update content

src/docs/articles/taytay-sings-the-budget-blues.md

@@ -17,20 +17,21 @@ Information Act requests, where the information sought is in electronic format.
 If we remove the limitation about it applying in legal proceedings then we're left with a definition
 that captures all of us: `discovery [...] where the information sought is in electronic format`.
 
-In this article I'm going to demonstrate how you can use the e-Discovery tools in Describo as part
+In this article I'm going to demonstrate how you can use the e-discovery tools in Describo as part
 of your work. You'll see how to extract information and reason about it. You'll also learn how to
 interpret a dataset through the lens of another. Finally, you'll learn the meaning of the title.
 
 ## Audience: about you
 
 Does this sound familiar?
 
--   You've collected a bunch of primary source materials in electronic form (newspaper articles,
-    manuscript pages etc) and are now embarking on a journey to find out what is contained in them.
+-   You've collected primary source materials in electronic form (newspaper articles, manuscript
+    pages etc) and are now embarking on a journey to find out what is contained in them.
 
--   Maybe you've found yourself looking at that folder of Microsoft Word documents on your company
-    fileshare. The documentation has captured all of the information about a project that the
-    company worked on but with only the filenames to go on, it's more data than it is information.
+-   Or maybe you've found yourself looking at that folder of Microsoft Word documents on your
+    computer or company fileshare. The documentation has captured all of the information about a
+    project that the company worked on but with only the filenames to go on, it's more data than it
+    is information.
 
 -   Or maybe you've just received the fifth revision of that 150 page policy manual for final
     checking and verification. You've already read it closely at least two times (it got harder and
@@ -64,11 +65,11 @@ contained in the documents.
 
 In the following screenshot you'll see that 4 documents have been selected for interrogation. Note
 that the Budget Overview document is a Microsoft Word document and the other 3 are PDF documents.
-Within seconds the assistant has provided a page count (371 pages) and a detailed summary.
+Within seconds the assistant has provided a page count (71 pages) and a detailed summary.
 
 <ImageComponent src="/images/articles/taytay-sings-the-budget-blues/budget-image1.webp"></ImageComponent>
 
-What else we can discover? In the following screenshot we have identified the largest contributors
+What else can we discover? In the following screenshot we have identified the largest contributors
 to revenue and the largest costs whilst also drilling down into the details of the excise
 contribution to total revenue.
 
@@ -112,7 +113,7 @@ augment your conversation with its own knowledge.
 <ImageComponent src="/images/articles/taytay-sings-the-budget-blues/taytay-image3.webp"></ImageComponent>
 <ImageComponent src="/images/articles/taytay-sings-the-budget-blues/taytay-image4.webp"></ImageComponent>
 
-And here's the conversation (note the licensing statement in the header):
+And here's the conversation (again, note the licensing statement in the header):
 
 <iframe
   class="border border-solid border-gray-400 p-2"

src/docs/guide/assistant-supported-discovery.md

@@ -27,7 +27,7 @@ For more information see:
 
 # Assistant enabled e-discovery
 
-Using the describo assistant you can interrogate your content to quickly establish the themes and
+Using the Describo Assistant you can interrogate your content to quickly establish the themes and
 narratives that you might want to describe.
 
 :::tip info
@@ -45,43 +45,43 @@ Get started by selecting the files and folders containing the content you wish t
 
 <ImageComponent src="/images/guide-discover/discover1.webp"></ImageComponent>
 
-Start the assistant. Notice that the left panel changes to show you the context (the files that will
-be interrogate and any that will be excluded - either an unsupported format or something went wrong
-when extracting the data).
+Start the Assistant. Notice that the left panel changes to show you the context (the files that will
+be interrogated and any that will be excluded - either because it's an unsupported format or
+something went wrong when extracting the data).
 
 Underneath the context are options for where to write the metadata you wish to create. The following
 images show you the options when one file is selected or more than one. In all cases, you can update
 the description of the main index. If one file is selected you can choose to update the description
 of that file. If more than one file is selected, you can create a new relationship and edit that
 description; or, use an existing relationship that is associated to the selected files.
 
-In this way, you can interact with the assistant and capture the information as it suits you.
+In this way, you can interact with the Assistant and capture the information as it suits you.
 
