I've added another heretical ;-) page [1] to the RESTWiki about the use of paths vs. query strings. It includes a semi-detailed taxonomic example and...
Jeff Bone
jbone@...
Mar 10, 2002 9:34 pm
2
This is a quick message to summarize where I think we're at on the discussion about naming, opacity, etc. In moving this over to rest-explore (hey, anyone...
Jeff Bone
jbone@...
Mar 10, 2002 9:40 pm
3
This is the most important conclusion of my paths vs. query strings analysis: "Where hierarchy exists in the model space, it is useful for hypertext...
Jeff Bone
jbone@...
Mar 10, 2002 9:55 pm
4
Your subject line upper cases REQUIRES but your quote uses the phrase "it is useful." ... That's the strongest terminology you can make because anything that...
Paul Prescod
paul@...
Mar 10, 2002 10:35 pm
5
... Yes. I already realized this and modified the statement on the Wiki page. Unf., once an e-mail is sent, it's rather difficult to go back and edit. ;-) jb...
Jeff Bone
jbone@...
Mar 10, 2002 10:59 pm
6
Paul's responses Paul1: Jeff says that hierarchical URIs are important for relative URIs. This is true. But relative URIs are themselves only a convenience ...
Paul Prescod
paul@...
Mar 10, 2002 11:32 pm
7
Paul: as a proposed rule of engagement, let's use the Wiki for stable positions and the list for debate and providing pointers to the Wiki, what do you say?...
Jeff Bone
jbone@...
Mar 10, 2002 11:47 pm
8
... Okay. ... You agree that it is a statement of fact. My sense is that it wasn't a fact that was obvious from the page. I thought it should be. ... That's...
Paul Prescod
paul@...
Mar 11, 2002 12:00 am
9
... Wonderful! We agree that people should have this responsibility! That seems like a change from your current position. ... And if I need global...
Jeff Bone
jbone@...
Mar 11, 2002 1:01 am
10
... From: "Jeff Bone" <jbone@...> ... ;-) ... You mean SMTP doesn't have DELETE?...
They can't all be jewels. These are the jokes, folks, these are the jokes... ;-) jb...
Jeff Bone
jbone@...
Mar 11, 2002 8:18 am
12
... I think that's one pitfall with trying to make names meaningful. For instance I've been thinking about /robots.txt. What if I want a robots.txt for just...
Paul Prescod
paul@...
Mar 11, 2002 9:54 pm
13
... No, we agree on what your position is. And now that I've confirmed it we can agree that we disagree. Can you read the quote that started all of this and...
Paul Prescod
paul@...
Mar 11, 2002 9:58 pm
14
... Option 1: the client has a priori knowledge of the query parameters. Then they don't need HTML (or WRDL or WSDL). Option 2: the client needs to be told...
Paul Prescod
paul@...
Mar 11, 2002 10:20 pm
15
... I'm suggesting that where it is appropriate, there are benefits to both the client and the owner of the information space for the owner taking on this...
Jeff Bone
jbone@...
Mar 11, 2002 10:39 pm
16
... Option 3: the client has a priori knowledge of how to construct complete URI, not restricted to query parameters. This is equivalent to the infospace in...
Jeff Bone
jbone@...
Mar 11, 2002 10:44 pm
17
(NB: I'm posting this to rest-discuss as well as rest-explore because it's a reasonably fundamental question that may be useful in helping even newcomers to...
Jeff Bone
jbone@...
Mar 11, 2002 10:50 pm
18
... Here's my definition of navigation. You are provided a URI which addresses a resource. I don't care whether it is on a napkin or a billboard or in a web...
Paul Prescod
paul@...
Mar 11, 2002 11:13 pm
19
... Nothing breaks. Things start to break if the *only way* to find out how to get information out of a web-based service is through generated URIs. The first...
Paul Prescod
paul@...
Mar 11, 2002 11:21 pm
20
(NB: I'm crossposting this one as well because I think a detailed response to this might be useful both the the neophyte and the enlightened. Warning: this...
Jeff Bone
jbone@...
Mar 11, 2002 11:21 pm
21
... No contest. But --- *what do you call the first step?* Given a name, you access something. From that something, you might navigate *to* something else...
Jeff Bone
jbone@...
Mar 11, 2002 11:40 pm
22
... I have never, ever, ever said this in the course of our discussion. This is ghost of your own creation haunting your perception of this debate. I have ...
Jeff Bone
jbone@...
Mar 11, 2002 11:42 pm
23
... I don't see how you can square this with your admission that I could solve your problem with hypertext but the hypertext it would be "extraneous." ... ...
Paul Prescod
paul@...
Mar 12, 2002 12:20 am
24
NB: REST heresy? Or REST clarification? I've added a page to the Wiki exploring the implications of applying the axiom of opacity too strongly. It results...
Jeff Bone
jbone@...
Mar 12, 2002 12:43 am
25
... Your assertion seems that I'm claiming that it's possible to "get away without hypertext blah blah" in the general case. I do not believe that to be...
Jeff Bone
jbone@...
Mar 12, 2002 12:49 am
26
... That's what I meant to say. "The part of your agument which I dislike is where you say that there exist circumstances where the existence of consistent,...
Paul Prescod
paul@...
Mar 12, 2002 1:04 am
27
I agree with Paul, mostly, but from my investigation into URI opacity, I basically concluded that; http://foo.com/bar?a=b implies the existance of a resource...
... This is all true. However, if in a hyptertext representation of http://foo.com/bar/b I encounter <a href=".."> I can infer --- indeed, the hypertext...
Jeff Bone
jbone@...
Mar 12, 2002 2:05 am
29
I don't know why rest-discuss seems to be cc:ed on every message. What's the point of having two lists if we're going to have the same old arguments on...
Paul Prescod
paul@...
Mar 12, 2002 3:00 am
30
Paul --- please --- carefully read and understand what I'm saying, okay? More than 50% of this dialog has been me explaining to you that you've been too hasty...
