1. Shoot images
2. Get home and import images to laptop
3. Check images have imported
4. Put a copy of images elsewhere
5. Wipe camera memory card (As needed)
6. Select and work on images (Which can take anything from a night to months depending on the project)
7. Once happy with images, import them to lightroom catalogue which lives on the drobo
8. Periodically delete images from laptop
9. Occasionally run out an up to date copy of the lightroom catalogue to an external drive

Here’s a very crude diagram representing the steps. There’s probably a better way of showing this, and there’s certainly a better looking way, but it’s late and I am tried. Areas in green represent times when the images are ‘safely’ being stored somewhere. When these green zones end it is representative of either deletion or hardware failure, with the time the media may exist on that medium quantified above:

There are a couple of problems with this workflow:

There’s very little automation
People are forgetful/lazy and I am no exception. I often forgot or skipped step 4 (A backup step), particularly for small projects. This is particularly dangerous when you see what step 5 consists of (Wiping the memory card). Once those images are gone from the memory card they only exist on my laptop. If it were to get lost, stolen, or broken they would be gone for good.

There’s not enough redundancy
As stated above, if I skip or forget step 4, once the card is wiped, the next chance my photos have a chance of being in more than one place is once they are copied to my drobo, and even that duplication will only last for as long as I leave them on my laptop. The startling lack of redundancy in the entire workflow is a disaster waiting to happen.

There’s not enough urgency
Inbetween shooting the images and ‘securing’ the images on the RAID array many months can pass. The backing up process (not that it is much of a backing up process) comes at the end of the chain when it should start as soon as possible. My data shouldn’t be exposed to risk for as long as it is.

Part of the reason for this is not just to do with me being lazy, but with my reluctance to engage with my image repository, with good reason… My Lightroom catalogue is gargantuan. There are numerous threads and posts on the internet discussing how big a lightroom catalogue can get, with it not being uncommon to have them reach the hundreds of thousands of images. Mine is around 30,000 images and it is feeling sluggish, particularly when accumulating metadata for me to search on. A few months ago I tried to operate on a per project basis, generating a new catalogue for each project/client, but this fast became messy, especially when trying to pull together examples of my work across a range of projects. Barbarically, this 30,000 total includes dud images that are of no use or value.

With this workflow rethink I set out to address the following problems:
-The lack of redundancy and the high level of risk my photos are exposed to.
-The archiving/distribution of my images in Lightroom to improve accessibility.
-The quality of the images that are ultimately stored, reducing duds.

So with that in mind, here’s the rethought workflow:
1. Shoot images
2. If particularly far from home, copy images to iPad
3. Get home and import images to laptop
4. Check images have imported
5. Backup to Livedrive of laptop automatically starts
6. Backup to time machine drive automatically starts
7. Wipe camera memory card (As needed)
8. Wipe iPad (If applicable/as needed)
9. Select and work on images (Still anything from hours to months)
10. Once happy with images, import them to appropriate lightroom catalogue which lives on the drobo
11. Backup to Livedrive of drobo automatically starts
12. Periodically delete images from laptop

The difference between the two diagrams is very noticeable. There’s a lot more green, which means I’m going to sleep easier at night. The number of layers means there’s a lot more time between things going wrong in which the problem can be fixed. There’s also been a shift in where the backing up process starts. The new flow no longer waits until work on a project is complete to make the data safe, instead securing offsite both the raw files and the processed ones. Now, when my drobo fails (for real this time), I should have time to source a replacement and grab my backup from livedrive.

There are two additional pieces of hardware for this rethink, the iPad and the time machine drive. The iPad is particularly redundant (meant in the nicest possible way) as it only serves to duplicate data for a few hours, or at most, days.

The iPad camera connection kit costs £25 and let’s you plug an SD card straight in. For the D90 at least the iPad will store RAW files (but display/utilise JPGs for apps/email). Deleting the images is a bit more cumbersome, as irritatingly, there is no ‘delete all’ option in the native iOS UI. Instead you have to mark the images and then delete them. I have found that tapping and holding an image, then dragging a finger over the rest in a zig zag pattern is a relatively quick way of selecting the images in bulk. You can clear a few hundred images in a couple of minutes. Another way, is to plug the iPad in to the computer and delete the images that show up in the mounted portion of the drive. I found I had to plug my iPad in to a windows machine I don’t sync with to get the drive to mount (I got that tip off Mikemurphy).

The time machine drive is actually the old 500GB which my drobo had filled up. After a quick format and the purchase of an external drive enclosure it was ready to go.

The real key to this new workflow isn’t just the fact that backup is brought to the front, it’s the fact it’s automated. Aside from failing to plug in the time machine drive, I can’t forget to back up anymore. And even if I do, providing I have Internet connectivity, livedrive will carry the data offsite for me. The only problem I can see here is extended trips abroad, with patchy Internet access and no time machine drive. In the past I have taken a laptop and small external drive with me, duplicating each days worth of photos across both and storing them in separate locations. This system has worked well, and I am considerably more religious about making those daily back ups when on the road.

