In case you are interested, a short thread with some screenshots of my last progress. Sadly it's done in other platforms, but perhaps it ignite some ideas.
@arquitextonica So you're doing this stuff in grasshopper?
Are the color codes representing different dictionaries?
What exactly is a dictionary and what are it's uses? @topologic Do you think this will be achievable in Sverchok in a few weeks?
Sorry for the questions, but until I try it, I'm just guessing what is what and how and why...
@JQL said: @topologic Do you think this will be achievable in Sverchok in a few weeks?
Fingers crossed. I teach full time and work on this evenings and weekends. It is going well. Just a lot of methods to wrap and test in sverchok. Once done this will only be available for Linux for the time being. Windows has TopologicGH and TopologicDynamo.
@JQL said: @arquitextonica So you're doing this stuff in grasshopper?
Are the color codes representing different dictionaries?
What exactly is a dictionary and what are it's uses?
Yes. Everything is done in Grasshopper (evidently the sketching and modelling is done in Rhino).
The color codes are representing dictionaries and/or the result of some queries. At the end, I use the colors to visualize the differen groups of spatial objects I achieve in each attempt.
Dictionaries for my is where the POWER and the real problem lies... @brunopostle sumarised it really well.
@brunopostle said: @topologic said:
This comes down to a question about 'what architecture is'; saying that a wall can be any-thing introduces a huge amount of complexity that in itself constrains action later-on. Whereas accepting that floors are horizontal, and that walls are vertical and follow a 2d path brings real flexibility - I can put a door, window, shelves wherever I need to.
When we think of BIM, we generally think of wonky upside-down buildings, but the IFC specification actually has all this great stuff for describing 'ordinary' buildings: IfcStorey, IfcWall, IfcDoor, IfcOpeningElement - if we use these things then it will be possible to have the software check the design completely, and to automatically generate clear drawings that can be used to construct it. A frustration is that useful ideas that would make this easier, like the *StandardCase building elements, are being dropped from the IFC standard.
Dictionaries for me are information storage units be it whatever we want. The more trivial information pieces I´m working with right now are "space programs". As soon as you use them in combination with topologic entities, knowledge emerges. If I have spaces with kitchen program added to them and spaces with wc program added to them, it would be fairly trivial to make a rule stating that IF a wall exists that is adjacent to both these programs it would have a strong priority to host water instalations, should be thicker, hollow and be cladded in ceramic tiles on both sides...
My goal now is to break the chains of pragmatical programmed spaces and wander into the conceptual-phenomenological side of space.
@JQL said:
I have thought many times in acquiring Rhino, just because of this...
It´s not open source but their price-licensing model amazes me and makes me a fan-boy for ever. The edu-license allows for professional use.
I´m a PhD candidate, but also work professionally and my edu license is fully usable not only for me as an individual but also for me as an employee provided my computer is only accessed by me...
For me is like having an architectural swiss-army-knife in my pc. Having said that, as soon as I finish my thesis, I´m plunging into FreeCad and Blender...
Thanks, I saw this but lost in the format of twitter (never used it) and translation.
Finally found the 'self-translated summary' and half-way through it, inspiring :D Some random thoughts in the interim:-
How high level systems (or abstration?) other than spatial interests / programmatic requirements, e.g. sensational quality of space and form be abstracted and programmed (likewise heritage value, poetic meaning) ?
Design paradigm - what are the similarities between modelling of architectural thought processes with the functional and OOP paradigm
If I understood @paullee correctly, I'd also really like to read the answers to this questions @arquitextonica could give.
One thing I struggle to understand on the Topologic approach is direct control.
The benefit from direct modelling is that it is, well... direct.
For people that dominate the tecnique it's like sketching, painting, physical model building, sculpting or any other expression. It's wysiwyg. Everything else that requires abstraction before action requires the other part of the brain to work. Architecture will always be a mix of both. So I definetelly see how a rule based approach as @topologic is great, but what happens to the model once you start modelling it directly.
Example: You want that corner window to be precisely like this, and you can model it in 1h, but then the corner changes it's dimension and you have to directly model it again. However, as the dimension of the opening stems from the base geometry that is built by topologic and a cell system, your direct modelling is lost. As that window has been directly modelled and might be out of the preset rules, the opening is also lost.
Is there any place for an hybrid workflow where topologic knows where that direct change to the model has happened and is able to replace it's position, or anyway adapt it to the new place?
Can it, at least, somehow be programmed to check for these direct changes to geometry?
I imagine, after a point, there will be a lot of changes, and in all my years of working on 3D models, I've never seen the benefit of jumping from one software to the other, or from one model to the other, loosing the possibility of iterating back to the root of the model's geometry. Even on construction documentation, the overall shape of the building is changing, and all the associated info should change with it, including all analysis.
