Re: Discovery/Affordances (Issue-32/Issue-57)

hello ted.

On 2013-06-14 10:09 , "Ted Thibodeau Jr" <tthibodeau@openlinksw.com> wrote:
>There's a strong blur between HTTP (HyperText Transfer Protocol)
>and HTML (Hypertext Markup Language) happening, and I am concerned
>that a similar blurring may be happening between LDP and LD (whether
>expressed as Turtle or otherwise).

not sure what that general blur between HTML and HTTP is that you claim is
happening, but let's look at concrete cases.

>For instance --
>On Jun 14, 2013, at 12:01 PM, Wilde, Erik wrote:
>> HTML clearly allows you to expect image/*, but then again when
>> you request some URI you GET whatever the server decides to
>> serve at that point in time, and maybe the link changed over
>> time from serving image/* to text/plain.
>That's completely inaccurate.  Unless you substitute HTTP for
>HTML.  But that substitution isn't valid throughout the thread
>nor post from which I pulled this snippet.

could you say why you think it's inaccurate?
http://www.w3.org/TR/html401/struct/objects.html#adef-src-IMG says that
img/@src "specifies the location of the image resource. Examples of widely
recognized image formats include GIF, JPEG, and PNG." i admit that (for
simplicity) i omitted the fact that you might link to HTTP, or HTTPS, or
even FTP or anything else, and then it all depends on the client whether
it can even follow the link. but once you follow it, i am wondering why
you think that HTML does not allow you to expect some image/*
representation from that interaction?

>Further, analogizing the new things to the old does nothing but
>confuse, when those old aren't kept clear and distinct.

that's a matter of taste. i agree that analogies have advantages and
disadvantages, and also that it's a matter of personal preference (and
background) whether they are useful or not. on the other hand, we're not
building something from scratch here. we're building something in the
context of the biggest information system that ever existed, so i would
claim that there is something to be learned from looking at how it works
today.

>Validating HTML is *entirely* distinct from determining whether
>a server is HTTP-compliant, and vice versa.  A fully-compliant
>HTTP server *may* not deliver *any* HTML content!

yes, sure, and nobody ever said differently in this thread (i think).

>Validating Linked Data is and should be distinct from determining
>whether a server is LDP-compliant, and vice versa.

nobody ever talked about "validating linked data" generically, as far as i
can remember. i think we all agree that this is about LDP.

cheers,

dret.

Received on Friday, 14 June 2013 17:31:14 UTC