Re: Issues with draft-ietf-atompub-format-04
On Jan 31, 2005, at 03:11, Asbjørn Ulsberg wrote: On Sun, 30 Jan 2005 22:06:23 +0200, Henri Sivonen [EMAIL PROTECTED] wrote: So how many European sites besides the EU have the resources to provide translations of the *same* content in multiple languages at the same time? The company I work in (the Norwegian Broadcasting Company), for instance. And most definately the BBC. And many other broadcasters, including RN (Radio Netherlands) which broadcasts and publishes content in many different languages. Don't any Finish broadcasters publish or broadcast content in both Finish and Sami for instance? There are TV programs in Swedish, but they are not translations of Finnish content. Both the Finnish and Swedish branches have news, but they have separate editorial processes. I expect the Sami radio to have separate editorial staff as well so that their content is not a mere translation of the content any Finnish-language channel. How many of those can't provide multiple feed links and really want to stuff everything in a single feed? All of the above, I'd presume. If it can be stuffed into the same feed, everyone would of course rather do that than create separate documents for each language. Why? The feed takes longer to download that way and you pay more for bandwidth. Aren't language alternatives within a feed over-engineering. No, not at all. It makes it possible to do for those who want, but don't have any implications on other producers. All features involve a cost at the consumer side. -- Henri Sivonen [EMAIL PROTECTED] http://iki.fi/hsivonen/
Re: Issues with draft-ietf-atompub-format-04
Robert Sayre wrote: I made that mistake because the draft in front of me is organized quite differently than the one in front of you. It was unclear to me as well. -- Anne van Kesteren http://annevankesteren.nl/
Re: Issues with draft-ietf-atompub-format-04
Robert Sayre wrote: So I can not include MathML in the TITLE of my weblog? I do not see why this restriction is necessary. Nope. Can any aggregator display it? I wonder if Shrook users are filling Graham's inbox with requests for MathML in their titles. Addressed in a separate thread. In Europe there are lots of different languages. It does not make sense to provide a feed based on language negotiation since feed aggregators do not support that. There are lots of different languages in America, too! I thought the only official language was en-US? Either Atom should provide support for multiple languages or we should address something like feed language negotiation in the specification. If what you say is true, then aggregators don't support multiple alternatives at all right now. Content Negotiation is covered by RFC2616. I believe most support content negotiation. However, I do not believe they send out an Accept-Language header. I think version should be dropped. We can always add it later. True. But that does not help current aggregators. At least, they will not reject new feeds. They will just found out they can not parse them at some point. The namespace will change. Only for the elements that are changed? -- Anne van Kesteren http://annevankesteren.nl/
Re: Issues with draft-ietf-atompub-format-04
On Jan 30, 2005, at 19:06, Anne van Kesteren wrote: In Europe there are lots of different languages. It does not make sense to provide a feed based on language negotiation since feed aggregators do not support that. So how many European sites besides the EU have the resources to provide translations of the *same* content in multiple languages at the same time? How many of those can't provide multiple feed links and really want to stuff everything in a single feed? Aren't language alternatives within a feed over-engineering. -- Henri Sivonen [EMAIL PROTECTED] http://iki.fi/hsivonen/
Re: Issues with draft-ietf-atompub-format-04
--On January 30, 2005 10:06:23 PM +0200 Henri Sivonen [EMAIL PROTECTED] wrote: So how many European sites besides the EU have the resources to provide translations of the *same* content in multiple languages at the same time? Pretty common in Quebec. We see English and Spanish in the US from Texas to California. California has voter guides in seven languages. It isn't limited to goverments, UBS's site is in four languages and the San Jose Mercury New has editions in Spanish and Vietnamese. How many of those can't provide multiple feed links and really want to stuff everything in a single feed? Good question. The answer probably depends on how much client software allows you to select a preferred locale. All browsers do, so they could easily do that with Atom feeds. Locales aren't just language. You could offer English in US, UK, and Australian versions. I was completely mystified about what the Aussies might mean by footy tipping. wunder -- Walter Underwood Principal Architect, Verity
Re: Issues with draft-ietf-atompub-format-04
