Salta al contenuto principale


Ink and Switch wrote a nice blogpost recently talking about "Malleable Software" inkandswitch.com/essay/malleab…

It resonated with me a lot. The idea of building software that has an on-ramp that keeps on-ramping, guiding a user to not only become an expert, but to make the software their own, really resonated with me.

It's felt for a long time that Gnome has gone in the opposite direction. Trying to make something easy and clean, but end users aren't part of that journey.

in reply to Christine Lemmer-Webber

I don't just want to focus on the negative: if not Gnome, what *should* we use? Well, I'm not sure anything else is currently ideal. Maybe it's time for something else.

I particularly like the bit about the above essay that talks about how Hypercard helped its users grow to become experts. Topical, and thoughtful.

This is the kind of direction I want to see from software, one where users are part of the story: software that grows with users, users that grow with software.

in reply to Christine Lemmer-Webber

What seems to be very hard is making something that works well for perpetual beginners or intermediates but still is malleable.
in reply to Jan D

@simulo I think programs which have come with good defaults and a good initial landing but are extensible do exist. Browsers (well, historically), spreadsheet software, etc are good examples.

Blender is an interesting piece of software that used to be notorious for being hard to use, and is in a categorically challenging software space, but is now considered given its domain relatively comfortable to pick up. But it's incredibly customizable, and many people become developers by using it

in reply to Christine Lemmer-Webber

@simulo I have been impressed by seeing how many artists who are not programmers first have developed tools in Blender using geometry nodes especially. Really cool to see. And many other artists have *become* python programmers by starting out with small customizations
in reply to Christine Lemmer-Webber

@simulo Necro, but: I just remembered another example of early end-user development that's largely forgotten nowadays - MS Access forms.

Custom data entry user-tailored to the specific task, extensible with VB if you need more!

in reply to Christine Lemmer-Webber

@Christine Lemmer-Webber sugar (from the OLPC project) was quite nice, and it's a bit more recent than hypercard, but I'm afraid it's also dead, right?

(also quite targeted towards children rather than general users)

in reply to Christine Lemmer-Webber

I love this piece. Thanks for sharing it. I'd not heard of Ink & Switch, and I'm going to enjoy browsing their work. Their spreadsheet ideas are really fascinating.

People seem to be pushing LLMs as the saviour to this problem. I think Lisp and Prolog are probably better answers.

Christine Lemmer-Webber reshared this.

in reply to Christine Lemmer-Webber

fosstodon.org/@AdrianVovk/1146…

GNOME force strikes again...

Questo sito utilizza cookie per riconosce gli utenti loggati e quelli che tornano a visitare. Proseguendo la navigazione su questo sito, accetti l'utilizzo di questi cookie.