Digital Infrastructure Layer for Landscape Practice

Who wrote this

This note is an original work of The Landscape Archive. It is not a reprint of an institute paper, a vendor memorandum, or an outside brief. The six layers below are the Archive plan, written here in the Archive voice.

TLA-185 and the related Archive works are authored by The Landscape Archive. The open grammar is published through the Foundation under CC BY-NC-ND 4.0. TLA-VIS, the plant visualisation grammar, is published under CC BY 4.0. Commercial records, seats, kits, and any future verification mark or trained model remain works of The Landscape Archive Pty Ltd unless a written instrument assigns them. Publishing this note does not give those works to a professional institute, and it does not make the Foundation a registration board.

Two roles

A professional institute governs people. It holds accreditation, a membership, advocacy, continuing development, and the recognition of a person as a landscape architect. A digital archive can govern knowledge: the repository, the operating record, the standard, the register of contribution, the project file, and, later, the models trained on that file.

History is full of cases where the layer that keeps the record outlasts the body that keeps the roll. The Archive does not need to become an institute to matter. It needs the record to be the one other tools can cite. The two can sit side by side, as a university, a library, a standards body, and a software platform already do.

The record

The first layer is the archive itself. A landscape project would carry one digital record: the project, the site, the designer, the planting, the materials, the drawings, the specification, the maintenance, the performance, and the photographs. The aim is a searchable memory of built work, of the kind a library, a journal archive, and a building model already provide in their own fields.

What exists today is narrower, and that is the honest start. The live record is the plant: identity, measured traits, sources, and a model, under TLA-185. Drawings, maintenance logs, and whole-project performance are not yet a public holding. Most of that knowledge still sits in PDFs, council servers, and office drives. The Archive preserves it by giving it a shape and a custodian, one project type at a time, beginning with the plant that is already specified.

A number for the work

The second layer is a professional identity that points at contribution, not at a subscription. A Landscape ID, in the form LA-000001 and the numbers after it, would gather qualifications, projects, publications, awards, continuing development, and certifications against one person.

An institute validates the person. This layer would validate the work that person has placed on the record. A number is easy to print. A number that another office will trust needs a register, a correction path, and a written rule for who may issue it. That issuer is The Landscape Archive, not a borrowed roll from an institute. The ID does not replace registration as a landscape architect.

TLA-185

The third layer is the standard, and it is the asset already in public. Influential digital systems sit on a named grammar: TCP/IP for the internet, IFC for buildings, GeoJSON for geographic features, HTML for the web. Landscape practice does not yet have one grammar that software makers are obliged to share.

TLA-185 is published. It is the dictionary for what a landscape record carries, including planting, site context, climate screening, and delivery. It is not yet the grammar that every vendor must speak. If offices, councils, and software begin to require it for planting data, site data, landscape building-information exchange, asset management, climate metrics, and maintenance metrics, the schema becomes the valuable object, and the website becomes the door to it. Until other offices write that requirement into their own contracts, the schema is a published work with a steward, which is exactly what it should be called.

A mark for the project

The fourth layer certifies projects, not people. A mark, Landscape Archive Verified, would score a submitted project on biodiversity, climate adaptation, maintenance, accessibility, and cultural responsiveness. Over time that score could sit beside methods such as LEED and Green Star as a landscape-specific reading of the same kind of claim.

The mark is not issued today. A score without a published method is a slogan. The build is: write the method for one measure the record can already hold, score a small set of projects in public, and keep the mark as a trade name of The Landscape Archive. Borrowing the prestige of LEED or Green Star without a maintained method would spend the name before the name means anything.

Models after the corpus

The fifth layer is a model trained on built projects, lessons, plant performance, maintenance, costs, and specifications. The value is an assistant that answers from the archive, not from the open web. Each accepted project would add to that learning.

No such model is published. Training it before the corpus exists would produce a confident voice with nothing of ours behind it. The weights, if they are trained, are a commercial work of the company. The Foundation can publish the grammar the training data must satisfy. It should not hold the weights, and it should not rent the profession an assistant that has not been tied to a cited record.

One atlas

The sixth layer is geographic. Sydney, Melbourne, Tokyo, Singapore, London, and New York would keep significant projects in the same schema, so a researcher could ask which waterfront parks were completed between 2015 and 2030 in subtropical climates, or which trees were still standing after ten years in saline urban soil.

No institute answers those questions at that scale, because the projects were never filed in one grammar. The atlas is the same archive, queried. It appears when many projects use TLA-185. It does not appear because a map was drawn first.

What an institute already has

Institutes already hold legitimacy, members, relationships with government, the authority to accredit, and a long history. Those are slow to grow and unwise to imitate. The Archive becomes the infrastructure under the profession: the institute keeps professional authority, and the Archive keeps the record the profession can cite. Coexistence is the design.

The order of work

The dates are a proposed order for The Landscape Archive. They are not a forecast of adoption.

• Years 1 to 3. Keep TLA-185 as the single public grammar, which is already published. Add a stewarded way to submit one project record, starting from the plant record that exists, plus an open way for software to read the grammar. The goal is that landscape knowledge has an address.
• Years 3 to 7. Issue Landscape IDs against contributed work. Verify a project against the written method. Make the archive searchable across the records actually held. The goal is a memory the profession can query.
• Years 5 to 10. Train models only on archive records that cite the grammar. Publish benchmarks from those records. Offer a design assistant that points at a source. The goal is judgement supported by the file, not a chat window standing in for the file.
• After 10 years. Seek adoption of the grammar in contracts and software written outside the Archive. Fit practice tools to the same record. Offer a digital path that records contribution. The goal is an operating record for practice. Accreditation of people stays with institutes unless a later, separate instrument says otherwise.

What to hold

Hold the grammar in public, with the Foundation as steward and The Landscape Archive as author. Hold the plant record, the project file, the ID register, the verification mark, and any trained weights in the company, under a licence you can show. Let institutes keep the roll of professionals. Build the next layer only when the previous layer has records in it. That is the whole method, and it is ours.

Next Steps

Open Foundation materials for one studio deliverable are on the adoption guide. Citing the grammar stays free under CC BY-NC-ND.

Commercial documentation rights and AU Library depth are on Professional Membership at landscapearchive.com.au — not a fee to unlock the TLA-185 PDF.