Hi everyone,

Is there a function in php to check if the date is in the RFC-822 date
format? I would greatly appreciate your help.

prashan

--- In rss-public@yahoogroups.com, "prashan"
<prashan_wanigasekara@...> wrote:
> Is there a function in php to check if the date is in the RFC-822 date
> format? I would greatly appreciate your help.

What language are you coding in?

Hi,

I try to be precise in a generic and popularized RSS definition for non
coders.

RSS is an XML dialect. It can handle external namespaces declared by "xmlns"
attributes.

But, hum, do you consider RSS as "namespace" in itself ? Or is it incorrect
to use this word.

In fact, I try to explain "namespace", in an RSS/Atom and modules/extensions
view, describing it as a group of elements all possibly used in a specific
goal, and needing to be declared to avoid polysemic confusion.

But how would you explain that RSS is a namespace... but that you don't have
to declare it the usual way...  (with "xmlns") ? Is the "<rss
version=2.0>... </rss>" section enough to say that the RSS namespace is the
basic namespace I the newsfeed ?

Thanks

secou

* secou <secou@...> [2009-02-02 16:30]:
> RSS is an XML dialect.

“RSS is an XML vocabulary.”

> It can handle external namespaces declared by "xmlns"
> attributes.

“It can incorporate other vocabularies if they have a namespace.
Namespaces must be declared with `xmlns` attributes.”

> But, hum, do you consider RSS as "namespace" in itself? Or is
> it incorrect to use this word.

No. Yes.

> In fact, I try to explain "namespace", in an RSS/Atom and
> modules/extensions view, describing it as a group of elements
> all possibly used in a specific goal, and needing to be
> declared to avoid polysemic confusion.
>
> But how would you explain that RSS is a namespace... but that
> you don't have to declare it the usual way...  (with "xmlns") ?
> Is the "<rss version=2.0>... </rss>" section enough to say that
> the RSS namespace is the basic namespace I the newsfeed ?

See above.

Regards,
--
Aristotle Pagaltzis // <http://plasmasturm.org/>

I've always found this mildly problematic.  Ideally, you should be
able to define a namespace for RSS and prefix the elements, but I
imagine that would break 99% of RSS clients, who aren't expecting RSS
elements to be prefixed at all.

--- In rss-public@yahoogroups.com, Aristotle Pagaltzis <pagaltzis@...>
wrote:
>
> * secou <secou@...> [2009-02-02 16:30]:
> > RSS is an XML dialect.
>
> “RSS is an XML vocabulary.”
>
> > It can handle external namespaces declared by "xmlns"
> > attributes.
>
> “It can incorporate other vocabularies if they have a namespace.
> Namespaces must be declared with `xmlns` attributes.”
>
> > But, hum, do you consider RSS as "namespace" in itself? Or is
> > it incorrect to use this word.
>
> No. Yes.
>
> > In fact, I try to explain "namespace", in an RSS/Atom and
> > modules/extensions view, describing it as a group of elements
> > all possibly used in a specific goal, and needing to be
> > declared to avoid polysemic confusion.
> >
> > But how would you explain that RSS is a namespace... but that
> > you don't have to declare it the usual way...  (with "xmlns") ?
> > Is the "<rss version=2.0>... </rss>" section enough to say that
> > the RSS namespace is the basic namespace I the newsfeed ?
>
> See above.
>
> Regards,
> --
> Aristotle Pagaltzis // <http://plasmasturm.org/>
>

* scamden <sterling@...> [2009-02-03 01:25]:
> Ideally, you should be able to define a namespace for RSS and
> prefix the elements

1. The syntax which you might use for this actually undeclares
    the prefix. You can’t actually declare a prefix for the null
    namespace.

2. What would be the point? The elements stil aren’t in any
    namespace.

Regards,
--
Aristotle Pagaltzis // <http://plasmasturm.org/>

If you wanted to use the null namespace for some other purpose, then
you could prefix the RSS elements with a properly declared namespace
prefix (in an ideal world, and in standard XML grammar).  But RSS has
traditionally been used without any prefixes, and clients expect it
that way.  I'm not asking for a change, I'm just pining for the
fjords.

