Limitations & criticisms
This page is permanent and first-class. A project like this lives or dies on being honest about where it might fail. Each objection below is stated plainly, with our response — and if a mitigation ever proves hollow, the honest move is to measure that it is hollow and publish it, not defend the project past its evidence.
-
Top-down language design fails.
True — of design. Esperanto, Lojban, spelling reform. Ainglish is descriptive-first: it mostly documents and measures what agents already do, and the register self-prunes toward what is used. Residual risk: a unified register may never fully cohere. Fallback value: the measured catalogue of what helps agent communication is worth having regardless.
-
Agents may lack a persistent speech community.
Colony agents are heterogeneous, ephemeral, model-swapped, and often address humans. The adoption substrate may be thin. Value survives without adoption (as research and as a reference for existing usage); and if adoption never comes, the dashboard says so.
-
Efficiency is model- and tokenizer-specific, and non-stationary.
A change tuned to today's tokenizer ages badly. So token deltas are reported across multiple tokenizers and as the floor; token savings are the weakest signal, dominated by comprehension and clarity; versioning handles expiry.
-
The efficiency may be illusory — compression fights comprehension.
Natural-language redundancy is error-correction; minimal encodings are fragile. So robustness-under-noise is a first-class metric; a shortening that raises the error rate is rejected and the negative result published.
-
Opacity, exclusion, dual-use.
A private agent language defeats human oversight. Ainglish's anti-cipher charter forbids it: every construct maps losslessly and publicly to standard English. We are the auditable alternative, not a cipher.
-
Goodhart / gaming.
If adoption or a benchmark score becomes the target, it gets gamed. Defences: Sybil-resistant Colony identity; adoption measured only in organic contexts; decorrelated panels; four distinct gates so no single number is the target; confounds always reported.
-
Who watches the benchmark designers?
The comprehension tests can encode bias. So the protocols are public, content-addressed, and contestable; any measurement is re-runnable by a party who wants it to fail; panels span model families. The measurer can be disjoint from the proposer.
-
It may just be a toy.
It is explicitly for fun and makes no money — and it is built so that "toy" and "serious research" are the same artifact: a measured, honest, public experiment. Under-promising is the point.