The optional language element is for an entire feed. I'm delighted
to find that this doesn't work for some people. Many people,
especially outside the USA, write in two or more languages in one
blog. Would it make sense to extend the option tag to also work as
an item element? The default language would be the one for the
channel, but an item's language element could override.

On the writing end, blogging tools could offer an optional checkbox
for second, third languages when editing a post.

On the reading end today, you can select those posts that you can
read.

On the reading end tomorrow, your newsreader will be able to run the
posts you can't read through Babelfish or some other machine
translation.

Inspired by a trip to http://blogspot.net in Vienna.

Hi

On Mon 02-Jun-2003 at 04:43:23PM -0000, Phil Wolff wrote:
> The optional language element is for an entire feed.

It can also be used with items and this has always been the case with
RSS 1.0.

Also all elements can use the xml:lang attribute and if we ever get
around to allowing element to be repeated you could do things like
this:

   <dc:title xml:lang="en">Foo</dc:title>
   <dc:title xml:lang="de">Foo</dc:title>
   <dc:language>en</dc:language>
   <dc:language>de</dc:language>

(For backwards compatability we wouldn't want <link> and <title>
elements to be repeatable)

Chris

--
Chris Croome                               <chris@...>
web design                             http://www.webarchitects.co.uk/
web content management                               http://mkdoc.com/

I came across some feeds using the 'und' language tag.  Presumably, to indicate
that the language is 'undetermined'.  A little further digging reveals there's
also a 'mul' language tag for indicating multiple languages.  What's missing (in
RSS), however, is a way to indicate /which/ languages are present if multiples
are used.

So, in the context of a feed that specifically /knows/ that multiple languages
are present it seems reasonable to use the <language>mul</language> markup.  If,
howevever, a feed knows that multiple languages /might/ be present without
knowing which ones it might be reasonable to assume that using the
<language>und</language> tag is appropriate.

As in, a site knows it will regularly include a mix of languages versus a
posting forum that knows many different languages might be present but not which
ones.  The specs seem to say it's best not to express 'und' unless the protocol
forces you to express one.  RSS does want a language tag to be present so
leaving it out is, technically, not an option.  But are there a significant
number of situations that will truly never 'know' what default language is
present?

I'm not sure this will help clarify things either way.   Knowing that a feed
'most frequently' uses a given single language is usually what most users would
need.  That being a way for speakers of that language to find that feed.  That
also being a way for programmatic translators to understand /where/ to start in
the translation.  Should a feed contain an item in a language other than that
indicated for the channel the item could indicate it on it's own.

Is there a huge need for expressing multiple languages in the context of the
channel's definition?  Or is having 'just one' language enough?  If it's not
'enough' then what's the most RDF-friendly way to express multiple languages
being present?

Since the ISO639 specs support the 'mul' code it sort of confuses things.
Without using 'mul' it might be possible simply contain multiple codes within
the <language/> element.  Although using which container style would certainly
be open for debate.  But since the 'mul' code exists it seems appropriate to use
it.  The question then becomes how to include additional markup that indicates
which langugages are present?  What would the XML markup for this look like?

I suppose this also falls over into the FoaF area of interest.  As in, a person
might speak multiple languages.  So how woudl it be proper to express this?
Which also leads into the 'reads' versus 'speaks' and degrees of fluency
issues...

Anyway, just thought I'd bring it up.  I'm open to suggestions on what would be
considered 'best practices' on marking things up with language code info.

-Bill Kearney

und and mul are both considered bad things in the RFC 3066 world, though
they are supported for protocols that insist on a language tag (forcing the
use of und), or which support only one (forcing the use of mul). See section
2.3 points 5 and 6.

In general it's preferred that you list multiple languages (as is done with
HTTP headers) or use a granular system to pick out which portions of a
document are in which language (as is done with XML and HTML).

> und and mul are both considered bad things in the RFC 3066 world, though
> they are supported for protocols that insist on a language tag (forcing the
> use of und), or which support only one (forcing the use of mul). See section
> 2.3 points 5 and 6.

That's certainly clear from reading the spec.  Consider that RSS expects the
expression of a language tag.  If a language isn't known then what?  The
original RSS spec made indication of language a required element.  The use of
Dublin Core for the language tag in 1.0 doesn't state clearly one way or the
other (I'd vote for continuing the requirement).  God knows what random level of
optional/required status is offered in the other forks of RSS.

> In general it's preferred that you list multiple languages (as is done with
> HTTP headers) or use a granular system to pick out which portions of a
> document are in which language (as is done with XML and HTML).