--- In rss-public@yahoogroups.com, Aristotle Pagaltzis <pagaltzis@...>
wrote:
>
> * scamden <sterling@...> [2009-02-03 01:25]:
> > Ideally, you should be able to define a namespace for RSS and
> > prefix the elements
>
> 1. The syntax which you might use for this actually undeclares
>    the prefix. You can’t actually declare a prefix for the null
>    namespace.
>
> 2. What would be the point? The elements stil aren’t in any
>    namespace.
>
> Regards,
> --
> Aristotle Pagaltzis // <http://plasmasturm.org/>
>

* scamden <sterling@...> [2009-02-03 03:00]:
> If you wanted to use the null namespace for some other purpose,
> then you could prefix the RSS elements with a properly declared
> namespace prefix (in an ideal world, and in standard XML
> grammar). But RSS has  traditionally been used without any
> prefixes, and clients expect it that way. I'm not asking for a
> change, I'm just pining for the fjords.

I tried that with Atom, which *is* in a namespace, and even then
most clients broke: http://plasmasturm.org/log/376/

The situation is greatly improved now among the popular
aggregators but I have no doubt that lots of fringe software is
still broken in this respect.

Regards,
--
Aristotle Pagaltzis // <http://plasmasturm.org/>

I think there's two scenarios that exist with RSS vis-a-vis
namespaces.
-RSS elements in RSS documents
-RSS elements in other XML documents

1. RSS elements in RSS documents

These RSS element are not in a namespace. This inevitably brings
about debate. Some suggest that there is some sort of null XML
namespace, but nobody has ever shown me where this is defined. Please
do call me on this.

2. RSS elements in other XML documents

Nobody has ever made any formal proposal on this. The closest we ever
came was PSS, which I'm unsure if it was a joke, but Dave Winer never
moved it forward.
http://www.ibt4im.com/?guid=20030615083037

I think it's time to proposal PSS for adoption by the RSS Advisory
Board in order to clarify use of RSS in other XML documents.

Thoughts and criticism welcome and wanted.

Randy
http://www.therssweblog.com


--- In rss-public@yahoogroups.com, "secou" <secou@...> wrote:
>
> Hi,
>
> I try to be precise in a generic and popularized RSS definition for
non
> coders.
>
> RSS is an XML dialect. It can handle external namespaces declared
by "xmlns"
> attributes.
>
> But, hum, do you consider RSS as "namespace" in itself ? Or is it
incorrect
> to use this word.
>
> In fact, I try to explain "namespace", in an RSS/Atom and
modules/extensions
> view, describing it as a group of elements all possibly used in a
specific
> goal, and needing to be declared to avoid polysemic confusion.
>
> But how would you explain that RSS is a namespace... but that you
don't have
> to declare it the usual way...  (with "xmlns") ? Is the "<rss
> version=2.0>... </rss>" section enough to say that the RSS
namespace is the
> basic namespace I the newsfeed ?
>
> Thanks
>
> secou
>

Dave Winer's original suggestion of a portable RSS format is now
available here:

http://www.rssboard.org/news/186

The original is gone from Harvard's blog server.

Reading that post and Randy's PSS 0.9 draft, I agree that we need a
way to use RSS elements in other XML dialects. But I'm not convinced
that the simplest way to do this would be to create a new format.

I think we should consider adding a sentence to the end of the
Extending RSS section [1]:

"Using RSS elements in other XML dialects requires the namespace
declaration "http://www.rssboard.org/rss-namespace/". This declaration
MUST NOT be used in an RSS document."

The preceding paragraph currently states that RSS elements "are not
themselves members of a namespace" for backwards compatibility.

I like this suggestion. Unless someone objects, I will propose it to
the advisory board. I'll give this board a couple days to vent.

