XMLHttpRequest in Internet Explorer 7 - no more ActiveX
Joe Attardi
I was reading over at the IE Blog the other day
[http://http://blogs.msdn.com/ie/] and read some interesting, and
encouraging news. According to Sunava Dutta, an IE Program Manager,
starting in IE7 XMLHttpRequest will be a plain JavaScript object. It
will still include the Microsoft.XMLHTTP ActiveX component for
compatibility, but IMHO this is a very good shift for IE.
What do you think?
More information at:
http://blogs.msdn.com/ie/archive/2006/01/23/516393.aspx
--
Joe Attardi
Carl Vondrick
9:09 AM Joe Attardi wrote:
> I was reading over at the IE Blog the other day
> [http://http://blogs.msdn.com/ie/] and read some interesting, and
> encouraging news. According to Sunava Dutta, an IE Program Manager,
> starting in IE7 XMLHttpRequest will be a plain JavaScript object. It
> will still include the Microsoft.XMLHTTP ActiveX component for
> compatibility, but IMHO this is a very good shift for IE.
>
> What do you think?
I am glad that they are adding an object that will make Ajax a standard
- this will cut down on the wild if else statements.
But, honestly, I will not be able to use this for the next 10 years. I
know of people who are still using IE4 -- getting them to upgrade to IE7
will be painful.
Hopefully this marks a turning point to universal standards (and not
local standards).
Carl
Richard Cornford
> Joe Attardi wrote:
<snip>
>> ... IE7 XMLHttpRequest will be a plain JavaScript object. ...
<snip>
> But, honestly, I will not be able to use this for the next 10
> years. I know of people who are still using IE4 -- getting them
> to upgrade to IE7 will be painful.
I would wager that the number of individuals using IE 6 with scripting
disabled exceeds the number still using IE4 by orders of magnitude. And
it is likely that in ten years time users will still be electing not to
employ optional web technologies like scripting.
If a system has been designed to produce an acceptable outcome for IE6
users who have withdrawn scripting (and/or ActiveX) then it should
produce results that are (more or less) as acceptable for users of IE4.
Richard.
Thomas 'PointedEars' Lahn
> I was reading over at the IE Blog the other day
> [http://http://blogs.msdn.com/ie/] and read some interesting, and
> encouraging news. According to Sunava Dutta, an IE Program Manager,
> starting in IE7 XMLHttpRequest will be a plain JavaScript object.
No, it will not. It will still be a host object accessible through
_JScript_.
PointedEars
Randy Webb
> Joe Attardi wrote:
>
>> I was reading over at the IE Blog the other day
>> [http://http://blogs.msdn.com/ie/] and read some interesting, and
>> encouraging news. According to Sunava Dutta, an IE Program Manager,
>> starting in IE7 XMLHttpRequest will be a plain JavaScript object.
>
> No, it will not.
Yes it will.
Although the posted URL is incorrect:
<URL: http://blogs.msdn.com/ie/ >
<quote>
In IE7, XMLHTTP is now also exposed as a native script object
</quote>
> It will still be a host object accessible through _JScript_.
Read the blog. Its a "native script object".
--
Randy
comp.lang.javascript FAQ - http://jibbering.com/faq & newsgroup weekly
Javascript Best Practices - http://www.JavascriptToolbox.com/bestpractices/
Thomas 'PointedEars' Lahn
> [...] Joe Attardi wrote:
>> I was reading over at the IE Blog the other day
>> [http://http://blogs.msdn.com/ie/] and read some interesting, and
>> encouraging news. According to Sunava Dutta, an IE Program Manager,
>> starting in IE7 XMLHttpRequest will be a plain JavaScript object. It
>> will still include the Microsoft.XMLHTTP ActiveX component for
>> compatibility, but IMHO this is a very good shift for IE.
>>
>> What do you think?
>
> I am glad that they are adding an object that will make Ajax a standard
Nonsense. Microsoft does not set Web standards (alone), how much they ever
would like to. And you want to get informed about the "Ajax" marketing
buzzword before you make any further statements regarding this topic:
<URL:http://en.wikipedia.org/wiki/AJAX>
> - this will cut down on the wild if else statements.
It will not. It will still be a host object, and even if it will not be
one but a part of the JScript language instead, it would not be backwards
compatible which would force any competent developer to do a feature test
on runtime before. "The wild if else statements", as you call them, are
necessary and will still be necessary, in order to provide for graceful
degradation.
> But, honestly, I will not be able to use this for the next 10 years. I
> know of people who are still using IE4 -- getting them to upgrade to IE7
> will be painful.
