How do you make your product better? Simple. Ask the people who use it. Ask your customers.

That’s the thesis of an article by Michael Fitzgerald in today’s NYT: How to Improve It? Ask Those Who Use It

There’s a companion article about Amazon’s Mechanical Turk, which relies on humans to provide the bits of artificial intelligence that software is not yet capable of delivering, and another one about how researchers at the University of Michigan have proven that human’s can’t really multitask efficiently.

Multitasking is going to slow you down, increasing the chances of mistakes,” said David E. Meyer, a cognitive scientist and director of the Brain, Cognition and Action Laboratory at the University of Michigan. “Disruptions and interruptions are a bad deal from the standpoint of our ability to process information.”

Here’s the connections I see

• It makes sense for companies to reach out to their clients, and get product design help from their clients.   That means talking directly with clients, and crucially, being willing to hear criticism.   It also means that you need to figure out a way to actually get your client involved.   In the software business, that means exposing elements of your product as a service and as an API.
• Mechanical Turk has proven that folding people into even the most technical software processes actually works.
• End users clearly want to participate because they are getting overwhelmed with multitasking requirements.

Where end users can string together the APIs from 3 or 4 web services into one unified composite application (or mashup), they have created an improved product.   The customer is doing the innovation.   And, they have also reduced complexity in their life.   Instead of having to multitask as they manually co-ordinate information across silos that might be as disparate as a spreadsheet, LinkedIn, Xing, SalesForce, Gmail and BigContacts, they enjoy the benefits of having one highly customized solution take care of it all for them.

Teqlo presented at Office 2.0 Under the Radar today. Jeff Nolan did the presentation. My job was to drive the slides and do the live demo.

So far, we’re the only live web based demo I have seen. Scary thing to do. The great thing is that Teqlo worked flawlessly. We also too the opportunity to demonstrate our new builder interface. I’ll be blogging more about that over the next couple of days.

People at UTR

• Talked to Jason Hoffman, CTO of Joyent last night at a VIP shin-dig. Joyent’s Accelerator product is doing an amazing job supporting the huge group that Twitter is undergoing.
• Stowe Boyd is here. But he isn’t talking to anyone. He’s become a full blown Twitter junkie. To me, Twitter is more proof that the day of the widget really is here. If you want to grow your company fast - develop a service and figure out how to let end users embedded that service in their web sites. Stowe, BTW, thinks that Sun should buy Joyent. Would be nice to see. Sun could learn a huge amount from Joyent.
• Jeb Boniakowski and Byron Binkley from Proto presented in the same session that Teqlo was in. Proto is the coolest desktop app I have seen in a long time. It’s a mash-up engine based on Excel. Very easy, very powerful. I hope they add a SaaS version.
• Paul Freet from Big Contacts is here. It was nice for me to be able to meet Paul in person. We’ve been working with them closely at Teqlo.
• Rob Hayes, from First Round Capital, asked Michael Schuster from System One about their platform. Rob said “Platform companies are really great unless they suck… and they suck if, when you ask them, what can you do with this, they answer anything. Platforms have taken off when people adopt the platform via a killer app. So what’s the killer app?”. Michael’s answer was rock solid. System One is a beautiful wiki with a spoky powerful search box. In SystemOne, as you write your Wiki post, the search box shows you all the related emails, web docs, files, in your environment. Michael’s answer was simple: “Our killer app is search”.
• To me, Michael was obviously correct. People within an enterprise will start to use System One because it helps them be more productive. In the Enterprise 2.0 space, adoption is critical. The judges at Under the Radar were asking about that, more than anything. Personally, I think that an increase in productivity is one of three key factors for adoption. Second is recognition. Finally, you have to give people that WOW moment that they have the power to publish
• I bumped into fellow Enterprise Irregular, Julia French, and talked with her about this issue of adoption for Enterprise 2.0 technology.   She added a forth requirement to the list of requirements for Enterprise 2.0 adoption - “People need obvious templates to use as a starting point”.

Big Contacts offers a dedicated “global address book”. It is an amazing tool. You can access the address book in the context of the Big Contacts CRM application. You can also access via an API. That means you can mash it into other applications.

Today, Ajaxian has an article about 37 signal’s new product: Highrise. Highrise is also being positioned as a global address book.

I’ll leave you to judge which company offers a more useful CRM application - I guess it depends on your needs. As you are making your decision, the Big Contacts populated demo is certainly worth checking out.

In both cases, while the CRM application is cool, the broader concept is much more interesting.

The Great Refactoring

One way to think about this is a spreading of Service Orientated Architecture Service Oriented Architecture; charging from behind the firewall out into the open Internet. The delivery mechanism is REST. And all the old complexities of data migration and data synchronization are being solved by letting end users pick one service provider to give them key global services. An end user could say, for example, I want:

• Big Contacts to deliver my universal address book
• SXIP to be both my prefered authentication protocol and access control vendor
• Cogenz to be my bookmarking service
• Gmail to be my mail client
• Facebook to be my social network
• DabbleDb to be my database in the sky

If software is going to be provided as a service, it make sense to refactor some services out, and get rid of the synchronization problem.