Randy
http://www.therssweblog.com


--- In rss-public@yahoogroups.com, "rcade" <cadenhead@...> wrote:
>
> Dave Winer's original suggestion of a portable RSS format is now
> available here:
>
> http://www.rssboard.org/news/186
>
> The original is gone from Harvard's blog server.
>
> Reading that post and Randy's PSS 0.9 draft, I agree that we need a
> way to use RSS elements in other XML dialects. But I'm not convinced
> that the simplest way to do this would be to create a new format.
>
> I think we should consider adding a sentence to the end of the
> Extending RSS section [1]:
>
> "Using RSS elements in other XML dialects requires the namespace
> declaration "http://www.rssboard.org/rss-namespace/". This
declaration
> MUST NOT be used in an RSS document."
>
> The preceding paragraph currently states that RSS elements "are not
> themselves members of a namespace" for backwards compatibility.
>

rcade wrote:
> I think we should consider adding a sentence to the end of the
> Extending RSS section [1]:
>
> "Using RSS elements in other XML dialects requires the namespace
> declaration "http://www.rssboard.org/rss-namespace/". This declaration
> MUST NOT be used in an RSS document."

My concern is that, no matter what you say in the spec, you can be sure that
people will start using that namespace in regular feeds. And when they do, I
suspect many feed readers are going to break.

I think we can possibly lessen that breakage by choosing a namespace URI
that is already in use in the wild, so there's some chance that feed readers
will already support it. The one that I've seen most often is:
http://backend.userland.com/rss2

Before this amendment is proposed to the board, I would like to run a few
tests to see how big an issue this is and if the choice of namespace URI
makes any real difference.

Regards
James

--- In rss-public@yahoogroups.com, "James Holderness" <j4_james@...>
wrote:
> I think we can possibly lessen that breakage by choosing a namespace
URI
> that is already in use in the wild, so there's some chance that feed
readers
> will already support it. The one that I've seen most often is:
> http://backend.userland.com/rss2

I look forward to seeing the results of your test, but since that URI
was used to define a namespace for RSS 2.0, won't that be received as
an encouragement to do exactly that?

Also, that URI lacks a trailing slash, and it's not under our control,
so we can't educate anyone who requests it in a browser on the proper
and improper use of namespace declarations in RSS.

We've had good success thus far in directing implementers to proper
use, through the spec, profile and the work with the Feed Validator
crew. The spec is clear on the lack of a namespace in RSS.

I will wait till you report back before I do anything.

Randy
http://www.therssweblog.com

--- In rss-public@yahoogroups.com, "James Holderness" <j4_james@...>
wrote:
>
> Before this amendment is proposed to the board, I would like to run a
few
> tests to see how big an issue this is and if the choice of namespace
URI
> makes any real difference.
>
> Regards
> James
>

rcade wrote:
> I look forward to seeing the results of your test,

Results now up on my blog:
http://www.xn--8ws00zhy3a.com/blog/2009/02/a-namespace-for-rss

> but since that URI
> was used to define a namespace for RSS 2.0, won't that be received as
> an encouragement to do exactly that?

I wouldn't think so. I may be wrong, but I was under the impression that
Dave experimented with it briefly in his feed, but never formally proposed
it be used for anything.

> Also, that URI lacks a trailing slash, and it's not under our control,
> so we can't educate anyone who requests it in a browser on the proper
> and improper use of namespace declarations in RSS.

I don't see the value of a trailing slash - the Atom namespace doesn't have
one and that has never been an issue AFAIK. The fact that it's not under the
board's control is a valid point though.

However, given that the namespace is in use in the wild and is conciously
supported by at least a few feed readers, I think creating a new namespace
assigned to the same set of elements is a bad idea. We could end up with the
same chaos we have with the Yahoo media rss extension, with different people
using different namespaces, and it'll end up an interoperability nightmare.

Regards
James

Thanks James!

There is only 7 clients that support the userland namespace and
wouldn't support a new namespace. I would go with a new namespace and
a bold warning against using the namespace in RSS documents. In fact,
I'm putting a warning on this thread, just in case.

    !!!WARNING!!!!
In no way should anybody interpret anything in this thread or any
spec changes resulting from this thread as permission to use a
namespace for RSS 2.0 elements in an RSS 2.0 document.
    !!!WARNING!!!!

Further, I wouldn't require the use of the namespace, but rather
suggest the use of the namespace when and only when it helps embed
RSS elements in other documents.

