W3C home > Mailing lists > Public > public-appformats@w3.org > March 2007

Re: XBL2 Comments

From: Jonas Sicking <jonas@sicking.cc>
Date: Mon, 26 Mar 2007 19:56:48 -0300
Message-ID: <46084FB0.1090406@sicking.cc>
To: Ian Hickson <ian@hixie.ch>
Cc: public-appformats@w3.org

These things take a long time to reply to, sorry about that. Especially 
when writing this from the beach of Buenos Aires :)

Ian requested that I split my comments into ones that are important to 
me as an implementor, and ones that are not. I also made a category for 
ones that are more comments that I think benefit users for one reason or 
other, and a category for clarification requests (i.e. comments not 
requesting functional changes to the spec, but editorial ones)

== Requests for clarification (or at least I think they are) ==

 >> 1.4.2)
 >> Why are selectors case insensitive?
 > That's a question for the CSS working group.

Since you're only referring to prefixes, please state so in the text. 
However I disagree that the spec actually says so. The spec for the 
@namespace rule does indeed say that it is case insensitive. However the 
selectors spec doesn't really say anything one way or the other.

 >> 3.6)
 >> What functions should be called if the xblEnteredDocument function
 >> implementation moves the bound element to a different document?
 > An xblLeftDocument and an xblEnteredDocument. These methods are called
 > when everything else is stable. So nothing would happen until the
 > first xblEnteredDocument function returned, then, after everything has
 > settled down, the UA would fire an xblLeftDocument, and if nothing
 > changed in that time, an xblEnteredDocument. Does that make sense?

I did realize that since the node is now in a new document it'll be a 
completely different binding that gets attached to it.

It may still be a good idea to state the xblLeftDocument can be called 
even when the bound element is positioned inside a document in this case.

 >> 4.1)
 >> What is "xml:base data"? Same thing as the xml:base attribute?
 > It's the data resulting from the use of xml:base. So in your
 > implementation, probably just the attribute, yes.

I'm not following what the difference is. In any case the term "xml:base 
data" is not defined anywhere in the spec. I suggest either using the 
term "xml:base attribute" or defining the term "xbl:base data" properly 
in the spec.

 >> 4.4.1)
 >> "Matching of the elements to the selector is done without taking into
 >> account the shadow tree in which the content element itself is found".
 >> Are other bindings on the bound element taken into account? In general
 >> this part feels too vaguely defined, what if there are implicitly or
 >> explicitly derived bindings, are they taken into account?
 > I don't understand the question. Could you give an example that you
 > believe is ambiguous?

Since it says "without taking into account the shadow tree in which the 
content element itself is found" it sounds to me like some other shadow 
tree is taken into account. My question is which ones. I take it no 
bindings for the current explicit parent needs to be taken into account, 
i.e. you can act as if the child is an explicit child of the element 
that you're currently creating the shadow tree for.

Let me see if I can come up with some better wording that makes this 
more clear. Suggestions of course welcome :)

 >> Same thing with <inherits> inside <content> and <content> inside
 >> <inherits>. In other words, what does the following three templates
 >> do? <inherited><content/></inherited>
 > The <inherited> element contents are used when there is no base
 > binding. If a node is assigned to a <content> element that doesn't end
 > up in the final flattened tree, well, it doesn't end up in the final
 > flattened tree.

So in other words the explicit children are distributed to the "DOM Core 
tree", independent on if parts of that tree is then hidden in the 
flattened tree. This would be good to state explicitly.

 >> 4.5)
 >> The first two paragraphs make it sound like the DOM is mutated, which
 >> should not be the case.
 > I'm not sure how to fix this. Do you have any suggested text? Note
 > that there is an additional paragraph there now that should makes
 > things a bit clearer.

Add a sentence somewhere in there that says something to the extent of: 
"These steps described here do not change the actual bound document or 
the shadow trees. They are simply describing the process that would 
generate a tree that is like the final flattened tree. The actual DOM 
nodes in the bound document and in the shadow trees are left intact"

 >> 4.7.1)
 >> It is unclear if when allow-selectors-through is true and an explicit
 >> child of a bound element is inserted in a binding using <content> if the
 >> parents of the <content> take part in parent chain as far as selector
 >> matching goes.
 > I don't understand what interpretation could lead you to assume that
 > it does. Could you elaborate on why you read it that way?

There's nothing in the text to directly suggest it. However I think it's 
easy to get the impression that the > operator works on the flattened 
tree since many other things operate on the flattened tree, and the > 
operator does partially too.