 <div class="flex flex-col space-y-1 lg:flex-row lg:space-x-1">
     <ImageComponent src="/images/guide-discover/discover2.webp"></ImageComponent>
     <ImageComponent src="/images/guide-discover/discover3.webp"></ImageComponent>
     <ImageComponent src="/images/guide-discover/discover4.webp"></ImageComponent>
 </div>
 
-In the next two images we see a conversation with the assistant and the associated metadata that has
+In the next two images we see a conversation with the Assistant and the associated metadata that has
 been created.
 
 <ImageComponent src="/images/guide-discover/discover5.webp"></ImageComponent>
 <ImageComponent src="/images/guide-discover/discover6.webp"></ImageComponent>
 
-In some instances you might just want to use the assistant to uncover some information and then be
+In some instances you might just want to use the Assistant to uncover some information and then be
 able to take away a PDF of the conversation. This is possible by selecting `Download as PDF`.
 
 <ImageComponent src="/images/guide-discover/discover7.webp"></ImageComponent>
 
 The first thing you will see is a dialog specifying which licence to apply to the PDF document. The
-default selection enables anyone to use the work and / or creat derivatives provided that they
+default selection enables anyone to use the work and / or create derivatives provided that they
 include appropriate attribution. You can control that in the dialog.
 
 <ImageComponent src="/images/guide-discover/discover8.webp"></ImageComponent>
 
-And the document (the final page/s will always be the context so you know what was interrogate in
-that session).
+And following is the document with the licence in the header (the final page/s will always be the
+context so you know what was interrogated in that session).
 
  <iframe
   class="border border-solid border-gray-400 p-2"

src/docs/guide/transcribing-content-assistant.md

@@ -19,37 +19,42 @@ accurate because of their level of detail or specificity. You should not rely on
 independently confirming their accuracy. The Services and any Outputs may not reflect correct,
 current, or complete information.
 
+For more information see:
+
+-   [Assistant enabled e-discovery#interacting-with-the-assistant](/docs/guide/assistant-supported-discovery.html#interacting-with-the-assistant)
+-   [Tips for using the Assistant](/docs/guide/prompt-engineering.html)
+
 :::
 
 # Using the assistant
 
 As you are transcribing your content, you are also marking up entities that are found in that
-content. That markup, as seen in the previous tutorial is stored in the metadata and as data
-attributes in the transcription.
+content. That markup, as seen in the previous page is stored in the metadata and as data attributes
+in the transcription.
 
-The next step is developing and enriching your description of the content by using the assistant to
+The next step is developing and enriching your description of the content by using the Assistant to
 interrogate the text; potentially from differing perspectives.
 
-In the following images, the assistant is first used to summarise the text and describe the main
-topics. Then, acting as an anthropologist (ie prompting it to assume a particular role), the
+In the following images, the Assistant is first used to summarise the text and describe the main
+topics. Then, acting as an Anthropologist (ie prompting it to assume a particular role), the
 assistant is again asked what the main topics are. After that, the assistant is asked to assume the
-role of a social scientist to explain the narrative with respect to any cultural symbolism.
+role of a Social Scientist to explain the narrative with respect to any cultural symbolism.
 
 <ImageComponent src="/images/guide-transcribing-content-assistant/assistant1.webp"></ImageComponent>
 
-In each case, the assistant provides a detailed answer with evidence from the text to support its
+In each case, the Assistant provides a detailed answer with evidence from the text to support its
 commentary.
 
-As the conversation is happening, all or part of it can be transcribed into the description fields
-on the right hand side. In addition, defined terms can be created from this content (or whatever you
-wish to define yourself) and associated to the file.
+As the conversation is happening, all or part of it can be transcribed into the description fields.
+In addition, defined terms can be created from this content (or whatever you wish to define
+yourself) and associated to the file.
 
 When we navigate back to the describe tab we see the descriptions and defined terms that we added
 previously.
 
 <ImageComponent src="/images/guide-transcribing-content-assistant/assistant2.webp"></ImageComponent>
 
-Moving onto another image we can again engage the assistant to describe the text and when we go to
+Moving onto another image we can again engage the Assistant to describe the text and when we go to
 add defined terms, not only can we create new ones but we can also look up existing terms that have
 been defined in the metadata. Notice the dropdown in the image shows the defined terms we created
 for the first page?