The Global ARCHICAD Community

Stay informed. Get help. Share your knowledge.

GDL language and objects, API, Add-ons, Script…

Moderators: ejrolon, Barry Kelly, Karl Ottenstein, LaszloNagy, gkmethy

What is your opinion about this Wish?

No votes
Not Important
Not Needed
User avatar
By Tomek Piatek

I've been watching this thread for a while. A little of my personal background before I chime in: I used to be a programmer, I have a degree in computer science, and after many years in the profession I decided to re-qualify. I went back to school to study architecture and I am now on my way towards professional registration in that field.

I completely agree with Braza that architects should not have to become programmers. They can choose to but shouldn't be forced to. While programming gives one the ultimate power over software it is a far too low level of an activity for architects to be engaging in. In the past architects did not have to know how drawing implements were made or how to make their own, they did not have to build or modify their own drafting boards and so on. It's the same with computers. They are just tools. I don't have to know about the innards of a tool to be able to use it well. Sure, light scripting here and there can help with some repetitive tasks but full on programming is not the answer.

The problem that I see currently is that software is just not good enough to allow its users to fully express themselves without the need to learn some low level tech skills, such a programming. While I try to stay open minded and embrace positive change, I don't want to have to change myself because the software is deficient. We need better software so that we can get on with better design and documentation.

We need to lower barriers not put up new ones.
User avatar
By Jp1138
I don´t know if architects are becoming programmers or not - probably somewhat yes in big projects, less in smaller ones. The benefits of programming could be used in all kinds of circunstances, but I don´t think it will become mandatory if that´s what you are saying - even in big firms the lead architects would probably know nothing of the sort and have others do it for them, and there´s a job oportunity there, at least for a while, till the AIs take out that part of the job too :lol:

I was just sying tha opening the API is a win/win situation for everyone.

Making a visual GDL editor could be fine if it´s well done, I´m not against it or something, but maybe the effort could be put somewhere else. And in the end it would still be coding, just "visually", and would require a similar effort to get things done if you are trying to do parametrical objects.
User avatar
By Tomek Piatek
Oh, I completely agree with you. GS should open up the API, publish documentation and generally entourage the wider audience to participate in the development of the platform, thus assuring its continuing future. If they don't they're writing themselves out of existence.

At the end of the day I don't really care if a graphical, but fully parametric, object editor is a GS product or someone else's, I just want one so that I don't have to deal with GDL, which is an antiquated coal-powered steam engine as far as I'm concerned ;-)
By Braza
@ Jp1138
Jp1138 wrote:at least for a while, till the AIs take out that part of the job too :lol:

You bet! :wink:

@Tomek Piatek

I think you have made a very good move to architecture. The future for humans in this high tech changing times is on, among others things, Arts and Creativity in general.

If the software doesn't allow us to do what we want, then we have this forum to express our opinion. And if enough people want the same, and Software companies hear their voices, I am sure we will get it done.
User avatar
By leceta
I consider coding as a matter of "literacy" in the informational age we live on.
Coding for an architect is (becoming) what drawing signifies for the profession at Renaissance, as Ludger Hovestadt says. Perhaps he is referring to the idea of the Architect in his most ambitious meaning, a creative, aware of his period/aesthetic, someone who has "something to say", who has "a point", who try to make the more mundane within a structure of a narrative... But his assertion is still valid, even if the daily routine of an architect is very other. Ready-made software can constrain (more than it is already constrained) the "creative space" of our profession. And I consider "creative" work even the planning of information workflow in a teamwork. Ready-made software constrains, coding + API expands "creative space".
Becoming a "programmer" is a very very different thing to "knowing to code". Ask it to a software engineer, or any professional programmer.
I don't know where you live people, but here in Spain, we are far too many architects. We cannot pretend to everyone be working as traditional architects. It is chimeric. Experience tells a very different thing.
Instead, there is a place in the market for specialized architects who does very specialized work in a team-worked chain of work.
Finally, if you know how to code, you still have a lot of room in other sectors if our industry doesn't have place for you, architect.
By Braza
Hi leceta,

I would like to make an analogy with coding and the ancient ability to use ink pens, or the use of velum paper on blue prints. 40/50 years ago any architect that didn't master these technics would have real problems in their day to day practice. The times have changed, and today lots of young architects simply laugh when I tell them how we used to work in that time. Coding will be the same. Today your work in this particular organic shapes is really easy with algorithmic design. I am sure you and many other talented architects have fryed too many neurons learning these technics. But as it has been said, AI is on the corner, and in a few years you will just need to scan a hand made sketch and feed it in an AI system and it will automatically give you the best technical/commercial solution to achieve your creation.

