....is a project I'm working on.

Born out of the habitual desire to have an easy way to answer the questions "what can I do with...", "how can I get...", and "what does... do to...", I have decided to compile a database of all sound changes, and create a tool to search any part of it. Because the Index Diachronica, for all its advantages, isn't indexed, and it notably doesn't talk about things like frequency of specific sound changes, which I feel should be an important part of this project.

There is currently a prototypical interface at http://173.31.182.88:8080/grimm/search.php (I know; I've been meaning to register a domain) which has a small database of some Ancient Greek sound changes.

There are three boxes: source, result, and conditioning environment; you can search one or more of these at once. Multiple phonemes can be entered in each box, separated with commas or spaces. There are tables of IPA consonants and vowels, which allow one to more easily enter special characters. There are also modifier letter boxes which allow one to search for e.g. "tʰ dʰ tʃʰ dʒʰ kʰ gʰ" by simply entering "t d tʃ dʒ k g" and checking the box.

TODO LIST:

1. Fix the search function so ordering works correctly (currently searching for st brings up ts, which is not desirable.)
   
2. Fix the ordering of results as per the footnote.+Results are now ordered by how exact of a match they are wrt having modifiers, having surrounding consonants, and being MOA/POA matches. The latter two aren't distinct (yet?)
   
3. Make the modifier checkboxes work.
   
4. Add GET compatibility so people can make requests using Pylaut or whatever software (assuming pth13g don't want me to rewrite the program into Python instead)
   
5. Fix the search function so pre/post modifiers work (e.g. aspiration—are there any more of these that contrast?)
   
6. Implement editing
   
7. Make the whole thing look pretty; I would appreciate anyone's help on this one as right now I am using minimal CSS
   
8. Come up with a better name as GUSCD is not very aesthetic - pthag has suggested Grimm, which I approve of
   
9. Sort sound changes by number of families they are attested in
   
10. other things which I am not remembering/will add later
    

This is very cool and I'd be happy if some nice day the whole Index Diachronica would be searchable this way. For API interoperability, you may want to make JSON output possible, I suppose. Python has Ways™ to read that.

I went with option (c) above. The source/result/environment search now works fully, and one can make GET requests, like so: http://173.31.182.88:8080/guscd/action.php?source=e&result=i&env=n

I will be away on vacation for the next week, and unable to access the computer that the server is running on, and I don't have time to write all of the editor functions before then. So if anyone wants to help transcribe the Index Diachronica, or add their own [potentially more reliable] sound changes—which I would appreciate very much—here is a Google spreadsheet with which to do so.

As the curator of the Index, I'd love to help any way I can.

By way of news related to the Index, chri d. d. has created a searchable online version of the (as well as rendering the latest version in XeLaTeX so it is now Unicode-compliant), and a reddit user is working on a database version using JSON. I also hope to release Version 11.0 sometime this weekend.

I've seen the searchable Index. I'm hoping to provide more functionality, though—for example, allowing one to search source and/or result, showing the most relevant changes first, grouping changes by the change itself rather than by family, allowing editing, etc.

Editing is, I think, fully implemented. You can add new entries, and you can edit entries that show up in your results list.

I will probably start moving stuff over from the spreadsheet soon.

http://173.31.182.88:8080/grimm/search.php