@paullee said:
1. How high level systems (or abstration?) other than spatial interests / programmatic requirements, e.g. sensational quality of space and form be abstracted and programmed (likewise heritage value, poetic meaning) ?
Don't completely understand your question but my approach is fairly straightforward. Space-objects qualified with properties. Then using querying and algorithms to operate on this information (hence the "informational" adjective).
Design paradigm - what are the similarities between modelling of architectural thought processes with the functional and OOP paradigm
I'm really struggling there because I lack the software/coding background. I would have liked to develop this spaces as objects with properties that produce effects when this objects collide/interact with each other but that would have led to two "problems".
1. The objects could be too concrete soon. What is the difference between a stair and a stairwell?...
2. If every effect is predetermined, we would actually be removing the possibility of emergence (as in cybernetic emergence)
With my approach, as I remove "automatic generation" I am displacing the need of operational imput towards the user, thus maintaining agency and control.
What Topologic may assist solving your problems ?
Topologic has been a godsend. It not only deals with geometries in an amazingly clean and powerful way, but also the possibility of assigning these informational properties to the geometries, converts them in informational objects which we can operate upon.
Simple example found by one of my students. If I have a kitchen space and a living room space, the overlapping result should be a dining room...
@JQL said:
One thing I struggle to understand on the Topologic approach is direct control.
The benefit from direct modelling is that it is, well... direct.
...but what happens to the model once you start modelling it directly.
Example:...
Is there any place for an hybrid workflow where topologic knows where that direct change to the model has happened and is able to replace it's position, or anyway adapt it to the new place?
Can it, at least, somehow be programmed to check for these direct changes to geometry?
I imagine, after a point, there will be a lot of changes, and in all my years of working on 3D models, I've never seen the benefit of jumping from one software to the other, or from one model to the other, loosing the possibility of iterating back to the root of the model's geometry. Even on construction documentation, the overall shape of the building is changing, and all the associated info should change with it, including all analysis.
This is also one of my MAIN hurdles to tackel. IMHO generative design surrenders user agency completely. That I despise. I want control. But then there is this direct modelling approach as you name it. My position is that this directness is actually a lazy way of naming a thought process that is so hard to explain that architects have decided to stop trying to do it anymore.
I believe the keypoint lies not in the directedness but in the iteration across the design sequence. I believe Christopher Alexander abandoned computers in his late work because the machines couldn't cope with the complexity of the sequences. Daniel Davis says in his PhD thesis that visual scripting sequences have too rigid structures to allow this kind of iterative design processes you and I seek.
My postulate, is to resort to human agency to solve this iterative processes. Make an effort to structure the code/tools in smaller "objects/operations" that can be rearanged to accomodate this spatial events (the not-based-on-rules window)...
As I said many times, my goal here is never to criticise but to make some questions regarding the conceptual/human/designerly purposes and mindframe of your amazing work. Cheers!
Hi @ReD_CoDE and thanks for chimming in. I talked a lot (too much, too willingly and too blindly) under this topic. That looks a very smart name and I doubt I said something as smart as that. :D
Could you please refer to which part of what I said you would describe like that?
What would be the tools for such multidimensional graph/network? arangoDB or protégé ? is IFCowl good enough for that? All Dictionaries would represent an ontology? beside to the IFC one..
I am blind too..
@JQL As far as I see in this picture you referred to "layered" graphs, graphs that are in different layers and have relations with other upper and lower graph layers too
However, I understand that for many of you these things are too advanced, and out of your knowledge and experiences :)
@JQL As far as I see in this picture you referred to "layered" graphs, graphs that are in different layers and have relations with other upper and lower graph layers too
However, I understand that for many of you these things are too advanced, and out of your knowledge and experiences :)
That wasn't a picture I referenced. I did the sketch to try explaining @topologic what I was thinking about. He said it was perfectly achievable with Topologic.
@ReD_CoDE said: @JQL what you're referring to is called "multidimensional graph/network"
Sorry, but NO.
Multidimensional graphs are actually possible with @topologic, as edges can be qualified deeper with dictionaries containing properties and articulated knowledge, but haven´t been discussed here yet.
As a side note, when you confuse to perspective-sketched plans with layered graphs I have to tell you the concesdence of your ulterior comment is unwarranted.
Thanks for the demo Wassim.
So to go forward and add some electrical plugs to topologic I will need to go 'multidimensionaly graphed! :-)
Room and Space is already available, but on the next dimension, we would add some furniture to the room, another dimension would have the water-tap and its supply, a next dimension: the electrical-power-plug and its supply..