Thus my question, what's the least painful way (rdf-wise) to properly express
that a feed, in the context of the channel element, possesses more than one
language?

This is entirely different than how should a feed express that items are
alternates of each other based on their language variation (as in, this one's
english, this one's the exact same thing translated into japanese).  The two
questions are related, of course, but are not exactly the same.

A channel being able to properly express multiple languages affects how
directories (like Syndic8) would present the feed to users searching based on
language.  That would also affect whether or not a reader program would "know"
that it has to translate the contents of the feed in it's entirety should the
channel-wide language markup be different that what the user expects.  Think
'babelfish on a handheld'.

The use of language markup within items, however, affects how an RSS reader
program would present items to the user once having /found/ the feed.

So I'm after solutions to different questions.  One is how should a channel
properly markup that it consistently uses more than one language?  This leads to
the question of how should items express their language?  How should they
express their language when more than one is expressed from with the channel?
Then, finally, how should the feed indicate which items are 'the same' and
simply translations of another item?

I suppose this leads to the idea that /if/ you express that a channel has more
than one language then you /must/ express which language is being used on /all/
subsequent items within the feed.  If only one language is expressed it's safe
to assume items are in that language unless they individually state otherwise.
As in, if the feed's in Spanish and one item is in English, only that one item
needs to express a language tag within it.  I suppose the channel itself could
"get away with" not expressing that it contains mutliple languages since the one
item could be consider an exception to the otherwise consistent nature of the
feed's default language.  This as a way to avoid the problem of added markup
bulk.

As in, if the feed's usually in a given language then it's probably rather
unecessary to markup the channel with multiple languages.  This would force all
subsequent items to state which of those multiples they used.  This to allow
that a feed regularly in one language but on occassion using another doesn't
need to make that exact statement within the channel during those occassional
occurences.  The item's own statement of using a language would seem to cover
that situation 'effectively enough' for most situations.  Otherwise some sort of
'default' language logic would have to be applied and marked up as such.  Then
we'd get into the problem of how to say which language is the default.  I don't
know that there's enough situations where this would be warranted.

-Bill Kearney

Hi RSSers and sitescoopers --

I developed Sitescooper a few years back ( http://sitescooper.org/ ),
which scrapes news sites, blogs etc. and renders them down to Palm-format
output.  I haven't been using it much myself recently -- I've been getting
more into RSS and reading updates via mail (using rss2mail) that way,
instead of syncing them to my Palm and reading them there.

Recently, I've been running into blogs without decent RSS feeds (ie. short
or missing descriptions or content:encoded parts).

As a result, it occurred to me that Sitescooper could do with an RSS
output mode, which would deal with (a) getting around crappy RSS
0.91-style feeds, and (b) the sites that don't have RSS output at all
(although that's stepping on NewsIsFree's toes a little ;).  It's also a
handy way to scrape into RSS, given that sitescooper has

   - (a) lots of site descriptions which should mostly work (although a few
     are suffering bit rot now),

   - (b) uses the .site file format -- a simple format for rules on how to
     scrape "stories" from news sites effectively,

   - (c) has good caching mechanisms, and

   - (d) pretty good support for wierdness like HTTP redirects and
     authentication.

So anyway -- after some hacking, the CVS version of sitescooper now
supports scraping into RSS 2.0.  Some fruits of this can be seen at
http://sitescooper.org/rss/ .  Each .xml is accompanied by the relevant
.site file.

I don't think it's quite ready for a release just yet, but I thought I'd
let you all know about it and get some feedback ;)

--j.

Luka Sherron

www.ecriteria.blogspot.com

http://groups.yahoo.com/group/blogging_community

Luka Sherron wrote:

> I know things are touch and go for rss where complete saturation is
> concerned but I found a feedreader that may actually inspire more
> active participation
>

Why will it inspire more active participation?  Or is this a spam? :)

--
Help Support NewsMonster Development!  Purchase NewsMonster PRO!

     http://www.newsmonster.org/download-pro.html

Kevin A. Burton, Location - San Francisco, CA, Cell - 415.595.9965
        AIM - sfburtonator,  Web - http://www.peerfear.org/
GPG fingerprint: 4D20 40A0 C734 307E C7B4  DCAA 0303 3AC5 BD9D 7C4D
   IRC - openprojects.net #infoanarchy | #p2p-hackers | #reptile

jm-rss@... wrote:

>So anyway -- after some hacking, the CVS version of sitescooper now
>supports scraping into RSS 2.0.  Some fruits of this can be seen at
>http://sitescooper.org/rss/ .  Each .xml is accompanied by the relevant
>.site file.
>
uh... so support RSS 1.0 and get back to us! :)

