Useful work should leave something behind.

Initiatives, research, projects and collaborations built around practical value rather than personal visibility.

Fig. 1

James Richman at work, November 2025. Printed in the register’s green ink.

Problem

Not every project has to be optimized for a transaction. Some work is worth doing because it creates knowledge, opens access, supports a community, or makes a useful idea easier for other people to use.

Approach

Each initiative is recorded in the same fields as this one: the problem, the approach, the people involved and what came out of it.

People involved

James Richman, CEO of OTLEN. Collaborators are named on the records they work on.

What came out

This site is the home for that category of work: a charter in five articles, and a register that opens with this record.

The charter · Articles 1 to 5

Five articles, written before the first initiative.

Article 1

Practical contribution

The most useful initiatives tend to be concrete. A resource people can actually use. A research project that clarifies a confusing topic. A tool that removes friction. A collaboration that gives capable people a better way to work together.

The test each one has to pass
Form	Useful when
A resource	Someone uses it without needing to ask the author.
A research project	A confusing topic is clearer afterward for people outside the project.
A tool	A step disappears from someone’s week.
A collaboration	Capable people work together on better terms than before.

Article 2

Knowledge as infrastructure

Good information compounds. A clear explanation can save thousands of people from repeating the same research. A well-documented process can outlive the person who created it. The goal here is to publish assets that remain useful after the original project has moved on.

The case on record: an open archive of protein structures

19717
structures in the Protein Data Bank when it opens at Brookhaven National Laboratory. [1]
199624hours
Genome project leaders agree in Bermuda to put human sequence data in the public domain. NHGRI’s 1997 policy sets release within 24 hours. [2]
20210.96Å
median backbone accuracy of AlphaFold2 at CASP14, a model trained on structures from the Protein Data Bank. [3]
2022200million+
predicted protein structures, free to use in the AlphaFold Protein Structure Database. [4]
2023200,069
entries in the Protein Data Bank archive on 11 January 2023. [5]
2024Nobel
Prize in Chemistry: half to David Baker for computational protein design, half to Demis Hassabis and John Jumper for protein structure prediction. [6]

Every structure deposited since 1971 became training data for a model nobody could have planned in 1971. Dots are placed by year: fifty years of deposits, then four milestones in four years.

Article 3

Collaboration

Interesting work often emerges between disciplines. Technologists understand systems; operators understand constraints; domain experts understand what matters in practice. Bringing those perspectives together can produce something neither side would have designed alone.

Technologist: Each remittance file carries a reason code for every denied line.
Operator: The team works denials in the order they arrive, by hand.
Domain expert: Some denials are worth appealing, and some are not.
Built together: A queue ordered by what each denial is worth recovering.

Illustrative example from healthcare revenue operations: one denied claim, three views, one thing none of them would have built alone.

Article 4

Focused experiments with public value

Not every initiative starts as a foundation-sized project. It can begin as a first prototype, a dataset, a guide, a workshop, or a shared resource. The standard is usefulness first, scale second.

ISO 216 paper sizes, used as a metaphor. Each sheet is exactly half the one above it, and each is still a complete sheet.

Article 5

A record of intent

This site will document initiatives in enough detail that the reader can understand the problem, the approach, the people involved, and what came out of it. The emphasis is on substance rather than ceremony.

Record no. 001Opened · not yet

Problem

Approach

People involved

What came out

The next record stays blank until an initiative has a result to report. Record 000, filled in, is the form at the top of this page.

Useful work compounds when it becomes something another person can pick up and use.

This site is a starting point. It grows with original articles, projects, research, tools and case studies as each initiative earns a record.

CEO of OTLEN · Singapore · October 2026

Sources · 6

Where the archive figures come from.

RCSB Protein Data Bank. PDB history: established in 1971 at Brookhaven National Laboratory with 7 structures. rcsb.org
NHGRI. Reaffirmation and extension of NHGRI rapid data release policies (Bermuda Principles, 1996; release within 24 hours, 1997). genome.gov
Jumper J et al. Highly accurate protein structure prediction with AlphaFold. Nature. 2021;596:583–589. doi.org/10.1038/s41586-021-03819-2
EMBL-EBI and Google DeepMind. AlphaFold Protein Structure Database, over 200 million predicted structures. alphafold.ebi.ac.uk
wwPDB. PDB reaches a new milestone: 200,000+ entries, 11 January 2023. lists.wwpdb.org
The Royal Swedish Academy of Sciences. The Nobel Prize in Chemistry 2024, press release, 9 October 2024. nobelprize.org