So I would suggest that it was explicitly stated that that is not the case.

 >> 4.10)
 >> "Sheets are always walked from the innermost shadow scope to the
 >> outermost shadow scope". How does this interact with sheets having
 >> different CSS origin levels, i.e. some being UA sheets and some
 >> being author sheets?
 > CSS already defines that, no?

My question is, does the above paragraph take precedence over the CSS 
cascade or not? I.e. do you first use all the origin levels at the 
innermost scope and then all at the second innermost scope? Or do you 
within each origin level first use the innermost scope and then the 
second innermost?

I guess it makes more sense to within each origin level first use the 
innermost scope and then the second innermost. If this is already what 
it defines, this would be good to clarify maybe?

 >> 6.2)
 >> "When a handler element is invoked by the object passed to the
 >> addEventListenerNS() method, the user agent must check that the
 >> event in question was forwarded to a handlers element from a bound
 >> element, and that that handlers element is the parent node of the
 >> handler element."  When is that not the case? Especially the parent
 >> part since I'd assume that removing a handler from the handlers
 >> element would call removeEventHandlerNS.
 > I'm not sure, but I imagine XML Events or other such wacky mechanisms
 > might do it.

But XML Events are not handler elements (Well, they can be I think, but 
not in the sense that is discussed here, i.e. ones in the xbl 
namespace), so the above can't apply to them as they're a separate spec.

 >> Does not setting the phase attribute cause the handler to be
 >> potentially triggered three times for the same even? I.e. during
 >> bubble, capture and default phases?
 > It can never trigger both during bubble and capture, since the event
 > is only ever registered for one or the other. It can never trigger
 > during both bubble and target or capture and target, since the target
 > element is never given the event in capture or bubble phase.

Where does it specify this? Note that even in DOM Events spec you can be 
notified both during bubbling and capturing if you call 
addEventListenerNS twice. Additionally, XBL events doesn't map directly 
to DOM Events due to the default action phase.

I guess the real question is, what phase should you be registered for if 
there is no phase attribute set.

 >> I don't understand the example for hasBinding at all, it seems to be
 >> missing context.
 > Not sure what makes you think that...?

Basically I don't understand the example at all and I can only vaguely 
guess what it means. But it's vague enough that it doesn't really help 
me at all.

I'll write a new proposal for section 6.8 that should make things 
clearer but not change what I think the spec is trying to say now.

== Comments foremost due to what I think is best for users (due to 
logical behavior, smaller spec, or otherwise) ==

 >> 1.4.3)
 >> Why is tab not considered whitespace?
 > They are.

Sure, now they are ;)

 >> 3.2.1)
 >> Why are xbl-PIs inserted in the DOM in error? That runs contrary to most
 >> things where the current state of the DOM is what matters, not how that
 >> state was reached. Scripts are so far the only exception to this. Same
 >> thing applies to dynamic changes to the PI.
 > Because otherwise interfaces implemented by elements in the document
 > could change dynamically. There is already a separate API for doing
 > this in dynamic environments (loadBindingDocument()).

Interfaces implemented by elements can still change dynamically since 
elements will change which selectors they match and can therefor match 
new element-attributes or get applied different CSS rules.

If you want to break the model of allowing dynamic modifications to the 
DOM I think you should have a reason for it other than "there are other 
ways of doing the same thing", as is the case for <script> elements and 
xmlns attributes.

And come to think of it, .loadBindingDocument is not a full replacement 
since it is syncronous, which in general is a bad idea for authors to 
use. So if an author want to apply a set of bindings asynconously they 
could do it by adding a PI. Hey, AJAX is the buzzword of the day right ;)

 >> 4.3.1)
 >> Why can xbl:text not appear by itself as a shorthand for 
 > What's the use case for that? Why not just use <content>?

This is not so much a request for a feature, but a request for a 
simplification to both the spec and the implementation. Since xbl:text 
can appear on either side the implementaion will have to deal with both 
parts of xbl:text anyway. I would imagine the implementation will have a 
generic 'reader' part that reacts to whatever is on the right hand side 
of '=', and a generic 'writer' part that writes to whatever is on the 
left hand side of the '='. This would work fine even when xbl:text 
appears by itself.

As the spec is written now the parser will specifically have to look for 
xbl:text and disallow that, even though it's likely that the rest of the 
implementation would deal fine with it.