Kevin

--
Help Support NewsMonster Development!  Purchase NewsMonster PRO!

     http://www.newsmonster.org/download-pro.html

Kevin A. Burton, Location - San Francisco, CA, Cell - 415.595.9965
        AIM - sfburtonator,  Web - http://www.peerfear.org/
GPG fingerprint: 4D20 40A0 C734 307E C7B4  DCAA 0303 3AC5 BD9D 7C4D
   IRC - openprojects.net #infoanarchy | #p2p-hackers | #reptile

Kevin A. Burton said:
> jm-rss@... wrote:
>
> >So anyway -- after some hacking, the CVS version of sitescooper now
> >supports scraping into RSS 2.0.  Some fruits of this can be seen at
> >http://sitescooper.org/rss/ .  Each .xml is accompanied by the relevant
> >.site file.

> uh... so support RSS 1.0 and get back to us! :)

Damn, I thought all that stuff was over :(   Guess I'd better get
hacking...

--j.

Kevin A. Burton said:
> uh... so support RSS 1.0 and get back to us! :)

OK -- it's now 1.0, viz. http://sitescooper.org/rss/ .  How's it look?

--j.

Kevin A. Burton said:

uh... so support RSS 1.0 and get back to us! :)

OK -- it's now 1.0, viz. http://sitescooper.org/rss/ . How's it look?

-- Help Support NewsMonster Development! Purchase NewsMonster PRO!
http://www.newsmonster.org/download-pro.html
Kevin A. Burton, Location - San Francisco, CA, Cell - 415.595.9965
AIM - sfburtonator, Web - http://www.peerfear.org/
GPG fingerprint: 4D20 40A0 C734 307E C7B4 DCAA 0303 3AC5 BD9D 7C4D
IRC - openprojects.net #infoanarchy | #p2p-hackers | #reptile

RSS Links is a collection of links from the announcement of RSS by
Netscape in March 1999 through the announcement of the RSS 1.0
Proposal in August 2000.

RSS Links includes links that relate to the development of the RSS
specification and participants, not generally usage or implementation
links.

The collection is available as a zipped set of RFC-822-style plain
text files:

   http://bitsko.slc.ut.us/2003/06/rss-links-0.3.zip

Seperate from the collected links is a list of selected links that the
author believes to be more substantive in the discussion.

The selected links can also be viewed online at:

   http://bitsko.slc.ut.us/cgi-bin/rss-links

A small Python script is included that has functions for reading the
collection of links, selecting URIs, and formatting lists of links.

RSS Links is released in the Public Domain.

Version 0.3 is the result of four link gathering sessions, including
entry of date, author, and subject information for all of the
"selected" links.

More links, comments, and more field information is welcome!

   -- Ken MacLeod
   ken at bitsko.slc.ut.us

Quick & dirty XSLT stylesheet to convert Outline Processor Markup Language
(OPML) format blogrolls into approximate Open Content Syndication (OCS).
Does the links bit, doesn't do the alternate format list (there isn't the
info).

http://dannyayers.com/2003/06/opml2ocs.xsl

I'm sure this has been done before, but I couldn't find one googling and was
in a hurry.

Cheers,
Danny.


----

http://dannyayers.com

Probably most of the people on this list are already well aware of the
work being done over at Sam Ruby's website on a next generation data model
and syndication effort for weblogs.

Its an interesting project, and so far one characterized by productive
conversation, and friendly agreements.  Would do well to have wide
community support, at least among the narrow field of people working on
syndication.

So, without further ado:
http://www.intertwingly.net/wiki/pie/


Kellan

--
"we reject kings, presidents and voting. we believe in
rough consensus and running code." [david clark (MIT)]

kellan@...

Pointers here:

http://backend.userland.com/2003/06/24#a302

Dave

Hi everyone. I hope I am posting to the right people, heh. I just
stumbled across this group while looking for some answers. I was
playing around with this fase4 rdf parser-

http://www.fase4.com/en/rdf/

I was really enjoying what I could do with it. Problem was, I had
just been toying with it for like two days, then I updated to the
most current version of php (on my home test server) and now I can't
seem to make it work anymore.

I found a link to this site from your links, so I assume someone
here might be somewhat familiar with this (I guess its called an
aggregator? parser? class? sorry for my ignorance on the subject)
and maybe someone can help me get back to toying with it again? I'm
hoping maybe there is some line of code I can change or something?
to make this thing work with the newest version of php?

