To get better compliancy with RSS, we upgraded our feed to RSS 1.0. Please refresh your feeds, test and report any problems in the comment section. On other site news, the GTK+ browser Dillo is now working better with our comment.php pages. The GnomeFiles.org feed also saw an update recently, it is now formatted in lovely HTML (it will remain in RDF version 0.9 though, as it will the NMC feed, because their creation php code is hairy).
Updated RDF/RSS News Feed; Please Test
About The Author
Eugenia Loli
Ex-programmer, ex-editor in chief at OSNews.com, now a visual artist/filmmaker.
Follow me on Twitter @EugeniaLoli
45 Comments
I was instructed to do so the way I did.
BTW, I can’t read your code, please send me an email with it, and I will investigate more (I am new on the RSS stuff).
the rss reader for the dreambox tell me its not valid
rss data. Yahoo, Slashdot etc . are normally accessable.
I am not sure its the dreambox or your rss file.
BTW, RSSOwl will also say that it’s not valid, but it works fine with it. Other validators say that our feed validates fine. I guess RSS is a lot like CSS in terms of support. ๐
Does it give you an error message? If yes, please send it to me.
“Not an rss document”
i guess it spelld wrong it should be
“Not a rss document”
but i am having spelling problems too
I have added:
<feed>
<name>OsNews.com</name>
<url>http://www.osnews.com/files/recent.rdf</url>
</feed>
to the feed.xml file…is the url correct?
;
Yes, the filename is correct. Please note that our document IS an RSS, but its filename is a .rdf (for legacy reasons with our readers). Maybe that’s the problem? BTW, what’s Dreambox?
I made a rss/xml file and uploaded it to my own server.
Added it to the rss-reader feeds.xml file and it works perfectly. I think the problem is in the header of your file…
I don’t quite understand what you mean or what you did or what Dreambox is (satellite receiver??). I am quite lost of what you are trying to do. Please be a bit more verbose and please explain to me what do you mean about the header. What header exactly? What part of the RSS?
Or do you mean about the fact that our server sends it as plain/text instead of the RSS mime type? This is a problem on our end indeed, but it is not a bug, it is a warning (I have asked our admin to fix it a month ago, but he doesn’t have time). If Dreambox chokes because of this though, then it’s a dreambox bug.
The Dremabox is a Linux driven satelite/multimedia box.
It runs on LInux 2.6 and has an optional dvd and hardrive.
The rss reader is a plugin you can download form a webpage.
I am using the Hydry (greek;)) image wich enables a lot of plugins etc. possiblities.
You can watch tv, satelite, dvd, mp3, ogg, mpg, ts etc on a dreambox, its got an email client and a webbrowser…its COOL
Sounds good. However, I would also suggest you email the author of the dreambox RSS module to fix the problem btw. If you think that the problem is on the header of the file and not on the actual code of the RSS, then he must start supporting text/plain, cause I can’t get hold of our server admin to fix it in time (in fact, the mime type is not even mandatory).
I made some changes to your feed and it now works perfectly
with the dreambox.
http://212.13.209.220/osnews_test.xml
BTW, I tried that xml file with Liferea, it completely choke! It ain’t valid. If your RSS module works with that thing and not with our current RDF file, I suggest that you really file a few bug reports..
I think i found the problem
Items are noted as:
<item rdf:about=”http://www.osnews.com/story.php?news_id=8384“>
But if i remove everything except the <item> tag its works…
I will email them with this problem.
Yep thst is the problem, i left all headers as they where but only removed the rdf info after in the item tags.
212.13.209.220/osnews_test.xml
It works now. You could make a seperate xml file for dreambox users and strip out the stuff after item with an php-script.
๐
rdf:about is a required attribute of RSS 1.0. Check here:
http://web.resource.org/rss/1.0/spec
If your RSS reader chokes on that, it’s your RSS reader’s bug.
I think so…
Is there not a way to have an ignore tag like in javascript when the rss-reader is only 2.0 comp.?