Note that as the spec is currently written xbl:attr="xbl:text=xbl:text" 
is allowed, ensuring that the rest of the could would have to be able to 
deal with this. So it's literally just the attribute parser that would 
have to make an exception for xbl:text.

 >> Why are not all explicit text descendants (rather than just children) be
 >> taken into account when xbl:text appears on the right hand side of 
the '='.
 > Because doing that could result in broken renderings (e.g. losing text
 > direction information if one of the children is an <html:bdo>
 > element) in cases the author did not consider.

If the author didn't consider <html:bdo> elements then with the current 
behaviour he'd miss out on parts of the text which seems equally bad.

The main argument I can see against copying descendants is actually that 
it may be harder on implementations to do in a performant way, though 
for mozilla that shouldn't be a problem.

 >> does xbl:template > my:B match the B element in the following binding
 >> <binding><template><my:B/></template></binding>. If yes, that doesn't
 >> match what is stated. If no, why not?
 > It does not, because the ">" combinator is defined to consider the
 > bound element to be the parent of the shadow tree, not the <template>
 > element. This allows you to make stylistic choices based on the
 > attributes and other facets of the bound element (if you set
 > allow-selectors-through, anyway).

I see the usefulness when allow-selectors-through is set, but when it is 
not set it makes things inconvenient for the author since there is no 
way to match things that are in the root of the shadow tree. As far as 
selector matching goes it will look like all elements in the root of the 
shadow tree are disconnected orphans.

 >> 4.7.3
 >> The dual meaning of the :bound-element pseudo class seems like a bad
 >> idea from a usability point of view. It also makes it impossible to
 >> match a bound element inside a shadow tree. Why not split it up into two
 >> selectors?
 >> What is the usecase for matching all xbl-bound elements in general?
 > It isn't clear that matching a bound element in general is a useful
 > thing. That this is possible at all is mostly the result of defining
 > what should happen in the absense of a bound element, rather than the
 > result of an actual use case.

I suggest we remove that feature since it seems just to add unnecessary 
bloat to implementations. Additionally, it seems harder for authors to 
use the selector since something like
:bound-element{border: red}
would not only add a border to the bound element that the current 
binding is attached to, but also add borders to all bound elements in 
the bindings shadow tree.

 >> 5.4)
 >> Properties from one external binding object needs to be forwarded to
 >> [external] deriving binding objects. Otherwise using the .baseBinding
 >> property is fairly useless in that it only contains the properties of
 >> the immediately inherited binding.
 > Not sure I understand this either.

Say that you have three bindings, A -> B -> C, with all inheritance 
being explicit. If the implementation in A overrides a function, 
setColor, where the implementation needs to call the overridden function 
you would do this using something like

setColor: function(col) {

However this would with the current spec only work if setColor was 
implemented directly on B. If the function were inherited from C then 
the above would have to be this.baseBinding.baseBinding.setColor. This 
is not a good idea since it makes inheritance of implementation less 
useful and is not how programming languages are usually designed.

In order to fix this we should say that forwarding is done from one 
external object to the next in the inheritance chain. And then just say 
that the bound element just forwards to the first implementation object.

 >> 6.9)
 >> Why the random number used for eventPhase during default-action phase?
 >> Couldn't a more rememberable one be used?
 > It's not a random number (it's the string "xblD" in ASCII read as a
 > big-endian 32bit integer) but it _is_ an arbitrary number, in that it
 > shouldn't really matter what its value is -- people aren't expected to
 > ever look at it. The only way for code to execute when it is in the
 > default phase is for a <handler> to have phase=default-action, so code
 > always knows whether phase has this value or not.
 > The reason for the crazy number was to avoid clashes with anything
 > else.

Waving a crazy sign in front of someone's eyes saying "don't pay 
attention to this sign" is bound to get them to focus all their energy 
on this sign, or at least get them annoyed. I would much prefer to use a 
more sane number. Say 0x100 or 0x101. If for no other reason simply 
having crazy numbers like that in specs is bound to confuse people, and 
I suspect there will be places where you call a common function from 
different handlers and pass the phase as an argument.

 >> Wouldn't it make more sense for the default phase to go from the
 >> document node to the targeted node? This since the final action will
 >> be on the innermost node, i.e. the targeted node. Additionally, this
 >> allows the binding to implement a default behavior for the bound
 >> element without having to worry about what inner elements it is
 >> using and their default behavior.
 > Well, you want it to go inside-out because that's how default
 > activation behaviours work in DOM3 Events (though this may not yet
 > have been actually written down anywhere). But yeah, I'm still looking
 > at this. Implementation experience will be helpful here.