Iff Internet Explorer version 7 is ever to be released for _public_ download
and at least for all Windows versions currently in use. Knowing Microsoft,
this will not happen.
> Hopefully this marks a turning point to universal standards
Knowing Microsoft, it will not.
> (and not local standards).
There is no such thing as a "local standard". The very property of a
standard is that it is widely accepted and implemented; for a Web standard
to deserve its name this means at least that it is implemented in different
user agents.
PointedEars
Joe Attardi
Thanks for that catch Randy. I must have been out of it yesterday to
not catch that error in the URL! Sorry about that :)
VK
Joe Attardi wrote:
> Hey all!
>
> I was reading over at the IE Blog the other day
> [http://http://blogs.msdn.com/ie/] and read some interesting, and
> encouraging news. According to Sunava Dutta, an IE Program Manager,
> starting in IE7 XMLHttpRequest will be a plain JavaScript object. It
> will still include the Microsoft.XMLHTTP ActiveX component for
> compatibility, but IMHO this is a very good shift for IE.
>
> What do you think?
It is not a shift, but another step forward towards wannabes in the
concern of IXMLHTTPRequest. Do not forget that it was Microsoft
proprietary extension for Microsoft proprietary technology (data
bunding). But people loved it so much that lesser than in one year all
others had to spit on standards and implement their own equivalents -
at the best of their particular abilities. That alone forced them to
implement some other Microsoft-proprietary tools like
onreadystatechange handler.
W3C (first time in many years) had to return to the regular order of
things then instead of inventing some standard out-of-head they have to
describe the existing one. They did not like so much that the relevant
part of specification is announced but still not finished.
In this concern the "standard way" term is not applicable to
IXMLHTTPRequest / XMLHttpRequest. Actually by going straight and fair
IXMLHTTPRequest *is the standard*, and XMLHttpRequest's of all kind are
"standard emulations".
Nevertheless Microsoft most probably will not enforce on the community
their ActiveX technology, and it will continue to keep JScript
compatible with previous versions and with other ECMAScript
implementations (JavaScript).
Do not forget that Microsoft is full member of TC39-TG1 - Dynamic
Scripting Language group in ECMA International. So it's signed the same
papers and under the same chapter as other group members. I'm using
this opportunity to remind who is entitled to decide (and currently
deciding) the future hold of ECMAScript standards:
BEA Systems, Inc. <http://www.bea.com>
IBM, Inc. <http://www.ibm.com>
Intel, Inc. <http://www.intel.com>
Macromedia - out by being purchased by Adobe
Mainsoft, Inc. <http://www.mainsoft.com>
Microsoft, Inc. <http://www.microsoft.com>
Mozilla Foundation <http://www.mozilla.org>
Convenor: Brendan Eich (Mozilla Foundation)
Thomas 'PointedEars' Lahn
> Joe Attardi wrote:
>> I was reading over at the IE Blog the other day
>> [http://http://blogs.msdn.com/ie/] and read some interesting, and
>> encouraging news. According to Sunava Dutta, an IE Program Manager,
>> starting in IE7 XMLHttpRequest will be a plain JavaScript object. It
>> will still include the Microsoft.XMLHTTP ActiveX component for
>> compatibility, but IMHO this is a very good shift for IE.
>>
>> What do you think?
>
> It is not a shift, but another step forward towards wannabes in the
Wannabes like you?
> concern of IXMLHTTPRequest. Do not forget that it was Microsoft
> proprietary extension for Microsoft proprietary technology (data
> bunding). But people loved it so much that lesser than in one year all
> others had to spit on standards and implement their own equivalents -
> at the best of their particular abilities. That alone forced them to
> implement some other Microsoft-proprietary tools like
> onreadystatechange handler.
Nobody was forced to do anything. It seemed like a good idea and there
was no standard at the time, therefore it was implemented. And it was
implement without ActiveX. If what you said was true, it would have not
been.
> W3C (first time in many years) had to return to the regular order of
> things then instead of inventing some standard out-of-head they have to
> describe the existing one. They did not like so much that the relevant
> part of specification is announced but still not finished.
Utter nonsense.
> In this concern the "standard way" term is not applicable to
> IXMLHTTPRequest / XMLHttpRequest.
Nonsense. If Microsoft would push this to be a W3C standard, it is not
unlikely that it becomes one.
> Actually by going straight and fair IXMLHTTPRequest *is the standard*, and
> XMLHttpRequest's of all kind are "standard emulations".
Nonsense.
> [Microsoft is a full ECMA member]
This is completely irrelevant regarding the current host object or what
will probably become an allowed extension of the ECMAScript standard by
Microsoft JScript in IE7.