MHO,

Randy
http://www.therssweblog.com


--- In rss-public@yahoogroups.com, "James Holderness" <j4_james@...>
wrote:
> Results now up on my blog:
> http://www.xn--8ws00zhy3a.com/blog/2009/02/a-namespace-for-rss
>

On 5 Feb 2009, at 05:20, James Holderness wrote:

> rcade wrote:
>> I look forward to seeing the results of your test,
>
> Results now up on my blog:
> http://www.xn--8ws00zhy3a.com/blog/2009/02/a-namespace-for-rss

FWIW, SimplePie used to support the Userland namespace, but dropped
support for it seeming it was used nowhere and does have a cost. It
appears (looking with grep�) to have been introduced for 1.0 Beta 2
and removed for 1.0 RC1 (i.e., the last release with support was 1.0
Beta 3 Bugfix 2). However, these versions oddly seem to match
element's expanded names case-insensitively (I'm sure I thought there
was some good compat. reason for this at the time). Furthermore,
versions prior to 1.0 Beta 2 (though maybe not 0.8, but I never had
anything to do with that) will have worked with all but the prefixed
variations (due to not using a namespace-aware XML parser).

To my knowledge, nothing broke by moving to a namespace-aware parser,
and nothing broke removing support for the Userland namespace.


--
Geoffrey Sneddon
<http://gsnedders.com/>

On 4 Feb 2009, at 04:57, rcade wrote:

> Reading that post and Randy's PSS 0.9 draft, I agree that we need a
> way to use RSS elements in other XML dialects. But I'm not convinced
> that the simplest way to do this would be to create a new format.
>
> I think we should consider adding a sentence to the end of the
> Extending RSS section [1]:
>
> "Using RSS elements in other XML dialects requires the namespace
> declaration "http://www.rssboard.org/rss-namespace/". This declaration
> MUST NOT be used in an RSS document."

To give my opinion as an implementer: do we really need another feed
format? Effectively, aggregators have support RSS 0.90, RSS 2.0, Atom
0.3, and Atom 1.0 � do we really have to add RSS-2.0-in-a-namespace to
that list?

Although we wouldn't have to support it as a feed format per-se, what
we would have to do would amount to supporting it for everything
except as a root element (for the sake of supporting, e.g.,
{http://www.rssboard.org/rss-namespace/
}title within an Atom document). I'm also dubious about that "MUST
NOT" � history has taught us content creators often ignore specs, and
I would not be surprised if we had to add support for it as a document
format (though with support for it within Atom that'd mainly be an
already sunk cost).

Furthermore, I'm doubtful of the real need for RSS-in-a-namespace:
what does RSS-in-a-namespace give us that Atom can't already do?
Unless a good answer can be found to that question, I'm against
creating (what amounts to) another XML format.


--
Geoffrey Sneddon
<http://gsnedders.com/>

You don't and nobody has to support the userland namespace. I don't
have any stats, but I'm betting that almost nobody uses it.
Thanks,

Randy

--- In rss-public@yahoogroups.com, Geoffrey Sneddon <foolistbar@...>
wrote:
>

On Wed, Feb 4, 2009 at 9:20 PM, James Holderness <j4_james@...> wrote:
> Results now up on my blog:
> http://www.xn--8ws00zhy3a.com/blog/2009/02/a-namespace-for-rss

I suspect many of those results need to be considered with care: the
two I know about are Firefox, which for reasons beyond the ken of
mortal man considers anything in any namespace for which
.indexOf("http://backend.userland.com") == 0 to be RSS (or more
strictly, considers it to not be in any namespace when translating
from {uri}element to its internal table of known prefix:element
things, so that {http://backend.userland.com/catsAreFurry}title is a
generic title-thing with the generic lumped handling of all various
RSS titles), and Thunderbird, which we desperately hope will be using
the same parser as Firefox for 3.0, but which currently considers the
namespace URI for anything thrown at it to be the namespace URI of the
first element with the localname "channel", so it will happily parse
"RSS" in the http://example.com/goats namespace, unless you have a
{http://backend.userland.com/rss2}channel or
{http://example.com/foopy}channel or {}channel coming first in the
document, in which case it will ignore later elements in any other or
no namespace (that horrible-sounding behavior looks to be an
accommodation to support a single parser for both 0.91+ and 0.90
without having to explicitly check whether the first channel is in the
0.90 namespace or in no namespace, rather than an attempt to support
any particular 2.0+ namespaces).