On 30 Jan 2005, at 8:06 pm, Henri Sivonen wrote: So how many European sites besides the EU have the resources to provide translations of the *same* content in multiple languages at the same time? How many of those can't provide multiple feed links and really want to stuff everything in a single feed? Aren't language alternatives within a feed over-engineering. I'm going to agree with Henry here. Sites will have separate feeds for separate languages. No one will use this feature (publishers won't trust software at the far end to support it, users will want absolute control), and it requires a lot of new stuff in the client, and intermediaries, to select which language to use. Graham smime.p7s Description: S/MIME cryptographic signature
Re: Issues with draft-ietf-atompub-format-04
On Jan 27, 2005, at 22:39, Robert Sayre wrote: Anne van Kesteren wrote: So I can not include MathML in the TITLE of my weblog? I do not see why this restriction is necessary. Nope. Can any aggregator display it? I expect Gecko-based aggregators to support MathML eventually. After all, once you support XHTML content in a Gecko-based aggregator in a non-bogotic way (XML DOM to XML DOM copy with platupus filtering), you get MathML for free. -- Henri Sivonen [EMAIL PROTECTED] http://iki.fi/hsivonen/
Re: Issues with draft-ietf-atompub-format-04
Henri Sivonen wrote: On Jan 27, 2005, at 22:39, Robert Sayre wrote: Anne van Kesteren wrote: So I can not include MathML in the TITLE of my weblog? I do not see why this restriction is necessary. Nope. Can any aggregator display it? I expect Gecko-based aggregators to support MathML eventually. After all, once you support XHTML content in a Gecko-based aggregator in a non-bogotic way (XML DOM to XML DOM copy with platupus filtering), you get MathML for free. We are not here to standardize eventually. This discussion is a waste of time. Robert Sayre
Re: Issues with draft-ietf-atompub-format-04
At 13:01 05/01/26, Eric Scheid wrote: It's only clear what's going on when the reader juxtaposes the two sections, and realises that the concept named 'type' in section [3.1.1] is not the same concept named 'type' in section [3.5.2]. Without that juxtaposition, the reader might well never realise that 'type' != 'type' and conflate the two concepts. Even you made that mistake just now, and you're the editor of the document. Pity the poor reader. Looking at it from a usability of specifications p.o.v., it doesn't hurt to have cross references. I agree this is a problem. Either we find new names for the attributes so that each element has a different attribute, or we put pointers to the other 'type' attribute(s) in each section about a type attribute (and ideally also a table somewhere that shows all of them). If we don't do it, confusion will be guaranteed, and we will know we are the ones to blame. Regards,Martin.
Re: Issues with draft-ietf-atompub-format-04
Robert Sayre wrote: * 3.5.1 rel Attribute Why are the only values defined alternate and related? I have implemented via for a long time on my personal weblog and some aggregators have even implemented support for it. I consider it to be quite useful. Write a Pace. I would support it. Yes, please. - Sam Ruby
Re: Issues with draft-ietf-atompub-format-04
On Tue, 25 Jan 2005 09:59:08 +0100, Anne van Kesteren [EMAIL PROTECTED] wrote: This was about it. I hope it is of some use. I share all your concerns and issues and hope they will be addressed properly before the format is finalized. -- Asbjørn Ulsberg -=|=-http://virtuelvis.com/quark/ «He's a loathsome offensive brute, yet I can't look away»
Re: Issues with draft-ietf-atompub-format-04
On 26/1/05 2:49 PM, Robert Sayre [EMAIL PROTECTED] wrote: [3.1.1 says TYPE is one thing] [3.5.2 says TYPE is the opposite] Ah, you're right. Still don't see how it's vague, though. It's only clear what's going on when the reader juxtaposes the two sections, and realises that the concept named 'type' in section [3.1.1] is not the same concept named 'type' in section [3.5.2]. Without that juxtaposition, the reader might well never realise that 'type' != 'type' and conflate the two concepts. Even you made that mistake just now, and you're the editor of the document. Pity the poor reader. Looking at it from a usability of specifications p.o.v., it doesn't hurt to have cross references. e.
Re: Issues with draft-ietf-atompub-format-04
Eric Scheid wrote: Ah, you're right. Still don't see how it's vague, though. It's only clear what's going on when the reader juxtaposes the two sections, and realises that the concept named 'type' in section [3.1.1] is not the same concept named 'type' in section [3.5.2]. Without that juxtaposition, the reader might well never realise that 'type' != 'type' and conflate the two concepts. Even you made that mistake just now, and you're the editor of the document. Pity the poor reader. I made that mistake because the draft in front of me is organized quite differently than the one in front of you. Bring this issue up again after the next draft if you still think it's worthwhile to embark on a campaign to rename an attribute. Robert Sayre
