The Future of Human Data Interaction | Hacker News Reader

The Future of Human Data Interaction

tomtunguz.com

67 pointspeter_dee14 comments

Bret is releasing the next video in this series on the 28th of May.

http://worrydream.com/#!/MediaForThinkingTheUnthinkable

Could someone please explain to me why these tools are not commercially available?

This is the desktop version of the app Bret Victor created: http://vimeo.com/64895205

The tools would make my live so much easier, I can not start to express how much science needs this. Why did no startup jump on implementing them, like "Light Table" [1]? Alternatively, why did Victor not release it commercially? He would not get rich, but I guess he does not need to earn money anymore. His tools could benefit humanity, from cancer research to safer cars.

I really wonder, what the reason is why Mr. Victor never published any of his tools. It makes me sad to see such genius applications and all I can do is look at them.

[1] http://www.chris-granger.com/2012/04/12/light-table---a-new-...

EDIT: Thank you @williamcotton for watching the summary (I didn't) Mr. Victor mentioned he will make it available on github, his account is: https://github.com/worrydream This is great news.

You can always email Bret yourself and ask - his email contact is available at worrydream.com.

I'm also interested in knowing why. A lot of his demonstrations are really cool and I'd pay a dollar or two just to play around with them as they are, so I imagine people would pay a lot if they could buy full featured packages.

He says at the end of the video that the tool will be made available on Github sometime soon.

Hrm... It looks like these talks are pretty recent. It seems like he is still exploring interaction and visualization and is creating some pretty fantastic work. That doesn't mean he is ready to or has something to sell.

There's a big difference going from a demoable app to production-ready software.

So much so, that it's exceptionally rare for a single individual to do both.

The worst managers don't see this, and mistake a well-made prototype for a finished app.

But as Elon Musk has said, "The first step is to establish that something is possible; then probability will occur."

This. it's really hard to ship, it's even harder to ensure all the marketing/support/commerce stuff you really don't want to do as an interaction developer is also running.

Personal case examples:

1) Recently announced a plugin mentioned by some top designers - basically brings some new interaction ideas to graphic design tools. It's taken about 4-ish months of evenings to make a functioning prototype. Then another couple months to design the website/do user interviews/etc. Despite the flattering mentions, I still freak out a bit that not enough people will buy it when I launch.

2) Made a desktop app that had more of an 'interaction-first' approach, and inspired by folks behind Balsamiq. After 2 years of doing code, hiring a coworker, etc. It was 80% done, customers couldn't wait for me to ship, but some just-hired manager didn't get it and canned the project, along with my employment. Just downloaded what the old stuff they still ship, it has 2012 copyright info.

Also, it hasn't been encouraging when going to conferences like microconf, where the focus and advice leaned towards "find that b2b niche and make a saas app, you're gonna regret otherwise"

HCI guy here, who gave a fair share of demos at conferences etc. to illustrate new concepts. Those demos are usually held together with duct tape, and would require TREMENDOUS work to turn into production ready apps.

Additionally, HCI researchers (group in which I'll include Bret Victor, even though he doesn't publish at the main conferences etc., sadly) are more interested in creating those examples to illustrate their research, but not in building commercial stuff (which is a very different type of work).

I actually was at Victor's talk "Stop Drawing Dead Fish", and at the end an audience member asked if he could release it. He said that if he released it, no one would get any use out of it, because it was really built only with that talk in mind. (which, of course, is not surprising– that's the nature of his work)

> ' He would not get rich, but I guess he does not need to earn money anymore.'

You're greatly overestimating things here, my friend :)

I'm sure Bret would call himself a designer rather than an HCI researcher, and its true that many HCI researchers sneak design ideas into HCI conferences (Are we designers? Are we scientists? Let's just through in some pointless vigor in what is otherwise a good design paper!). Sorry, this is just one of my pet peeves, and its a problem we have in my own field (PL) also.

I agree most of our prototypes are held together by duct tape and do not represent ideas that are ready to go into production. The paper/idea is the artifact, not the prototype/demo, which exists to promote the idea. Ten years from now (more or less) the best ideas will be integrated into products by some entrepreneur who has the tenacity to make it all work for reals. E.g. we hope this happens with LightTable.

Here is another example from Microsoft - http://www.youtube.com/watch?v=5KSZuHGTcC8

Logically we should be seeing higher and higher levels of abstraction or languages, but unfortunately the process to get there is extremely messy...

Just take a look at asm.js or NoSQL. 10 years ago I used to work in C and when things like asm.js come out it makes me shake my head and wonder about progress.

It is not fair to call it going backward, as it really is a fix (on top of fixes on top of fixes...) trying to solve the eternal problem of incompatibly. And many of todays beautiful apps are built on an unholy amount of such fixes. But that gives me hope.

Evolution is a highly messy and inefficient process but we will get to Bret Victor style interfaces soon enough...after wading through a whole lot of shit.

That demo from Microsoft is nothing at all like Victor's demo; it seems to be a drawing-based interface to make the kinds of canned charts one can do in products like Excel.

Let's try another Microsoft research demo:

http://www.youtube.com/watch?v=01Xyoh-G6DE

Actually, I have older stuff from 2007 with videos embedded in a paper (I need to extract the videos to youtube):

http://research.microsoft.com/pubs/179365/mcdirmid07live.pdf

Its a long journey from here to there.

The description is "The Future of Human Data Interaction" and the MS demo fits the bill.

Reply on news.ycombinator.com