Talk:SOAP

From Wikipedia, the free encyclopedia

Cleanup Taskforce article This article has been improved by the Cleanup Taskforce to conform with a higher standard of quality. Please see its Cleanup Taskforce page for more details on this process, and possible ideas on how you can further improve this article!

The articles says: "Typically, SOAP is about 10 times slower than binary network protocols such as RMI or IIOP. Of course, this is not an issue when only small messages are sent."

That second sentence is not true, is it? I suggest it be deleted entirely. If it is going to stay, it wants to be accurate. For example:

"Typically, SOAP is about 10 times slower than binary network protocols such as RMI or IIOP. Of course, this is not an issue when only small messages are sent and they are sent and received infrequently."

Joaquin


Since SOAP is no longer an acronym, would it make sense to move this page to SOAP (or something like "SOAP (computer protocol)")?

I'd like to know (and article should probably state) why/when this became no longer an acronym. Elf | Talk 22:30, 26 Aug 2004 (UTC)
As far as I know it is because it's neither a simple nor (just) an object access (it's capable of transmitting any XML message) protocol. --Ondrejsv 07:36, 7 July 2006 (UTC)
Unfortunately I haven't been able to find any record of the discussion that led to the name change but I can tell you what I've heard. Until version 1.1, the recommendation (to split hairs, it isn't technically a "standard") was called "Simple Object Access Protocol". This name never did really fit: it has never had anything in particular to do with accessing objects, and even the initial version was never particularly simple, especially when compared to its progenitor XML-RPC (and don't even get me started about the various SOAP add-ons like WS-Security, WS-Transaction, WS-ReliableMessaging, and WS-KitchenSink). I gather there was at least one faction in favor of renaming it "Services Oriented Access Protocol"[1], but instead the expansion was dropped completely and in version 1.2 of the recommendation it is simply called "SOAP". -Saucepan 06:11, 27 Aug 2004 (UTC)
Further to the above, I've repointed the all-uppercase SOAP redirect to this article from the disambig page. This article is currently the most likely intended destination for someone looking for "SOAP" with all uppercase, and has prominent disambiguation links for those cases where I'm wrong. If nobody objects, in a day or so I'll finish the job by moving this article to SOAP and leaving behind a redirect. -Saucepan 00:41, 4 Sep 2004 (UTC)
I think something should be added to the article explaining why it's now "formerly". It's what caught my attention, yet there was no reason for stating it. RedWolf 22:03, Sep 21, 2004 (UTC)
That caught my eye too. But I think its enough explanation to specify when the SOAP specification started saying "It's no longer an acronym." That's Version 1.2. I added that info when I rewrite the introductory paragraphs for readability. Isaac R 05:13, 16 Apr 2005 (UTC)
Might SOAP now be considered a backronym for Service-Oriented Architecural Pattern? CptJoker 03:59, 13 June 2006 (UTC)

Contents

[edit] SOAP type information

Where's the embedded type information in the example? All the on-the wire SOAP transactions I've seen recently have embedded type info. -- The Anome 22:07, 22 Oct 2004 (UTC)

That would be RPC-style SOAP, in which there's no document schema but the type information is inlined. It tends to be used by Axis and tools based on it (mostly Java). The example in the article is based on a capture of .NET traffic and uses Document-style SOAP. In Document-style, the WSDL includes or refers to a schema, and the sender assumes the recipient has the schema and so knows the types already. (If you are thinking that this sounds like a bit of a mess you are quite right IMO.) I guess the article could use some explanation of the different styles of SOAP.
I used Document-style messages for the examples just because they were the sample messages I had laying around. My current feeling is that RPC-style type info would tend to clutter the example and confuse readers unused to XML schemas, detracting from the goal of presenting a simple, high-level overview. (This is the same reason I left out the XML Byte Order Mark.) Saucepan 22:54, 22 Oct 2004 (UTC)

Somebody help to include the more common variants in the main page please? A reference to a more detailed discussion elsewhere would help too. 203.9.186.134 03:54, 28 September 2005 (UTC)

I am wondering if this (SOAP) will make a resurgence as Microsoft has patented the Serialization and Deserialazion of XML streams to Program Objects (JAVA specificly... little revenge for SUN winning the JAVA case vs Microsoft?). Might be worth a discussion. 67.171.245.44 23:01, 27 November 2006 (UTC) Frankmon

[edit] "All protocols"???

