anthony galvin


I'm a technolgist at TodayTix, the posts below are about work, but probably don't reflect my employers opinions. If you want to connect about work related topics, then this is probably the best place to do it. Work related tweets @workoworko

Some things I learnt at R/GA...

A few weeks ago I left R/GA after nearly 4 years, working first as a Technical Team Lead and then as a Technical Director. I’ve learned 1000s of tiny lessons and quite a few big ones over the last few years. Here’s some key ones.

1. Multi-discipline collaboration from day 1
Complex (digital) projects require talented people from many disciplines. Casting the right team can be a challenge, but once on the ground (usually at R/GA in a war room), collaboration needs to be rapid, open and without (too much) ego.

2. Code scamps (aka prototypes) are essential
Just about working software is the best way of explaining an idea. If you can sketch in code, getting something up and running no matter how hacky then you’re moving forward. The things you learn by doing this early are nearly always invaluable as long as you’re prepared to throw away more than you keep - being over invested in the scamp is going to get in the way of iterating.

3. Really good QA helps produce really good products
The internet is everywhere and on everything (Brad Frost, also ex-rga has some great slides about this). I’ve worked on projects where the QA engineer is testing on 20 devices. Obviously there are real benefits in test automation, but there’s no shortcuts - at somepoint somebody is going to need to tap through your app on all the devices, again and again and again.

4. Work smart and work hard
The industry has a long hours culture and sometimes working hard (late) is the way to a briliant product. But it’s not the only way. From a technology point of view investing in smart work; automation, auto-scaling, developer tools that work is way better than just staying late.

5. Access to tools and permissions matter
The battle for better tools and services is constant. If you constantly put a barrier in front of getting things done then you kill the speed at which innovation can happen and in an agency environment slow projects die a slow death. If it takes a 48 hour helpdesk response to get the CI box back online or multiple paper based forms to spin up a cloud service then the velocity of your crack innovation team is being severly hampered.

6. Location doesn’t have to be important
There’s a clamour to have EVERYONE IN THE WAR ROOM ALL THE TIME. But collaboration doesn’t really work like that. Clearly facetime matters, and at the right stage of the project it really is worth having everyone sat in the same room. But having highly motivated, talented people pulling in the same direction is way more important than having them sat in the same office (or even timezone).

#work #rga #agency #lessons #retrospective

2014-08-05 17:16:17 GMT permalink

Last day at R/GA London

Today is my last day at R/GA. It’s been an intense and entertaining 3.5 years. Everyone says, “the people here are amazing”. But it’s true R/GA people are talented, hardworking and (borderline) alcoholics.

I remember a couple of days after I started having to be part of a sprint review for a major web app project. There seemed to be hundreds of people on the call from all over the world plus a ragtag bunch of us in London. The review went OK, but it seemed like madness - the ambition of the project was unbelievably high and I couldn’t work out how we were going to deliver the work or get the various random people on the phone to agree about anything. I really wasn’t sure that I’d made the right decision to leave my cosy client side tech role.

But somehow we did it. The work was hard but the results were awesome, even if perhaps the client didn’t quite understand what we’d built - which is probably true for most of my R/GA projects.

What I will remember most fondly are those times when we aimed high, put our foot on the gas and delivered something that seemed almost impossible on day one.

There’s way too many people to call out individually but a big thanks to Patrick and the tech team here in London who’s hard work and talent made me look good on a daily basis and in particular the project teams on Pearson and Getty.

Thanks - it’s been fun.

This is a slightly extended version of my ‘all London’ email to the fantastic folks at R/GA London.

#work #rga #goodbye #future #email

2014-07-18 21:56:00 GMT permalink

Your API is your brand

Recently I was asked to take a look at an API as part of an advertising awards entry that we were making. As with most awards entries the assessment I was making wasn’t just about the functional elements such as the available methods, approach to content negotiation and compliance with open standards, but also the way the API was presented. Was there good documentation and tools as well as an active an easy to engage developer community.

All those elements were there, but there was something else less tangible and probably more important that I was really looking for. What did the API really say about the brand? This assessment is more subtle, but gets to the heart of how and why an organisation is choosing to expose services to third-parties.

- How does the sign-up process work
- Can I have a limited number of API calls without having to sign-up, so that I can test if this is for me 
- Which programming languages are the code examples available in (the choice of languages can say a huge amount about the brand)
- How often is the API updated, is this a live project
- Is there an easy way to submit bugs or feature requests
- Which developer tools are being used to share code and examples (can I do a pull request on GitHub?)