In regard to the lightroom catalogues I’ve decided to have one catalogue per year. This means the the number of images in the catalogue will be finite, as dictated by January 1st. I’ve broken the old monster catalogue down and my new catalogues are as follows:

Aside from nicely illustrating my desire to capture an increasing number of images as time goes on, these catalogues are a lot easier to work with. Note, I don’t set out with the goal to beat last years total, it’s just sort of happening on it’s own. It’s looking like this year may be the start of my plateau. Sorting in to the above catalogues also served as effective spring cleaning measure, deleting nearly two thirds of my library, consisting of those images considered too similar/uninspirational/less than perfect. Things could further be improved by archiving every year on to an external drive, this is something a giant catalogue just wouldn’t allow.

So that’s my new process. It’s a lot better than the old one, and I’m lucky I had the scare of losing everything, without actually doing so. If you’re the kind of person who hoards data and memories (or work) on drives and machines, I’d urge you to think about how much you would lose if x, y AND z all broke tomorrow.

It all started in about 2008, when I was getting sick of having photos across several external drives, any of which could fail at any given moment. I decided to buy a drobo, which is a user friendly mirrored RAID array. With the drobo I could safely amalgamate all my existing external drives in to one unit, which would protect me against single drives failing. Brilliant!

There is however an obvious problem with this: When the drobo unit itself fails/is destroyed, everything is gone. Fully aware of this risk, I got my self an offsite/online backup solution called mozy. All the important files on my drobo would also be uploaded onto the mozy servers, and if the worst happened i would be able to download my files from the cloud. I figured I was set and started uploading my media to the mozy servers, knowing full well it would take a few weeks for the initial upload.

For a few months the system worked well. After the slow process of that initial commit (which took about 2 months!) the incremental backups made everything much quicker.

Then one day, for a reason I forget, I unplugged my drobo from my computer for a while. I continued to use the computer in this time, and when I plugged the drobo drive back in mozy started doing a fresh commit instead of an incremental backup. After some digging in their forum, and some emails to their support, it seemed the system wasn’t set up to deal with temporary volumes (like external hard drives). I dont know if this is still the case, but for me, at the time, it was a deal breaker. Very annoyed, I sent a strongly worded email, demanded compensation for my wasted time and cancelled my mozy subscription, vowing to look in to an alternative backup solution ‘soon’.

Years passed.

In those years the drobo soldiered on like a trooper. A drive failed and it didn’t bat an eyelid. I performed one or two routine upgrades to expand capacity and all the while I invested more and more data in to this single point of failure. Every time the thought of an online backup came in to my head it was always put on the to do list. It was a big job to compare them all, make sure they did what I wanted, test them and finally commit to one. I wanted to be able to dedicate an appropriate amount of time to such a decision! There was however never an ‘appropriate’ amount of time to look in to it. In the world of offsite backups, even a hastily made ‘wrong’ decision is better than no decision.

So here’s how it all went wrong: My drobo was running low on space, so I’d just bought a fresh TB HDD to swap with one of the old 500GB ones. Knowing that the drobo usually takes a weekend to rebuild/transfer parity data to a new drive I’d held off for a weekend I knew I wouldn’t use it, as it can sometimes get quite sluggish during a rebuild. The morning I was due to fly to Paris I opened the chassis, ejected the 500GB drive and positioned the 1TB. Each bay in the drobo has a spring loaded clasp which securely fastens the drive in place. I pushed the drive in, the clasp sprang shut and all the lights went off briefly before flickering back on, all red. I’ve since reasoned that the power lead must have been teetering on the edge of conductivity, and the jolt from the closing spring caused it to temporarily disconnect.

So there I was, about to catch a plane, and staring at a bank of red lights which represented a total annihilation of my digital presence. Photos, video projects, university work, music…. You name it. I tried the age old adage of turning it off and on again. The feedback I received from the status lights indicted “This drive has failed and needs to be replaced” on all 4 drives.

The river of expletives ran fast and furiously, forming a waterfall in record time, culminating on my floor in a foul pool of stagnant words.

I put everything down and left for the airport. My head swimming with thoughts as to how often and how many of my files I had be backed up in the last few years. The numbers were depressingly low on both counts.

I tried to forget about it, and did a semi decent job of enjoying my time in Paris. Every once in a while I’d take a photo I really liked and it was like twisting a knife as the thought of “Where will this photo go when I get home” haunted me.

Upon my return I plugged the unit in, and this time the lights were slightly different. Three were solid red, with one blinking red, which suggests that “3 drives are low on space, and one has failed”. So the state of affairs was as follows:

I figured the best thing to do was to put the old drive back into bay 1 and hope drive 2 rebuilt itself. Alas, it didn’t work, and every light was still some degree of red. I poured over the forums/FAQs looking for any glimmer of hope. This was my best bet – “I added a new drive and my Drobo device doesn’t seem to see it. What should I do?”

After ejecting all my drives and reseating them all the lights weren’t red anymore. They were flashing orange/green to indicate they were rebuilding.

It was alive. The drobo was repairing the damaged drive. It took an agonisingly long time to show up when I plugged it in to a computer, but as soon as it mounted I knew my data was safe.

For now.

I immediately moved all my project files to a portable drive as the drobo rebuilt, then upgraded to that terabyte drive. Each rebuild took around 2 days.

Having been given a second chance, offsite backup became my top priority. After sending out a few feelers I did what I should have done long ago, and went on a recommendation from a friend. Livedrive is now my offsite backup provider. They have a number of options, and I’ve opted for their backup + ‘briefcase’ plan. I can back up an ‘unlimited’ amount of data and store up to 5TB in a folder shared between machines. When compared to the 3.2GB limit of my current dropbox account I become giddy at the thought.

