ORIGINAL ARTICLE · PERSONAL OBSERVATIONAL STUDY (N = 1)
A personal observational study of one engineer: methods, materials, results, and limitations after seven years of fieldwork.
- 1Independent; Pune, India.
- *Correspondence: hello@aarav.me
- Received
- 14 Apr 2026
- Revised
- 02 May 2026
- Accepted
- 17 May 2026
- Published online
- 23 May 2026
Abstract
Background. The subject (UK, male, 25y) is a software engineer and occasional writer based in Pune, India, who has produced public software artifacts and written work continuously since 2019. No prior longitudinal description exists.
Objective. To describe the subject's work, writing, methods, and observable preferences in a single accessible document, in lieu of the more common “about page.”
Methods. Self-report, supplemented by version-control history and a directory of all shipped artifacts since Jan 2019 (n = 27). Findings are cross-referenced where possible (Section 3).
Results. The subject demonstrates a consistent preference for long maintenance windows over short ones, for the typographic over the decorative, and for first-person essays over technical writeups. Output rate is approximately 1 substantive artifact / 6 weeks (Fig. 1).
Conclusion. The subject is, to the available evidence, a normal but somewhat patient engineer. Replication is welcomed.
Keywords: software · long-form writing · typography · maintenance · Pune · personal websites
1. Introduction
The genre of the “about page” on a personal website has, since approximately 2004, settled into a recognizable form: a first-person paragraph, three social links, occasionally a photograph. This document is an attempt to take the same material seriously enough to format it as a paper [1, 2].
The motivation is straightforward. The subject has been producing work for some time, but has not previously described it in any one place. This document is therefore both an archive and an observation, written in the first person where appropriate and in the third person where the third person seems funnier.
2. Materials
All artifacts described in this study were produced by the subject between
January 2019 and the date of writing (Table 1). Source is held in
version-controlled repositories at github.com/aarav;
written work is held at aarav.me/writing. The subject is the
sole author of all listed items unless otherwise noted.
| Year | Artifact | Form | Status |
|---|---|---|---|
| 2025 | Threadwell | Web app | Operational |
| 2024 | Fieldnotes | Desktop app | Operational |
| 2023 | Marginalia | Static site | Archived |
| 2023 | Three weekend tools | CLI | Archived |
| 2022 | Earlier (consolidated) | Various | Withdrawn |
3. Methods
The subject reports a working method that can be summarized as follows. Each artifact begins as a written brief of no more than 300 words, revised over approximately one week. Implementation follows, typically in 4–8 week sprints separated by 1–2 weeks of explicit non-work (Fig. 1). Maintenance is treated as a first-class activity and allocated 30–40% of available time, in accordance with [3].
Tooling, where it can be observed, is conservative. Editors: Neovim and VS Code. Languages: TypeScript, Go, Python. Databases: Postgres, SQLite. Hosting: a single small Linux VM since 2021, unreplaced.
4. Results
Over the period of observation, the subject produced 11 software artifacts (Folder I, archived elsewhere) and 14 published essays. Of the software artifacts, 8 remain in active use; 3 have been withdrawn. Of the essays, all 14 remain online; 4 have been substantively revised at least once (Table 2, supplementary).
A consistent stylistic preference is observable: long maintenance windows (median: 26 months from first commit to last), restrained typography (two typeface families per project, no exceptions), and a refusal to ship decorative animation. The subject reports this is intentional.
5. Discussion
The subject's pattern of work is unremarkable in any individual particular but unusual in the aggregate. The combination of (a) sustained low-tempo output, (b) long maintenance, and (c) a willingness to write about the work afterward is, in this author's experience, rare enough to be worth recording. Whether it generalizes is left to the reader.
Limitations. Self-report; N = 1; no control. The author and subject are the same person; the reader is invited to apply the usual discount.
6. Conclusion
The subject is, on the evidence available, a normal but somewhat patient engineer. He may be reached, in either capacity, at the address given in the title block. Correspondence is welcomed; replies are slow.
† References
- Krug, S. (2014). Don't make me think, revisited. New Riders.
- Bringhurst, R. (2004). The elements of typographic style. Hartley & Marks.
- Hunt, A. & Thomas, D. (1999). The pragmatic programmer. Addison-Wesley.
- Newport, C. (2016). Deep work. Grand Central.
- Mehta, U. (2024). On the long maintenance period. aarav.me/writing/011.