Hacker News Reader

Comment by kyro | Hacker News Reader

This is a wonderful example of why you shouldn't shoot down concepts. When that designer unveiled his idea for Phonebloks, the majority of the feedback was criticism about how it wasn't technically feasible, how the designer was being unrealistic and downright naive for not understanding the mechanics [1].

But that's what concept design is about. You create what you envision to be a progressive solution, and find ways to make it technologically feasible. You would think that after all of the stories of Jobs demanding devices be smaller, sleeker, thinner, more responsive, being told it'd be near impossible to do, and seeing him actually pull it off, that people would be a bit more open to new design solutions.

[1] https://news.ycombinator.com/item?id=6385105

View on HN

The way in which you shoot down something matters greatly.

When I saw phoneblocks, I was instantly reminded of designers who work without regard for their medium - a great concept with zero thought given on how to make it a reality, or even awareness of the compromises the medium will inevitably impose. If you're a front-end developer, at some point you were probably given an awesome-looking PDF, then expected to make a website/app that is a 1:1 representation of it. Of course, that never ends up happening and you have explain why it could never be a reality.

I'm really excited to see where this goes, but whatever it ends up being, I can guarantee you it won't look much like that phoneblocks video.

> "whatever it ends up being, I can guarantee you it won't look much like that phoneblocks video."

Yes, because concepts evolve. Your guarantee isn't some bold bet you're making, and I assume it's supposed to show how impractical the initial design was. But this is standard across all creative industries, from tech, to fashion, to auto: You start with what many times seems like a farfetched idea and you evolve it into a consumer product that retains your initial vision as much as possible.

> "If you're a front-end developer, at some point you were probably given an awesome-looking PDF, then expected to make a website/app that is a 1:1 representation of it. Of course, that never ends up happening and you have explain why it could never be a reality."

I'm going off topic, but as a designer and a front-end developer, I get frustrated when the developed product isn't a 1:1 representation (when sending out the work to get coded). The end result should be a pixel perfect match of the artwork. Nothing is impossible translating an image to a website/app.

> a designer and a front-end developer

Could that be because you subconsciously make your designs using knowledge of the technical limitations of the web?

This is very different than Phonebloks, which was and is totally impractical. This seems to be for peripherals only, not the core system.

> And really, skies the limit when it comes to the modules that can be attached to an endo. Hardware keyboard, faster processor, bigger battery, or something different altogether (taser gun perhaps?). Anything you can dream up.

Emphasis mine.

Hardly seems like just peripherals.

Adding an ADDITIONAL faster processor is proven tech. Heck, Nintendo was doing that with $70 cartridges 20 years (e.g. the FX ship used in games like Starfox essentially overrode the built-in processor, but the main CPU in the console was still used for basic IO.

It's when you're talking about actually REMOVING that built in processor that things get really interesting.

Well, there's no particular reason that part couldn't be swapped too, sort of like swapping the motherboard in a PC and then connecting everything else you already had. Yes, that would be essentially upgrading to a "new" phone but if the standard held you could still use the screen, camera, memory and so forth which isn't possible now (potentially later upgrading those bit by bit as need be).

From the official Motorola Blog:

> "A module can be anything, from a new application processor to a new display or keyboard, an extra battery, a pulse oximeter--or something not yet thought of!" [1]

So it doesn't seem too-far removed from the original Phonebloks concept. And you can't really tell me that had this new version been presented as the original, we wouldn't have seen the same feedback.

[1] http://motorola-blog.blogspot.com/2013/10/goodbye-sticky-hel...

The definition of what is "core" to a computer is nebulous at best. Nothing stops you from having a SOC mainboard with some flash memory and an embedded processor that can run on its own, with peripheral attachments to add in beefier hardware as need be. That is what PCI is for, after all.

I think that may be one the largest failures of modern computing - we never built a hardware platform where just socketing (especially in a plug and play way) new processors, memory, busses etc was fluid and feasible. The performance penalties were always huge (albeit understandable with a monolithic virtual memory abstraction if you don't account for varying degrees of locality).

I think everyone got hung up on the implementation details. For example, the engineers saw pins and they all saw blood in the water.

Reply on news.ycombinator.com