I’m still in the process of setting up livedrive the way I want it, but it wouldn’t be at all unfeasible to store all of my files across all my machines in the cloud (as Alex does) and enjoy access to them regardless of where I am.

So this is a very lengthy blog post telling you to ask yourself: If my laptop/external drive/whatever was to break today, how much would I lose? If you chose to go with livedrive, here’s my referral link

Because even if you have 5 redundant drives on your property, you never know when rioters might decide to set fire to your flat live on national television

The installation process is clunkier than Google+ (which just works) but I think the video quality on FB is better.

I am of the opinion that both seem inferior to Apple’s Facetime in terms of the wider user experience. Having to video conferencing/chat while tethered to a computer, even a laptop, is like being forced to use a landline after having had a mobile. With Facetime on the mobile device it’s right there when you need it, not at home where you need-to-arrange-a-time-to-make-sure-everyone-is-at-their-computer-with-a-glass-of-water-and-have-that-interesting-thing-they-want-to-show-you-to-hand.

The social networks (though it is with some reservations I put Google+ under this umbrella, as it has yet to prove itself) will need to put this feature in their mobile applications for people to really start using it. That said, video chat is the future that every ‘forward thinking’ tech company seems to be pushing for, but few consumers actually want. When it comes to sci-fi tech we want teleporters, flying personal transport (cars/hoverbikes/hoverboards) and replicators. I don’t want to be able to see how amazing your holiday destination is, I want to beam right over and have a Mojito too. I don’t even like mint, but I want to have the option of moaning about it on the other side of the world within 30 seconds. Everyone needs to stop dicking about with webcams and get on with it.

The facebook video platform is powered by skype, a company which is currently being bought by Microsoft. Considering Microsoft took an equity stake and solidified advertising contracts with facebook a number of years ago, it’s looking like their relationship is growing ever more solid. It’s an excellent alliance to help Microsoft gain something over google online, as facebook is THE dominating power on the (Western) social web right now, a party google desperately wants in on.

While google+ is generally getting positive feedback (unlike GoogleWave and Buzz) it’s got a long way to go before it catches up with facebook. If any social networking site can do it, this could be the one, but right now the only thing that might topple facebook is their own momentum. They seem most vulnerable to tripping over their own feet as they race to stay ahead. They came close earlier this year after exposing a little too much personal data and quickly backpeddled. Another poorly thought out move of that nature could see the company landing in some legal hot water. At least they’re in bed with a partner who’s no stranger to such things.

As a side though, I’m not so sure the Microsoft/Nokia purchase will help them stop the runaway train that occurs when glass meets aluminium and is garnished with the silhouette of a certain fruit, but MS is trying their darnedest to fight every tech battle going.

In their favour, my Xbox360* is still working.

(*My second, replacement Xbox360)

The five computers i decided you’d need were:
A laptop, a fast one for when you need to set up a mobile power house.
A tower, essentially a home server, big on storage (and grunt), doubling up as a gaming machine.
A laptop, a not so fast but highly portable, for when you need to get some light weight work done.
A compact computer, to serve media to the TV, likely streamed from the “big on storage” tower above. A very basic box.
A server, for hosting and web tinkering, always on, accessible to the world.

I decided on this around 1998 and without knowing it, my overactive young mind had anticipated the netbook revolution and media centers like the apple TV.

Now it’s not uncommon for a collection of similar items to get names assigned to them, so as to make differentiation easier (Unique identifiers are very useful things like that). I can’t remember if my naming convention came first and I worked the computers into it or if i imagined the computers first and the naming convention happened to work out nicely. Regardless, the convention is exactly the kind of thing you’d expect a 12 year old boy to come up with: My computers are named after Thunderbirds.

Now the convention does fall down in places, technically Thunderbird 4 lives inside Thunderbird 2 and doing that with computers is just silly, but for the most part, it holds together, and it holds together well. Both the rockets are laptops, thanks to their slim profile, the server is a 24/7 space station, the size and function of the Thunderbird matches nicely with what each machine does. What more could you want from a naming convention i hear you ask? Well for one, i do have more devices than fit into this schema. One thing my young self failed to anticipate was just how quickly mobile devices would mature.

When I was 12, your average mobile phone looked something like this. The Matrix was released in 1999, and in school, the phone in this film was considered to be the hottest handset on the block. Mainly because it shot the microphone down your face. A friend of mine acquired one (A Nokia 8110) a number of years after, as the film was being butchered by its sequels, and it was quite a beast. From what i remember, the charger alone was pushing a kilogram. The saddest thing though was that the mic slider didn’t actually work like in the film. Their handset had been modified. Nokia realised this was the only thing people cared about, and refined the model into, the nokia 7110 (Nokia’s own naming conventions boggle the mind, as do Nikon with their DSLR camera range, but i digress. Immensely). The point of this paragraph is to say that i missed a trick, and my naming convention is not particularly extensible, to mobile devices, or any additional computers i may have. Now watch as i segue this blog post beautifully back on track…