And what you comfortably forgot to mention again is that Microsoft Corp.
is a full W3C member as well, as that allows you to distinguish between
Microsoft ("good guys") and W3C ("bad guys") in your own twisted way of
percepting reality. Figures.
,-<URL:http://www.w3.org/Consortium/Member/List>
|
| [...]
| # Merck & Co., Inc.
| # Mercurial Communications Inc.
| # Microsoft Corporation
^^^^^^^^^^^^^^^^^^^^^
| # Ministerio de Administraciones Públicas
| # MITRE Corporation
| # Mitsue-Links Co., Ltd.
| # Mobileaware, Ltd.
| [...]
PointedEars
VK
Thomas 'PointedEars' Lahn wrote:
> VK wrote:
> > It is not a shift, but another step forward towards wannabes in the
>
> Wannabes like you?
"wannabes" *doesn't* mean at all "loosers".
But if this is too upseting for your ears then "other browser producers
targeted to take leading positions on the browser market".
> Nobody was forced to do anything. It seemed like a good idea and there
> was no standard at the time, therefore it was implemented. And it was
> implement without ActiveX.
Totally right. But it did not and it can not make the original
(IXMLHTTPRequest) non-standard. Same way if one paints a copy of
Jokonda, it doesn't make the original worthless :-)
> > W3C (first time in many years) had to return to the regular order of
> > things then instead of inventing some standard out-of-head they have to
> > describe the existing one. They did not like so much that the relevant
> > part of specification is announced but still not finished.
>
> Utter nonsense.
Presuming I'm developing a new browser (non-Gecko based) with
XMLHttpRequest support, what official paper should I follow to make it
standard-compliant?
http://www.w3.org/.... ???
http://www.ecma-international.org/....???
> > In this concern the "standard way" term is not applicable to
> > IXMLHTTPRequest / XMLHttpRequest.
>
> Nonsense. If Microsoft would push this to be a W3C standard, it is not
> unlikely that it becomes one.
But it did not. Microsoft doesn't want to puch ActiveX by itself on
anyone. But they are targeted to puch Internet Explorer as hard as they
can - which is not a crime, averyone wants the same with their own
browser.
> > Actually by going straight and fair IXMLHTTPRequest *is the standard*, and
> > XMLHttpRequest's of all kind are "standard emulations".
>
> Nonsense.
Why? Please be sure that I'm not a Microsoft defender.
But taking the situation as it is: side A invented a technology, side B
implemented this technology later on another basis. Neither of both
implementations are part of any international standards. One says "B is
the standard one no matter what". I'm still asking why?
> And what you comfortably forgot to mention again is that Microsoft Corp.
> is a full W3C member as well, as that allows you to distinguish between
> Microsoft ("good guys") and W3C ("bad guys") in your own twisted way of
> percepting reality.
In my presumably twisted reality there are no "good guys" and "bad
guys". Microsoft, Inc. is not an Empire of Evil and Mozilla Foundation
is not a Jeday Union. it's all for teens.
The only evil (as I'm not tired to repeat) is an absolute dominance of
one force on the market - and in this domain I do not care who could it
be: W3C, Mozilla or Microsoft. Any absolute dominance leads to
stagnation, and end-users are forced to eat the only one available
meal... or stay hungry.
Also there is a cardinal difference between ECMA International and W3C.
ECMA Internation is not appointed / self-appointed to keep and develope
ECMAScript. It is just a broad range standartization bureau. They can
write for you (based on provided specs and samples) ECMAScript,
microchips know-how or TV set user manual, just gimme money.
Therefore TC39-TG1 group members activity and communication to each
other is the crutial and the only force giving the trend to the further
ECMAScript development. Plus of course our (user and developers)
feedback and complains.
<http://msdn.microsoft.com/library/en-us/dnclinic/html/scripting07142000.asp>
is dated July 14, 2000 but became actual again.
Thomas 'PointedEars' Lahn
> Thomas 'PointedEars' Lahn wrote:
>> VK wrote:
>> > It is not a shift, but another step forward towards wannabes in the
>> Wannabes like you?
>
> "wannabes" *doesn't* mean at all "loosers".
^^^^^^^
Perhaps you should learn more about the English language before you use it.
(Does it not make you think that even me, who is a speaker of English as
foreign language, say that?)
A wannabe is (by all accepted definitions) someone who wants to be and is
often pretending (and often failing) to be someone or something that he is
not (from: want to be -- [colloq.] wanna be). A _loser_ is (by the same
definitions) someone who loses or someone who seems to be born to lose.
Both terms are used derogatory, so you are simply not making any sense.