The most important area to understand is the value that the API is offering to developers and third-parties, and by extension what is the expected return value the business can hope to receive.

Empowering developers though your API is a form of co-development with people you might not know so well. Choosing the playing field (the services) carefully is an important way of shaping the development direction.

There’s clearly an assumption within some organisations that “having an API” (public or private) is enough, but just showing up is no longer a winning strategy. The way companies engage the developer community is a pure expression of branding through doing (show, don’t tell). If done in the right way it empowers others to realise your brand expression through their own art, copy and code. Which seems like something that is worth investing in.

#work #branding #api #words #awards

2014-06-18 10:23:00 GMT permalink

Casting Teams: Digital agency agglomeration

In Evan Davis’ provocative documentary Mind The Gap, he argues that a key attribute that makes London such a successful city is the way that a huge mass of people with diverse talents are able to combine to produce ever more effective and powerful alliances. This effect, known as agglomeration, is a key reasons for the success of start-up communities such as silicon roundabout.

Digital agencies increasingly require the effects of agglomeration to innovate on behalf of clients. Brands are looking to ‘innovate to differentiate’, working with agency partners on ever more complex campaigns, co-creating connected products or pushing the web through responsive builds and smarter mobile development.

These kinds of briefs require an agency to line up diverse teams, with previously ‘non-standard’ skill sets. This isn’t just a production problem. The ‘mad men era’ copywriter+visual creative team isn’t an atomic unit in the world of prototyping, hardware hacking and auto-scaling cloud based solutions. Small project teams need to be comprised of specialists from a wide range of disciplines who quickly need to form productive teams. These teams need to live in the medium. Presenting ‘mobile first’ creative in a PSD is clear warning sign for agencies and clients.

So how do agencies find a way to crunch together multi-discipline teams at short notice. The 3 martini lunch era is a distant relic, and ever more efficient agency businesses are loath to run ‘a large bench’ of specialists. Conversely keeping control of freelance costs exerts financial pressure in the opposite direction. Staffing an effective and innovative team in 2014 is a challenge.

One strategy is to develop a t-shaped skills culture, where people continue to have a strong specialism (and craft) but also have a wide exposure to complementary skills. An agency of designers who code and hardware hacking project managers.

Agency frogger is seen as a threat to business continuity for agencies, but recruiting talent who have been exposed to a wider variety of projects and environments is a potential shortcut to a more diverse and flexible team. Encouraging staff to share, talk and present at events inside and outside of the industry is another way of exposing teams to external influences.

This isn’t an easy problem, but it is one that agencies need to solve to avoid producing out of date, commoditized work that is no longer effective. In many ways the type of skills required are the same, but the tools and landscape have changed. A digital agency that continues to show copy in Word and present ideas using PDFs is already suffering from having the wrong teams in place. The future is already here, it is just unevenly distributed.

#work #agency #agglomeration #teams #talent

2014-04-03 11:07:00 GMT permalink

Music to code by

What does work sound like.

What should work sound like.

If you walked the floors of a successful, creative digital agency with your eyes closed could you tell if ‘things’ were getting done.

I noticed that @crashposition had a couple of playlists called “Music for coding and thinking” - which got me thinking about the way developers use music.

One of the 12 items on the Joel Test is a quiet working environment, without distractions. For many developers, what this means in practice is a quiet environment where they can put on some oversized headphones to filter out interruptions and side conversations - and create their own carefully controlled musical accompaniment to coding.

In most cases developers are trying to create or maintain ’flow’. Flow is for many coders the thing that gives them most satisfaction - the point where almost effortlessly working code seems to shoot from the brain through the keyboard and onto the screen without interruption. Coding downhill, with your feet off the pedals.

For many developers a day with good flow is a ‘good day at the office’.

Flow can be deceptive - extended periods of head down development without collaboration or an alternative viewpoint can result in a spaghetti code pile up of insane proportions. I’ve worked on plenty of projects where someone sat up all night writing code poetry  - 'fixing everything’ - because they were in the zone - only to leave the team saddled with the Object Oriented equivalent of doggrel.

Collaboration is a key attribute of a good developer, something that is especially true within an agency environment, but also true for anyone who isn’t just a 'lone wolf’ developer. Which means how effective is 'headphones on, head down coding’? Finding the right channel and balance of communication is important - traditionally IRC has provided the right level of persistence and real time for most developer discussion, but increasingly new tools such as HipChat and Campfire are finding ground. This seems to be especially true with more distributed remote teams.