My lack of extensibility got me thinking further. I have a friend who names his computers after elements, not much room for growth with that system (Unless you’re in the habit of creating new elements between winning noble prizes), but it’s got enough scope that the names available will almost certainly outnumber personal computing devices. The thing i liked about my own naming convention was that the original Thunderbird was able to map so closely to the corresponding computer’s function. I asked and my suspicions were confirmed, my friend hadn’t quite gone as OCD with his naming convention. While the machine known as Rubidium would react rather violently with water, this was simply a coincidence, and the element name had nothing to do with the machines function. I performed a quick interogation of a few geeky friends to see what other conventions were kicking about:

I think it’s safe to assume that you have to be pretty geeky to get to the stage where you need to have a naming conventions, and as a result, there are some massively geeky naming conventions. The fact that i’ve written all this (and the fact that i STILL think it’s quite interesting) probably makes me the biggest geek of them all

Thanks to those who took part in my very brief survey. Leave a comment or @the_defiance me if you have an interesting one of your own to add to my sexy bordered table! People are coming up with new ways to make office tech departments more nonsensical all the time…

“Guys, Cookiemonster is full and Big Bird is still in the shop. Can we PLEASE get Kermit back in the green before Miss.Piggy explodes?!”

If you’re trying to return these result via the API it’s important to know what you’re getting, and try to appreciate why there are differences.

Let’s take a page with a like button, for sake of our example, the IMDB page for “The King’s speech”. You’d like to think they have a pretty decent implementation of the like button seeing as they were one of the early adopters and fb liked to use them as an example in their documentation. Running the URL through the “Facebook URL linter” tells you what facebook thinks of the pages metadata. It breaks out the opengraph tags (<meta property=”og:type” />) and puts them in a nice table:

http://developers.facebook.com/tools/lint?url=http%3A%2F%2Fwww.imdb.com%2Ftitle%2Ftt1504320%2F

This linter is a handy way of checking you’ve implemented your OG tags correctly/in a way facebook can understand.

The bit i’m particularly interested in at the moment is towards the bottom of that page, the ‘like’ button. At time of writing, the like button says it has “11,395 likes”. These are live links, so it will have changed when you click on it. Immediately below this button is a link to the page as seen by the Graph API: https://graph.facebook.com/?ids=http://www.imdb.com/title/tt1504320/.

Right now, this link states bold as brass that the page has 1,239 likes, or nearly a tenth of what it ‘should’ according to the button. If you go via the restserver.php script you get a much more helpful breakdown of what’s going on: http://api.facebook.com/restserver.php?method=links.getStats&urls=http://www.imdb.com/title/tt1504320/

Which returns:
<share_count>3450</share_count>
<like_count>3099</like_count>
<comment_count>4846</comment_count>
<total_count>11395</total_count>

So that’s where the number on the button comes from! It’s actually a combination of likes, shares and comments. No wonder it’s such a big number! At least now we know how it’s made.

Those of you good at countdown will realise that these numbers can’t be arranged to generate the 1,239 likes as reported by the graph API. So what’s that number got to do with anything? Well, according to this lengthy (but excellent) post on the matter, this is the actual number of people who have liked the link, as reported by the automatically created fanpage for the link in question.

Why the discrepancy? Could it be to do with some of these referring to ‘share’ links that can also be included on webpages? I wish I knew. Don’t try to understand facebook logic, that’s impossible. Instead… only try to realize the truth: There is no logic.

If you can explain it for me with a comment or email it would be much appreciated. Also if you know which API call will return a set of IDs/usernames of those who have liked a button on a 3rd party site I would be even more appreciative, as that’s what i was originally trying to do when i got sucked into this wild goose chase.

Last night I attended my first facebook developer garage event Here are some of the main things covered concerning changes to the facebook platform:

-3rd party IDs replacing UIDs
http://developers.facebook.com/blog/post/431
Previously, FB happily issued the unique ID of users so that applications could target content at individuals walls and the like. Some rogue developers had discovered a market to sell these UIDs for spam purposes. To combat the selling of FB UIDs, facebook are looking to replace them with a new kind of unique number, generated from a combination of the user ID and your application ID. These would be useless for sale (unless somebody cracks the generation algorithm). The switch over is still being worked through, but is reported to be ‘coming soon’.

-GET vs POST
A bit of a rushed point, but potentially a very important one. To avoid private data (like the above 3rd party access token) being left in the URL, all session data is going to be sent/requested via POST. This change was said to be coming into effect March 10th, but I’ve been unable to find any official documentation on this. Depending on how existing applications are handling REQUEST data, this could be what facebook refer to as “A breaking change”.

-Geo targeting content
It was said that the graph API (not to be confused with the open graph protocol!) can be used to deliver content based on geographic location. Again, documentation is sparse, and the point was rushed (the speaker was running overtime) but I think this is to do with the ability to filter a status update via region, so I think this is just a programmatic way of doing what is supported natively on the status update field.

-Test user API
http://developers.facebook.com/docs/test_users/
FB has beta released an API just to churn out test users. There are some limitations on them (for obvious reasons they can’t like pages, or post on real users walls, but they can interact with other test users). Not sure if this is going to be more useful than a manually created test user account, but fb is said to be looking to delete those once this test user API is made officially. By then they hopefully will have implemented the ability to create them via the fb developer UI rather than having to write your own scripts to generate and manage them.

