Hacker News Reader

Comment by bergie | Hacker News Reader

Exactly. I'm only using my laptop when I'm actually writing code. All the rest of the computery stuff (like HN) happens in a comfortable chair with a tablet.

Eventually either somebody invents a great way to write on these devices, or I'll finish my visual programming project (see http://noflojs.org). Then (and only then) I can see my laptop usage starting to near zero.

View on HN

That looks interesting, most of the visual programming tools I have seen or used thus far are unusable or inefficient compared to writing code but that actually looks like something I could see myself possibly using.

When I think about a program in my head I basically construct it as a huge mental diagram anyway and then translate it into code. Perhaps detailed visual design is the way to go.

Visual programming has been around since the 70s and has never really taken off because it requires lots of screen real estate. The conciseness and semantic density of text will never be beaten.

You don't discuss philosophy with flowcharts; why should you code that way?

Well , we have vastly increasing screen resolutions as demonstrated by the new iPad not to mention easy zooming etc.

I'm not necessarily suggesting that you entire program will be converted to a diagram overnight but it might be an interesting way to declare general structure.

I've designed algorithms via OmniGraffle similar to this, though had to manually write the code later. It's an interesting way to do things, and it does seem to make designing complex things easier.

Yet almost every software project I've been to has started with boxes and arrows on a whiteboard. What if these boxes and arrows would be the executable application? And what if you could go to a running application, see how it is connected, and how data flows between components?

Reply on news.ycombinator.com