Wednesday, 25 March 2009

Company blogging

As part of The Team's new website - which has a long way to go, believe me! - we'll be surfacing opinion via a company blog. I've cobbled together some guidelines which are shared here in case they're of value to anyone else.

The important thing here is that companies shouldn't look to create guidelines that are too restrictive or lengthy - they won't get read, and will probably put people off. Rather the best thing to do (IMHO) is create a light framework and a system of support. That's what I've tried to do here.

I reckon this is also a lot better than nothing. It can be intimidating writing in a public forum for the first time, and not knowing whether there are any rules of the games can be off putting by itself.

Obviously this is written from an agency perspective but most of the guidelines apply in any corporate context.

Tuesday, 24 March 2009

Ada Lovelace Day

Today is Ada Lovelace Day, which is a day when we celebrate women in tech. I'd personally like to see more women in tech, particularly in programming, predominantly because I believe in the power of diversity in general. I believe that the best way to solve problems is to bring as many different viewpoints and backgrounds into the equation. It combats what I see as the natural human tendency to seek out those who agree with you, to seek the comfort of the echo-chamber, which is often a path to mediocrity. And diversity tends towards a more interesting workplace.

So, I have a conundrum. There are two women in particular that I find incredibly inspiring. Which do I pick? No problem - I pick both!

Kathy Sierra is one of the most inspirational speakers I've ever seen, and ran a fantastic blog until 2007 (still worth looking back through it - all very relevant). She worked for Sun - obviously a male dominated organisation - for several years, and took on the tremendous challenge of making technology accessible to, and understandable by, everyone. She participated in one of this year's most interesting panels at SXSW, about presentation techniques. It hasn't been uploaded to the SXSW youtube channel yet, but it's well worth watching out for. She doesn't have her blog anymore (long story), but you can follow her on Twitter.

The other woman up there with Kathy is danah boyd (her choice not to capitalise). She's done a lot of research about how teenagers use the web, and represents the voice of reason and clarity in a world where people make far too many sweeping generalisations. She also participated in one of the best panels at SXSW ("Everything I Needed to Know About the Web I Learned from Feminism"). Her recent dissertation (Taken Out of Context: American Teen Sociality in Networked Publics) is required reading if you'd like to engage with this important audience. Her blog is awesome, and she's also on twitter.

Yay, ladies!

Monday, 23 March 2009

Mrs Sew&Sew

I've been a project manager for a loooooong time now, almost 13 years if you include my marine insurance days. And while it's been a lot of fun - no, really! - I've often wondered what it might be like to be a creative on a web project.

Well, now I know! We were recently approached by a client - the Imperial War Museum - to apply some social media foo to encourage families to visit the museum over the Easter Holidays. They're currently running an exhibition on Children in Wartime, so we thought it might be fun to have someone tweeting and blogging from that era, to tie into the real life exhibition.

They've got an astonishing range of material in their archive, including some from a campaign from during WWII called 'Make Do And Mend'. So we've managed to get our hands on loads of videos, posters, knitting patterns, photos and some pretty amazing books. The campaign will roll out over the next four weeks, up to the end of the holidays. Should be fun!

Here are some links to her Twitter page, her Blog and the museum's Youtube channel (srsly there are some fab videos on there, well worth checking out).

As for who's actually doing the writing...well, we've got to retain some sense of mystery, haven't we? ;-)

Open source show and tell

I'd just like to announce an event that's being jointly hosted by my previous employers, Osmosoft, and my current company, The Team. It's an open source show and tell, and I'm thrilled to confirm several high calibre presenters will be in attendance. So if you're interested in what's happening in projects such as Drupal, Ubuntu, TiddlyWiki and oh so much more, come along!

You can sign up on the Upcoming page if you like; attendance is free, and there will be beers afterwards (maybe even during!).

Wednesday, 4 March 2009

VRM: Collectively immense

After last week's VRM meeting, I've been pondering the merit of my Personal RFP model. Again, the notion of a broker (especially a powerful one, such as Google) was unpopular, but no-one suggested a better way of getting to a point where a network of individuals can support the model.

So I've been mulling over a variation on the theme. Could we create a social network which replaces the broker?

I think we could. I've looked at the roles the broker would've played and tried to figure out if the abstract notion of a social network could play these roles instead;

- access to large volume of vendors
- filtering RFPs on the way TO vendors
- filtering RFPs on the way FROM vendors back to potential customers
- aggregating recommendations and ratings

I believe that a social network could manage all of these roles using the same mechanisms as wikis (with their low cost of repair, and therefore low level of vandalism), combined with Amazon's model of recommendation (where recommendations planted by those with vested interests are lost in the collective voice of genuine feedback). And a crowd sourced system of classification.

Neither wikis nor Amazon's recommendation system are perfect. But they have all proven to be Good Enough.

Let's look at those roles, starting with accessing large numbers of vendors. We all have our favorite vendors, but without an immediate incentive I can't see customers alone getting enough vendor data into a system to get sufficient critical mass. But perhaps we could reach that point by scraping data from somewhere? Could be a variety of sources, starting with the Open Directory (which is somewhat out of date, but hey you've got to start somewhere). This, combined with a way of allowing vendors to edit their own details (with a system of take down and lock down in place, a'la Wikipedia), could work.

Filtering on the way in and out of the system could be handled algorythmically. They could be based on location (given radius), keyword (e.g. carpenter) and rating (minimum that the customer is interested in). Perhaps some level of semantic analysis if we're feeling clever. My suggestion here is that we start with something basic and open source-it, so that we can constantly improve results based on what we've learned, as well as drawing in talented people.

Actually I envision that this whole project would need to be open sourced. No security-by-obscurity here!

For the recommendation engine, I've already highlighted Amazon's model as being the most effective. And I think customers would be far more likely to contribute back their reviews and ratings to a system which is neutral than to one which is commercial.

Which leads us to a business model. Hey, it's 2009, we don't need no stinkin' business model! Only (half) joking. But seriously, I believe a platform like this could attract advertisers. I'd prefer to avoid charging customers and vendors for using the system, certainly at the beginning. Customers will never be charged of course, but I think vendors would need to see the value in the system before accepting a small charge towards upkeep of the project. I'd prefer to keep investors out of it, if possible.

So the final stage in this very rough analysis of using a social network as a broker is to rationalise it against the evolving VRM principles. Does it tick all the boxes?
  1. Relationships are voluntary.
  2. Customers are born free and independent of vendors.
  3. Customers control their own data. They can share data selectively and control the terms of its use.
  4. Customers are points of integration and origination for their own data.
  5. Customers can assert their own terms of engagement and service.
  6. Customers are free to express their demands and intentions outside any company’s control.
Boxes 1, 2, 5 and 6 are ticked.

Re: 3, if a customer shares their data with a system, and allows that system to share the data with multiple vendors based on a publicly available algorythm, does this count as still being in "control"? What if the terms of this use are defined by the network of customers, rather than just by the individual? Could we evolve the system further down the line so the customer can define their own algorythym? I like to think so.

Point 4 is a little tricky. Customers would obviously be the point of origin, but given that the filtering and forwarding is happening on a distant server, is the customer still a "point of integration"? My feeling is that this obeys the spirit of the principle, if not the letter. Again, it would be an open system and subject to the same scrutiny as the best open source project - which I think would give the customer the transparency they need to decide whether to trust the system in the first place.

Lots to mull over.