Or is that the current problem, a good rss-reader would igone the rss 1.0 info??
That tag is not to be ignored, it is to be used. It’s your RSS dreambox reader’s bug that can’t parse that attribute. The tag IS needed, it’s a required attribute and it’s used by other RSS readers correctly! It is not to be ignored or removed. The only solution is to have your RSS reader fixed.
I just updated the rss reader but it seems they only support
rss 2.0
Eugenia is it ok when i put your recent.rdf on my sever and make changes to it so that users of a dreambox (and there are a lot) can read the osnews feed? I will try and ad your website (gnomefiles.org too) to the official feed list.
I will let you konw..
Sure, but might be also an idea to ask support for 1.0 feeds too. I mean, most sites on the web still have 0.9 still…
in Firefox it looks ok. I noticed a few days ago that it wouldn’t show new titles in the right order, I think it showed them alphabetically, but now they’re sorted by date/most recent.
Keep up the good work!
Actually, this is was firefox bug. I filed a bug report 10 days ago but it is still not fixed.
it does look fine now, I’ve compared the bookmark with the page and it looks in the exact same order..
Have you tried it recently, I mean since you’ve been working on OSNews’ RSS feed?
Of course it looks fine now. It now uses the new RSS 1.0 feed and so the bug is not there anymore, as the bug happens with the 0.9 RDF feeds. The bug is still reproducible with the gnomefiles and nmc feeds for example, because I won’t be upgrading those.
No. I have never used those, I don’t see the point as they provide the same service, and I know of no one who uses them.
No problems with my self-build MagpieRSS based PHP reader. With exception of strange item [ http://www.vilgots.lv/osnews.png ] at 5:47 GMT+2.
The feed doesn’t validate at feedreader.org: http://www.feedvalidator.org/check.cgi?url=http%3A%2F%2…
– Feeds should not be served with the “text/plain” media type [help]
– Your feed appears to be encoded as “iso-8859-1”, but your server is reporting “US-ASCII” [help]
– line 48, column 195: description contains bad characters (2 occurrences)
Also, why RSS 1.0? While a good attempt to re-introduce RDF and the Semantic Web back into RSS, it seems like the choice of formats is becoming less popular in the face of the tide of 2.0 and the spin-off of Atom.
If you want to change the mime-type without getting the server admins involved you can do this in PHP:
header(‘Content-type: text/xml’);
(or whatever mime-type you want to send)
I can’t do that, because the .rdf page is not a php dynamic page, it is a static file generated by php. I don’t want it to be dynamic for load reasons.
Regarding the feedvalidator, except the mime stuff which as I said I KNOW about and was not a reason to re-mention it (the second warning is a result of the first one), the third one is a real one that I have no control over. It depends on the stories text each time. I clean up a lot of such characters before a feed goes live, but there are always characters to be missed. So, no matter what feedvalidator says, the feed is as good as it gets right now.
http://www.washington.edu/webinfo/htaccess/
You can probably add a simple .htaccess in the directory to override the mime type.
The feeds from gnomefiles are displayed backwards, instead of showing the last entry firt it does the other way around (at least using Firefox 1.0-pre)
blogger supports both atom and rss, although google is slowly standardizing on atom.
the feed works fine with akregator btw.
No problems here.
I’m really nitpicking here, but in an ideal world, you wouldn’t put raw html code as CDATA (enclosed in <![CDATA[ ]]>). When doing that, you don’t give a clue to the parser about what kind of data it is. If I were to do things the way they are supposed to be done, I would display that CDATA in its raw form, not as HTML as it is the case in gnomefiles’ RSS feed.
XML namespaces should be used instead.
<description><![CDATA[ foobar ]]></description>
would become
<description><html:div>foobar</html:div></description>
for instance.
Anyway, it works in the real world:
http://www.gnomefiles.org/files/gnomefiles.rdf“ rel=”nofollow”>http://dev.skamp.net:8080/news/rss.php?feed=