feat: publikacja danych o urzadzeniach i bledach jako cytowalnych zbiorow - #268
Merged
Merged
Conversation
The compatibility list is the one thing this project has that nobody else does - twenty vendors, every row carrying a link to that vendor's own published statement - and it existed only as a file in a git repository. A tool that wanted to consume it had to know about GitHub, and an assistant answering somebody's question had no stable URL to cite. Checked before assuming: docs.msgwing.com served no JSON at all, and the site declared no Dataset markup anywhere. Both files are now published under /data/ and both pages declare Dataset, with distribution pointing at the file rather than at a description of it. That is what lets a crawler, a tool or an assistant tell the difference between a page backed by structured data it can fetch and prose it has to parse - which is the same question the owner raised about AI discoverability, answered with the asset we actually have rather than with more markup on prose. Copied rather than moved. data/ stays the source of truth, because it is what every generator reads and what CI checks; docs/data/ is a mirror that cannot drift, since one script writes it and --check fails when the two differ. The JSON is round-tripped so a stray formatting difference in the source cannot make the published copy look changed when it is not. llms.txt now cites the stable URL rather than a GitHub blob link, which was the only address an assistant had until today.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Lista kompatybilności to jedyna rzecz, jaką ten projekt ma, a której nie ma nikt inny — dwadzieścia producentów, każdy wiersz z linkiem do opublikowanego oświadczenia tego producenta.
I istniała wyłącznie jako plik w repozytorium git.
Sprawdzone przed założeniem:
docs.msgwing.comnie serwował żadnego JSON-a, a serwis nie deklarował znacznikaDatasetnigdzie.github.com/.../blob/main/data/...docs.msgwing.com/data/devices.jsonDatasetOba pliki są teraz publikowane pod
/data/, a obie strony deklarująDatasetzdistributionwskazującym na sam plik, a nie na jego opis. To jest to, co pozwala robotowi, narzędziu albo asystentowi odróżnić stronę opartą na danych, które da się pobrać, od prozy, którą trzeba parsować.To ta sama sprawa, którą właściciel podnosił przy AI — odpowiedziana zasobem, który naprawdę mamy, a nie kolejnymi znacznikami na tekście.
Kopiowane, nie przeniesione
data/zostaje źródłem prawdy, bo to z niego czytają wszystkie generatory i to jego sprawdza CI.docs/data/to kopia, która nie może się rozjechać: pisze ją jeden skrypt, a--checkwywala się, gdy się różnią.JSON przechodzi przez pełny cykl parsowania, żeby przypadkowa różnica formatowania w źródle nie sprawiła, że opublikowana kopia wygląda na zmienioną, gdy nie jest.