Selective Attention as a Metaphor for the Research Process

Watch.. (if you’re still under your Internet-proof rock and have somehow managed to miss it…)

As I see it, there are two ways of missing the gorilla in the research process.

The intentional

First, the intentional. You throw your Harry Potter invisibility cloak over the big monkey and choose to ignore it. 

This could be entirely your fault. Because a certain strand of evidence doesn’t back your theory,  or you dislike the work of another author so don’t read it, or just don’t bother to do your searching very well.

Or, it could be the fault of your environment. Maybe your institution doesn’t have access to the right materials, or you don’t have the funding to purchase them. Or, you just don’t have the time to give it the attention it deserves.

The unintentional

The cape is on the ape before you knew there was going to be an ape.

It might be the thing you hadn’t considered. Much like one of those choose your own adventure books, your research could take you down a particular route and you entirely miss other options. How many pages down the Google search do you go?

Or, it could be about the technology. Whether it’s the search algorithm prioritising results above others or your organisation/ISP/government filtering certain content from you.

An argument for high quality research

All of which clearly points towards planning the research method being just as important as choosing the topic when it comes to my upcoming dissertation. Time to open that textbook

How to host a website on Dropbox

Built yourself a website and need somewhere to host it for free? There’s a really quick way to do this with your Dropbox account and a neat little web-app I discovered over the holiday called Pancake.

As the image above explains, just hook up Pancake to your Dropbox account, create a folder for your site and you’re done. Pancake will give you a URL, and there’s your website.


Ideology vs. commercial success

We’ve recently been learning how mentors can help and/or hinder our progress through the startup maze. Here’s what we’ve discovered about picking and choosing their advice.

Like many other early stage tech companies we have mentors helping steer our work in the right direction. If you’re in a similar position I’m sure you do too, or if you’re thinking about getting in to the startup world I’d recommend you get some. You can’t be an expert at every aspect of your business, but you can find people who’ll help you learn pretty fast.

That said, I’ve become increasingly conscious of the need to pick and choose the advice we take; it’s easy to be awestruck by ‘older & wiser’ influences and to forget the reason you started the company in the first place.

Specifically, the advice I’m currently struggling with is that you should split what you believe to be right from doing what will make you the most money.

Whilst I appreciate our mentors are there to guide our company to success, I strongly believe that success is about more than just a big number in a bank account.

I’ll give you an example from the software license world. From a purely commercial perspective annual contracts are preferable to rolling monthlies. Your accountant will tell you it’s about financial security – once the customer is in you know you have that guaranteed income for the year.

But, there are countless examples of successful companies who ignore this advice. FreeAgent and 37Signals to name just two- great products that our team couldn’t do without, excellent company ethos, businesses doing very well. In short, the kind of company we would want to emulate.

The real way to keep a customer is to make a product they love, and offer it on terms they aren’t challenged by. For us that equates to free trials, and easy entry/exit routes. It doesn’t mean they have to promise to pay you for a whole year even if it turns out they decide your software wasn’t right for them after 3 months.

Making your customers sign for a year may mean you don’t lose money if they hate it and stop using it early, but what you do end up with is a customer who can’t get out of their contract so are stuck complaining about paying for something they don’t want.

To my mind, designing a contract to keep people even if they want to leave is a negative starting point. Of far greater value than keeping one client for the minimum of a year is losing one early that is still positive about their experience with you. This particular product wasn’t for them, but others may be and the experience was still good enough for them to recommend you to others.

If, like us you’re starting-up or doing the freelancing thing, one of the main reasons you chose this route was to not have a boss and to run things the way you want. You don’t always have to agree with your mentors.


Whole school markbooks

It’s a fairly open secret that one of our current build projects is an online markbook (or gradebook if you’re that way inclined). We’ve been working closely with a group of partner schools on the design, and as we head towards the next phase of testing we’ve started discussions about roll out.

I wanted to share a few of the common requests I’ve seen that we think may be a mistake if you are also planning to roll out any centralised marking of this style.

Mistake 1: Do everyone at once

A central markbook is quite a shift from the individual books/Excel documents you have now, but moving the whole school in one hit is a big change. You want to trial it with a small group first, make them your champions, the evangelists that can help you deploy at larger scale.

They’ll also help you figure out any technical issues in a nice low profile way before hundreds/thousands are let loose on it…

Mistake 2: Do everything at once

Yes, you could force your staff to enter every single mark they ever give in to your shiny new central markbook. But, do you really need it all? It might help you keep tabs on who is doing what, but is there really a lot to be gained?

Why not start smaller- add the data you can really learn something from? Is there really any value to students in making your teaching staff enter every single homework grade?

Give me a reason to enter the data. Whole school assessments, levelled pieces of work- these have a useful analysis reason to be centralised.

There’s a great (unattributed) quote about not fattening a pig by weighing it. You should look that up for data discussions.

Mistake 3: Do everywhere at once

With all this now online, you could share it with students and parents and broadcast general trends to the world but you don’t need to immediately. Give your staff a chance to get used to it before adding the public pressure.

You may even find your wider stakeholders aren’t interested in the same things you are. For example, in one conversation I discovered that a majority of parents were not interested in how their child compared to the average of their year group in any measure. Just because you like the pretty graphs doesn’t mean everybody does. Ask questions.


It’s a consistent message from us, but one people keep asking us about. The best way to manage any major change you think your institution needs to undertake  is to start small, ground

up. It’s not as easy as broadcasting the orders from the front, but it’s the right way.