This is a log of repairs. It is not a guide, and it is not a shop. I write down what a book looked like when it arrived, what I did, and what the result looked like six months later — which is the part nobody writes down, and the part that turns out to matter most.
I do not sell repairs through this page and I do not take commissions. There is no contact form. There are no photographs, because I found that photographs made me stop noticing.
Recent entries
Linen spine linings, and why the third attempt held
· 8 h over three days · a 1931 poetry volume, cloth case
Two failures before the one that held. The difference was not the adhesive; it was that I stopped pressing the lining while damp and let it dry under almost no pressure, then pressed it once, cold.
Rust migration in a steel case: eight years of the same damage
· a folio with a steel plate in the board
The rust had moved along the grain of the board rather than across it. I describe the pattern because I did not expect it, and because the only fix I found was to stop trying to neutralise it.
A tide mark on a 1908 almanac, and the decision to leave it
· no intervention on the staining
The staining had become structural: the boards bent along it, and flattening them would have broken the text block at the fold. I wrote down what I left alone.
Wheat paste made twice a week, and the reason I stopped making it in bulk
· adhesive · small batch
Twelve litres a week became two litres, twice. The loss in time was smaller than the loss in quality from paste that had sat overnight in a cold room.
Numbers from the bench
| Entries kept | since 2020-10-11, 214 logged jobs |
|---|---|
| Jobs I would not repeat the same way | 39 |
| Repairs checked again after six months | 88 |
| Repairs that failed the six-month check | 14 |
| Books turned down | 22 (I keep a reason for each) |
Small changes at the bench
- Job 197 — 2026-11-09 Added the release-paper detail that broke attempt two
- Materials I keep — 2026-11-09 Recorded that the 45 gsm linen is now used far less often
- job 197 written up; six-month check scheduled.
- job 164 passed its six-month check.
Older entry worth knowing about: pressure-sensitive tape, and the four years it took to undo.
What this page is for
For me, mostly. Secondarily for anyone who repairs books and wants to know that a method that works in a textbook can fail on a specific volume for a specific and describable reason. I write those reasons down even when they are embarrassing.