What else is needed? Here are some ideas:

• File storage, complete with versioning and access control
• Global shared To Do lists
• Mapping services
• Printing Services
• Approval and Process Services
• Invoicing Services
• Payment Services

Many of these already exist. But now, with the advent of mashup platforms, such as my company, Teqlo, it is easier to understand how these services will be consumed by end users.

History is filled with the serendipitous confluence of major trends. These new, highly focused web services, combined with a Web 2.0 inspired desire by end users to take control of applications is about to meet up with platforms that let end users assemble their own applications. The result will change the way people work with the web. They will replace PCs with Personal Servers, assembled from refactored services and run on mashup platforms. This is going to be fun!

Great news. The Under the Radar folks are offering a discount for readers of this blog. Office 2.0 Under the Radar Discount.

I am personally really excited about this conference. There will be some great companies demoing. Here is the full list:

Approver | Blogtronix | Brainkeeper | Cogenz | ConceptShare | ConnectBeam | Diigo | EditGrid | Firestoker | InvisibleCRM | Koral | Longjump | Mashery | My Payment Network | Proto Software | Scrybe | Sitekreator | Slideaware | Smartsheet | Spresent | Stikkit | System One | Terapad | Teqlo | TimeSearch Inc. (Calgoo) | Tungle | Vyew | WorkLight | Wrike | Wufoo | Xcellery

As most readers of this blog know, I work for Teqlo. We are have been working hard these last few weeks to put together a series of really interesting new things to show you. If you like Teqlo now, I think you will be blown away by whats going to be shown at Under the Radar.

Does your CEO know what a blog or a wiki is? Does he or she care? Should they? Last week, Robin Fray Carey asked the Social Media Collective “Does social media need a Frank Luntz to take our words and translate them to the middle American middle person?” The answer, obviously, is no.

Big companies are slowly starting to adopt Enterprise 2.0 technology. But, selling technology isn’t the answer. And selling technology misses the point.

It is my belief that if you are trying to help people, you need to tell them what problems you are going to solve, and not necessarily how you are going to solve the problems. Too much of social media / Web X.0 is a discussion about technology - not enough is about basic human needs and business problems.

We all want to express ourselves, be listened too, respected for our ideas and receive rewards and recognition for our effort. Humans are social creates. We want to have friends. We need social contact. Faced with the mechanistic processes of production, confronted with the rigid order of creating goods and services, we naturally layer on top a blanket of social interactions to warm our days. Water cooler gossip, lunch with co-workers, meetings for the sake of recognition.

To get work done, we need help organizing and simplifying complexity. To compete in today’s global economy, we need help archiving the goal of constant innovation and constant improvement. To work well, we try to reuse. We need help avoiding the temptation to constantly reinvent the wheel.

The end results, the kinds of communication, the entertainment and personal fulfillment, the recognition and the efficiency: these are the things that drive the adoption of Social Media / Web 2.0, not technology.

In the 1980s, and early 1990s, Silicon Graphics had the coolest workstations around. They cost $20K and up. They produced amazing 3D graphics, and are still used for professional quality movie, TV and video editing.

The company was worth Billions at one point, but ended up filing for Chapter 11 bankruptcy protection last year. They are now out of bankruptcy, and have evolved to into a company focused on delivering “a complete range of high-performance server and storage solutions”.

SGI went from rock star to failure because it focused on workstations. The advent of the PC based graphics card totally disrupted their market. And, as high end video editing software such as the three Autodesk products mentioned above, became available on Linux machines, the market moved completely away from SGI. BTW, if you are interested in what kind of tools you would need to make the jump from a iMovie YouTube special to a real Hollywood blockbuster, check out these two pdfs:

visual_effects_and_compositing_product_information_brochure.pdf (pdf - 311 Kb)

Autodesk Workflow Solutions Brochure (pdf - 1372 Kb)

This didn’t have to happen to SGI. SGI had the skills to dominate the graphics card business, rather than ceding the market to NVIDIA. But SGI was focused on distributing through only one sales channel.

There is a lesson in this for Web 2.0 companies as we move into the era of the recombinatorial web; hedge your distribution channels. Provide your service as a destinatioin site, as an API and as embeddable widgets. If you are using today’s best approaches to systems architecture, adding the extra distribution channels is so cheap it amounts to a free option.

Look at YouTube. YouTube started as a company that made video widgets that you could embed within your MySpace site. The YouTube destination site was less important that the MySpace distribution channel. Of course… the balance has since changed.

In an age of hyper connectivity, with network effects just waiting to take hold and launch, how many distribution channels do you have?

There have been some interesting reactions to Ning’s new business plan. Last week, Ning switched from a hosted apps company to that provides an instant social network.