Your words remind me an old professor who told us that in a city there are lots of Buildings. Most part of them are in silent, some can scream, a few can talk, but very very few of them can sing. That is our job as architects: Try to make buildings sing.
User avatar
By leceta
"But as it has been said, AI is on the corner, and in a few years you will just need to scan a hand made sketch and feed it in an AI system and it will automatically give you the best technical/commercial solution to achieve your creation."

I don't agree with this prediction. What you say will probably never happen. I have the intuition that "Intentions", in the broad sense of Christian Norberg-Schultz, will probably never be captured by a computer/logical system. Still, computers are invaluable tools for "assisting" the designer. I prefer to think that AI will "assist" not "replace" the designer.

On the other hand, I don't think intentions are captured in a sort of mistical superiority of the creator/genius´s sketch. Behind a great building, there is a hard-working architect. I want to

If you were consequent with the future you describe, you as an architect should be quite concerned about the future of (your) profession. Maybe you should invest time now learning to code, it seems that the times we will be forced to change of profession because of irrelevance are "near the corner". AI is coming... ; )
Last edited by leceta on Thu Mar 05, 2020 1:37 pm, edited 1 time in total.
By DGSketcher
Braza wrote:But as it has been said, AI is on the corner, and in a few years you will just need to scan a hand made sketch and feed it in an AI system and it will automatically give you the best technical/commercial solution to achieve your creation.
Sorry, I disagree with the myths of AI. We were talking about this kind of thing decades ago and I see little evidence to support its viability other than in some SciFi fantasia. Feel free to submit to machine control but the reality is there are currently 7 billion people on this planet who all want and need something different. The climate is changing bringing its own existential challenges including dwindling resources. The autonomous vehicle myth is falling apart as the reality of AI failing to adapt to human behaviour is resulting in numerous accidents & fatalities. Architecture will always be about the creation of environments for people, which will rely on the empathy of the designer to interpret that need, viability and enable the project. Such solutions will not fit within an algorithm.
So to get back on track you need to enable the designer with flexible and intuitive tools to model the environment in the same way as a potter will mould his clay. We have primarily communicated need through images not words, the visual design process of stretching shapes etc will always be at the core of what we do, not having to sit down and write bits of code. I'll keep saying it, the original Sketchup was visionary in its ability to move the design process into 3D using limited simple & flexible tools. We have more powerful tools in AC today, but we still lack the ability to treat the collection of elements as a cohesive block/component. If this were addressed then much of the need for visual or coded GDL would be eliminated.
User avatar
By Moonlight
To all all,

Guys although we can discuss all that we want about AI, but I think that we are diverging from the main topic which are and (as much as I see it):

a. Do we really want a clumsy Graphical GDL interface with GDL's inherent limitations just so that each one would be happy in his/her happy safe space ?


b. Do we want to have a real BIM program prepared for the algorithmic era of the AEC that is already here ?

And with the available resources at GS disposal you can only choose one.

Personally, I choose B, the good news, with an open API we can work to get A.
By Braza
@ leceta

I am not saying that AI will replace the designer. It will replace the cad monkey... or better the poor guy that have to do that complex math, code, geometry or whatever to achieve the handmade napkin sketch of the Architect/Designer.

@ DGSketcher

Perhaps I didn't express myself clearly. What I wanted to say by
"... you will just need to scan a hand made sketch and feed it in an AI system and it will automatically give you the best technical/commercial solution to achieve your creation." is not that you leave to the "machine" to create it for you. I was referring to the example video that Moonlight posted. In that case, the architect could just draw a sketch or create a clay model (like in the auto industry) > scan it to feed an AI system that analise the geometry > and according to structural/commercial parameters it would give all the detail pieces needed to build the "puzzle". But I agree that architects/designers have to have the most flexible tools to achieve their vision.


The AI subject came in discussion to support my opinion that its not a good move to an Architect to learn how to code (GDL, Python, XML, or whatever) when in a short term AI system will learn all of them in a matter of hours and use them perfectly every fraction of a second. I will never have the chance to compete with this. So please don't say we have to learn GDL, python, C++, HTML, XML, uffffff! to make architecture. I know you can do a lot of stuff with it... but man our brains has limits. :)
  • 1
  • 4
  • 5
  • 6
  • 7
  • 8