I think the most useful experience here will be user experience rather 
than implementation experience. I can definitely see arguments for going 
both ways, but generally outside in feels more useful for xbl.

 >> 6.10)
 >> Does the :focus handling also apply to :active? I.e. can there be a
 >> chain of :active elements?
 > If you can get the CSS working group to agree on what :active means
 > _outside_ of XBL, let me know and I'll update the XBL2 spec to
 > elaborate on what it should do when XBL is involved!

Given how browsers today treat it I think it makes sense to say that 
there is a chain of :active elements too, and possibly revise the other 
way once the CSS group makes up their mind.

 >> 2.5)
 >> Why [doesn't] removing the locked attribute reset the list of included
 >> children?
 > Why should it?

Because the DOM generally behaves according to its current state, 
independent of how it got to that state. To put it the other way, why 
would the user remove the locked attribute if he didn't want the element 
to behave unlocked?

Additionally, this doesn't match the steps in 4.4.1

 >> And does the includes attribute have any effect when locked is true?
 > Yes. See later in the spec, where this is covered in detail.

What is the use case for the includes attribute on locked insertion 
points? It seems confusing to the user that children can be removed even 
though the element is locked. Granted, we have to define something, but 
it seems to me that the simplest solution, i.e. locked means locked and 
the user is fully in charge of managing the list of inserted children, 
works just as well.

 >> 4.4)
 >> What is the use case for sticky insertion points?
 >> Aren't locked insertion points enough?
 >> Additionally, this doesn't match the steps in 4.4.1
 > I've removed the offending paragraph, which was out of date.

Sticky insertion points are still there according to the definition of 
the setInsertionPoint function (first paragraph after the numbered 
steps). I would recommend adding a first step that says to throw if the 
element is not locked.

I also noticed that step 3 in this list make it sound like the child has 
to be a DOM Core child of the bound element for the binding. Shouldn't 
you be allowed to specify the insertion point of any of the explicit 

 > In the case of your example with the <h1>, why would you put an <h1>
 > again in the template? Consider a <button> binding. You wouldn't put a
 > <button> in the implementation of <button>, right?

I can't really come up with a use case for this off the top of my head, 
we need user experience here I think.

 >> 4.9.6)
 >> Why does references in SVG pointing to a bound element use the first
 >> element in the nodes shadow tree? That seems contrary to other behavior
 >> where references to a to a bound element still point to the bound
 >> element, but rendering renders both the bound element and it's
 >> "flattened children". Maybe this is only desired in SVG when the bound
 >> element is an SVG element.
 > It is so that you can do things like define bindings that define
 > gradients or fonts or whatnot.

It is still weird to me that just the first child is the referenced one. 
This runs contrary to other uses of XBL where all the children of the 
<template> are rendered. Would it be possible to say that the bound 
element should render as a <svg:g> element making the shadow tree under 
it what is effectively rendered? Or would that extra <svg:g> affect how 
rendering behaved in undesirable ways?

== Comments that are important to me as an implementor ==

 >> 2.13)
 >> Seems very complicated that <style> elements affect the bound element.
 >> Which binding to apply, and thus which <style> elements to apply won't
 >> be known until after style resolution, requiring a second pass.
 > Correct. However, it is really required for many use cases.

I'll have to run this by our layout folks to see how big the performance 
hit is going to be. Potentially we won't really know until we implement 
this part.

 >> 3)
 >> "the binding must be applied such that to any running scripts it appears
 >> that the binding was applied immediately". When is that in relation to
 >> mutation events? It may be better to say that the binding is applied
 >> before the mutating function returns. It seems like this only applies to
 >> element attribute attached bindings? If so it would be good to state
 >> explicitly
 > Yeah, this is something that there are known issues with. I'd actually
 > really like implementation feedback on this. I've fixed this a bit.

"When mutation events are to fire, they must fire after the binding 
being applied." This can be tricky actually. For example when removing a 
node from the tree the mutation event is fired before the node is 
removed. However selector matching would be really hard to do until 
after the node is removed. So not until then it is possible to know 
which bindings to add or remove.

Somthing like "To mutation events it must look like bindings are applied 
at the same time as the mutation occurs. This means that mutation events 
that fire before the mutation will not see the bindings applied, whereas 
mutation events that fire after the mutation will".