-No more FBML (By the end of 2011:Q1… supposedly)
http://developers.facebook.com/roadmap
According to the “canvas and page tab“ section of the roadmap, FBML will no longer be an option for new apps by the end of this quarter. This will coincide with full support of iframed apps in tabs. It’s worth nothing that this was originally meant to happen at the end of last year, but has been pushed back. As a result, this vague date is still susceptible to change. Once this change occurs, from what I can gather there will be a freeze on the FBML tags and existing FBML applications. There will be no further support, but they should still continue to work (providing they are working when the freeze occurs). Unsure of how cut and dry this freeze/transition period will be, or how much time there will be to play with iframe tabbed applications before they become mandatory. More details as I get them. This should mean that all documentation produced by facebook moving forward should be relating to new methods/in the new style, which has (touch wood) been a lot better than previous efforts on documentation. (I still think they’d do well to open it up as a wiki though).

-Updated beginners guide
http://developers.facebook.com/docs/guides/canvas
This page has recently been updated to try and wrangle the basics of app development into a concise guide. It’s worth a re-skim just to quash any past misconceptions or gospel truths you might have held about app development. I’ll be reading it now, as I’m pretty sure my knowledge of the platform has been heavily bastardised over the years by numerous hacks and workarounds. It’s important to remember how to get hacky, but also important to know the ‘preferred’ way of doing things, just in case a known workaround has been rendered obsolete or (gasp) come to be officially supported.

UPDATE:
The slides from the original presentation by @Iskandar are now online. Well worth checking out for a number of additional points I forgot to note/cover/weren’t particularly important to my current projects.

The event was well worth the ticket price (£7), for the content alone (there were additional talks, but i have only covered the tech points here), but also for the fact that it’s easy to drink and eat more than £7 worth of beer and pizza. Given London prices, you’ll have made your money back after your first gulp

Jan

-Enjoyed playing in the snow.
-Moved into a new flat.
-Went to pro photographers rights demonstration.

-Generally tried to stay warm, but sometimes ventured outside to play with traffic.

-Left my job after 18 fun filled months.
-Pottered about London/London zoo inbetween jobs.

Apr

-Started new job.
-Had fun watching marathon runners trundle past the flat.
-Did some headshots for a friend.

-Got older.
-Enjoyed the start of good weather.
-Played with a newborn.

-Got asked to shoot 2 weddings by friends (most of the photos aren’t online, that’s up to the couples ).
-Did very little else but prepare and process wedding photos.

-Tried to make the most of the weather by getting OUTSIDE.
-Went to a stag do.
-Ate too much free ice cream at the Ben and Jerry’s sundae festival.

-More socialsing and outdoorsness.

Sep

-Shot some friends in a stage show.
-Went to Marrakech.

Oct

-Went to a rally to protest against potential cuts to science funding.
-Shot a friends acoustic set.
-Got a new camera!
-Set up a facebook fanpage to try and force myself to upload work more regularly/get feedback.

Nov

-Shot a French Soirée.
-Went to Seville (and scandalously haven’t uploaded any photos to flickr yet!).
-Saw Nadal serve up a storm.
-Started a new job (Yes, again…)

Dec

-Head shots for a friend.
-Kitten shots for a friend.
-New computer to replace my loyal (but horribly slow) mac mini.
-Another play in the snow.
-Put together a snow video.
-An early Christmas morning bike ride.

Slow snow from stuart on Vimeo.

So that was that. I certainly managed to come good on my 2009 new years resolutions. A few more popped into my head though:

-Watch 100 films over the course of the year
I read about an online chum doing this and thought i’d give it a go. It was surprisingly harder than I thought it would be to watch 2 movies a week and do everything else i usually do, but here they are. I had more relaxed rules than Billy, the films could be ones i had seen before, and i’d guesstimate somewhere between a third and a half of them were. So how do you know i’m not cheating and padded it out with films i know but didn’t actually watch? You don’t, but if i were going to cheat i’d have put some considerably better films in there. Have you ever seen Tank Girl? Don’t bother. I had hoped IMDB would have an API so i could pull some stats on these and work out how many hours i’ve spent watching them. Sadly they don’t, so i’ll have to guesstimate it to be about 200 hours, or a little over 8 solid days of film watching. That’s a lot of days. I won’t be doing this again next year, by the autumn it started to feel more like a chore than anything else. I did watch some great new films though, but i think i’ll just stick to recommendations to watch as and when from here on.

-Learn to solve a rubik’s cube
Decided to ask for one for my birthday, along with a book on solving it. It only takes a few days to learn the patterns/tricks. I can do one in a little over 5 minutes, which is about 5 minutes off the world record. Now those guys are smart/fast.

-Carry on logging expenses
At the start of 2009 i began logging my expenses, and managed to do it for a whole year:

It has become such second nature to me i forgot to mention it as a resolution… but i carried on. When you log your spending with the level of OCD that i do you have a very insightful asset. I’ve now got two years worth of train tickets, lunches, shopping sprees, presents, charitable donations… everything. Now that i can compare this year to last i see that my outgoings went up by quite a bit. Buying cameras, laptops and not living at home will do that. Luckily, moving jobs isn’t all bad, and my income went up to cover the extra spending. After much analysis i’m starting to think it might be most cost effective to live at home. I will need another few years of data before i publish these findings though.