The nitpicker in me doesn't like the statement "SOAP can be run on top of all the Internet Protocols...". All protocols??? NTP? Ping? I think the statement needs to be qualified somehow, but I don't know enough about SOAP to do it myself. Something like "SOAP can be run on top of all protocols that support exchange of text bundles..." Isaac R 05:13, 16 Apr 2005 (UTC)

[edit] Cleanup

I tried to read through the article from the perspective of someone trying to learn what SOAP is, and I think we need to work on it quite a bit. The article dives into too many technical details and provides very little in terms of general definition and explanation -- it needs to focus more on answering the question "what is SOAP?" David 11:51, 2005 Apr 16 (UTC)

  • I was looking at this article literally twenty minutes ago thinking exactly the same thing. I'm new to Wikipedia, and so I hesitated to Be bold in updating pages and tag it for cleanup. In any case, I totally support this. Pojo 16:11, 16 Apr 2005 (UTC)
    • Did my rewrite of the opening hurt or help? Don't spare my feelings! Isaac R 20:52, 16 Apr 2005 (UTC)
      • I think it's a start but honestly, I feel like the whole opening section (before "Contents") needs to be obliterated and rewriten in layman's terms. Pojo 14:35, 18 Apr 2005 (UTC)
        • Always a desirable goal, but difficult to achieve in this case. Take the very first sentence, which says that "SOAP is a light-weight protocol"? What's the layperson's term for "lightweight protocol"? Bear in mind that you can't just leave the concept out -- it's a key part of the definition.---Isaac R 17:20, 18 Apr 2005 (UTC)
          • The light-weight part is actually fairly controversal inside the technical community -- many people, especially REST supporters, characterize SOAP as heavy-weight. I'd suggest taking "light-weight" out of the introduction and putting it under history, something along the lines of "SOAP was designed to be a light-weight protocol" or "SOAP's designers intended for it to be a light-weight protocol", without trying to prejudice the article either way. In any case, I don't think that we need to target laypeople who are completely ignorant of computers, so much as people who are familiar with computers but not with SOAP. David 17:42, 2005 Apr 18 (UTC)
            • "so much as people who are familiar with computers but not with SOAP" I agree, that seems fair. There's no point trying to explain SOAP to someone who's still looking for the any key. :) Pojo
            • If "lightweight" is controversial, then it certainly needs to be removed from the definition. But I don't think we necessarily have to find a place for it elsewhere. Material on the controversy might improve the article, but I don't think adding it is an immediate priority. ---Isaac R 18:44, 18 Apr 2005 (UTC)

I redid the beginning of this article, adding a brief "layperson" definition. There is no getting around the fact this is a technical subject that requires a certain foundation of information. However, I think a brief non-technical introduction gives the best of both worlds. Let me know what y'all think. --RobertDaeley 20:05, 2005 Apr 19 (UTC)

  • Thanks -- I think that's a good start for all of us to build on. David 00:06, 2005 Apr 20 (UTC)
  • I strongly disagree. I really hate that definition. It's one of those vague pseudo-explanations that don't really improve anybody's understanding of a topic. Robert, please don't take this as a criticism of you're writing skill. You did your best to do something that's just not possible: explain a concept, SOAP, without reference to any of the other concepts SOAP builds upon. Bad idea. At least with the old version, people could identify concepts they weren't familiar with, and follow hyperlinks to improve their knowledge of them.---Isaac R 22:52, 21 Apr 2005 (UTC)
    • No worries, Issac -- I figured this was just a starting point. :) --RobertDaeley 17:15, 2005 Apr 30 (UTC)
    • I've rewritten the introduction to try to make it clearer. The first sentence answers the question "what is SOAP?" directly, and the rest puts SOAP in a broader technical context rather than diving down into minor details. I've also corrected the history of SOAP. Let's use this as a starting point, and see if we can refine it even a little more. David 02:40, 2005 Apr 22 (UTC)
      • Good stuff, but I have a quibble. You say "soap is a standard", instead of "soap is a protocol". I think the more specific word is better, even if it does have the taint of computerese. ---Isaac R 03:04, 22 Apr 2005 (UTC)
        • Thanks for the kind words. I chose standard not to be less technical, but to avoid quibbles -- the SOAP standard defines both transmission protocols and message formats; in fact, I'm not sure if we could call it a single protocol in any case, since it has so many variations, like RPC, asynchronous. David 14:45, 2005 Apr 22 (UTC)

I'm trying to cleanup the article so it is easier for the average reader to understand. I think that the introduction is okay now, although still a little heavy on the technical jargon. The example section seems out of place. Any suggestions? Ben Babcock 05:26, 30 Apr 2005 (UTC)