This is definitely something that would be good to have tested in the 
test suite so that we can be sure that implementations pay attention and 
give feedback for what's hard to do.

 >> 3.5)
 >> It is unclear to me if it currently is allowed to call the
 >> xblBindingAttached/xblEnteredDocument functions after all bindings have
 >> been constructed and attached for a subtree. This is vital for styling
 >> performance since we don't want to have to return to the main event loop
 >> for each bound element during layout.
 > I wouldn't attach during layout. I would just make a note of what
 > bindings need to be applied, and when you have finished your style
 > pass, apply those bindings and then do layout again.

But then it would only be possible to style up to the bound element 
since before bindings are attached it is impossible to know what 
children the flattened tree will have. In XUL practically every element 
has a binding so this would render one element at a time, which of 
course is unacceptable.

 >> Why does setting the [xml:base] not cause a mutation event?
 > No mutation events must be fired at all during the entire cloning
 > process, because there didn't seem like a good reason to have them
 > fire and it would allow performance improvements to block them.

Hmm.. Saying that no mutation events at all fire during the above is an 
interesting idea. However it's only taking us part of the way. At the 
very least you should say that userdata handlers are not notified of the 
cloning either. I'm worried that there are other events or notifications 
that would normally fire that you'd want to avoid firing too. I don't 
see any other solution than that the spec contain a comprehensive list 
of which notifications are blocked, but that seems hard or impossible 
considering future specs.

This seems like it could be hard to implement if existing APIs always 
fire mutation events. It turns out in mozilla that we currently have a 
flag for not firing them when performing certain mutations, however this 
is a recent flag, and it's one that I've been wanting to remove.

I wonder if this is an optimization that is really doable, though it 
would be really nice if it was.

 >> Why the waste of CPU to forward all attributes, including ones that
 >> haven't changed when any attribute is mutated.
 > That's an implementation detail. The spec doesn't require you to
 > implement this inefficiently if you can think of better ways to do it
 > than have the same effect. :-)

I can see two ways of doing this, both less performant that simply 
updating the changed attribute.

Either you could do the simple thing and update all attributes whenever 
any of them are changed.

Or you would have to keep track of which attributes in the shadow tree 
have been manually changed and only update those in addition to the 
changed attribute.

Additionally, it seems better for the user to only blow away minimal 
manual changes to the shadow tree whenever possible.

 >> 5.2)
 >> What is the purpose of the xblImplementations property? It seems to
 >> defeat the purpose of hiding the shadow tree etc in that it can make
 >> users dependent on the internals of a binding.
 > It returns the external object, not the private object. How does it
 > defeat shadow tree hiding?

We should remove the ability to access the implementations by index, 
otherwise people might do things like .xblImplementations[5].setColor(...)

 >> Finally I have two concerns without proposed solution:
 >> The browser will behave differently if an element is implemented using
 >> XBL or using a native implementation when that element is further bound
 >> by a document author. In one case <inherited> will render content and in
 >> one it won't. Something that might help this is to let <inherited> in
 >> the base binding render the contents of the remaining (so far without
 >> insertion point) children. I.e. let there be an implicit binding with
 >> the shadow tree <template><children/></template>.
 > I intend at some point to work on a spec that defines how various
 > elements should appear to be implemented at the UA level in terms of
 > XBL.

Fair enough. Another thing I realized this shows up in is the 
.baseBinding property. It would be good if we for the least derived 
binding made that return the element itself to make a browser that 
implement an element using xbl and a browser implement an element using 
native code work as well as possible with the same binding.

== Comments that are less important to me as an implementor ==

 >> Is it really a good idea to implicitly apply bindings into the
 >> document in which they live? This seems wasteful at the least. Might
 >> be better to only do this when the <xbl> is not the root element?
 >> Even then you could require an xbl PI containing just a fragment
 >> identifier. A use case might be if want to override the look of <h1>
 >> by wrapping some elements around it. You could do that using a
 >> template like <template><div class=...><div
 >> class=...><h1><content/></h1></div></div></template>.  However that
 >> would create a recursive situation if the binding applied to the
 >> <h1> in the shadow tree.
 > I could see an argument for special-casing <xbl> being the root
 > element, but that seems like a can of worms that we'd want to avoid.

