This is Signal vs. Noise, a weblog by 37signals about design, business, experience, simplicity, the web, culture, and more. Established 1999 in Chicago. Follow us on Twitter for more information on our products.

What if you had a typo? Just kidding. I guess you could always just grab the marquee tool and paint over your png.

I’m a big fan of pen/pencil and paper. I wonder how many UI designers out there prefer paper over the screen.

The best deal, I suppose, is to use both paper and screen depending on needs ;]

That’s just silly. What if you need to revise it and republish? (At a more fundamental level than ‘a typo’—logic or flow changes, for example.)

Did I spend too much time on this? http://bit.ly/7TDxh1 Perhaps. But that document had a lifecycle of a couple months, off and on (through 2 product versions.) It was invaluable for the duration of that project, and was revised probably 3 or 4 times a week, especially early on as marketing and product decisions were still being decided.

This project only got (maybe) 20% of my time for its duration, too—without accurate and up-to-date documentation, my ability to keep any of it in my head would’ve seriously suffered. And yes, I know - ‘get real - work in code.’ Wasn’t practical for that project. Not enough time, mixed codebases (HTML + old AOL -proprietary ‘Rain Man’ screens.)

I continue to respect many of the decisions that 37signals makes, but likewise grow weary of vapid ‘worksforme’ half-baked suggestions passed off as insight (which is “an accurate and deep understanding.”)

And now I know I’ve spent too much time on this comment…

I agree with Bryce. I’ve come to regard the “getting real” method as a solid addendum to the new school of agile development and consider it recommended reading for contemporary developers.

However, I’m aslo a fan of the scientific method – http://www.sciencebuddies.org/mentoring/project_scientific_method.shtml.

But what is the benefit of getting rid of it? I made one of these yesterday in about 4 hours. I’m not really “afraid” to throw it away, but I don’t see a reason to get rid of it. In a few months if I want to make a new one, it’ll go by real fast if I have the last iteration ready.

I guess it matters what the diagram is for. Mine was for a customer to show on powerpoint to other potential customers. That sure has a different set of requirements than a quick diagram to show a coworker. But in that case, I would have just used a whiteboard.

I don’t get it. What’s the value to junking it over having it in a folder?

that’s fine if you’re not doing IA for a $1M+ project. We worked on flows and wireframes that went thru 3 rounds of reviews

(client’s workflow, not mine)

I disagree. Put some KB files in my TERRA -byte hdd is always welcome.

The disargreement among the commenters highlights the importance of understanding the context.

The disargreement among the commenters highlights the bullshitness of this blog.

The fact you comment in a blog you think is bullshit, highlights the amount of spare time you could spend on more usefull things.