Phil Ringnalda wrote:
> I suspect many of those results need to be considered with care: the
> two I know about are Firefox, which for reasons beyond the ken of
> mortal man considers anything in any namespace for which
> .indexOf("http://backend.userland.com") == 0 to be RSS

That makes sense. I'm pretty sure that the Universal Feed Parser used by
Planet Venus does something similar, and I know for a fact that Snarfer does
(although slightly more restrictive).

I didn't mean to suggest that the one userland namespace that I tested was
the only one that would work with those clients. The point was to determine
what percentage of clients had explicit support for a particular namespace
URI (or set of URIs), and which clients were ignoring the namespace URI
altogether.

The URI I chose for my test was just the one that I'd encountered most
frequently in the wild.

Regards
James

* Randy Morin <randy@...> [2009-02-03 04:25]:
> I think it's time to proposal PSS for adoption by the RSS
> Advisory Board in order to clarify use of RSS in other XML
> documents.

What need would it serve?

* rcade <cadenhead@...> [2009-02-04 06:00]:
> Reading that post and Randy's PSS 0.9 draft, I agree that we
> need a way to use RSS elements in other XML dialects.

For what purpose?

Regards,
--
Aristotle Pagaltzis // <http://plasmasturm.org/>

Geoffrey Sneddon:
> nothing broke removing support for the Userland namespace.

I know that feeds using the Userland namespace did exist in the past, and I
know of at least one that still exists[1]. Unfortunately there's no easy way
to determine how widespread the usage is today (for me at least).

It used to be possible to find such feeds with a simple Google search, but
they removed that functionality a while back. I've often been tempted to
develop my own web crawler specifically for gathering this kind information,
but then I remember how lazy I am and go back to taking a nap.

Regards
James

[1] http://www.jsfcentral.com/?feed=rss

The need is to embed RSS in other XMLs. Currently you cannot do this within any
element that has a default namespace.

Randy

--- In rss-public@yahoogroups.com, Aristotle Pagaltzis <pagaltzis@...> wrote:
>
> * Randy Morin <randy@...> [2009-02-03 04:25]:
> > I think it's time to proposal PSS for adoption by the RSS
> > Advisory Board in order to clarify use of RSS in other XML
> > documents.
>
> What need would it serve?
>
> * rcade <cadenhead@...> [2009-02-04 06:00]:
> > Reading that post and Randy's PSS 0.9 draft, I agree that we
> > need a way to use RSS elements in other XML dialects.
>
> For what purpose?
>
> Regards,
> --
> Aristotle Pagaltzis // <http://plasmasturm.org/>
>

Randy Morin wrote:
> The need is to embed RSS in other XMLs. Currently you cannot do this
> within any element that has a default namespace.

I think Aristotle was looking for something a little more specific than
that. Why do we need to embed RSS in other XML vocabularies? I too would
like to hear at least one compelling use-case.

Atom has had embedding functionality from the start, so if such a concept
really is useful, there must be someone out there doing something
interesting with it. Where are all these cool embedded Atom applications?

Not that I'm saying they don't exist - I just don't know of any. I think it
would be helpful to the discussion if we had something more concrete to
review.

Regards
James

On 2/5/2009 5:05 PM, James Holderness wrote:
> That makes sense. I'm pretty sure that the Universal Feed Parser used by
> Planet Venus does something similar

Something different but with more amusing possibilities: UFP thinks that
http://www.rssboard.org/stuff/things/backend.userland.com/rssRULEZTEHUNIVERSE
is a proper namespace URI for RSS - it just wants the string
backend.userland.com/rss somewhere within the namespace URI, beyond that
it's happy to let a thousand namespace URIs for RSS bloom.

--- In rss-public@yahoogroups.com, Aristotle Pagaltzis <pagaltzis@...>
wrote:
> What need would it serve?

RSS 2.0 is one of the most widely supported XML dialects in existence.
That puts a lot of code out there that handles RSS elements and a lot
of brains accustomed to structuring data as RSS items. (When I create
MySQL tables for web content, I end up using item element names as
table fields -- I've become wired to define everything in terms of
title-link-description.)

