Skip to main content

Hey FME'ers

Since I'm off on vacation for a while, I wanted to post a challenge that could run for a couple of weeks, and here it is: Scientific Workspaces.

In short, create a workspace that illustrates an aspect of scientific theory or discoveries. For example, here's my entry: a workspace version of the Cern Supercollider!

It's a sort of GeometrySmasher that searches for elusive Lutz-Murray geometry particles! I added a BaguetteFilter transformer because, well, this!

The Rules

1. Your workspace can illustrate any scientific fact, theory, discovery, or whatever. Maybe something to do with quantum mechanics, Darwin's theory of evolution, or black holes? Use your imagination.

2. Your workspace doesn't necessarily need to run or do anything (mine doesn't), just illustrate your scientific concept. Having said that, bonus points if it does do something interesting.

3. That's about it. Have fun and produce something brilliant (I know you will). When I return from vacation I'll surely be in the mood to hand out generous amounts of reputation points to you all!

Regards

Mark

Does that first transformer actually say "BaguetteFilter"? Is that in 2017 beta?


Does that first transformer actually say "BaguetteFilter"? Is that in 2017 beta?

Sadly, no, it's not a real transformer. Neither is the Accelerator (which would be cool) or the GeometrySmasher. I just renamed the transformers on the canvas.


How about a new transformer called the 'BlackHole' and anything that passes into it disappears forever and cannot be observed ever again ;)


How about a new transformer called the 'BlackHole' and anything that passes into it disappears forever and cannot be observed ever again ;)

I think it's called the Terminator ;-)


A thought generating model:

The ThoughtGenerator transformer requires:

1) Previous Experience,

2) Sensation,

3) Reality

Passing these three variables into the ThoughtGenerator will generate the most valid and rational thought.

For example, sensation being 'touch', the reality being 'fire burning on wood' and the previous experience being 'when I touch fire it hurts'... should generate the thought of ''I shouldn't touch that fire, it's hot!"

:)


Here is my submission - Normal Distribution. The workspace actually works and really demonstrates how 100,000 features (or as many as you set in Creator) are distributed among the final Junctions.

 

 

To illustrate the distribution within the workspace as required by the rules, I added empty annotations (not to scale).

The workspace: normaldistribution.fmw

Upd: The picture will also go to my FME Art Gallery :)


Brilliant idea, quite hard to think of something though! Time to ponder this over.


My submission is a dangerous one! Nuclear Fission of Uranium 235.

The workspace runs to create pictures of the element diagrams of isotopes that appear in the workspace. However it takes about 20 minutes to run, as Barium has a half life of 18.3 minutes. The log file gives little facts as it goes through the stages.

 

 

Control rods were used to disable some transformers to stop the reaction getting out of control.

nuclear-fission.fmw

elements-inspector.png


My submission is a dangerous one! Nuclear Fission of Uranium 235.

The workspace runs to create pictures of the element diagrams of isotopes that appear in the workspace. However it takes about 20 minutes to run, as Barium has a half life of 18.3 minutes. The log file gives little facts as it goes through the stages.

 

 

Control rods were used to disable some transformers to stop the reaction getting out of control.

nuclear-fission.fmw

elements-inspector.png

Really cool!


Not quite scientific, but I've always thought of FME as the (geo) data equivalent of the Transmogrifier from Calvin & Hobbes, so in hommage to them...


As far as discoveries go, the value of Pi is right up there with the wheel, sliced bread and actual pie. In all my geometry programming it's come up more than a few times, so I do appreciate it's constantness.

One rough way of estimating Pi is by throwing sticks randomly on to a floor with measurement lines, and counting how many of the sticks actually cross a line. I've made a simulation of this estimation in FME, letting you throw a number of sticks and returning the calculated value of Pi. It's a bit rough but the value does get in the general area of Pi itself.

 

 

As it is the best place for pies in The Netherlands, I've made all this Pi calculation happen in Limburg. I also shaped the workspace like a Pi, but sadly not the tasty one.


Some great ideas everyone. I'll make this the last call for submissions before I close the question tomorrow and hand out reputation (maybe a blog post too).

 


Hey folks. OK, I wanted to wrap this up with a final submission of my own. Here it is: the Quantum Tester

Forgive me if my scientific facts are wrong, but in Schrodinger's Cat (thought) experiment, until you open the box the cat is both alive and dead simultaneously.

Here I am testing whether parks in Vancouver contain dog areas. While I am merely testing the data, and not viewing the results, all parks are simultaneously dog parks and not dog parks:

However, when I add an Inspector (or Logger) transformer to the workspace, then the wave function is collapsed and there is now only one outcome for each park:

Basically this custom transformer uses the FeatureReader to read the master workspace itself and check if an Inspector (or Logger) is being used. To try it out, open the template, change the 2DForcers to Inspectors, SAVE THE WORKSPACE, and then run it.

Like I said before, there are some great submissions here and you can expect to find some reputation points heading your way shortly.


As far as discoveries go, the value of Pi is right up there with the wheel, sliced bread and actual pie. In all my geometry programming it's come up more than a few times, so I do appreciate it's constantness.

One rough way of estimating Pi is by throwing sticks randomly on to a floor with measurement lines, and counting how many of the sticks actually cross a line. I've made a simulation of this estimation in FME, letting you throw a number of sticks and returning the calculated value of Pi. It's a bit rough but the value does get in the general area of Pi itself.

 

 

As it is the best place for pies in The Netherlands, I've made all this Pi calculation happen in Limburg. I also shaped the workspace like a Pi, but sadly not the tasty one.

Fantastic. I was reading about this recently. Supposedly the "roll-a-coin" games at a penny arcade (you'll have to Google it, I can't seem to paste a link) are also a good way to calculate pi. Who knew!?

 

 


Not quite scientific, but I've always thought of FME as the (geo) data equivalent of the Transmogrifier from Calvin & Hobbes, so in hommage to them...

@redgeographics you should submit this one for your next certification. ;)

 


Hi Mark (@mark2catsafe), do you think you could keep this one open? Perhaps a section in the Knowledge Center? Just for those who want to share this kind of workspaces.

 


Hi Mark (@mark2catsafe), do you think you could keep this one open? Perhaps a section in the Knowledge Center? Just for those who want to share this kind of workspaces.

 

The question is still open for sure, so feel free to post new answers here. I just won't be giving reputation rewards. But I do like the idea of saving the best challenge answers in a knowledgebase page and will see if I can set that up.

 

 


Hey folks. OK, I wanted to wrap this up with a final submission of my own. Here it is: the Quantum Tester

Forgive me if my scientific facts are wrong, but in Schrodinger's Cat (thought) experiment, until you open the box the cat is both alive and dead simultaneously.

Here I am testing whether parks in Vancouver contain dog areas. While I am merely testing the data, and not viewing the results, all parks are simultaneously dog parks and not dog parks:

However, when I add an Inspector (or Logger) transformer to the workspace, then the wave function is collapsed and there is now only one outcome for each park:

Basically this custom transformer uses the FeatureReader to read the master workspace itself and check if an Inspector (or Logger) is being used. To try it out, open the template, change the 2DForcers to Inspectors, SAVE THE WORKSPACE, and then run it.

Like I said before, there are some great submissions here and you can expect to find some reputation points heading your way shortly.

This amuses me as I recently experienced a bug that was only triggered if there was an inspector connected in the workspace. No inspector the workspace ran, with inspector (not even necessary to be on the same stream) workspace crashed.

 

 


Reply