City Planning - Layout

[PriscilaCoghlan]

Name: City Planning

Purpose: Fight the squared tiles

Description: Given that you have the chance to create a world from scratch, who will be inhabited with creative and innovative people, why don’t transition from the classical colonialist grid in city planning to an intermediate between that grid and the organic design of medieval cities. I believe that, even if you are innovating in your area of expertise, you should also focus on the design on this new world, encouraging discoveries and new experiences of living in a city that otherwise could not be possible.

I think that choosing the hexagon as a figure and developing from that onwards the results hold a lot of potential. The city can develop itself as a fractal and create an interesting kaleidoscope full of possibilities, for the ones that prefer standard and traditional settings as for the adventurous and more creative ones. Inspirations (images from the internet):

Regarding one question that I received about the difficulties of building in not orthogonal tiles I have to say that it doesn’t necessarily have to be more difficult or impossible. You only have to see the construction in Europe, you will find that almost all the constructions are built in non-orthogonal spaces and that the results contribute to what we call in urbanism “the promenade architecturale”. This promenade has the unique quality of “the discovery” which is lost with orthogonal grids, as every turn is not premeditated and gives the observer a “surprise”.

What’s more, non-orthogonal tiles can contribute to a more diverse and morphologically rich city, stimulating their users to think out of the regular box at the time of building infrastructure. If there is one thing we all are tired to see in our cities is the same squared buildings extruded just from their squared lands to make as much profit as possible, neglecting not only the people that is going to inhabit those spaces but the impact that they make in the cities’ landscape, all the building look alike and the cities lose their appeal.

Of course it would be also a great idea to not make all the districts alike, and vary the way that the "apples" are divided inside, creating bigger and smaller tiles as needed.

Below I attach the research in the matter (I just did screenshots because I am short of time right now, but I think that the concept is clear)

1. Different ways of subdividing the apples in geometrical generating figures, which allows for continuos equal subdivisions:

2. The city planning proposed zomming out:

Different districts:

Also as you can see in the images the city can be organized in poles, which will be dedicated to, for example: music, arts, films..... etc... , creating a "multicentric city" such as (images from the internet): La Plata, Argentina

or Canberra, Australia

or Charles de Gaulle from the Arc of Triumph, Paris, France

People who visit medieval towns for the first time or contemporary suburbs which have an organic framework, often feel lost and disoriented, which can lead to anxiety, fear, and sense of uneasiness. Even though one of this example has an unplanned growth and the other is planned, you can see the dangers of a completely framework-free city, where streets can develop odd parkours and lead to people not visiting them as a result. Lastly this provokes excluded and segregated zones where land price diminishes.

The grid can appear to people a great choice from the efficiency standpoint, engineering, easiness to decodify it for the pedestrian, between other reasons. I believe that those advantages can be discussed as in a grid such as the ones in the colonial cities, which you can see in almost every city of the American continent, every point of the grid, every intersection holds the same value or priority. When you analyze what happens in the cities where priority has been given to certain points in the space, you can see how that has led to some interesting and diverse spaces, which make the city more exciting in counterpoint to what a homogenous grid can offer.

In this sense the grid can improve or degrade the urban environment.

My proposal is an intermediate between these two situations. In the one hand, it gives an organic and non-standardized growing possibilities’ framework and on the other hand it keeps the space apprehensible and easy to use for the visitor.

3. Here you can see that the zone can be devided and present also the tiles that have more squared shape, each of this zone are radiocentric within themselves but contribute to the multicentric city in the whole. Besides this, there are also the "green spaces" to be designed and planned, which will serve as public places for the community and that I will post at a latter stage.

The tiles can be separated from others with some type of low green wall, like bushes:

I am aware that maybe this is a huge complication for the developing team but I believe that it is worth your extra time. And maybe it is still compatible with the sale that you started, as the tiles that you sell in this first round could be the squared tiles shown in the picture above. Leaving the geometrically more complex ones for the next round. Below you can see quick pictures of buildings that I designed in a non-squared tiles, and you can see that the possibilities are endless. proyect 1:

proyect 2:

It's a matter of perspective, it's not the circumstance but how you position yourself in it.

