these are examples of my diagrams, they don't mean much without explanations but just to get the picture....

program is not generated. especially not in service to diagrams. it should exist as a result of someone actually needing to use a building.

graphic standards has typical programs readymade for all kinds of building types, if you need a place to start, but really the whole point of rem and all of the others you mention is not to make diagrams but to reconsider programmatic arrangement in useful and interesting ways that can lead to a building typology that exceeeds expectations. it ain't about looking cool.

your teachers should be teaching this to you.

start with a building, look at how it is going to be used, reconsider the value of traditional functional arrangements in light of modern requirements, add splunky text to post-rationalise, and you are done. voila, c'est fini. simple.

dick in the box! hah! (alf-like laugh)

kay, now go and build it!

You have mapped a series of existing relations in your diagrams.
Architecture intervenes in space - it makes changes.

What reality do the diagrams relate to?
How would changes in the diagrams affect reality?
For example, how would a change in program or site affect your diagrammatic view?

Be critical of your diagrams. Why are they a useful way to represent reality?
Conventionally, we draw in section because it reveals things that a plan doesn't. What does your diagrams reveal that we wouldn't see in conventional drawings?

Analytical diagrams alone can't produce architecture. At some point you need to make some propositions. You need to intervene in the world revealed by your diagrams.

I disagree with jump that you should post-rationalise.

rem does good diagrams

have you ever seen rem's dick in a box?

have you seen s,m,l,xl?

chax, this is going to sound terribly old-fashioned, but here it is:

I don't think your drawings are diagrams, I think they are charts.

There is a great web page I just found - I image googled "bubble diagram" which is the little hand-drawn adjacency sketchs that were my intro to how to design a building (23 years ago - I told you it would sound old-fashioned!) - that explains how hand sketching, combined with some architectural conventions, can develop into a building form. Here it is.

I suggest doing some work with your hands. Like jump said, programming comes from the needs of the users of the building. You diagram the needs, not generate a program from a diagram. And don't brush it off as "plain stupid analysis" - that's what diagramming is: analysis. Once you have a indepth understanding of the diagram, you can start to understand what the form might be - and very likely it won't "look" anything like the diagram.

Check out some Tufte. The diagram master.

What?! How can you start with a diagram and get program? That's backwards!

Shouldn't you be forgretting diagrams for now, generating a program list, and then developing analytical diagrams of that program list while you tweak, refine, and massage it according to input from your analysis?! I'm lost. A program is like the easiest part to come up with. What is the use of your people? What will the users need? What direction does your concept push the programming in? Is there are certain part of the building you want to draw more attention to than others? What should be the biggest part? What should be the smallest part, or even non-existant?

I used to make a list, cut out pieces of paper to approximate the size of each chunk of program, and swap 'em around in places to start physically diagramming.

Yeah i agree with Myriam, although i usually start with the site, and the given program. I'm a modeling-type of person so i create simple boxes in FormZ and just start moving 'em around. Typically I'll create 10-20 arrangements and then take a little bit of time to think about each of the arrangements, and pick out the best. Then start carvin' into them, and revisit your diagrams, create new ones and try to combine the ideas and connections generated by the diagrams with the physical layout of the site, and program

Are you thinking about something along these lines? Because I think the program was established before the diagramming stage.

What is the use of your building** (not people!)

I don't know the degree of relevancy but as i was looking for joseph beuys' diagrams, i stumbled into this piece. i don't all get it, but sounded useful.
Every diagram is commutative.

and there is this by rudolph steiner.

just some more to throw in the mix...

3wi(se)men sez it better than i

i was being fascetious about the post-rationalisation; cuz it seems this sort of assignment very nearly requires it.

who asks students to make diagrams without buildings or functions? what a thing to mis-teach a student of architecture.

There are other ways to teach than simulating real-world projects. Some studio projects emphasise particular skills or processes, not because those are necessarily the only skills/processes or even the best skills/processes. A well-rounded architectural education is something that happens over the course of five years or so, not in one semester.

I'm sorry agfa, but by the 4th year, someone should have been taught something other than one single procedural methodology.

Maybe if this were his/her FIRST semester...!

Yeah, i agree, this is my third year now of a five year BArch degree. The first year is basically art theory, basic analytical and critical thinking. So myself, and the rest of my class, are working on our fifth studio project now, and presented our schematic design on monday. By this time in studio, most people that didn't take architecture seriously have dropped it (although there are still a few that are clinging), but i think that i can say that most of my class has developed the necessary skills to begin laying out a project. Perhaps our analysis may be flawed, and our drawing techniques can honed, but i think that most of us have a clear understanding of general program layout, site and concept development, and other skills to take that "first step."

