3 ms·
alright, I'll attempt to make a substantive comment on this. Aside from the drama, I think their approach did not work well because visual design is fundamenta
by Jack000 7y ago
alright, I'll attempt to make a substantive comment on this.
Aside from the drama, I think their approach did not work well because visual design is fundamentally not a constraint satisfaction problem. If it were, it would be possible to design something aesthetically pleasing by mechanistically following a few "design rules" (eg. balance, repetition, contrast etc) yet this is generally not the case, barring limited exceptions (eg. templates). The general rules behind visual and color design are sort of intentionally vague and in most cases requires some intuition to apply properly. We're certainly a lot closer to automating this intuition now than 2010, but there's still a long way to go.
- asperous 7y agoVisual design is about messaging. You can have a page look messy, professional, punk, clean, etc. You can accomplish that all with templates, no need for ai. The grid was a content manager at its core, with prebuilt templates. It basically just lost to squarespace.
- Jack000 7y agoright, but templates are just.. templates. Someone had to have designed those templates in the first place. Based on my understanding of thegrid's approach, the problem was that they did try to automate the design process itself and it just didn't work very well. If they just used templates they may well have succeeded. I do think it is possible to automate certain aspects of the design process, now that we have convnets.
- asperous 7y agoI agree with you, I think if you look past the marketing all they really did was make templates with parameters that computers tried to play with. I think their approach tended to not work better then humans picking the templates and parameters, but the bigger problem was just that they weren't as good as squarespace and their marketing hype wasn't enough for them to compete.