I didn’t start doing this because i needed to budget, i just figured it couldn’t hurt to keep tabs and it makes for handy reference at times. Being able to look at how much holiday x or y costs has helped me (and other people) plan how much money to change up. It’s also nice to know how much money is there if i had to find it/felt the urge to save more. Filtering down by category, i can see the ridiculous amount i’d save if i stopped drinking and the similarly large wodge of cash that i’d get back if i made my own lunch. Currently i quite like drinking and eating out, but it’s nice to know if push came to shove, that money is pretty easy to retain.

The app is called ixpensit and it’s handled updates and different handsets without a problem. It even lets you email your data as a .csv file, which is good for backing up and if you wanted to do your own graphs off of it. There’s a free version too, i think it limits you to 30 transactions so it’s very much just there to get a taste.

Conclusion:
As suspected, 2010 didn’t really top 2009. It was certainly a very busy year, maybe too busy, or busy in the wrong way. Either way, i preferred 2009, so 2011 needs to be more like that. Maybe i’m just getting old and grumpy (grumpier) though.

My photography got a bit better, mainly through a keener eye on how to process images, but it’s not made enough progress for my liking. I am hard to please like that. It’s good though, the day i stop spotting problems will be the day i no longer feel the need to improve, so if that ever happens you can have a ride on my yacht, because by that point i would probably have to be the best photographer in the universe. There was a flash under the Christmas tree, so that may yield some interesting experiments. It should also make macro photography a bit more accessible to me.

I didn’t travel nearly as much in 2010, and while my trip to Seville was planned and organised in a week, and lovely as it was (and it really is a very lovely place i want to revisit) it didn’t manage to knock a shuttle launch in Florida off my mind. They were very different trips, Seville was by far the tastier and more relaxing. Florida was buzz buzz buzz, and that wasn’t just the wildlife. This year, providing work clears my holiday request, i’m hoping to hit 3 continents, two of which i have never set foot on. Then there will only be two left i’ve not been to! (Though i don’t much fancy going to Antartica, pituresque as it can be… it’s a bit desolate. Iceland is much more appealing).

For 2011… the resolutions are pretty simple:
-Keep pushing the photography, even if it’s just to maintain the standard it’s at, but hopefully improve it. The new camera is already helping with that as i’ve got a lot more flexibility with it.
-Take (and pass) a driving test… because not being able to rent cars is starting to be a pain.
(-Given that i moved job twice in 2010 i should probably make a resolution to not quit this one… I shouldn’t really have to make a resolution about that though! Quitting jobs is not something i particularly enjoy anyway!)

I guess it’ll be another 365 days before i can mark myself!

Full set here.

http://techcrunch.com/2010/12/16/is-yahoo-shutting-down-del-icio-us/

They want to focus on what they consider their core products, as stated by the CEO to be: Search, email, the homepage, mobile and advertising.

Last time I checked, every single one of those ‘strengths’ was synonymous with the g word. The same word that’s been granted access to the Oxford Dictionary. I don’t remember the last time somebody said they were going to Yahoo something.

Yahoo also acquired flickr, the photo sharing site, in 2005. Clearly somebody had the good sense to spot the social shift in the web and make these acquisitions, presumably with the intention of growing into that space but instead they were left to stagnate. The Facemashbook launched in 2004 and now Zuck’s the second youngest billionaire in the USA, with no clear rival (in the Western world) to his service, so clearly there’s some mileage in this space and Yahoo were IN this space at the same time facebook was rising to dominate it. The funny thing is, they’re STILL making social acquisitions but it’s as if they’ve written off the motherland and are trying to get their foot in the door elsewhere. Out of their last 5 acquisitions to date, 3 were social media based, and 2 of those were outside of the USA, in Jordan and Indonesia:
http://en.wikipedia.org/wiki/List_of_acquisitions_by_Yahoo!

It’s clear Yahoo have seen Google get bigger and better at the things they themselves used to be good at, taken it to heart, and decided to try and claw their way back in instead of running off to frolic in the wide open field that is the social web. I’m sure facebook would play nice, they a need a new net giant friend given their on-going scuffles with the big G (http://www.highposition.net/article/google-and-facebook-butt-heads/4562673).

Google have shown that they don’t seem to understand what people want out of social projects. Buzz and Wave were both a disaster (though I quite liked Buzz). Google are made up of engineers, they do highly functional products like search, email, browsers, ‘phones’ and most recently, cloud based operating systems. Are these things fun? Not really, to steal a phrase “They just work”. Social projects need to be fun more than they need to be useful. The ones that are both are the real hits. Yahoo could have been focusing on developing the great social products they bought for the last FIVE YEARS, but instead it sounds like they’ve been infighting about which direction to move the company. A civil tug of war at a corporate level is not pretty, and it seems like the laggards are winning as they’ve also made redundancies on the flickr team earlier this week.

*slow clap*

Way to ignore an opportunity for so long guys, and even more insultingly, spit in its face because “it’s not working”. It was working fine, that’s probably why you bought it. Thing is you have to keep innovating on it… these things aren’t just heirlooms you acquire and expect to retain their value.

There’s an even longer rant about it from this photographer:
http://thomashawk.com/2010/12/an-open-letter-to-carol-bartz-ceo-yahoo-inc.html

Good luck Yahoo, watch out though, I’ve just had an amazing idea that’s going to knock you out of the game for good. I will be starting work on a project I’ve codenamed “web portale“ this weekend. People go to a page and click on links and stuff. It’s going to be immense.