UPDATE: Explore this proposal in 3D https://youtu.be/32-bcET-XLc

[abarmat]

I like it, it's like the tiles in Civilization.

[PriscilaCoghlan]

Thanks @abarmat ! I assume that Civilization is a game, I will google it right now and see what is like =)

[lkngtn]

I wonder if this needs to be a rigid design element rather than simply a natural emergent property. How big are land tiles intended to be? And can parcels of land be sold smaller than the individual tile?

I'm wondering because regardless of if tiles are squares or hexagons, a city zone could be purchased in square tiles by a community of developers and parceled out in hexagonal pieces. By doing it that way, you could have some "cities" which have this hexagonal promenade effect, and some that don't.

I think the system should allow for as much flexibility as possible, rather than promote a specific city design or architecture.

EDIT: Also just wanted to be clear, I would much rather see more organic city layouts like this than endless grids.

[dinomiha]

This is wonderful

[HeadClot]

This is pretty amazing. Got to admit.

[ASINNV]

I agree with @lkngtn, it should be as flexible as possible. Land should be able to be partitioned and sold off by the owner in any size and/or shape they wish for any price they wish. In time, as great cities and hubs of activity emerge, land values in these areas will skyrocket creating an interesting economy much like the one we have in the physical world. Also, the ability to partition land in any shape and size will lead to imperfect segmentation in some cases where less-desirable slivers and arcs of oddly-shaped land will have lower value.

[archnomics]

I second with @lkngtn and @ASINNV

We need to adopt a specific design of tile and specific constraints and limits. If you'd ask me, a 3D cube would be best suited complete with land (ex. tropical), water, air, and its specific microclimate (randomly generated).

There are lots of patterns in urban planning. If the devs would hear me out, "Procedural Generation" will produce these cubes organically just like how the real world was designed. Physics will need to be introduced to create specific limits.

You're introducing something like a Second Life combined with Habbo Hotel but on a VR platform. Habbo Hotel which was from early 2000s, has generic rooms you can choose, and some virtual furniture you can buy, sell, and place on your room. I'll be writing this on my own proposal soon.

[aninaser]

@Yuyujin it’s certainly an intriguing idea, and I certainly agree with procedural generation, but I have a couple of questions about the three-dimensional cube. What benefits do a three dimensional parcel of land bring, especially one as rigid as a cube? Would moving from one face to another not be jarring and unintuitive? And how would these cubes of land be interconnected? Personally I do like the procedural generation and a three dimensional land organization system does intrigue me, but I question its merits.

[ASINNV]

I don’t see the need for three-dimensional parcels, however it is worth noting that if you don’t put a limit on the sky, the owner could theoretically construct, for example, an infinitely tall skyscraper. This might make the purchase of additional parcels completely pointless and lead to far fewer parcel purchases in general. One piece of land could effectively serve as an infinite number of worlds (however small or vertical they may be), separated by some amount of Y-space. That is unless you plan to cap a parcel’s server-load. How were you planning on limiting a parcel’s server-load?

Three-dimensional parcels is one simple, somewhat-claustrophobic way of doing it. Then again, even three-dimensional parcels don’t account for the nature of the world within—for instance, some worlds in a three-dimensional parcel might take tons of server-space and processing power due to high-res textures, wild particle effects, realistic physics systems, high-traffic, etc. whereas others might be a simple sunset skybox with a pool-chair and one user.

So I take that back, three-dimensional parcels don’t solve that problem. But that leads me to wonder, what is your solve here? I’m quite interested!

Sincerely,

Adrian

On Jul 22, 2017, at 5:22 PM, Aunteek Naser notifications@github.com wrote:

@Yuyujin https://github.com/yuyujin it’s certainly an intriguing idea, and I certainly agree with procedural generation, but I have a couple of questions about the three-dimensional cube. What benefits do a three dimensional parcel of land bring, especially one as rigid as a cube? Would moving from one face to another not be jarring and unintuitive? And how would these cubes of land be interconnected? Personally I do like the procedural generation and a three dimensional land organization system does intrigue me, but I question its merits.

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/decentraland/proposals/issues/2#issuecomment-317219469, or mute the thread https://github.com/notifications/unsubscribe-auth/AHwQ0R3mEHXffN7gXMjf8Nut0eXBgEunks5sQpJRgaJpZM4OWHeE.

