Merge branch 'spec_rewrite' into dev

docs/README.md

@@ -26,7 +26,7 @@ extra_css: [extra.css]
         </div>
         <div class="col-md-4 text-center">
             <br><br>
-            <img src="druids.svg" alt="" class="img-responsive">
+            <img src="seqcol_abstract_simple.svg" alt="" class="img-responsive">
         </div>
     </div>
     <hr>

docs/recipes.md

@@ -0,0 +1,36 @@
+
+
+## Operations enabled by seqcol, organized by input
+
+### Seqcol digest as input
+
+* **seqcol digest -> sequence digests**: Given a seqcol digest, sequence digests for all contained sequences can be retrieved by the `/seqcol/:digest/:reclimit` endpoint by setting the `reclimit` to 1.
+* **seqcol digest -> sequences**: Given a seqcol digest, sequences themselves for all contained sequences can be retrieved by the `/seqcol/:digest/:reclimit` endpoint by setting the `reclimit` to 2, or omitting it.
+* **seqcol digest -> metadata**: Retrieved by using the `metadata` endpoint.
+* **seqcol digest -> aliases of seqcol**: Does the API offer this? Should be consistent with refget.
+* **seqcol digest -> metadata of seqcol**: Provided by the `metadata` endpoint.
+* **2 seqcol digests -> assessment of compatibility**: Provided by the compatibility function of the seqcol protocol. First, retrieve the annotated sequence digests for each sequence collection using the *Retrieval* function, then use the *Comparison* function to assess compatibility.
+* **seqcol digest + sequences -> validation claim**: Use the *seqcol algorithm* to compute the digest of the set of sequences, and then use the *comarison* function to validate (or, if you validation requires strict identity, just confirm that the digests match).
+
+
+### Sequence collection as input
+
+* **sequences -> refget digests**: Individual sequences can be converted into refget digests using the refget algorithm
+* **sequences -> seqcol digest**: Sequence collections can be converted into seqcol digests using the seqcol algorithm. To obtain a sequence digest, follow the algorithm outlined above, sequences -> refget digests -> seqcol digest.
+
+### Refget digest as input
+
+* **refget digest -> seqcol digests containing this sequence**: Use the `containing_collections` secondary endpoint. 
+* **many refget digests -> seqcol digests containing all of these sequences**: Repeated application of `containing_collections` endpoint?
+
+## Coordinate system as input
+
+* **coordinate system -> seqcol digest**: If you have only a coordinate system (lengths with no sequences), you may still compute a seqcol digest using the seqcol algorithm, because the sequences themselves are optional. Coordinate systems can therefore be uniquely encoded, retrieved, and compared using the seqcol protocol.
+
+
+## Related operations: What seqcol does *not* do
+
+* **sequences -> sequence names**: This is the refget reverse lookup.
+* **alias -> seqcol digest**: Will the API offer this?
+* **seqcol digest -> other seqcol digests compatible at some level**: The database does not store comparison flags, so if you're searching for compatible collections you'd need to retrieve and compute the compatible yourself using the comparison function.
+