Thanks for reading this, and I hope I'm not keeping you fine folks
from something more productive,

Patrick

Hello Patrik

i used fase4-rdf every time an PHP vesion,

what ist your old/new PHP version, because
the XSL parserfunction and also XML Dom changed.

if you dont find anything write am PM to me.

mfg franz@...

Am Mittwoch, 25. Juni 2003 07:26 schrieb Patrick:
> Hi everyone. I hope I am posting to the right people, heh. I just
> stumbled across this group while looking for some answers. I was
> playing around with this fase4 rdf parser-
>
> http://www.fase4.com/en/rdf/
>
> I was really enjoying what I could do with it. Problem was, I had
> just been toying with it for like two days, then I updated to the
> most current version of php (on my home test server) and now I can't
> seem to make it work anymore.
>
> I found a link to this site from your links, so I assume someone
> here might be somewhat familiar with this (I guess its called an
> aggregator? parser? class? sorry for my ignorance on the subject)
> and maybe someone can help me get back to toying with it again? I'm
> hoping maybe there is some line of code I can change or something?
> to make this thing work with the newest version of php?
>
> Thanks for reading this, and I hope I'm not keeping you fine folks
> from something more productive,
>
> Patrick
>
>
> To unsubscribe from this group, send an email to:
> rss-dev-unsubscribe@egroups.com
>
>
>
> Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/

Franz Wieser <franz@w...> wrote:
>
> i used fase4-rdf every time an PHP vesion,
>
> what ist your old/new PHP version, because
> the XSL parserfunction and also XML Dom changed.
>


hmmm.... well I'm not sure what my old version was.... possibly 4.0
something? The version I'm using now is 4.2.3

On my home machine I use this php dev package

http://www.firepages.com.au/devindex.htm

thanks for the insight,

Patrick

(copying FOAF list for info)

Hi all

Just a heads-up that I've spent a little time lately updating the FOAF site
at http://rdfweb.org/ -- nothing dramatic, just cleaning out some obsolete
docs/links, and actually writing in the front page weblog. I also fixed a bug
in the MailMain patch I'm using to generate RSS feeds from our mailing lists[1]
which might be of wider interest. I've set up new lists for the foafnaut and
foafcorp efforts, begun a tidyup of the Wiki and have been making an effort to
write up interesting developments in the weblog as they happen. There's a huge
backlog of course, but even this week things have been busy, with
I18N discussions (lots re Japan + foaf-a-matic translations in 8(!)
languages[3]),
XML::FOAF, Spring v.1.3.1, FOAF Checker and a 'howto' for using the foaf:weblog
property. There are various (funky) RSS feeds linked from the front page of
the site, pointers to IRC logs and suchlike should you care to keep an eye on
our discussions.

The FOAF 'home page' at http://rdfweb.org/foaf/ is shrinking, for
maintainability,instead I intend to use the wiki and weblog for links to
resources, news,
howtos etc. I've also linked a bunch of the more frequently accessed FOAF
links from the front page alongside the weblog, for easy access. Needs a bit
more work on navigation, duplication-avoidance and content, but I'm getting
there.
I've also been making a start on the long-awaited FOAF Specification cleanup,
which will (see discussion on FOAF list at [2]) introduce a more
fine-grained way of representing the stability of individual FOAF terms. Instead
of worrying whether 'FOAF' itself is stable, finished etc or not, you will get
to worry about whether individual things like foaf:depiction, foaf:weblog,
foaf:aimChatID etc have stabilised, using a 'stable','testing','unstable'
terminology borrowed from Debian. We're building a floating ship here, it's
good to know which planks we can safely step on... ;)

That's about it really. Just wanted to let RSS folks know that things are
moving along. So, next stop is the spec. I plan to have something interesting
for you all to read by mid July, or a damn good excuse why not... ;)

take care,

danbri

[1] the RSS mailman patch trail goes:
http://lists.w3.org/Archives/Public/www-rdf-interest/2003Feb/0047.html
http://rdfig.xmlhack.com/2003/02/09/2003-02-09.html#1044813421.381747
https://sourceforge.net/tracker/index.php?func=detail&aid=657951&group_id=103&at\
id=300103
-> http://rdfweb.org/2003/06/mailman-rss/rsspatch for my patch'd patch.