  • I'm sorry to be harsh Ben, but I'm afraid I don't much care for your work. Did you notice the convesation (just above on this talk page) about "protocol" versus "standard"? And no, your cutting necessary technical terms does not make the explanation more accessible -- just harder to follow. Also, your assertion that SOAP is still an acronym is mistaken -- as described elsewhere in the article! ---Isaac R 05:45, 30 Apr 2005 (UTC)
I did notice that conversation. The criticism is well-deserved, as I am not familiar with SOAP. The W3C defines it as a "lightweight protocol" however and I chose to go with that definition. I kept the last paragraph where it mentions that the acronym no longer applies, even though it was originally an acronym.
The crux of the matter is, of course, the technical terms. Obviously this isn't done all in one shot. Perhaps keep the entire introduction very simple, but then move the necessary technical information to the body of the article? If we do that, then the average person can get a basic understanding of what SOAP is, while those who want to continue can read the entire article. . . .Ben Babcock 06:00, 30 Apr 2005 (UTC)
Ben did an excellent job using simple, clear language, but unfortunately his rewrite introduced some inaccuracies and removed some essential information (for example, the main reason people care about SOAP is the fact that it's part of web services, so the intro needs to mention web services somehow). I've reverted for now. David 11:22, 2005 Apr 30 (UTC)
I've created a temporary SOAP page. We can edit this one until we can agree on the right balance between technical accuracy and plain language, then place its content on the actual SOAP page. So far I've just reorganized the bottom links alphabetize them.
I still think the introduction part of the article is too long and detailed, and that some of the information should be moved to a subsection of its own, so we can keep the introduction shorter and easier to understand at a glance. Ben Babcock 18:39, 30 Apr 2005 (UTC)
Okay, I've rearranged my temporary article. The first two paragraphs have been preserved because they contribute to a simple but accurate introduction. I moved the third paragraph into an "Overview" section, which also includes the "Transport Methods" section now. I included information about the way a SOAP message is structured, and then placed the examples as a subsection of that. Ben Babcock 00:36, 2 May 2005 (UTC)
Page updated to reflect the rewritten version, which has hopefully maintained the technical accuracy while making it easier to read. Ben Babcock 04:27, 7 May 2005 (UTC)

I'd like to see some criticisms mentioned. I'll write something and get opinions. 203.216.0.91 01:37, 26 July 2005 (UTC)

[edit] Search for SOAP

I'm new to this wiki stuff, but when I used the search box on the left to search for "soap" it only came back with one result for soap (as in bar of soap). It does return the SOAP entry if you search all-uppercase but I'm not sure most people will do this. I wonder if this situation could be improved by adding a suffix to the entry title. Alternatively should we alert the admin crew that the search is case-sensitive?


[edit] broken link

It just might not be my day, but the external link "Dave Winer's history of SOAP" appears to be broken.

- On 8 May 2006 the link is fine. Jebbo 17:26, 8 May 2006 (UTC)

[edit] URI PROXY etc; programming examples

The Perl POD for SOAP::Lite mentions URL, PROXY etc. I'll keep roaming, and hopefully add some usefull content later. (maybe a wikibook humm...) Supaplex 20:13, 16 June 2006 (UTC)

[edit] Circumvention of security?

SOAP's use of the HTTP protocoll to get through firewalls is praised in the article. I'm not a security expert, but isn't this a circumvention of one of the security mechanisms that administrators have to prevent code from being invoked on their systems from outside?

Would it have been deemed acceptable for CORBA or DCOM to use port 80 just because it is open? Any security experts here who could shed some light on this aspect of SOAP.

62.242.39.226 08:09, 27 June 2006 (UTC)

[edit] More abstract layers?

The very first sentence of this article says "providing a basic messaging framework that more abstract layers can build on" -- isn't this an oxymoron? Shouldn't it be more *concrete* layers that build on an *abstract* framework? --HeXetic 16:15, 11 July 2006 (UTC)

[edit] 'Simple Object Access Protocol'

From the article:

At the time of the creation of the standard, SOAP was an acronym for 'Simple Object Access Protocol'. Version 1.2 of the standard, which became a W3C Recommendation on June 24, 2003, dropped this acronym as it was considered to be misleading.

Can someone please add a comment in the article explaining why it was considered misleading? Kaimiddleton 17:08, 20 October 2006 (UTC)

[edit] Generic SOAP Client

From the External Links:
Generic SOAP Client is timing out .... Kaimiddleton 17:17, 20 October 2006 (UTC)

Retrieved from "http://en.wikipedia.org../../../s/o/a/Talk%7ESOAP_ce43.html"