This is how we'll catch the looters

I've got an idea that'd help us find these rotten rioting shits.

We've got thousands of pictures of them from CCTV, camera phones, news footage. All we need to do is connect those images to Facebook. It'd need a fairly chunky Facebook Graph app, and some face matching software.

Here's how it'd work:

  • We have a library of photos (criminals)
  • I connect my Facebook account to the app
  • The app scans my friend list, grabs a profile image for each, and tries to match it against the library of bad slags
  • If it doesn't find any in my immediate network, it starts to crawl my friend's public walls, searching for shitty crims
  • As soon as it gets a match it flags it as a potential convict 
  • It uses the matched profiles as seed points to continue crawling

You can assume that each robbing dickbag is associated to other robbing dickbags through a couple of degrees of separation on Facebook, so a couple of matches would unlock thousands of others.

Here's what it'd take:

  • Lots of crunching power
  • Something that'd match faces (maybe Picasa?)

Who's in?

Timesheets

People don't like filling in timesheets.

Some just find it distruptive and irritating. The more paranoid among us see it as some kind of invasion of privacy.

I know we all think we're different. We think our minds are unique, and the way we work isn't suited to time tracking. That's great, and I'm sure you're right, but at the end of each month your boss needs to know where your time went. It's very unlikely that he's trying to jam a stick in your spokes and distrupt your day.

Here are a few of the problems I've experienced:

  • Traditionally, filling in a timesheet is a retrospective activity. You enter your time after a task is complete, or at the end of each day/week/month. This is disruptive. When I finish a task, I want to start the next thing straight away.
  • Despite my best intentions, I never stick to my plan of filling in my timesheet as I go along. I get too busy, and I let it slip. It snowballs, and 3 hours later, I worked on 4 different things, and I'm not exactly sure how much time went to each project.
  • Too often, filling in the timesheet at the end of each day relies on memory. On a good day that's fine, on a bad day, that's not so fine. I'm conditioned to rely on external tools to help me remember things. To-do lists, phone numbers, email addresses — these aren't things I find it easy to remember these days.

So, what's the solution? Well, I don't know exactly, but I've got a couple of ideas. Over the next few weeks I'm going to try out a few things:

  • Log what I'm doing, not what I just did — The idea is to record tasks when I start them, not when they're finished. This should remove the need for me to remember anything.
  • Log everything I do — I'll log everything I do during a working day, no matter how routine or trivial. This is as much an experiment to see where my time goes each day, as it is anything else.
  • Categorise — I'll categorise every activity, probably through a basic tagging system so that I can report on my progress throughout the day/week. This will show me how much of my time can be billed, how much is routine activity, and how much is taken up carrying out personal tasks.

I haven't found a tool to do this that doesn't irritate me in some way, so I spent Sunday knocking together a basic Rails app that should do the trick.

It's nothing fancy, but it gives me a couple of things that should help me stick to the plan. Basically, I need to be able to pick the project I'm working on, enter the task, and tag it, all with my keyboard (my mouse slows me down), in a couple of seconds. The idea is to remove as much friction and interruption as possible and if it works, make the process part of my daily routine.

I guess I could just use a timer, but I want some automation. I want to have a report at the end of the day that shows me:

  • Time spent on each project
  • Time spent on each task within each project
  • How much of my time was spent on billable tasks
  • The effect of distractions during the day

Filling in timesheets is never going to be fun (despite the claims of several time tracking apps), but hopefully I'll be able to find a way that makes it easier for me to stay on top of things.

 

Brooklyn Bound

Some of you know already, some of you don't, some of you won't care. Anyway, I have some news.

UPDATE: My visa was approved, so it's official. I'm going :-)

I officially parted company with Fudge on the 25th March 2010, after ~5 years' service. There are lots of reasons, most of which I'll be keeping to myself. To over-simplify, my general outlook has changed, and Fudge doesn't give me what I need anymore.

I joined Fudge as it's sole developer about 5 years ago (becoming a partner a year later). Along with everyone there (see list below), I helped build Fudge and grow it into something pretty cool. We had a rediculous amount of fun, and learnt loads from each other. I've been lucky enough to work with some of the smartest people I've ever met.

Obviously leaving is a sad thing, and being one of the owners, not the simplest of matters, but it's the right time for me to move on.

So...

I'm selling my house, most of my posessions (with the exclusion of my bikes) and moving to New York.

Nyc_application_03

I'll be working with Area 17 in NYC and Paris. I've known most of the people there for a few years now. They're a lovely bunch.

I went out to NYC in January and spent some time with everyone at A17 to see if I'd fit in. They couldn't have made me feel more welcome. They have an awesome office dog called Simon. He pretty much sealed the deal.

If all goes to plan, I'll be leaving the country at the beginning of October this year. In the meantime, I'll be working for A17, from the UK (and Paris for a while in the summer).

