Computer Modeling Should Become a Popular Hobby

Agree broadly with everything in the article. However, I have one point to add:<p>&gt; Physical models that reproduce behavior are limited by the physics of the world, while computer models have much looser bounds.<p>This is also one of the weak points of computer modeling. The outputs are strongly influenced by modeling assumptions, and without feedback from reality, one can easily be deluded towards wrong models. The computer simulation encodes the <i>map</i> not the <i>territory</i>. This is particularly true in hard to model fields like economics &amp; finance. As long as one appreciates that, computer modeling is a powerful tool.<p>Regarding what would make modeling more accessible to computer users, I think the answer is an environment where users have access to the full live system (few boundaries) and can tweak it in real-time, with right feedback loops. Eg: Smalltalk, Lisp.

Recommend adding (1996) to title. That doesn&#x27;t take away from it, it just shows how prescient the thought was.

It is a shame that the field is pretty undeveloped still after 20 years. Only recently there have been popping up few consumer CADs (fusion360&#x2F;onshape), but those are still quite limited and of course do not satisfy free software user like me. Setting up decent physical simulations seems yet another whole endeavor even if you got your model done. There are few physics engines available, but using them seems to require fair amount of work even just for basics. Then comes the social sharong aspect that the article mentions as &quot;Shared federated simulation environments&quot;. While I don&#x27;t think that we need such fully immersive shared environment that the author envisioned, just being able to share models and simulations for exploration on the web would be nice. Being able to get nice photographic-like visualizations would be then the cherry on top this whole thing.

I&#x27;ve had a lot of fun with agent-based and discrete event simulations. Hook it up to a visual and it&#x27;s basically a game. Writing a basic simulation engine is also a lot of fun. Simulation is also the only application where I&#x27;ve felt comfortable with classical object-oriented modeling. I always find that object oriented programming forces me to deal with a lot of emergent complexity, but when you&#x27;re doing simulation modeling, emergent complexity is the whole point.

&gt; The fog surrounding economic projections, and the other uses and abuses of statistical evidence in politics, might lift at least a little.<p>The perpetual historical failure of academic economic modelers is a counterexample to this line of thinking; to think non-academically trained folks would fare better is laughable

I&#x27;d love to see more &quot;gamey&quot; and fun platforms for agent-based modeling (Sims, but with different types of goals to tweak!) and something for system dynamics modeling.<p>There&#x27;s been a few runs at both, but for how poor human brains are at intuiting things about non-linear systems, and how desperately we need to understand them, more things in this area would be great!

VRML became really good a couple of years after this with SGI&#x27;s CosmoWorlds. I built some incredible things with it, with my actual sister as if we had gone back to childhood with the LEGO bricks out. In fact my sister&#x27;s well paid career is entirely due to how impressive the VRML that fitted on a floppy disk was. Funny to think that you also applied for jobs by post back then.<p>Admittedly the texture rendering was not up to modern gaming standards but everything worked and existed on a beautiful imaginative plane - much like how a little bit of imagination is needed with LEGO.<p>But then we realised nobody else was on the trail. Our models were far in advance of anything published online. Only the SGI demos were cooler than what we had come up with. Then there weren&#x27;t many of those beyond the demo disk that came with new SGI machines (and needed the power of an SGI machine).<p>The tools - Cosmoworlds - were fantastic and not hard. I particularly liked the audio aspect, being able to have different music play in a scene and for the stereo to move around as you explored a scene. If I was a musician then I would have wanted my album played that way.<p>But no musician thought to put their traxxx in a 3D virtual world, online for the world to hear. They had distribution in those days so the required VRML viewer could have been distributed on CD to solve the problem of that ten minute download over dial up.<p>3D then became the cottage industry that it remained, only done in games companies or by specialists doing things like oil exploration, CAD or other things like TV adverts. Despite the wonders of CosmoWorlds it never became this accessible thing that people wanted to play with in a Lego like way.<p>After my sister and myself pushed VRML so far my sister ended up working in a world where the interactive exhibits she was to be building for one of the world&#x27;s greatest museums would only ever be flat and two dimensional. The portfolio of awesome interactivity that got her the job never made it past the bureaucracy, committees and bullshit of the regular world. It was a glimpse of a future that was never going to happen, a high note of creativity to never be reached again. We downsize our dreams, from 3+ dimensions to 2.<p>The problem wasn&#x27;t us or SGI&#x27;s CosmoWorlds. It was everyone else. We are not that adventurous as a species. On a given day the finest mountain tops with the most awesome views only have a small amount of people there, not seven billion people. Even though getting there needs no more effort than putting one foot in front of the other for half the day.

I wonder if there&#x27;s not a more basic reason for the observations of the author. When I code as a hobby, my main goal is to make sure that no effort is wasted. I get precious little free time to hobby code as it is, so every minute spent in front of a text editor that&#x27;s not driving forward to the desired goal feels almost physically painful.<p>The last thing I want to do in my off time is fiddle endlessly with all the endless layers of friction.<p>To hobby is to demand instant gratification. The land of hobby seems to want to be connected to the land of research. But research is only truly useful when the desire for instant gratification can be dispelled.<p>If I&#x27;m reading the author correctly, computer modeling <i>might</i> be that magical land where hobby and research can commingle. Where instant gratification can produce tangible research results.<p>I&#x27;m not convinced. Is it really a matter of tooling? Because it seems to me that any kind of modeling that would actually have research potential is going to have to be invented. So now my hobby involves research <i>and</i> invention?<p>Maybe after I retire.

Tangentially related: if anyone (or their kids) is bored with voxels, try Astroneer; it&#x27;s more like virtual clay. <a href="https:&#x2F;&#x2F;astroneer.space&#x2F;" rel="nofollow">https:&#x2F;&#x2F;astroneer.space&#x2F;</a>

I want to model my home in 3D, with measurements. What&#x27;s a good free alternative to autocad for that?

Years ago I read &quot;Growing Artificial Societies - Social Science from the Bottom Up&quot; by Joshua M. Epstein and Robert L. Axtell and was really motivated to do computer modeling as some sort of hobby.<p>For me at least the problem was once I got their original model up and running I wasn&#x27;t able to sustain the focus to construct a graphical representation that was compelling. So I never went back and debugged the stuff that I added to model (not seriously anyway).

Well I definitely spend most of my hobby time modeling 3D printed robots!

I think it has, just not in the way the author intends here. This article was basically written on the cusp of the game modding scene really taking off online, followed by the indie game revolution of the early 2000&#x27;s into what we see today as a glut of games being released. The popular computer modeling pastime is game development and exhibits much of what is talked about in the article.

Definitely needs a &quot;(1996)&quot; on the title.

Many games have a modding community. A lot of these mods are usually suprisingly detailed model packs.

Seems like the author was right; I&#x27;d look to Minecraft as the analogue for model trains.

For those that are developers&#x2F;programmers, you might like OpenSCAD where models are generated from code. I prefer it over knowing where&#x2F;when to use menu items.<p>As a bonus, the exports work with 3D printers.<p>Source: Printed parts of a hand as a hobby.