The power- plug can already have its roomlocation (dictionary) & positionwithintheroom (geometry = shell in the cell) (what topologic already could do). Then I need to know how I would implement the next dimension: the electrical power-supply-graph.
Would it just be a second instance of 'topologic' with an adapted behaviour (instead of geometry it would handle power ratings). Both topologic instances would run parallely and/or interact? what are the tools behind this topologic intelligence?
Is it understandable?
Lukas
Comments
In case you are interested, a short thread with some screenshots of my last progress. Sadly it's done in other platforms, but perhaps it ignite some ideas.
@arquitextonica I think he was referring to the TAD BIM Software. https://www.teamtad.com
Thanks @topologic, Yes @arquitextonica I mean TeamTad :)
p.s. @arquitextonica Can you point out where can have a look at your works ? Thanks :)
https://community.osarch.org/discussion/comment/4790#Comment_4790
about TAD..
@arquitextonica So you're doing this stuff in grasshopper?
Are the color codes representing different dictionaries?
What exactly is a dictionary and what are it's uses?
@topologic Do you think this will be achievable in Sverchok in a few weeks?
Sorry for the questions, but until I try it, I'm just guessing what is what and how and why...
Fingers crossed. I teach full time and work on this evenings and weekends. It is going well. Just a lot of methods to wrap and test in sverchok. Once done this will only be available for Linux for the time being. Windows has TopologicGH and TopologicDynamo.
I have thought many times in acquiring Rhino, just because of this...
Yes. Everything is done in Grasshopper (evidently the sketching and modelling is done in Rhino).
The color codes are representing dictionaries and/or the result of some queries. At the end, I use the colors to visualize the differen groups of spatial objects I achieve in each attempt.
Dictionaries for my is where the POWER and the real problem lies... @brunopostle sumarised it really well.
Dictionaries for me are information storage units be it whatever we want. The more trivial information pieces I´m working with right now are "space programs". As soon as you use them in combination with topologic entities, knowledge emerges. If I have spaces with kitchen program added to them and spaces with wc program added to them, it would be fairly trivial to make a rule stating that IF a wall exists that is adjacent to both these programs it would have a strong priority to host water instalations, should be thicker, hollow and be cladded in ceramic tiles on both sides...
My goal now is to break the chains of pragmatical programmed spaces and wander into the conceptual-phenomenological side of space.
It´s not open source but their price-licensing model amazes me and makes me a fan-boy for ever. The edu-license allows for professional use.
I´m a PhD candidate, but also work professionally and my edu license is fully usable not only for me as an individual but also for me as an employee provided my computer is only accessed by me...
For me is like having an architectural swiss-army-knife in my pc.
Having said that, as soon as I finish my thesis, I´m plunging into FreeCad and Blender...
Thanks, I saw this but lost in the format of twitter (never used it) and translation.
Finally found the 'self-translated summary' and half-way through it, inspiring :D Some random thoughts in the interim:-
Thanks :D
If I understood @paullee correctly, I'd also really like to read the answers to this questions @arquitextonica could give.
One thing I struggle to understand on the Topologic approach is direct control.
The benefit from direct modelling is that it is, well... direct.
For people that dominate the tecnique it's like sketching, painting, physical model building, sculpting or any other expression. It's wysiwyg. Everything else that requires abstraction before action requires the other part of the brain to work. Architecture will always be a mix of both. So I definetelly see how a rule based approach as @topologic is great, but what happens to the model once you start modelling it directly.
Example: You want that corner window to be precisely like this, and you can model it in 1h, but then the corner changes it's dimension and you have to directly model it again. However, as the dimension of the opening stems from the base geometry that is built by topologic and a cell system, your direct modelling is lost. As that window has been directly modelled and might be out of the preset rules, the opening is also lost.
Is there any place for an hybrid workflow where topologic knows where that direct change to the model has happened and is able to replace it's position, or anyway adapt it to the new place?
Can it, at least, somehow be programmed to check for these direct changes to geometry?
I imagine, after a point, there will be a lot of changes, and in all my years of working on 3D models, I've never seen the benefit of jumping from one software to the other, or from one model to the other, loosing the possibility of iterating back to the root of the model's geometry. Even on construction documentation, the overall shape of the building is changing, and all the associated info should change with it, including all analysis.
Don't completely understand your question but my approach is fairly straightforward. Space-objects qualified with properties. Then using querying and algorithms to operate on this information (hence the "informational" adjective).
I'm really struggling there because I lack the software/coding background. I would have liked to develop this spaces as objects with properties that produce effects when this objects collide/interact with each other but that would have led to two "problems".