However, the term "wannabe" definitely fits for you regarding software
development and being a hacker.
> But if this is too upseting for your ears then "other browser producers
> targeted to take leading positions on the browser market".
>
>> Nobody was forced to do anything. It seemed like a good idea and there
>> was no standard at the time, therefore it was implemented. And it was
>> implement without ActiveX.
>
> Totally right. [...]
You are such a hypocrite. You tell something utterly wrong, I tell you that
you are utterly wrong and tell you that the opposite is true and then you
say I am totally right, but ... something.
>> > W3C (first time in many years) had to return to the regular order of
>> > things then instead of inventing some standard out-of-head they have to
>> > describe the existing one. They did not like so much that the relevant
>> > part of specification is announced but still not finished.
>>
>> Utter nonsense.
>
> Presuming I'm developing a new browser (non-Gecko based) with
> XMLHttpRequest support, what official paper should I follow to make it
> standard-compliant?
> http://www.w3.org/.... ???
> http://www.ecma-international.org/....???
The question itself shows that you have not understood what you are talking
about.
You should not be allowed to develop anything except your personality, for
which there is great need indeed, for the time being. Given your posting
history, what you develop is inherently harmful to others.
>> > In this concern the "standard way" term is not applicable to
>> > IXMLHTTPRequest / XMLHttpRequest.
>>
>> Nonsense. If Microsoft would push this to be a W3C standard, it is not
>> unlikely that it becomes one.
>
> But it did not. [...]
And that is probably the fault of the W3C in your twisted version of
reality.
>> > Actually by going straight and fair IXMLHTTPRequest *is the standard*,
>> > and XMLHttpRequest's of all kind are "standard emulations".
>>
>> Nonsense.
>
> Why? [...]
Because a standard requires much more to be one, and the term "standard
emulation" is just as nonsensical as the rest of your argument (if it
deserves to be called so).
>> And what you comfortably forgot to mention again is that Microsoft Corp.
>> is a full W3C member as well, as that allows you to distinguish between
>> Microsoft ("good guys") and W3C ("bad guys") in your own twisted way of
>> percepting reality.
>
> In my presumably twisted reality there are no "good guys" and "bad
> guys". Microsoft, Inc. is not an Empire of Evil and Mozilla Foundation
> is not a Jeday Union. it's all for teens.
You are talking that now, but you are argumenting the opposite on numerous
other (probably also future) occasions, including the posting I am replying
to. You are a hypocrite par excellence.
> [ECMA vs. W3C]
Again you have no clue what you are writing about. Figures.
PointedEars
VK
Thomas 'PointedEars' Lahn wrote:
> VK wrote:
>
> > Thomas 'PointedEars' Lahn wrote:
> >> VK wrote:
> >> > It is not a shift, but another step forward towards wannabes in the
> >> Wannabes like you?
> >
> > "wannabes" *doesn't* mean at all "loosers".
> ^^^^^^^
> Perhaps you should learn more about the English language before you use it.
> (Does it not make you think that even me, who is a speaker of English as
> foreign language, say that?)
<offtopic>
A wannabe is who aspires, often vainly, to emulate another's success or
attain eminence in some area. It doesn't imply " vainly" (like
"looser") but allows it. So yes "wannabe" is a bit humorous but not
intentionally dispropagating, so I reserve my rights to use this word
in the situations I feel convenient for.
Thank you for your care about my language education, but I'm pretty
fine with it w/o your assistance.
</offtopic>
<ontopic>
204 No Content
</ontopic>
:-)
Randy Webb
Looser - less tight. Loose - not tight.
Lose - not win. Loser - one who loses.
And then you know his implications about pointing at looser.
But it is typically a loser that points out such things.
VK
Randy Webb wrote:
> Looser - less tight. Loose - not tight.
>
> Lose - not win. Loser - one who loses.
I be damned! A shameful typo...
"score adjusted", something like that? :-)
</offtopic>
Randy Webb
You learn fast Tonto :)
> </offtopic>
Joe Attardi
PointedEars --
I understand and respect that you have a lot of experience and
knowledge with these sorts of things. But why must you show such harsh
contempt for those who might not know as much as you, or who might have
the wrong idea about something?
Randy Webb
>> Although the posted URL is incorrect:
>
> Thanks for that catch Randy. I must have been out of it yesterday to
> not catch that error in the URL! Sorry about that :)
>
I don't think I am the only one that caught it, I think I am just the
only one to post the corrected URL :)
Randy Webb
>> Again you have no clue what you are writing about. Figures.
>
> PointedEars --
>
> I understand and respect that you have a lot of experience and
> knowledge with these sorts of things.