The curve makes excellent sense as a dock, and there’s still a surface which would nicely allow for traditional input methods (The keyboard and mouse). This would be a really versatile unit, those who just want to use it in “computer mode” aren’t going to be stuck, in fact they’ll have a nice big screen to work with, maybe even too big, but that would make it great for watching movies on/TV mode. Those who want to do more interactive work have the touchscreens at their disposal, and I can see a myriad of software being tweaked to take advantage of the new input method. I think the only things that would realy hurt it would be the prohibitively high cost of having such large touchscreens (I highly doubt a mass produced version would work off the projector/camera concept). It is a bit cumbersome at present, but even if it was shrunk down I still think the curved input could hold up nicely at a reduced size. I don’t know, maybe it’s nonsense, but I still like it.

From iWoz, Steve Wozniak’s autobiography, talking about the Apple II computer around 1977.

I got a migraine earlier this afternoon, which has meant looking at a computer screen has been very low down on my to do list. Instead, I decided to finish one of several half read books. The Wozniak autobiography is pretty good, if a bit rambling in places. I think one of the main themes that’s in there is the importance of the homebrew/enthusiast scene, and I’m seeing a lot of the beginnings of the app store in the story, 30 years before it was launched. Back then I guess it would have been considered to be made up of the same kind of people who contribute to open source projects now. Though of course instead of shared codebases it was Xerox’d circuit diagrams.

At the time, the Apple II was an easily accessible platform for what would now be loosely be described as developers (software programmers) or ‘entrepreneurs’ I suppose (hardware engineers). You can see the same scenario playing out with the iPhone (and now iPad) and the app store. Apple creates a great product as a solid foundation, and lets the geeks in their bedrooms make it really useful at no extra cost to the company. The initial product does some stuff, but it’s the extensibility that makes it worthwhile.

Unlike in 1977, this time round Apple has been able to position itself quite comfortably as a middle man. The software is revenue shared and if the hardware wants to be certified or in apple stores I suspect money has to change hands there too. Clever… the app store experience coming to the mac OS is only going to line their coffers further.

I can’t complain though, as I’m part of the problem. This whole post was written on my iPhone. It was even good enough to have “Wozniak” in the dictionary.

So what’s wrong with the below screenshots? On the left is the application, on the right is my confirmation email.

Did you spot the moonwalking bear difference? The application thinks my order will be ready at 2, but the email/servers that run the order system think the deadline is 2:15.

This is an absolute disaster.

Here’s how i think it happened:
-At 13:42 i opened the wagamama app and started my order. I remeber that because i thought “Excellent, i’ll make the 2 o’clock slot” (so i kind of knew this might turn into an issue later…)
-Wagamama requires you give at least 15 minutes notice from ordering to give them time to cook your food. The app asks you to select the time slot and restaurant before you start ordering. They offer slots every 15 minutes (see example below) so i selected the 14:00 slot.

-By the time i was done with my order, credit card details (which i would love to be remembered!?) and verified by visa (garaghh) it was 13:47 or so. Little did i know, crossing over the 13:45 minute mark would have DIRE CONSEQUENCES.
-The app tells me everything is peachy for 14:00 and i walk to Wags to collect.
-Upon arrival, at 2 on the dot, I’m told my order isn’t ready, they were working to 14:15. I show them the app at 14:00, they show me their ordering system at 14:15. This is madness.
-I check my confirmation email and sure enough, it differs from the app by 15 minutes.

It looks like somewhere, somebody dropped the ball when it came to keeping this information consistent. I can’t for the life of me remember the name of this problem, but it’s quite similar to read/write error that might crop up on a database if one part of a system is trying to modify data at the same time as another part. It isn’t too uncommon a problem either, so i suspect it might have caught a few others out (the teller did say this problem sometimes comes up with their system). It could have been spotted at the flow chart phase (I like to think there was one), all it takes it a pedant like me to pipe up asking “but what happens if…”. I guess the resident pedant (read: pro-active problem-spotting busybody) was off ill that day.

I can think of two ways to solve this problem. The user journey could be rearranged, moving the selection of a restaurant and collection slot to the end of the process, when all the fiddly inputting has been done. It’s not the best solution though, as it still runs the risk of crossing over the 15 minute threshold, it just reduces the window in which it can happen. Also, there might be some clever bits going on behind the scenes that require you to pick a restaurant and time first so they can say: “Sorry, this branch is out of stock for that dish” or “That particular item takes 30 minutes to cook”. It would also totally mess up any any documentation they have on user journey and break regular users expectations, not to mention require a fair bit of code hackery.

No, there’s a much more elegant way to solve this problem. The application would make another request to the order server just before displaying the final screen to the user, checking itself against the actual ordering system for mistakes. If the user has taken too long to input their order the application offers an automated (but entirely sincere!) apology and tells them they’ve been bumped to the next time slot. Minimal code hackery, just slotting a new bit in between two old bits instead of rearranging and botching old bits into new and unintended places (something i think developers generally hate being asked to do). Also, the user would have no clue anything had changed, no need for new designs or anything.

In the end I only had to wait 5 minutes more to get my food anyway, so it doesn’t really matter.

OmNom.

A relic from a time when if somebody said “compact flash”, they didn’t mean digital memory.