1. The objects could be too concrete soon. What is the difference between a stair and a stairwell?...
2. If every effect is predetermined, we would actually be removing the possibility of emergence (as in cybernetic emergence)
With my approach, as I remove "automatic generation" I am displacing the need of operational imput towards the user, thus maintaining agency and control.
Topologic has been a godsend. It not only deals with geometries in an amazingly clean and powerful way, but also the possibility of assigning these informational properties to the geometries, converts them in informational objects which we can operate upon.
Simple example found by one of my students. If I have a kitchen space and a living room space, the overlapping result should be a dining room...
This is also one of my MAIN hurdles to tackel. IMHO generative design surrenders user agency completely. That I despise. I want control. But then there is this direct modelling approach as you name it. My position is that this directness is actually a lazy way of naming a thought process that is so hard to explain that architects have decided to stop trying to do it anymore.
I believe the keypoint lies not in the directedness but in the iteration across the design sequence. I believe Christopher Alexander abandoned computers in his late work because the machines couldn't cope with the complexity of the sequences. Daniel Davis says in his PhD thesis that visual scripting sequences have too rigid structures to allow this kind of iterative design processes you and I seek.
My postulate, is to resort to human agency to solve this iterative processes. Make an effort to structure the code/tools in smaller "objects/operations" that can be rearanged to accomodate this spatial events (the not-based-on-rules window)...
As I said many times, my goal here is never to criticise but to make some questions regarding the conceptual/human/designerly purposes and mindframe of your amazing work. Cheers!
I'd be happy to be one of your students...
Thanks @arquitextonica for your detailed explanation. Maybe once you have a gap, show some examples to understand your works better ! :D
@JQL what you're referring to is called "multidimensional graph/network"
Hi @ReD_CoDE and thanks for chimming in. I talked a lot (too much, too willingly and too blindly) under this topic. That looks a very smart name and I doubt I said something as smart as that. :D
Could you please refer to which part of what I said you would describe like that?
Thanks in advance.
What would be the tools for such multidimensional graph/network? arangoDB or protégé ? is IFCowl good enough for that? All Dictionaries would represent an ontology? beside to the IFC one..
I am blind too..
@JQL As far as I see in this picture you referred to "layered" graphs, graphs that are in different layers and have relations with other upper and lower graph layers too
However, I understand that for many of you these things are too advanced, and out of your knowledge and experiences :)
omfg ! A multi-dimentional-layered graph representation in 2d sketch !
I am thinking of hosting a webinar to field all questions and have a discussion about the potential of all this. Is there interest?
What the hell are you guys talking about?
@topologic said:
Shoot the date. I'm in!
That wasn't a picture I referenced. I did the sketch to try explaining @topologic what I was thinking about. He said it was perfectly achievable with Topologic.
Wassim Jabi is inviting you to a scheduled Zoom meeting tomorrow Friday 19 February at 16:00 GMT (London time)
Join Zoom Meeting
https://cardiff.zoom.us/j/83477566475?pwd=WmVwOGZxQVhQMFFaeWN3cklpSlNoQT09
Password: Topologic
Sorry, but NO.
Multidimensional graphs are actually possible with @topologic, as edges can be qualified deeper with dictionaries containing properties and articulated knowledge, but haven´t been discussed here yet.
As a side note, when you confuse to perspective-sketched plans with layered graphs I have to tell you the concesdence of your ulterior comment is unwarranted.
@arquitextonica
I don't care about your nonsense comments
I mentioned that type of graphs, because they are popular today, but not in "average" solutions
Reminder that the Topologic Zoom meeting will start in 15 minutes at 16:00 GMT
Join Zoom Meeting
https://cardiff.zoom.us/j/83477566475?pwd=WmVwOGZxQVhQMFFaeWN3cklpSlNoQT09
Password: Topologic
Can't join / login via browser :(
And recording ?
Sorry to hear that! Yes we recorded it. Will share here in a few minutes once Zoom makes it available
Thanks for the demo Wassim.
So to go forward and add some electrical plugs to topologic I will need to go 'multidimensionaly graphed! :-)
Room and Space is already available, but on the next dimension, we would add some furniture to the room, another dimension would have the water-tap and its supply, a next dimension: the electrical-power-plug and its supply..
The power- plug can already have its roomlocation (dictionary) & positionwithintheroom (geometry = shell in the cell) (what topologic already could do). Then I need to know how I would implement the next dimension: the electrical power-supply-graph.
Would it just be a second instance of 'topologic' with an adapted behaviour (instead of geometry it would handle power ratings). Both topologic instances would run parallely and/or interact? what are the tools behind this topologic intelligence?
Is it understandable?
Lukas