Tag: meejacake


Lessons Learnt, Part 1

As I mentioned yesterday, we killed our startup. I’ve been thinking a lot about the things I’ve learnt from the experience, and after yesterday’s post about starting work immediately, something else has come to mind.

I’ll be the first to admint that I’m not a great graphic designer – though I do enjoy doing it. Similarly, my interface design isn’t great. However, I enjoy doing it. So I’d taken on that role in MeejaCake. One of the big things about MeejaCake was that we were studying the relationship between objects, and so we needed a good interface to allow people to add those relationships. For some reason, I went for the standard, Web 1.0 way of doing things, which was to have separate forms for each relationship and force the user to navigate between them. I knew this wasn’t the best way, but for some reason, I stuck with it. It didn’t work very well. I thought a lot about how I could improve the forms to make them work, but what I didn’t consider was a completely different way of doing the forms, which is what I should have done.

When I hit a brick wall in programming, often I’ll try and code the same problem a different way, to see if that produces the same errors or results. Most times, I’ll find I come up with a better solution than my first attempt. If I had approached the interface design like a programming problem, I would have come up with a much better interface, but for some reason i stuck with the one that didn’t work and kept trying new things to make it work.

My A-level Computing teacher always told us “There’s more than one way to skin a cat”. I should have applied the same rule to interface design. I know I will in future!

For those interested, what I believe I should have done is what the Amazon product does – which is instead of presenting you with a form for each relation, it shows you the entire page (i.e. the page with all the relations on), and allows you to simply add new ones, one at a time. This makes the back and front end code much neater, and since you can do everything with AJAX means the form is a lot more intuitive. Also, users don’t need to learn two layouts, one for the page and one for the form – once they see the page, they will immediately know how to edit it.

Lesson: learnt.


Death of a Start Up

Advice: if you get a good idea, DROP EVERYTHING. Do it. Do it now. Don’t sex, sleep, eat, drink, dream until it’s done. Why? Because the moment you think of the idea the window of opportunity starts to close. Unfortunately for us, the window closed just as we were thinking about leaving private beta. Someone else had the same idea – and that person was Amazon. No way in hell can we compete with them, so we’ve closed the startup. Which, though disappointing, means we were definitely onto something. We wish, however, that we had just done it 18 months ago when we had the idea.
Ah well, back to the drawing board.