Some potential uses for an RSS 2.0 namespace:

1. Adding podcasting support to other formats that lack it, using the
enclosure element.

2. Adopting the basic item elements -- title, link and description --
in a weblog backup format.

3. Associating a thumbnail image with other XML formats (such as OPML)
using the channel-image element.

4. Adding information from RSS items to weblogUpdates a.k.a.
changes.xml, the common update
format for weblog pings.

5. Embedding RSS in XHTML (see James Snell's blog post at
http://tinyurl.com/b9vaad ).

6. Whatever Sam Ruby and Tim Bray had in mind in 2003 when they
pitched an RSS namespace in 2003 (see http://www.rssboard.org/news/186 ).

More generally, though, the argument for using RSS as a namespace
elsewhere is the same for using any XML dialect elsewhere. Reusability
promotes interop and reduces the need to specify new behavior.

On 6 Feb 2009, at 04:32, rcade wrote:

> --- In rss-public@yahoogroups.com, Aristotle Pagaltzis <pagaltzis@...>
> wrote:
>> What need would it serve?
>
> RSS 2.0 is one of the most widely supported XML dialects in existence.
> That puts a lot of code out there that handles RSS elements and a lot
> of brains accustomed to structuring data as RSS items.

But "title" and "{http://backend.userland.com/rss2}title" _aren't_ the
same element, and the same goes for any other local name, and as such
for anyone using a namespace aware parser (which they already should
be seeming RSS can contain foreign content) they are totally different
elements.

> Some potential uses for an RSS 2.0 namespace:
>
> 1. Adding podcasting support to other formats that lack it, using the
> enclosure element.
>
> 2. Adopting the basic item elements -- title, link and description --
> in a weblog backup format.
>
> 3. Associating a thumbnail image with other XML formats (such as OPML)
> using the channel-image element.
>
> 4. Adding information from RSS items to weblogUpdates a.k.a.
> changes.xml, the common update
> format for weblog pings.
>
> 5. Embedding RSS in XHTML (see James Snell's blog post at
> http://tinyurl.com/b9vaad ).
>
> 6. Whatever Sam Ruby and Tim Bray had in mind in 2003 when they
> pitched an RSS namespace in 2003 (see http://www.rssboard.org/news/
> 186 ).

Atom already provides all this. Why do we need RSS too?

> More generally, though, the argument for using RSS as a namespace
> elsewhere is the same for using any XML dialect elsewhere. Reusability
> promotes interop

We already have a fair amount of interop. for Atom within other
foreign documents, but not for RSS. Why do we want to add something
else that is not already well supported? What do we gain by using RSS
and not Atom?

> and reduces the need to specify new behavior.

That's exactly what you're doing though, putting RSS into a namespace.
We already have Atom which handles all these use-cases fine.


--
Geoffrey Sneddon
<http://gsnedders.com/>

On 5 Feb 2009, at 19:23, Randy Morin wrote:

> You don't and nobody has to support the userland namespace. I don't
> have any stats, but I'm betting that almost nobody uses it.

This is completely missing my point: if we put something in the spec
it massively increases the likelihood of it being used in the real
world, and as such the likelihood of implementers needing to implement
it to support deployed content increases. Splitting up the feed format
market even more does not seem like a good idea.


--
Geoffrey Sneddon
<http://gsnedders.com/>

On 6 Feb 2009, at 01:46, James Holderness wrote:

> Geoffrey Sneddon:
>> nothing broke removing support for the Userland namespace.
>
> I know that feeds using the Userland namespace did exist in the
> past, and I
> know of at least one that still exists[1]. Unfortunately there's no
> easy way
> to determine how widespread the usage is today (for me at least).
>
> It used to be possible to find such feeds with a simple Google
> search, but
> they removed that functionality a while back. I've often been
> tempted to
> develop my own web crawler specifically for gathering this kind
> information,
> but then I remember how lazy I am and go back to taking a nap.

This sorta helps:
<http://google.com/codesearch?hl=en&lr=&q=%22http%3A%2F%2Fbackend.userland.com%2\
Frss2%22&sbtn=Search
  >.

That said, it mainly contains feed parsers, so it would appear a fair
number of feed parsers already support it, even if next to nothing
uses it.


--
Geoffrey Sneddon
<http://gsnedders.com/>