[2] http://rdfweb.org/pipermail/rdfweb-dev/2003-June/011267.html
[3] http://rdfweb.org/mt/foaflog/archives/000018.html

ps. I considered posting this from my non-work address, danbri@...
to be clear that FOAF isn't a W3C "thing" (although I hope W3C folk like it :)
However I'm subscribed to RSS-DEV as danbri@... I think, so am
announcing this as danbri@.... Having noted this, I'll carry on as
usual with danbri@....

pps. FOAF, being semantic web vapourware, has never been officially announced.
If you've no idea what I'm babbling on about, Edd's Dumbill's article from
a year ago is a good introduction, as well as being the thing that thrust
an ill-documented FOAF out into the wider world.  See
http://www-106.ibm.com/developerworks/xml/library/x-foaf.html or
the 'finding friends with foaf' sidebar link from http://rdfweb.org/

Hi,
Is that a way to search different companies and just return one news
for each company?
Thanks!
Ben

Hi Ben,

> Is that a way to search different companies and just return one news
> for each company?

Sorry, I can't really understand what you have in mind. Could you please be
a little more specific?

Cheers,
Danny.

Hi Chris et al,

"Chris Goad" <cg@...> wrote:
>The new BloggerPro version ("dano") generates an RSS 1.0 feed, whereas the
>previous version generated RSS 0.9x.  Good news, IMHO. I have a little test
> feed at http://rdfmapperexample.blogspot.com/rss/rdfmapperexample.xml
If you have any control over that output, you probably want to check it
out with the RDF Validator [1] - otherwise someone at Blogger would
want to. :)


[1] http://www.w3.org/RDF/Validator/


Regards,
Morten Frederiksen
(who is working on gml->rdf...)
---
A fool and his money are soon elected.
--
<URL: http://xml.mfd-consult.dk/foaf/explorer/?foaf=morten.rdf >

Hi Danny,
For example, I want to get only one news for below
each company:
IBM, Microsoft, ORACLE.

Is that a way to do that?
Thanks!
Ben
--- Danny Ayers <danny666@...> wrote:
> Hi Ben,
>
> > Is that a way to search different companies and
> just return one news
> > for each company?
>
> Sorry, I can't really understand what you have in
> mind. Could you please be
> a little more specific?
>
> Cheers,
> Danny.
>
>
> To unsubscribe from this group, send an email to:
> rss-dev-unsubscribe@egroups.com
>
>
>
> Your use of Yahoo! Groups is subject to
> http://docs.yahoo.com/info/terms/
>
>


=====
Ben

__________________________________
Do you Yahoo!?
SBC Yahoo! DSL - Now only $29.95 per month!
http://sbc.yahoo.com

Hi Chris et al,
"Chris Goad" <cg@...> wrote:

The new BloggerPro version ("dano") generates an RSS 1.0 feed, whereas the previous version generated RSS 0.9x. Good news, IMHO. I have a little test
feed at http://rdfmapperexample.blogspot.com/rss/rdfmapperexample.xml

If you have any control over that output, you probably want to check it
out with the RDF Validator [1] - otherwise someone at Blogger would
want to. :)

-- Help Support NewsMonster Development! Purchase NewsMonster PRO!
http://www.newsmonster.org/download-pro.html
Kevin A. Burton, Location - San Francisco, CA, Cell - 415.595.9965
AIM - sfburtonator, Web - http://www.peerfear.org/
GPG fingerprint: 4D20 40A0 C734 307E C7B4 DCAA 0303 3AC5 BD9D 7C4D
IRC - freenode.net #infoanarchy | #p2p-hackers | #newsmonster

> I was really enjoying what I could do with it. Problem was, I had
> just been toying with it for like two days, then I updated to the
> most current version of php (on my home test server) and now I can't
> seem to make it work anymore.
>

I might have a little insight.

You see my PHP RSS parser, Magpie, broke under PHP 4.3.2

I was seeing all sorts of weird behaviour; strange XML errors,
corrupted gzip encoding, and no seeming rhyme or reason to it.

Turns out that PHP 4.3.2 changed the handling of reading from sockets,
so if Fase4 is responsible for downloading the RSS feed (not just
parsing it) then that might be a good place to go looking for
problems.

In particular you're looking for code like:

$content = fread($fp, $large_number);

Which really should be:
while ( $data = fread($fp, $some_number) {
...
}

Not really a RSS issue, but a PHP one.

Good luck.

kellan

ps. Magpie 0.5.2 is now compatible with PHP 4.3.2

I have checked into the source forge repository for aggie a template for
merging several rss feeds into one.  This could be useful if you wanted
to particpate in a syndicate.  You could, for example, merge all your
blogroll into a giant rss feed.

I hope the next installer for aggie will incorporate this template.

The template can also be retrieved at
http://ransom.dyndns.ws/doug/template/ as well as source forge.

--
----
Doug Ransom
House for sale http://ransom.dyndns.ws/house