call these kinds of a Phweets “Coffee Talk” from the SNL skit where the Linda Richmond character would introduce a topic, such as “The Progressive Era was neither progressive nor an era” and then say “Discuss.”

This new feature is not formally launched yet, but is available for testing.  To create an “Undirected Phweet” sign in to Phweet.com and use this link: http://phweet.com/phweetup.cgi

Try it and give us some feedback about what you think this feature needs.

Dial-in to Phweet and use your phone or mobile Phweet Detail Step by Step with instructions.
Dial-in to Phweet Slideshow (turn ‘info on”) This is the same show below with instructions visible.

Currently dial-in is only available for US numbers. If you have a US number as your Skype CallerID that will work too.

Note if you put your dial-in number in more than one Phweet account it will stop working for it no longer knows which account you are calling in from.

We have many more solutions and choices on the way. My first SIP share uses Gizmo5 a SIP based client that embraces open telephony that has been around almost as long as Skype. The way we use it gives it new meaning. You won’t need to add any buddies or new contacts. You will just need an account setup. Then you will find the audio quality improves and you gain new opportunities to direct calls to where you want for just a few dollars.

Gizmo5 and Phweet Detail Step by Step with instructions.
Gizmo5 and Phweet Slideshow (turn ‘info on”) This is the same show below with instructions visible.

When you follow the complete set of instructions and if you enable this new Phweet client on auto answer you will find that accepting Phweets becomes “seamless”. “Accept Invitiation” and you are talking. However do be aware that if you are hosting the call and have Gizmo5 set this way when the Recipient accepts the call you may just hear them calling out “hey… are you there?” for there won’t be a ring. Disable the auto answer if this bothers you. Key point is when hosting a Phweet you don’t want to miss that call when they accept your request to talk.

For Advanced Users some things to note:

If you have access to any SIP number you can direct calls to that device. For example in your company you may actually have a SIP phone on your desktop and you probably don’t even know it. All you need is the SIP address for that phone and you can insert it in Phweet. Now you can direct calls that handset just by selecting the channel. Ask the IT guys.

I’ve shown this example using Gizmo5. Using it on your desktop costs nothing. Forwarding calls to many other SIP numbers is also free. The only time you pay is when the call has to be connected to a PSTN number (that’s your mobile or landline). One day maybe even your mobile provider will give you a SIP number. Why should you pay for redirecting a VoIP call to your mobile? It’s your handset don’t you just want the convenience? Note in the USA you are running minutes either way! SIP would be cheaper for the mobile operator!

Gizmo5 is also installed on many high end Nokia n-series handsets with WiFi. Similarly Truphone offers a similar service. This makes it easier to integrate using Phweet on your mobile. If you already use Truphone you can insert your SIP number into Phweet the same way we updated channels with Phweet. Unfortunately many companies that use SIP / VoIP related protocols remain closed. Examples of CLOSED TELEPHONEY are Skype, Rebtel, Jajah, Mobivox. We hope they will soon provide their users with SIP connection options so their users can embrace the numberless calling opportunity provided by Phweet. At Phweet we’d suggest you consider moving your “CallOUT” minutes to providers that embrace this philosopy. It won’t cost you anymore to use their services to make your long distance calls.

When you have added other Channel options to Phweet you will find that you can even “leave” a Phweet and return on anther channel. This is something I often do. Taking the call on my mobile and then switching to another phone or application.

Apart from embracing OPEN TELEPHONY, one of the key differences Phweet embraces is the idea of zero switching costs. We don’t want you to download new software to get you talking. We don’t want you to adopt or have to use a headset on a laptop if you hate talking that way. I know I like walking around outside when I’m talking. You like talking on your home phone or office phone? If you have a unique callerID you can simply add that to the channel page and “dial-in” to the Phweet session. I know a number of people that wait for their browser to “whistle” and then change channels to “dial-in” and simply have the number on speed-dial! This only works in the USA for now. Other countries will be coming. See also the iPhone with Phweet instructions.

When you set up a Phweet you create your own private telephone exchange that you then advertise (share) to those you want to talk to by sending them a Twitter message. This message and the PhweetURL it contains replaces the dial-tone. Just keep thinking about it. You don’t need a number to call anyone anymore. You just need to see their profile or a webpage with a little Phweet button on it. Need to see that? Check this out. Adding Phweet to your Twitter page.

iPhone and Phweet Detail Step by Step with instructions.
iPhone and Phweet Slideshow (turn ‘info on”)This is the same show below with instructions visible.

Note: I’ve tested the same basic approach on Nokia n-Series phones. Like the iPhone example (we’re stil an alpha!) the screens have not been optimized. On an N95 if you zoom the browser to 50% Phweet presents ok.

Some things to note.
When you send a Phweet it is a talk request. Depending on how the recipient manages their twitter notifications they may get it instantly (SMS) or there may be a delay (twitter apps) or even longer if it is a direct message and they check their email infrequently.

To be fast and responsive set up SMS notifications to your phone (Twitter limits this to US, Canada, & India for now). You can do this in settings / devices / SMS / on. Concurrently change your settings / notifications / “all @ replies”. Then even people you are not following can send you a message and potentially a call request. This should be the default.

All Phweets set up by “d” direct message are private. Only the host may invite additional people into the Phweet. Friends and Followers may ast the host to join public “@” phweets. That makes them more fun!

Dial-in works great when you receive Phweet however take care using dial-in when you send Phweets to others. If you miss the SMS beep notifying you of your friends acceptance they may be left on the bridge listening to music. Today I recommend solving this by downloading Gizmo5 installing on your PC, going to the profile and adding the SIP number to your channels. Then buying some minutes and setting Gizmo5 to call forward to your mobile. Now when you are not on your laptop or fail to answer the call there your mobile will ring. No need to ever miss that important moment when your Phweet request has been accepted. Plus now you have the freedom to take Phweet calls almost anywhere.

RIght now you must have a US “CallerID’ for this to work. We hope to add other countries in the future. If you have a US SkypeIn number you can also set that as your CallerID and then use Skype to dial-in to your Phweet session.

Download the podcast or play in pop-up.

The discussions included a broadranging technical disucssion that has been further fullowed up. Here’s two illustrations – Dan York tells us why you can use Phweet when Skype is blocked and TringMe’s CEO Yusuf Motivala shares more about Tringme’s traffic platform.

Squawk Box is Iotum‘s Alec Saunders’ daily soapbox where discussions range from a variety of issues around VOIP, video, communications and even wine!