If you want to avoid the inconsistency you could say that bindings are 
never implicitly applied. If you want to put the bindings in the same 
document as your main document you can do something like:

<?xbl href="#myBindings"?>

This is mostly a resource saving thing for me. It's unneccesary to have 
a bunch of bindings imported into a document if you're not going to use 
them anyway. Worst case I can always add a mozilla specific attribute to 
the <xbl> element that removes the implicit import if it turns out that 
it makes a differance.

 >> 3.9)
 >> What happens if a binding document has already started loaded and the
 >> user then calls loadBindingDocument with the same URI?
 > See 8.1.1.

My question was, what happens if the load has started, but not yet 
finished? Does loadBindingDocument wait with returning until the load 
finishes? If not, what does it return since there may not yet be a 
document created.

 >> 4)
 >> Wouldn't having more than one template element in a binding make the
 >> binding in error?
 > Yes, this is already defined.

"The template element used to generate a binding is always the first 
such element in a binding element" says that even if there are multiple 
<template> elements the binding is still applied. The "in error" section 
says that elements that are in error are typically ignored, however that 
conflicts with the above sentence since it says to use the first <template>.

 >> Why not insert an explicit child text child under the element when
 >> xbl:text is on the left hand side of the '='. That is more in line
 >> with how attributes are explicitly mutated when on the left hand
 >> side of the '='. It also makes it clearer and more easily
 >> implementable how the inserted child interacts with bindings on the
 >> element.
 > Because the element may have other children in the Core DOM that we
 > wouldn't want to blow away.

Attribute values are blown away the same way though. Note that this 
doesn't change the bound document, it's just the shadow tree that would 
get automatically mutated when explicitly requesting it.

 >> It is in some cases needed that one binding can get to the internal
 >> implementation object of another binding. One example of this is a
 >> tab-strip implementation where the binding for the <tabstrip>
 >> element needs to get to the internals of the individual <tab>
 >> bindings. One way to do this would be to make it possible for
 >> bindings from the same binding document to using some API be able to
 >> reach each others internal implementation object.
 > Why can't the <tabstrip> just use the <tab>'s official API?

That would force you to put a method like 'showAsSelected' in the public 
api of the <tab>, which would allow users to make multiple tabs be shown 
as selected. Basically what I'm asking for is something similar as C++s 
'friend' concept. However at this time I don't have a proposal for this, 
so I'm willing to wait until we have more implementation and usage 
experience before figuring this out. It is even something that could 
wait to a later version of the spec.

 >> 6.2)
 >> What does the sentence "However, their being in error does not affect
 >> the processing model described above" mean? Should addEventListenerNS
 >> still be called? If so, why?
 > Yes. In the case of the event attribute being in error, it makes no
 > difference. In the case of the phase attribute being in error, it
 > seemed pointless to have the UA check that the attribute had a correct
 > value before going on.

I guess if addEventListenerNS is called or not is really an internal 
issue, the user won't see a difference either way, right?

 >> The fact that explicitly inherited bindings are attached at the same
 >> time as the base binding forces all referenced binding documents to be
 >> loaded whenever a binding document is imported, whether the bindings are
 >> ever used or not. This might be good to state. In fact the imported
 >> document can not be considered loaded until this is done which means
 >> that .loadBindingDocument must wait until all inherited binding
 >> documents have been loaded.
 > This is an interesting point. I'm not really sure how this works with
 > mutations of the binding document, though. I've changed
 > loadBindingDocument() to require that all dependencies be fetched
 > (well, not all dependencies; I guess stylesheet @imports and such like
 > wouldn't actually be found unless needed later... this could be
 > messy).
 > What else needs changing?

First of all, please write the part "and any bindings defined by that 
document must be applied to matching elements in the document that 
corresponds to this DocumentXBL object" as a separate sentence. It took 
me 5 or so reads to figure out what "that document" refers to (sounds 
like the inherited bindings document now).

You probably also need to change when the bindings-are-ready counter is 
reduced, or state that loading of dependent binding documents should 
start right away causing the bindings-are-ready counter to be increased.

I don't see where you say that other dependencies (other than inherited 
bindings) have to be loaded. Did you mean to say that <style> sheets 
have to be loaded and such too? I'm not requesting that that be made the 
case, but it sounded like you were saying that.

For mutations that happen after the document is considered loaded 
obviously can't delay the point when the document is considered loaded. 
For mutations that happen before, I could see leaving this undefined. 
Alternatively state that mutations that happen before that point must 
delay loadBindingDocument from returning, but that seems like it could 
be very messy to implement.

/ Jonas
Received on Monday, 26 March 2007 22:57:08 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 20:50:07 UTC