So what to listen whilst coding, or rephrased, is there a musical shortcut to achieving ‘flow’. If it was that simple then I’d have one playlist or a couple of tunes to listen to on loop that would instantly transport me to the land of easy coding. If only it was that simple.

#coding #work #music #flow #playlists #development

2014-01-20 15:39:27 GMT permalink

Dear Legacy Code Creator...

I’m sure you didn’t mean to create this problem. It probably seemed to make a lot of sense at the time. The client was a pain, you were under pressure to deliver - producer over your shoulder, asking “is it done yet”. As for that Tech Lead, he couldn’t code review his way out of a paper bag.

A project that nobody cared about. Ship it. Move on.

And it shows. Because 2 years later we are staring at the code. Trying to fix up some problem or other. It’s hard to know where to start because, well, there’s not many comments, the commit messages are garbage and the little documentation that there is makes no sense.

I understand that you didn’t think anyone would download that app, but people have. And the client, despite what you thought, thinks this is quite a good product and would like to promote it, if only it wasn’t so buggy. So now it’s over to us.

It’s OK for you, and the rest of the team. You left. A contractor. Got fired. Who knows, but you’re not here now as a couple of us go toe-to-toe with your lack of error handling and crazy re-invention of the most common design patterns.

And you know what, I know that we could all write better code, that I don’t have the context of those meetings and planning session when it all made sense. I wasn’t there.

But next time just write a few bloody comments and maybe some documentation that doesn’t assume you know the app inside out already. And think about those of us who have to up pick your pieces…

#work #code #legacy #frustration

2013-10-15 11:18:03 GMT permalink

What we talk about when we talk about planning poker

How sizing stories (points, t-shirt sizes, days & hours) is more than just Numberwang.

For (digital) agencies who use agile techniques for managing the design and development process there are some added complexities hidden in the abstractions of the story sizing.

Depending on the nature of the development project, a typical agency team might run the gamut of disciplines, from Interaction & Visual Designers, working in Illustrator and Photoshop through to Python engineers coding in Vim and living in the command line. And each with a (subtly) different understanding of the complexities involved.

Planning poker is the technique of getting multiple people on the team to independently size user stories, whilst trying to avoid the risks of ‘anchoring’. It’s a great method for driving cross discipline understanding of those tiny assumptions that can threaten to derail a project further down the line.

Which looks something like this:  Sat in a circle, post-it notes scattered across the floor and walls, someone is describing a level understanding of a particular user story — it’s pretty simple on the surface “As a user I can sign-in to the platform, so I can register to use the service”. People scribble on the post it notes, using the Fibonacci sequence to size stories based on the relative complexity involved for the UX (interaction and visual design) and technology teams. After a few seconds everyone stops writing and holds up their post-it notes. It’s near the start of the meeting so people are still getting a little warmed up, even so the variance in the numbers is quite high. Not just between disciplines but also between team members who work in the same area. This is where the interesting bit starts.

The discussion that follows is the bit I find fascinating. The cases where someone in tech misunderstands the complexity of a design challenge or someone missing the difficulty of integrating with a third-party (“it’s just an API call”) is understandable — that’s why we are doing this as a cross disciplinary group. These are easy to flush out and correct.

The real area of interest is where people explain the assumptions behind the scores they’ve given. At this point we are into the details, and it’s great for the momentum of the project. Weeks of thinking, sketching and notes come tumbling out in the conversation — no amount of ‘annotated wires’ can accomplish the benefit of someone challenging why “that’s only a 3".

Capturing these assumptions is key — they make up the real scope and difficulty of a story. Often it’s also the risk factor that you’re gathering “I gave it a 21 because we just don’t know exactly how that’s going to work”. Whilst all forms of planning and project management are probably inherently flawed, you’re unlikely to get this level of collaboration and discussion whilst scoping out a typical waterfall build.

Agencies running agile are usually working with more constraints than a development shop. The client will probably have a fixed budget and timeline, so the only area of flexibility is scope. Whilst the aims of the project are (hopefully) going to be clearly documented and contractually agreed, these small instances of scope and the relationship between a ‘must’ and ‘could’ are the area of flex that teams are going to need deal with the reality of ‘unpacking a story’ and finding it’s more of a nightmare than a fairy tale. But that’s probably something to add to the backlog for another post…

#work #agile #planning #process

2013-09-30 11:39:00 GMT permalink

<< previous next >>