[FA] First draft of Chapter2/Page2 (#129)

* merged branch0 into main. no toctree yet.

* updated toctree.

* Updated the glossary with terms from chapter0.

* Second draft in collab w/ @schoobani. Added empty chapter1 for preview.

* Glossary typo fix.

* Added missing backticks.

* Removed a couple of bad indefinite articles I forgot.

* First draft of ch2/p2. Adds to glossary. Trans. guidelines moved out.

* Fixed missing diacritics, fixed the py/tf switch placing. Other fixes.

* Changed the equivalent for prediction per @kambizG 's direction.

* Redid ambiguous passage in translation per @lewtun 's direction.

chapters/fa/TRANSLATING.txt

@@ -0,0 +1,67 @@
+1. Use this style guide for Persian. https://apll.ir/wp-content/uploads/2018/10/D-1394.pdf
+    We deviate from this style guide in the following ways:
+      - Use 'ء' as a possessive suffix instead of 'ی'
+
+2. Don't translate industry-accepted acronyms. e.g. TPU or GPU.
+
+3. Translate acronyms used for convenience in English to full Persian form. e.g. ML
+
+4. Keep voice active and consistent. Don't overdo it but try to avoid a passive voice.
+
+5. Refer and contribute to the glossary frequently to stay on top of the latest
+    choices we make. This minimizes the amount of editing that is required.
+
+6. Keep POV consistent.
+
+7. Smaller sentences are better sentences. Apply with nuance.
+
+8. If translating a technical word, keep the choice of Persian equivalent consistent.
+    If workspace is mohit-e-kari don't just switch to faza-ye-kari. This ofc
+    does not apply for non-technical choices, as in those cases variety actually
+    helps keep the text engaging.
+
+9. This is merely a translation. Don't add any technical/contextual information
+    not present in the original text. Also don't leave stuff out. The creative
+    choices in composing this information were the original authors' to make.
+    Our creative choices are in doing a quality translation.
+
+10. Note on translating indefinite articles(a, an) to Persian. Persian syntax does
+      not have these. Common problem in translations is using
+      the equivalent of 'One'("Yek", "Yeki") as a direct word-for-word substitute
+      which makes for ugly Persian. Be creative with Persian word order to avoid this.
+
+      For example, the sentence "we use a library" should be translated as
+      "ما از کتابخانه‌ای استفاده می‌کنیم" and not "ما از یک کتابخانه استفاده می‌کنیم".
+      "Yek" here implies a number(1) to the Persian reader, but the original text
+      was emphasizing the indefinite nature of the library and not its count.
+
+11. Be exact when choosing equivalents for technical words. Package is package.
+      Library is library. Don't mix and match.
+
+12. Library names are not brand names, nor are they concepts, so no need to
+      transliterate. They are shorter namespaces so just use the English form.
+      For example we will transliterate 'Tokenizer' when it is used to indicate
+      a concept but not when it is the name of a library.
+
+13. You need to surround the entire doc with a <div dir="rtl">, also separately
+      for lists. They render in LTR with the current version of Huggingface docs
+      if you don't do this and the preview will look ugly. Refer to our previously
+      submitted pages.
+
+      Also with function names that include ellipses you need to surround them with
+      a SPAN. i.e. <span dir="ltr">pipeline()</span>
+      These are stopgap measures for the preview to render correctly.
+
+14. Put namespaces in backticks. Library names are shorter namespaces.
+
+15. Persian sentences follow the format Subject-Object-Verb. That means it is
+      generally expected that the sentences end in a verb. Again there is a lot of
+      gray area here and we need to apply nuance. But try to keep that in mind.
+      For example maybe don't end sentences in "هستند یا نه". Maybe, bc if you
+      do, that is your creative choice and we respect that.
+
+16. If a Persian equivalent for a word in the glossary has diacritics, they
+      are important and should be repeated in every occurrence of the word.
+      While we do want to facilitate our reader's access to the English source
+      material by introducing loanwords and transliterating where we can, we
+      also want to avoid confusion in the correct pronunciation of loanwords.

chapters/fa/_toctree.yml

@@ -12,6 +12,8 @@
   sections:
   - local: chapter2/1
     title: مقدمه
+  - local: chapter2/2
+    title: پشت صحنه خط‌تولید
 
 - title: واژه‌نامه
   sections: