You are viewing a plain text version of this content. The canonical link for it is here.
Posted to j-dev@xerces.apache.org by Ed Staub <es...@mediaone.net> on 2000/09/19 20:45:53 UTC
[VOTE (PLEASE!)] Xerces2 Requirements 11, 20, 31, 32, 34, 37
This is a call for a vote to finalize requirements 8 through 29 (with
exceptions) in http://xml.apache.org/xerces-j/issues.html. The process for
refining the requirements is described in
http://apache.org/xerces-j/issues-process.html.
Please read the attached proposal document closely.
Note that the words "shall" and "should" have precise "hardness" meanings.
"Shall" indicates a "hard" requirement which must be met. "Should"
indicates a strongly desired but "soft" goal.
In order for the requirements to truly be finalized, we need as much
participation in this refinement process as possible, especially from
committers. If only three people give +1's on an item, the consensus isn't
very meaningful.
Because there have been numerous misunderstandings based on the terse
language in the voting forms, I'm making it a lot more terse. Now you
_must_ read the attachment!
-----------------------------------------------
VOTING FORM:
11.
20.
31.
32.
34.
37.
-----------------------------------------------
-Ed Staub
Re: [VOTE (PLEASE!)] Xerces2 Requirements 11, 20, 31, 32, 34, 37
Posted by Ted Leung <tw...@sauria.com>.
-----------------------------------------------
VOTING FORM:
11. +1
20. +0
31. +1
32. +0
People in certain industrial settings may be prevented from moving up to 1.2 even in a year's
time. It may sound crazy to most of us, but there are people in the world who are constrained
to 1.1 for indefinite time frames. I personally don't have any clients like this at the moment, but
I've talked to people that had this constraint.
34. -0 I don't know what this support entails. Certainly we cannot add items to the infoset
because we would be outside the XML 1.0 rec.
37. +0 I think we all know my opinion on the DOM by now.
-----------------------------------------------
-Ed Staub
------------------------------------------------------------------------------
---------------------------------------------------------------------
To unsubscribe, e-mail: xerces-j-dev-unsubscribe@xml.apache.org
For additional commands, e-mail: xerces-j-dev-help@xml.apache.org
RE: [VOTE (PLEASE!)] Xerces2 Requirements 11, 20, 31, 32, 34, 37
Posted by Ed Staub <es...@mediaone.net>.
If the requirement is removed, then contributors (IBM or elsewhere) are free
to introduce dependencies on 1.2 features, such as the collections classes,
soft/weak/phantom references, etc. The parser may or may not run on 1.1;
removing the requirement says "we don't care if it runs on 1.1". That
doesn't sound like what you mean, Andy.
Do you think you would require this if it were asked next year?
because that's roughly when we really care about, i.e., when X2 is ready for
production and people are reliant on features which aren't in X1. I don't
think many people will care; or, if they do, they will be very acceptant of
the need and desirability of moving on to newer and more performant
platforms. Other factors will be driving them forward more than Xerces.
I don't see any compelling reasons to use 1.2 features either, with the
possible exception of soft references (for schema caching). But I hate to
close it off without strong evidence of need (for 1.1 compatibility).
-Ed Staub
-----Original Message-----
From: Andy Clark [mailto:andyc@apache.org]
Sent: Tuesday, September 19, 2000 5:31 PM
To: xerces-j-dev@xml.apache.org
Subject: Re: [VOTE (PLEASE!)] Xerces2 Requirements 11, 20, 31, 32, 34,
37
Edwin Goei wrote:
> Hmmm, I voted +1 to _remove_ the requirement that the parser
> supports Java 1.1. It sounds like you are voting for the opposite.
> Am I misunderstanding your vote?
I guess I misunderstood what the "remove" requirement on the
voting page meant. I like the following wording from your
next posting:
> 32. +1 Xerces1 works fine and is optimized for Java1.1, Xerces2
> should target Java2 and does NOT have to work with Java 1.1.
You can state that it "does not have to" work with Java 1.1 as
long as it doesn't say "will not" work with Java 1.1. By removing
the requirement altogether, you are not disallowing the use of
Java 1.1 in an implementation which is what I want to see
regarding this point.
With the current state of customer requirements, I would not
write a parser that does not compile and run under 1.1. And if
I were required to disallow the use of Java 1.1 by customers,
then I would -1 this item.
--
Andy Clark * IBM, JTC - Silicon Valley * andyc@apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: xerces-j-dev-unsubscribe@xml.apache.org
For additional commands, e-mail: xerces-j-dev-help@xml.apache.org
Re: [VOTE (PLEASE!)] Xerces2 Requirements 11, 20, 31, 32, 34, 37
Posted by Andy Clark <an...@apache.org>.
Edwin Goei wrote:
> Hmmm, I voted +1 to _remove_ the requirement that the parser
> supports Java 1.1. It sounds like you are voting for the opposite.
> Am I misunderstanding your vote?
I guess I misunderstood what the "remove" requirement on the
voting page meant. I like the following wording from your
next posting:
> 32. +1 Xerces1 works fine and is optimized for Java1.1, Xerces2
> should target Java2 and does NOT have to work with Java 1.1.
You can state that it "does not have to" work with Java 1.1 as
long as it doesn't say "will not" work with Java 1.1. By removing
the requirement altogether, you are not disallowing the use of
Java 1.1 in an implementation which is what I want to see
regarding this point.
With the current state of customer requirements, I would not
write a parser that does not compile and run under 1.1. And if
I were required to disallow the use of Java 1.1 by customers,
then I would -1 this item.
--
Andy Clark * IBM, JTC - Silicon Valley * andyc@apache.org
Re: [VOTE (PLEASE!)] Xerces2 Requirements 11, 20, 31, 32, 34, 37
Posted by Edwin Goei <Ed...@eng.sun.com>.
Andy Clark wrote:
>
> > 32.
>
> +1.
>
> I'm happy with removing this requirement, as will all those
> people "stuck" using 1.1 that would be locked out of using
> the XML parser. There's nothing in the parser that will not
> work unless coded for 1.2 and higher.
Hmmm, I voted +1 to _remove_ the requirement that the parser supports
Java 1.1. It sounds like you are voting for the opposite. Am I
misunderstanding your vote?
-Edwin
Re: [VOTE (PLEASE!)] Xerces2 Requirements 11, 20, 31, 32, 34, 37
Posted by Andy Clark <an...@apache.org>.
> 11.
+1.
Clarification text is fine.
> 20.
+0.
I want it known that I'm in support of XInclude and other
standards that operate on the XML Infoset (such as Schema).
However, people should realize that there are serious
problems with doing a complete implementation of such
standards in a streaming parser.
While these operations are useful "in stream", regardless
of whether they are before or after validation via the
validator pipeline stage, full support can only be done
on the complete document in memory. Which implies the use
of a document tree model and that the operation is
implemented using this tree, outside of the pipeline.
> 31.
+1.
Fine. However, having full support for operations on the
XML Infoset are going to need *some* kind of model. Unless
we state that we are only going to support those features
of XInclude, etc, that can be done in a streaming fashion.
> 32.
+1.
I'm happy with removing this requirement, as will all those
people "stuck" using 1.1 that would be locked out of using
the XML parser. There's nothing in the parser that will not
work unless coded for 1.2 and higher.
> 34.
[Q] Doesn't this one imply a tree model? Or are we referring
to the ability of the parser to make all of the XML Infoset
data available in some way. In other words, not losing info
that is part of the Infoset somewhere in the pipeline?
> 37.
+1.
We should make it clear how much of DOM we are supporting.
There are a lot of optional parts of the DOM specification
and we should state which ones we are going to support in
our implementation.
--
Andy Clark * IBM, JTC - Silicon Valley * andyc@apache.org
RE: [Xerces2] Remove requirement 37 on grounds of redundancy
Posted by Ed Staub <es...@mediaone.net>.
Ok, then lets fix 16.
I'll call for a vote after we get it nailed down.
How does this sound?
------------------------------
Change
> The parser shall support DOM Level 2.
to
The parser shall support the following modules of DOM Level 2
(http://www.w3.org/TR/DOM-Level-2):
XML
HTML
Events (Mutation Events only)
Traversal
Range
The parser will not support the following modules:
Views
CSS
CSS (extended interfaces)
User Interface Events
Mouse Events
HTML Events
The parser may support the following modules:
Stylesheets
------------------------------
-----Original Message-----
From: Arnaud Le Hors [mailto:lehors@us.ibm.com]
Sent: Wednesday, September 20, 2000 11:21 AM
To: xerces-j-dev@xml.apache.org
Subject: Re: [Xerces2] Remove requirement 37 on grounds of redundancy
Ed Staub wrote:
>
> Mike Pogue wrote:
> > 37. +0 (isn't this the same as DOM L2 mutation events, he foolishly
> asked?)
>
> Item 37 reads:
> The DOM should emit events to interested listeners when changes
are made to
> the DOM.
>
> This requirement is already covered by requirement 16:
>
> The parser shall support DOM Level 2.
>
> and by requirement 9:
>
> The parser should be upwardly compatible with Xerces.
>
> If there is no dissent, I would like to REMOVE requirement 37.
One problem with that is that requirement 16 is slightly bogus. DOM
Level 2 includes several modules, most of which are optional. In Xerces
1 we currently support DOM Core & XML, plus the optional modules:
Ranges, Traversal, and MutationEvents. On the other hand we do not
support CSS. So, requirement 9 means to me that people want
MutationEvents to be supported.
--
Arnaud Le Hors - IBM Cupertino, XML Technology Group
---------------------------------------------------------------------
To unsubscribe, e-mail: xerces-j-dev-unsubscribe@xml.apache.org
For additional commands, e-mail: xerces-j-dev-help@xml.apache.org
Re: [Xerces2] Remove requirement 37 on grounds of redundancy
Posted by Arnaud Le Hors <le...@us.ibm.com>.
Ed Staub wrote:
>
> Mike Pogue wrote:
> > 37. +0 (isn't this the same as DOM L2 mutation events, he foolishly
> asked?)
>
> Item 37 reads:
> The DOM should emit events to interested listeners when changes are made to
> the DOM.
>
> This requirement is already covered by requirement 16:
>
> The parser shall support DOM Level 2.
>
> and by requirement 9:
>
> The parser should be upwardly compatible with Xerces.
>
> If there is no dissent, I would like to REMOVE requirement 37.
One problem with that is that requirement 16 is slightly bogus. DOM
Level 2 includes several modules, most of which are optional. In Xerces
1 we currently support DOM Core & XML, plus the optional modules:
Ranges, Traversal, and MutationEvents. On the other hand we do not
support CSS. So, requirement 9 means to me that people want
MutationEvents to be supported.
--
Arnaud Le Hors - IBM Cupertino, XML Technology Group
[Xerces2] Remove requirement 37 on grounds of redundancy
Posted by Ed Staub <es...@mediaone.net>.
Mike Pogue wrote:
> 37. +0 (isn't this the same as DOM L2 mutation events, he foolishly
asked?)
Item 37 reads:
The DOM should emit events to interested listeners when changes are made to
the DOM.
This requirement is already covered by requirement 16:
The parser shall support DOM Level 2.
and by requirement 9:
The parser should be upwardly compatible with Xerces.
If there is no dissent, I would like to REMOVE requirement 37.
-Ed Staub
Re: [VOTE (PLEASE!)] Xerces2 Requirements 11, 20, 31, 32, 34, 37
Posted by Mike Pogue <mp...@apache.org>.
-----------------------------------------------
VOTING FORM:
11. +1 I think the wording on this is fine.
20. +1 I'm not a big fan of XInclude, but I think it's OK to implement.
31. -0 I'm not a fan of "thou shalt not" requirements. There are a
jillion ways
to NOT implement something, e.g. Thou shalt not directly support any
automatic
machine translation of the XML into Danish. But, since this is a
"should not",
it's not really binding, so it probably doesn't matter. Thus
the -0.
32. +1 for removing the requirement for Java 1.1 . I think it's time.
34. +1 There are enough people who actually need this, that I think it
should be
a requirement.
37. +0 (isn't this the same as DOM L2 mutation events, he foolishly asked?)
-----------------------------------------------
----- Original Message -----
From: Ed Staub
To: xerces-j-dev
Sent: Tuesday, September 19, 2000 11:45 AM
Subject: [VOTE (PLEASE!)] Xerces2 Requirements 11, 20, 31, 32, 34, 37
This is a call for a vote to finalize requirements 8 through 29 (with
exceptions) in http://xml.apache.org/xerces-j/issues.html. The process for
refining the requirements is described in
http://apache.org/xerces-j/issues-process.html.
Please read the attached proposal document closely.
Note that the words "shall" and "should" have precise "hardness" meanings.
"Shall" indicates a "hard" requirement which must be met. "Should"
indicates a strongly desired but "soft" goal.
In order for the requirements to truly be finalized, we need as much
participation in this refinement process as possible, especially from
committers. If only three people give +1's on an item, the consensus isn't
very meaningful.
Because there have been numerous misunderstandings based on the terse
language in the voting forms, I'm making it a lot more terse. Now you
_must_ read the attachment!
-----------------------------------------------
VOTING FORM:
11.
20.
31.
32.
34.
37.
-----------------------------------------------
-Ed Staub
---------------------------------------------------------------------
To unsubscribe, e-mail: xerces-j-dev-unsubscribe@xml.apache.org
For additional commands, e-mail: xerces-j-dev-help@xml.apache.org
Re: [VOTE (PLEASE!)] Xerces2 Requirements 11, 20, 31, 32, 34, 37
Posted by Edwin Goei <Ed...@eng.sun.com>.
Arnaud Le Hors wrote:
>
> -----------------------------------------------
> 32. -1. Unless we encounter a particular incentive to make it Java 1.2 I
> don't see why we should bluntly disregard Java 1.1 support.
The incentive would be ease of development and performance
improvements. Java 1.2 was a significant break from Java 1.1, which is
why some people decided to call it Java2. As already stated, the Java
Collections API is in 1.2 which has useful features as well as improving
performance. There are other features in Java2 that would be useful as
well.
-Edwin
Re: [VOTE (PLEASE!)] Xerces2 Requirements 11, 20, 31, 32, 34, 37
Posted by Arnaud Le Hors <le...@us.ibm.com>.
-----------------------------------------------
VOTING FORM:
11. +1. I'm afraid what "result in significant compromises, cleanliness
of the architecture" will be very subjective though...
20. -0. I certainly agree the design should allow for XInclude support,
but I disagree actual support is required. I see it more like a nice
thing to have.
31. -0. First, I find this question rather moot given that I don't think
we've clearly defined what "the core parser" is. Is the validator part
of core? Second, I agree things like a full DOM implementation shouldn't
be part of the core, however, we may need to have some tree structure as
part of the core to support features such as keyrefs in XML Schemas.
32. -1. Unless we encounter a particular incentive to make it Java 1.2 I
don't see why we should bluntly disregard Java 1.1 support.
34. +1. Even though there is no formal way of enforcing this I think
it's important to be aware of what the Infoset qualifies as important
versus non important (such as whether simple or double quotes were
used).
37. +1. But that's part of MutationEvents support (and we already have
in Xerces 1).
-----------------------------------------------
Arnaud Le Hors - IBM Cupertino, XML Technology Group
Re: [VOTE (PLEASE!)] Xerces2 Requirements 11, 20, 31, 32, 34, 37
Posted by Edwin Goei <Ed...@eng.sun.com>.
Ed Staub wrote:
>
> Because there have been numerous misunderstandings based on the terse
> language in the voting forms, I'm making it a lot more terse. Now you
> _must_ read the attachment!
Nicely done ballot attachment.
> -----------------------------------------------
> VOTING FORM:
>
> 11. +0 Writing a Java parser will be difficult enough, but this seems like a good compromise
>
> 20. Can XInclude be supported as a separate module or does it directly affect parser internals?
>
> 31. +1
>
> 32. +1 Xerces1 works fine and is optimized for Java1.1, Xerces2 should target Java2 and does NOT have to work with Java 1.1.
>
> 34. +0 I do not fully understand the InfoSet spec to know whether this is practical
>
> 37. +0 This should be left to the DOM module implementation of which there could be more than one.