I mean, by the time i graduate, or become certified, i need to fully understand all of these design implications, and the many steps from that first initial concept and development, to building the thing, i don't mean to be overly harsh, but getting stuck on that first part of development will show in every stage of a project.

and since when does my junk look like a toilet paper dispenser???

exactly.

more worrying is that a student of architecture is being taught that diagrams are an end in themselves, and worse, a starting point for developing program, as though use was something that emerges from the architect and not the circumstances of its context (cultural, physical, whatever).

crazy experimentation that disregards function and is filled with theory is very much ok with me. but it still requires rigour.

this seems to be a bit like cargo cult, mistaking diagrams as the source of great architecture when it is actually just a tool.

agreed, and nice reference Jump,
sometimes my ideas get too tangled when i try to approach a project in a strictly theoretical way, I think i'm learning to balance it now, but you just have to step back and look at the true function of a project, because ultimately, that is the singlular representation of your idea. Architecture requires knowledge of everything, theory is only a small part of a huge picture, and while theory does connect and influence your structure, so does the program, the circulation, your skin system, materials, site, mechanical systems, ecologically-intelligent building designs, and a seemingly endless amount of other design influences, and your combination and justification of these forces are what ultimately makes you an architect

threeewizmen, your focus, engagement and thinking about architecture as a third year student is impressive.
btw,
jump has a phd on the craft and a practicing architect in japan.

chax... get these books:
- Guallart's Sociopolis... plenty of diagrams relating rules for vertical and horizontal programming in it
- take a look at some of Ken Yeang's documentation.. any one of his folios will do

actually i am still working on phd, but should be finished writing dissertation by december and with parchment next spring. unless i dont make deadline ( a scary possibility ) in which case it will be another 3-6 months after that...but i am working on it!

take a look at Vidler's Diagrams of Diagrams , it should be on jstore .

Firs of all thanks for your responses, any kind of advice is helpful

Ok now, let me explain this some more.. The site that we were supposed to rebuild is a Railway Station which has been built in 1882. And the complete neighborhood is shit: crime rate is high, there’s tons of hookers and homeless people and gypsies and generally that is the part of the city with lowest standard… The territory that we’re working on has 400 square km. The major part of the territory are Main Railway and Main Bus station and behind them there is a lot of unused space (some old warehouses that are out of use for years, and old houses that are falling apart). Right behind those there is a river called Sava which is connected with Danube and the riverside is really beautiful but it is also completely unused – right now it’s full of factories, again warehouses and all kinds of different crap… so the location has endless potential and basically we’re starting blank cause the Railway and Bus stations are going to be relocated to another place and warehouses and shit are being torn down so I have 400 square km and I do have a list of different functions that people of this city need (the city of Belgrade, Serbia) but I have no fucking ideas how to distribute them in that space – so it’s not just one building it’s the whole neighborhood which is supposed to become the new city centre. So my idea was that every function is going to be a cube of different color and I thought that I could make a computer program that would disperse those cubes in 3D space (horizontally and vertically). The thing is that I need to make an algorithm by which the program will work by I don’t know which parameters I can take, cause it’s a blank space.. so I need to come up with parameters that will control this distribution - which of course sounds like a complete fiction and I have completely gone mad… So anyways I’m not trying to come up with a list of functions, I have that already, I am trying to come up with their distribution on the given location and when I have that I would take the spatial forms that are produced by movement of people, of traffic, insulation and other parameters and put them on those cubes and the section of these two - the cubes that would be within this forms would actually become the function of those buildings.

And we have been thought everything from different methodologies, traditional means of design, construction, static, mechanic, analytical geometry and different theories in mathematics, new technologies and construction systems and materials and what not... and on master studies we have 20 different studios and every one of them deals with different subject - we are exploring diagrams and we are trying to push them to their limits and to get from them as much as we can

visualcomplexity.com

if you didn't know about that you can buy me a drink

thanks rfuller those youtube link‘s ‘ve been very inspiring, and Orhan Ayyüce thanks for that commutative link – that’s kind of what I’m trying to do here – find some kind of mathematical formula (figuratively speaking) that would distribute those functions – you know, the abstract machine….

lots of good discussion already. i'm a diagram fanatic, so it's been fun.

