Marcos Caceres wrote:
Hello again,
This is a request for review of the *Widgets 1.0: Packaging and
Configuration* Spec prior to publication:
Some comments:
In section "1.2 Definitions":
| A widget resource is digitally signed is it contains [...]
"is it" -> "if it".
"6. Configuration Document":
| Authoring requirement:It is recommended [...]
Missing space after the colon.
The first example uses a "title" element, which should be "name".
The second example does not declare the namespace, but "6.1 Namespace"
says that this must be done.
"6.1 Namespace":
| [...]. If it is omited [...]
Missing a "t".
"6.2 Attribute Values and Types", "Boolen Attributes":
| [...]. A valid boolean value is must is a string [...]
Sounds strange ;-)
For the element types, some say "Contexts in which this element must/may
be declared" and some says "present" instead of "declared". The same word
should be used for consistency. (The word used in the HTML 5 spec is
"used" btw. That may be even better).
The name element is the only one that has "Expected children: None.". It
should be removed.
"7. Steps for Processing a Widget Resource":
| The steps listed as follows and are specified verify the
| zip archive its file entriesin detail in the following sub-
| sections.
"entriesin" -> "entries in". This sentence also sound extremely weird IMO.
Is it missing some commas or something?
"7.3 Defaults":
| Icons
| empty list of URIs
"Empty list" alone sounds better IMO, because if it's empty, there
certainly aren't any URIs.
"7.4 Parsing Rules":
| User agents may optimize any of the parsing [...]
This is pretty much a repetition of the first note under "7. Steps for
Processing a Widget Resource".
"Step 6 - Process the configuration document", first example:
| [...] walking an through an element's children [...]
The first "an" should be removed.
Also, in the same example there should be a "name" element, not "title".
In the schema, "title = element name" should be "name = element name".
Nitpick: Some headers (e.g. "4. Zip Archive") are bolder than others
(because of <dfn>'s I presume). Maybe this should be fixed for consistency.
--
David Håsäther