UPDATE: It looks like they have just switched the marketing - they still host a raft of cool services.

www.tribe.net had an instant social network service 5 years ago. The New York Times says that Cisco is about to buy Tribe.

The problem with Ning’s closed social networks is that your investment in your identity on the social network is not portable. For example, Filippo Corti has founded the a Ning social network called “World Wide Mac“. You can see Filippo’s profile here under PhilApple. The URL to the is http://macusers.ning.com/profile/philapple. macusers is the social network known as “World Wide Mac” and Filliop’s profile (the /profile/philapple part) is unique to that group.

You build up relationships and contacts at one site, and end up having to reproduce it all at the next site. Tribe.net had solved this problem, but quickly all the “tribes” started to look the same. The only way to solve that problem would be to let people customize each of the tribe pages much the same way people can customize their MySpace pages.

Or, put another way, basic Ning is just SocialNetworks are similar to mySpace page for a club. MySpace would only have to add a second type of page within their network to create serious competition for Ning. Instead of just “Myspace”, they could have “myspace” + “myclub”.

UPDATE: I should have done more research. Clearly… underneath the covers, there is much more to Ning that just the social networking capability. Check out I want to understand Ning’s architecture and how it works for details. Actually, as a sales distribution tool, the idea of packaging up those services within something that is easily understood - like a social network, is a really interesting idea.

There are many things I really enjoy about WordPress. Article preview is wonderful. The post editor is great. Inline comments is GREAT. However, there are some things I am stumbling over.

Right now, I am trying to work out why my my posts are being cut off within my feeds. I have configured the syndication to include full text, but it still seems to be delivering only partial posts.

I’ll be sure to post if I can work out what the problem was.

UPDATE:

It turns out that my old RSS feed, which was built within MT wasn’t right, and instead, WordPress works just. The issue has to do with what the feed template puts between

<description><![CDATA[  and ]]></description>

as compared with what it puts between

<content:encoded><![CDATA[ and ]]></content:encoded>

My old template within MT kind of cheated and put everything in the description tags.

My old MT template did the equivalent of this:

<description><![CDATA[<?php the_content() ?>]]></description>

WordPress currently does this:

<description><![CDATA[<?php the_excerpt_rss() ?>]]></description>

If you stop by this blog often, you will know that generally, I don’t write a commentary about what other bloggers are doing. I usually just try to describe what Web 2.0 / Enterprise 2.0 issues I am dealing with, and what ideas and insights fall out from there. A little while ago, I promised the folks at Dealmaker Media that I would participate in their Radar Relay. The Radar Relay is a series of articles designed to sum up what is going on in the world of Office 2.0 in the last week, and stimulate discussion leading up to the Under the Radar: Why Office 2.0 Matters conference.

I participated once already. See all the gratuitus links in How many ways can your service reach people?

But… that didn’t get me off the hook. Not even close. Apparently, a long winded article on the recombinatorl web just didn’t cut it. I was shown Richard MacManus’ post Best of Web Office This Week. Read/Write Web is one of my favorite sites out there, and Richard, of course, offered a great summary of the recent Office 2.0.

OK… they have me in a tough place here. Not is the Dealmaker team one of the best connected crews in the value, but Teqlo is going to be presenting at the conference. We do, btw, have some very cool new stuff to demo. It should be fun:

So… in an effort to get it right, here is the best of the Web Office news for the last few days:

• In a quick poll run on Read/Write web, most people didn’t think Google Apps Priemier represents a tipping point, but they thought one was coming soon.
• In this case, the crowd is wrong. Google Apps Priemier is a tipping point. But it isn’t because Google’s Docs and Spreadsheets rock. They don’t. At least, not yet. Google Spreadsheets are a great begining, but they will need to scale up in size, and add a range of functions before they can be used for serious business mash-ups, such as finding the best prices from a dump of competitive prices. I say that Google Apps Priemier represents a tipping point because of this pointed article by my fellow Enterprise Irregular, Jason Busch. Jason says simply: Vista, Office and Outlook 2007 are a Nightmare. Google Apps Priemer is a tipping point because it attaches a Service Level Agreement to Google at Your Domain version of Gmail. Will “real” companies use SaaS for mission critical stuff like email?…
• According to Larry Dignan, the SalesForce has just landed a 25,000 seat deal with Merrill Lynch. Anshu Sharma points out that there are probably some discounts to this deal. “Now, if Merrill Lynch actually were to pay this price [$500/seat/year], they would be down $150 Million a year. That would be more than 25% of Salesforce.com’s annual revenue.”
• On O’Reilly Radar, Brady Forrest writes about Google has released their usage stats for Google Gadgets. There are over 4,000 Google Gadgets. Many are viewed over a million times a day.

The recent news shows two simple trends.

1. Business software delivered over the web is starting to look more mainstream.
2. Widgets / Gadgets are getting big…. very big!