if you glean nothing else from this discussion, please garner one little nugget that's coming up in different ways:

diagrams can be used in analysis and can be great analysis tools. but diagrams that begin to define a building or a building's program are DIFFERENT diagrams and can't be direct results of your analysis. an analytical diagram won't translate directly to a result (no matter what mr hyper-rational ramus tries to sell).

this kind of projective diagram must convey a need or desire because it's about a possible future.

hmm. makes more sense now. it is an urban planning project more than architecture, no?

rem has been doing some nice projects of similar scale lately. have a look at latest 2 parter from croquis. lots of inspiration there. what you are doing sounds a bit like what oma did in toronto to convert an airforce base into a park (along with bruce mau), itself an idea emerging from his proposal at Parc de la Villette...both abstract works founded on mutability over time.

the most obvious recent version of similar approach can be found with mvrdv's KM3 and regionmaker. Something came through the office recently from them explaining their software. i didn't look at it too closely as i am a skeptic (sorry), but it is certainly out there and could be a good precedent for you. i imagine they have dealt with the same problems as you.

given this project though, it may be that a more radical approach is required. that is, not so much in terms of process and design of the algorithm, but in programatic content. urban revitalisation is very hard, and for a place as complicated and derelict as you paint a picture of, it could be that a better approach than programatic distribution would be to introduce program that works without humans as well as with...if that makes any sense (you know, kinda like co-opting the negative and making it positive). the not for human spaces could be as important as traditional urban functions and allow you to not be worried about filling kms of space with trad urban functions...just a thought, but an approach that is creative with program may be more powerful than an algorithm. or could inform it in terms of density/attractor distribution etc if the problem really is about distribution....i dunno, maybe you have already thought of all that and found the approach doesn't work...but personally i think cities tend to work for non-rational reasons, and the challenge of planning at large scale is how to capture the non-rational in a plan...

just thinking aloud at this point though. forgive me if it sounds absolute rubbish, cuz there is a good chance that is what it is. ;-)

yes, it’s a large scale project, kind of like what Koolhaas did with Almere – he defined the program, it’s distribution on the site and volumes that will hold that program, and than there was a competition and various famous architect got to work on various buildings that were part of the city (but within the boundaries of the general project that OMA did) so, like that, I need to define the program, its distribution and elementary volumes and than just focus on the form of one or two buildings

anyways, thanks jump, your thinking aloud helps.. yes, I'm going in that direction - co-opting the negative and I’d have to give some thought to those not-for-people-spaces i like the idea

when i call a building diagrammatic it is not a compliment.

IF you really, really, really wanna know where architects got this fascination w/ diagrams in the 20th century, check out a book by Christopher Alexander, called "Notes on the synthesis of Form". This should be the perfect starting point for doing diagrams.

All the morons that fell in love with the book,(Rem, Eisenman, et. al.) misrepresented the ideas, or took it upon themselves to claim diagrams as their contribution to architecture.

that's true enough, vado. but what building is diagramatic? certainly using them is not a bad thing?

i have the almere book in the office. i don't think there are many diagrams in it, per se. i think (apart from parc de la villette and downsview) when it comes to planning rem is pretty straight forward with designing volumes and space in a traditional iterative way, but with a higher programttic imperative adding a twist. at almere his major contribution was if i recall correctly to try and break out of the standard grid and to engage the waterside to densify the suburb/town. it is an amazing project if you consider that he made everything from near tabula rasa context, but it is important to realise he had some pretty big social goals in mind when he did the work. he was also very nearly fired from the project for having funny ideas. but it is going ahead still...and apparently succesful.

no offense chicarachitect but i seriously doubt rem stole chris alexander's ideas. chris is fetishistic about the diagram, rem and the others just use it as a tool. the tradition goes back much farther than the 60's in any case, certainly as far back as the enlightnement.

So are typical construction drawings diagrams, not revealing anything thats hidden but only whats physical?

Eisenman had some great diagrams. His El Croquis (if memory serves me, I haven't opened it in many a year) shows how the diagrams influenced the designs. 2D and 3D diagrams too.

trace: you think Eisenman's strategy make sense for a real building? Can you show me an example and a few words of explanation?

In 1997, there is an international competition for IIT student campus, most of well known architects are invited. Eisenman's scheme is a joke. It didn't solve any major IIT problem, just pure context line. From this competition, I know why the real estate developer likes Rem.

Sorry, I didn't check the date of this thread.