I'll be working on a few *side projects* too, so look out for that.

Thanks to...

It wouldn't be right if I didn't thank the people I've worked with at Fudge. Over the past few years, they've done some brilliant work, and regularly made me laugh so hard, I've been in physical pain. In alpabetical order (in an attempt to avoid any unintended offence):

So, good luck to everyone above. Some are still with Fudge, others have moved on.

And some special mentions go to...

Mike Byrne—For being perpetually upbeat, no matter what.
Ben Taylor—For making the last year loads of fun. We were quite a team.
Christian Brown & Gavin Bates—For tolerating my "opinions" on design all these years.
Matt Hall & Jim Neath—Incredible developers who've saved my life on many occasions.
Dave and Andy—For making all of this as painless as possible. 

Extra thanks to George, Andrew and Kemp at Area 17 for all their help and advice. See you soon.

And, thanks to Shivers for putting up with *a lot* over the past few months.

Subfolio

Subfolio

In their own words, "Subfolio provides an elegant, practical and customizable web interface to your file system". They're not wrong.

Subfolio is a product managed by my good friends over at Area 17. It was originally developed as a quick and easy way for them to display design work and as a centralised resource for their client files. It worked so well for them, that they decided to release it as an open source project.

What does it do?

Subfolio acts as a nice front end for your file system. It displays your directory structure and files, and automatically thumbnails images, giving you a really clean way to browse through your stuff. In addition to the basic functionality, it also provides 'enhancers' which give you slightly more complex interface components like featured folders, text blocks etc...

Installation is easy. Just upload it to the root of your domain and update the config file.

There's loads of info on their site, so have a look there for a full feature list.

Who's using it?

Lots of people. A few of Area 17's clients like it so much, they're using it internally (AOL amongst others). We've been using it at Fudge for ages, and it's been a similar story with our clients.

Get it

It's still in a private release, but instructions on how to get a copy are here, so just ask nicely — http://www.subfolio.com/04_download.txt

Thanks NYC

I just got back from a trip to New York with Shiv. I need to thank our friends at Area 17 for showing us around Brooklyn and generally looking after us.

  • Andrew/Kemp — Thanks for dinner, without which we wouldn't have sat opposite Alexa Chung for an evening (to the astonishment of Shiv).
  • Kemp — Thanks for letting us use your spare room all week.
  • Debbie — Thanks for scoring us free tickets to the Museum of Natural History. The planatarium exhibition is excellent (and narrated by Whoopie Goldburg).
  • Everyone at A17 — Thanks for the drinks, and making us feel so welcome.
  • Simon the Dog — The A17 office dog is awesome (and surprisingly well dressed).
  • The 4 year old 'ninja' boy — Thanks for chasing me down Franklin St. with a stick. Shiv was amused.

Amazon (almost) ruined my Mum's Christmas

That's a minor overstatement, but I'm sticking with it.

We all know Amazon had trouble shipping things this year, mainly due to the weather. They're big and important, but they don't control clouds, so I'll forgive them that.

I ordered my mum a lovely Kindle, a present she was sure to appreciate. It arrived late (3.30pm on Christmas Eve), despite me paying for next-day delivery. Because it arrived so late, I didn't have time to wrap it, so I planned to do that at my parents house shortly before the big hand over.

So, I opened the City Link delivery bag and this is what I found. Unbelievable:

The box was covered in mud. It was dripping wet, and it didn't smell good. Luckily, it hadn't damaged the Kindle.

Obviously, I didn't give it to my mum in the box. That would have suggested some kind of festive dirty protest.

It seems that some conscientious person dropped the box in a puddle, river or toilet and left it there for a while until it soaked up some extra Christmas cheer. They then decided that "it'll be alright" and popped it in the delivery bag.

Anyway, Amazon, I want a refund on my delivery charge, a grovelling apology, and some kind of voucher.

Thanks in advance and lots of festive love,
Phil Jeffs (Order #: 203-2023868-2649162)

Chrome for a Cause was brilliant

You might have missed it, but hopefully not. Between December 15th and 19th this year Google gave money to charity whenever participating users opened a new tab in Chrome. They found a way for people to donate by doing what they do every day – opening browser windows. Very clever.

All you needed to do was install the Chrome for a Cause extension and it started counting. At the end of each day, you decided which charity to donate your total to. The more tabs you opened, the more you could donate. The nice thing was, you could choose a direct action, not just a charity name. It's nice to know exactly where your donation will be spent:

Google_chrome_for_a_cause_extension

The cynic in me would call it a "we're not evil, we're still nice" marketing ploy, but who cares. It raised $1,000,000 in 5 days.

Here are the participating charities, and here's a bit more info about the campaign. I expect it'll come back soon, so install the extension and it'll notify you when you can participate again.