Don't be fooled by Thomas. He is a brilliant researcher. He can quote
specs that nobody gives a hoot about. That doesn't make him
knowledgeable though, only a good researcher.
> But why must you show such harsh contempt for those who might
> not know as much as you, or who might have the wrong idea about something?
The sad irony is that if he would drop his behavior problems and apply
some common sense and decency to his posts then he might return to being
pleasurable to read and converse with.
Joe Attardi
> only one to post the corrected URL :)
I'll keep my flame retardant suit on for a few days, just in case.
Thomas 'PointedEars' Lahn
>> Again you have no clue what you are writing about. Figures.
>
> PointedEars --
If you want to contact me, you should use private e-mail (use the Reply-To
address, not the From address).
> I understand and respect that you have a lot of experience and
> knowledge with these sorts of things. But why must you show such harsh
> contempt for those who might not know as much as you, or who might have
> the wrong idea about something?
Because unfortunately, it is not anyone, is not someone who just might not
know as much as me, neither is it someone who just might have the wrong
idea about something, nor is it just something:
<URL:http://groups.google.com/groups?as_ugroup=comp.lang.javascript&as_uauthors=VK&scoring=d&filter=0>
HTH
PointedEars
Martin Honnen
Randy Webb wrote:
>>> starting in IE7 XMLHttpRequest will be a plain JavaScript object.
>
> Yes it will.
> Although the posted URL is incorrect:
>
> <URL: http://blogs.msdn.com/ie/ >
>
> <quote>
>
> In IE7, XMLHTTP is now also exposed as a native script object
>
> </quote>
>
>> It will still be a host object accessible through _JScript_.
>
>
> Read the blog. Its a "native script object".
Current JScript engine details on Windows XP SP 2 before installing IE 7
Beta 2 preview are:
JScript Version 5.6.8820
JScript engine details after installing IE 7 Beta 2 preview are:
JScript Version 5.6.8833
Running the following WSH script with cscript
WScript.Echo('typeof XMLHttpRequest: ' + (typeof XMLHttpRequest));
shows
typeof XMLHttpRequest: undefined
with IE 7 Beta 2 preview installed so there is no native XMLHttpRequest
object in JScript.
The MSDN documentation here
<http://msdn.microsoft.com/library/default.asp?url=/workshop/author/dhtml/reference/objects/obj_xmlhttprequest.asp>
lists XMLHttpRequest as a host object of Internet Explorer not as part
of the JScript engine.
The check at
<http://home.arcor.de/martin.honnen/javascript/msxmlVersionCheck1.html>
indicates it is MSXML 3 behind the scenes in IE 7 (at least for
responseXML).
So whoever wrote that blog does not use the term "native object" with
the meaning defined in the ECMAScript edition 3 specification:
"4.3.6 Native Object
A native object is any object supplied by an ECMAScript implementation
independent of the host environment."
--
Martin Honnen
http://JavaScript.FAQTs.com/
Thomas 'PointedEars' Lahn
> Randy Webb wrote:
>>>> [According to Sunava Dutta, an IE Program Manager,]
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^[1]
>>>> starting in IE7 XMLHttpRequest will be a plain JavaScript object.
>>
>> Yes it will.
>> [...]
>> <URL: http://blogs.msdn.com/ie/ >
>>
>> <quote>
>>
>> In IE7, XMLHTTP is now also exposed as a native script object
>>
>> </quote>
>>
>>> It will still be a host object accessible through _JScript_.
>> Read the blog. Its a "native script object".
>
> [...]
> Running the following WSH script with cscript
>
> WScript.Echo('typeof XMLHttpRequest: ' + (typeof XMLHttpRequest));
>
> shows
>
> typeof XMLHttpRequest: undefined
>
> with IE 7 Beta 2 preview installed so there is no native XMLHttpRequest
> object in JScript.
>
> The MSDN documentation [...] lists XMLHttpRequest as a host object of
> Internet Explorer not as part of the JScript engine.
>
> The check at
> <http://home.arcor.de/martin.honnen/javascript/msxmlVersionCheck1.html>
> indicates it is MSXML 3 behind the scenes in IE 7 (at least for
> responseXML).
(You should not declare something as UTF-8 when it is not. And you should
include an appropriate `charset' parameter in your Content-Type header.)
> So whoever wrote that blog does not use the term "native object" with
^^^^^^^^^^^^^^^^^^^^^^^[^1]
> the meaning defined in the ECMAScript edition 3 specification:
>
> "4.3.6 Native Object
> A native object is any object supplied by an ECMAScript implementation
> independent of the host environment."
Thanks, I am still smiling.
PointedEars, amused