[aninaser]

@ASINNV I don't think servers are the issue, this is decentralized after all. That being said, I'm not sure how to tackle the computation load problem from a decentralized, blockchain, front either.

[gregrebholz]

With the tokenized LAND contract starting to take shape, I believe this could be implemented in Aetherian or other districts if they were willing to insert an off-chain component, without much additional support from DCL (see issue 2 below).

Initially, after terraforming, the district owner will have their address associated with all of the land parcels staked, as a grid of tiles and their cartesian coordinates. The district owner could develop and deploy a smart contract that maps the [x,y] grid into a series of interlocking hexagons or any other pattern, tokenized in the same manner as LAND, but likely revokable back to ownership of the district under some conditions. The district's contract assigns ownership of those hex parcels to those who staked land to the district, or by other criteria, for developing models in the hex tiles. Those hex tile owners would be responsible for posting their content to IPFS, and storing the reference in the district's contract. The district owner would run an off-chain webserver that pulls IPFS content (hex tile models) and slices the models to DCL grid locations.

From the DCL land contract's perspective, every client query for any grid tile in the district would be sent to the district's HTTPS endpoint, where the off-chain server would be responsible for returning square grids of models, sliced from the hexagon tile models uploaded by parcel developers. The DCL client is still rendering only cartesian land tiles, but the models are designed and uploaded in hex-tile segments. The off-chain server would also be responsible for inserting "public space" between hex tiles - paths with street lamps, benches, roads, or other objects which reinforce the theme of the district. Similarly, districts that wanted to have underground utilities like tunnels or a subway station could shift hex tile models to positive Z values, and return the underground from Z:0 to the "surface", and hex tile models above that surface. The boundary of the district would have to provide a sufficient elevation change to make this integrate with neighboring land tiles at Z:0.

Any of @prilink 's sub-divisible tiles could be implemented in this system, provided their coordinate system can be translated to the cartesian coordinates of DCL. The off-chain server would do the mapping, and the district contract would track hex-coordinate-to-developer ownership on-chain.

Potential problems with this plan include:

1. Scripting in the Silicon Age. Depending on things like the render distance of the client, how scripts are returned and run, etc. it may be problematic to "slice" hexes back onto the grid without breaking scripts. When scripts are introduced in the silicon age, the off-chain server may need to be adapted to provide scripts only in a grid pattern, while models are built in hex patterns. This complicates parcel development.
2. DCL client-side features, such as forced pathways between grid locations, or other design elements forced into the grid pattern, would break this approach. The client will need to render grids as seamless regions to allow this process to translate hexes to the grid.
3. District owner's private key would be in control of the entire district and all of its parcels. If the key is compromised, the content of the entire district could be redirected away from hex tile owners, without recourse. This is also true of any district that intends to manage all of their staked land instead of distributing it to other members. It would be advisable for the district owner to transfer ownership of the district to a secure, audited, multi-signature contract, and for signatories to use hardware wallets or other offline solutions to sign transactions to the LAND contract. The district's hex contract should be similarly managed.
4. Development and operation costs. The hex mapping smart contract is a relatively straightforward adaptation of the LAND contract, requiring only a different method of applying coordinates to hexes. The off-chain server, however, would require extensive development effort to retrieve models from IPFS, slice, and return them as grid tiles, and implement the "public spaces" previously mentioned. The off-chain server would require high availability and sufficient performance to return models to any DCL client within rendering distance of the district. To address this, hex owners could be taxed in Ether, MANA, or other token; requiring them to pay the district contract periodically to retain ownership. The district contract could send funds to a financial officer, to be sold on exchanges for the currency required to pay for hosting, and continued development.
5. Centralization. The vision of DCL was proposed as an eternal and immutable space in a decentralized network, allowing for true "ownership" of virtual real estate. Technically this remains true of the district as a whole, but inserting an off-chain server weakens this philosophy with respect to hex tile owners. Their ownership of a hex would be on-chain, but without the off-chain intermediary, there would be no plausible way to assert that ownership in DCL. I believe this is a reasonable compromise within a district, but others may disagree.

I can assist in the hex tracking smart contract, but lack A-frame or 3d modeling experience to load and slice models for the off-chain component.