My grandparents are in the process of moving house. They’ve lived in their current house for as long as i can remember and probably nearing a lifetime of mine back further still. People can acquire a lot of things in that time and they are no exception. I find some of the things are of more interest to me than others. This is one of those things, and oddly, the internet is pretty sparse when it comes to information about it, turning up a few ebay auctions of vintage camera gear (from which i can gather this isn’t worth much as a collectors item, which might explain the lack of information). If you can add any info about this please comment.

It really is a very compact flash unit when it’s all closed up, easily pocketable, but it’s all aboout the fan action on the reflector. Inside is a 15V battery and a 200 microFarad capacitor, neither of which have leaked, amazingly. There’s a handy conversion dial on the back, which i assume is only for reference as it doesn’t feel like it’s making contact with a potentiometer to alter the strength of the unit, but i could be wrong. Found no bulbs to test it with, or partner camera unfortunately. I’m not sure if you’d be able to use the bulb a few times, or if it was proper old school one shot per bulb. I think it would be the former, as there’s a “Test” button on the back. When a bulb is inserted it simply pushes back on the bit of red plastic above the “EJECT”, creating a button to force it out again.

I wish i’d taken a picture of it before i cleaned it up, rust was just starting to set in. Not sure if it’s 60s or 70s but it’s probably not been used for at least 30 years or so. No idea if it was actually sold by Boots from the highstreet.

It fits on my hotshoe, though i’m not sure how much use one contact is. I suspect with a little careful rewiring and new components it could be modified to work with LEDs, which would be brilliant (spot the lighting pun). I’ll add it to my list of projects i never find the time to start (or finish). Not sure what to do with it till then.. sit it on a shelf looking retro until i too bundle it into a box and forget about it for 30 years and my grandson can speculate making it work with a micro fusion reactor.

They don’t make them like they used to.

(In case you’re wondering why this is cool/clever, the left hand side of the game is flash and the right hand side is HTML5, two technologies which are supposedly battling it out for web supremacy, but here they are working together!)

It’s funny ‘cos it’s true. When i took an internal flight in the States last year i was amazed when the pilot announced there was in flight internet, and even wrote a high-on-jetlag blog post using it. It’s a novelty thing, for me, internet on the plane was new and amazing, but i’m sure after a few flights with the WWW you come to expect it. Still, i was just psyched i could get on facebook and chat on MSN with my travel companion at the other end of the plane (Yes, we’re geeks, the first thing he said when i signed in was “What took you so long?” (Verified by VISA is what took me so long)).

While in flight internet was a real wow moment, i’m terrible with my phone. If it’s not working lightning fast i do get frustrated but after watching that video i’m going to try and be a bit more patient with it.

Virgin Atlantic chose to pay the price with a pretty standard approach: By throwing money at the problem to buy some exposure.

Cebu Pacific decided to get creative and earn their exposure instead.

The earliest version of the Virgin Atlantic video I can find was uploaded on October 1st and has 45,696 views.

The earliest version of the Cebu Pacific video I can find was uploaded on September 30th and has 8,120,255 views.

So unless Virgin can rustle up 8,074,559 views in a day it looks like Cebu Pacific got lucky with their cute idea and cleaned the floor! Of course I’m being very cynical, and this doesn’t take into account all the views Virgin will be getting from the TV airings (more bought media) of this advert…

…Nah, sorry, i’m still going to have to chalk 1 up for creativity.

They’ve found some talking lions*. They live in the foothills of the Himalayas and they speak ‘Koro’.

“Languages like Koro “construe reality in very different ways,” Dr. Anderson said. “They uniquely code knowledge of the natural world in ways that cannot be translated into a major language.”

In an era of globalization, languages have been disappearing by the hundreds, edged out by English, Chinese and Spanish or suppressed by government practices. Of the 6,909 known languages, about half are expected to disappear in this century; every two weeks, the last fluent speaker of a language dies. This newest, with only 800 or so speakers, may be no exception.”

Both fascinating and sad.

In a way i don’t think it’s all bad that languages die out, but maybe i can just happily say that because English is at no risk of disappearing anytime soon. The way i see it, the more people that align to a common language, the more easily people will be able to share thoughts and ideas with those of insanely different cultures. The catch22 of that however is if the language of that culture is phased out, a lot of that culture might also be lost with it/suffer at the hands of the globalisation which brought the common language in the first place. Even just recording the language and conversations between the linguists might be a problem (though in this case I doubt it). Still, the observer effect is a tricky thing.

*In case you hadn’t worked it out, no, they’re not actual lions.

Those are not flares or artificial lighting or a Photoshop job. And it’s not a teleporting sequence from a sci-fi movie either. It’s the Kopps-Etchells Effect. This is how it happens, according to Command Sergeant Major Jeff Mellinger:

“Basically it is a result of static electricity created by friction as materials of dissimilar material strike against each other. In this case titanium/nickel blades moving through the air and dust. It occurs on the ground as well, but you don’t usually see it as much unless the aircraft is landing or taking off. The most common time is when fuel is being pumped. When large tankers are being fueled they must be grounded to prevent static electricity from discharging and creating explosions.”

via gizmodo.

One of my flickr contacts has been churning out some amazing black and white landscape photography recently (taken in Norway):

A robotic arm learning how to flip pancakes:

A robot learning to flip pancakes from Sylvain Calinon on Vimeo.

Flipboard, an ipad app that is a social media aggregator, in the style of a magazine.

And saving the best till last, the new TRON trailer: