Talk:Reverse Ajax

Untitled[edit]

Much of the content on this page is duplicated by the article on standard Ajax. Please link to that article and reduce the text to only the discussion of what "reverse Ajax" is, not Ajax itself.

This term is not notable[edit]

As far as I can tell from googling, only one author (who originally created this page) uses the term “reverse ajax”. All the references to it are either by this author, or links to his explanation. I suggest merging any unique information from this article into comet (programming), and then redirecting “reverse ajax” to there. Otherwise, this article is essentially original research (see WP:SOAP and WP:OR). --jacobolus (t) 20:56, 11 October 2007 (UTC)[reply]

Well anyway, I'm merginfsdfdsg into comet (programming) now. No one seemed to have any objections, and as far as I can tell the scope of the articles is identical (Comet as defined by Alex Russell describes the user interaction model, not the specific technology; all of these reverse ajax techniques fall under the umbrella of Comet, despite the misconceptions of the reverse ajax guy). --jacobolus (t) 23:54, 18 October 2007 (UTC)[reply]

I object! It is a notable phrase - Googling “reverse ajax”, including quotes, returns some 41,000 very relevant results. (My article on the term, which was certainly not the first on the subject, is far from the front page). In fact, the very last entry is a Japanese(?) page discussing the subject. If you read the articles, Reverse Ajax is a term encompassing many user interaction models which can be implemented in Ajax technology to achieve the equivalent of a server push. Comet is only one of the many interaction models, and to redirect is to do injustice to the other models. Reverse Ajax, I believe, stems from the version 2.0 release of the software package DWR produced by Getahead IT consultancy. The article discussing it in depth was published on the internet in April 2006. The original article, software and company are completely independent from me, the creator of the Reverse Ajax article. I decided to create this article after considering the many different interaction models for use in my own ajax based software. 212.124.252.210 (talk) 18:42, 28 November 2007 (UTC)[reply]

From the comet page - Comet differs from the original model of the web, in which a browser receives a complete web page in response to each request, and also from the Ajax model. Ajax and Ajax polling are also only mentioned in that article as a comparison. Reverse Ajax does not have to keep a channel continuously open between the server and browser...it is therefore not by necessity Comet. sprocketonline (talk) 19:13, 28 November 2007 (UTC)[reply]

That doesn't make the term notable. Find some significant coverage of the term in reliable sources. See WP:N --jacobolus (t) 21:02, 28 November 2007 (UTC)[reply]

To clarify, I don't think that this topic is interesting or useful independent of the Comet (programming) article, or perhaps an article about DWR specifically. It describes an "umbrella term" which is only actually used by DWR developers, or those describing DWR, and which is otherwise not in wide use. Currently, the only content of this article ends up being a comparison of Comet with polling (and PiggyBack, which as far as I can tell is also a term used only by DWR, and which is moreover directed at completely different application use cases), which can perfectly adequately be accomplished at the Comet article. --jacobolus (t) 21:19, 28 November 2007 (UTC)[reply]

Also, at least a few sentences of this article seem to have been cribbed from [1]. --jacobolus (t) —Preceding comment was added at 21:22, 28 November 2007 (UTC)[reply]

Reverse Ajax is NOT Comet. It is actually more polling or piggyback than comet. sprocketonline (talk) 19:25, 29 November 2007 (UTC)[reply]

OK, so more research - I found a good number of notable references. Katherine Martin's Java.Net article of March 2007. The oldest source I could find Bryce Nesbitt's Slow Load Technique which mentions Reverse Ajax more than 2 years ago and includes an example. I also added Michael Mahemoff's Ajaxian.com and a ServerSide.com link to the Reverse Ajax article. DWR might be the predominatly discussed implementation, but php mag introduces Xaja, a php based Reverse Ajax implementation. www.reverseajax.com is another example of its implementation. sprocketonline (talk) 20:04, 29 November 2007 (UTC)[reply]

I agree, the article currently does look like a comparison between three different technologies, which could just as easily be incorporated into the Comet article or in a Comparison article. However, this article is specifically about the use of any of these models specifically in Ajax (Comet can be used in any of hundreds of communication technologies). It is my hope that this article will grow and begin to move away from being a straight comparison, and look more about the specifics of these models in Ajax. (this however may take a long time due to the relative obscurity of the term). sprocketonline (talk) 20:04, 29 November 2007 (UTC)[reply]

P.S. I can 'crib' as much of my own work as I like. (It's Creative Commons anyways) sprocketonline (talk) 20:04, 29 November 2007 (UTC)[reply]

Comet is a term which specifically refers to an "Ajax"-like user interaction model happening in a browser, so I'm not sure what you're talking about. --jacobolus (t) 03:19, 1 December 2007 (UTC)[reply]

Both the Ajaxian and theserverside.com simply link to Jonathan Downes article, and excerpt little bits of it, without adding any new content. Neither link implies that the term "Reverse Ajax" is in wide use. As for the slow load technique article, it is an independent use of "reverse-Ajax", which does not agree with the definition given in this article. In fact, it seems to use the term to refer to exactly what the Comet article calls Comet. I think that this article should be merged somewhere, because it is neither notable enough, nor contains enough content, to stand on its own. If it sits like this for a while, I'll probably put it up on VfD. --jacobolus (t) 03:27, 1 December 2007 (UTC)[reply]

If there are no new responses to this discussion, or changes to the article which establish the notability of the subject, I will list it on VfD. --jacobolus (t) 01:52, 21 December 2007 (UTC)[reply]

I agree that this term does not appear to deserve a separate article, and should be deleted, but suggest waiting until after the problem with articles in this area and associated redirects are sorted out. Restepc (talk) 16:52, 3 June 2008 (UTC)[reply]

I disagree that the article should be merged with Comet, if at all then with Ajax with which it has a much closer relationship in my opinion. - 83.254.208.192 (talk) 13:47, 14 June 2008 (UTC)[reply]

How can a term not have a close relationship with its synonym? “Reverse Ajax” and “Comet” are two words for nearly the same thing (modulo the confusion of some users and Wikipedia editors). Answer in the section below, so those coming from the article page can find the discussion. —jacobolus (t) 23:21, 14 June 2008 (UTC)[reply]

Happy that you don't want to merge "World Wide Web" into the "Comet" article... yet. - 83.254.214.72 (talk) 17:07, 29 June 2008 (UTC)[reply]

What? World Wide Web is not synonymous with “Comet”, so that merger would make no sense. —jacobolus (t) 01:04, 1 July 2008 (UTC)[reply]

I have the feeling that you somewhat missed the sarcasm. --Damiens.rf 01:23, 1 July 2008 (UTC)[reply]

The point is that Comet is one implementation of Reverse Ajax. They are not synonymous. Sorry for my 2-year delay in responding. David Spector (talk) 19:14, 31 March 2010 (UTC)[reply]

Merger with Ajax[edit]

Go ahead and merge what you can. But this article should redirect to Comet, not Ajax. —jacobolus (t) 23:19, 14 June 2008 (UTC)[reply]

You clearly have a conflict of interest when it comes to Comet (see Comet talk page), in the past you have deleted and redericted other articles to Comet. - 213.115.160.70 (talk) 09:46, 21 June 2008 (UTC)[reply]

Actually, I merged them, and left redirects, which is exactly proper according to wikipedia policy. See WP:MERGE. From that page, merges are suggested when articles “Overlap - There are two or more pages on related subjects that have a large overlap. Wikipedia is not a dictionary; there does not need to be a separate entry for every concept in the universe.” —jacobolus (t) 11:51, 21 June 2008 (UTC)[reply]

No you didn't merge, you actually deleted information and redirected to your favourite meme. It is not only annoying to lose information but also that someone repeatedly redirects to an article that can not explain the topic as pointed out on talk page. -- 83.254.214.72 (talk) 16:05, 28 June 2008 (UTC)[reply]

Erm … no information was deleted. Everything said on this article (info about both polling and long-polling) is explained in great detail at the Comet page, which explicitly cites “Reverse Ajax” as an alternate name. A reader trying to figure out what “Reverse Ajax” is would have a better understanding after reading the Comet page than after reading this page. Since the two terms refer to essentially the same thing, perfect articles for the two would have >90% overlap. No good reasons have been provided why these articles should remain separate. —jacobolus (t) 01:03, 1 July 2008 (UTC)[reply]

You deleted at least the contect of 3 articles in the past and redirected the following pages to your favourite "Comet" meme: HTTP streaming, HTTP push, HTTP stream, Pushlet, Pushlets. In particular you did this against the objections and reasons provided from other people. I am concerned this will happen again. - 62.127.211.2 (talk) 10:21, 1 July 2008 (UTC)[reply]

I agree with jacobolus. It even says on the Comet article that reverse Ajax is another name for it. The two are essentially the same thing with only minor differences, so I don't see any reason not to merge. On the other hand, there is no reason this should be merged into the Ajax article, because Ajax and reverse Ajax have little to do with each other. — ^Fatal Error 19:25, 5 July 2008 (UTC)[reply]

Definitly merge. There's too many terms out there describing the same things. Creating a script that sends an HTTP request to a server after certain intervals is ultimately Ajax/Remote Scripting; reverse Ajax, Ajax Push, Full Duplex Ajax, Streaming Ajax -- these are all jargon. Creating a script that sends an HTTP request to a server that flushes out its response after certain intervals and finally closes the response after say 30 seconds is ultimately Ajax/Remote Scripting, too; Comet, Ajax Push, Reverse Ajax, Two-way-web, HTTP Streaming, HTTP server push -- all jargon. There's nothing new in either of these articles or methods that I see in the sources. I would swear there has to be over 100 terms describing the use of the XMLHttpRequest object, the IFRAME element, the SCRIPT element, the IMG element, the OBJECT (Flash, Java) element and so forth to create HTTP requests -- all of them seemingly differing in the littlest way, such as keeping a connection alive by using some sleep function, or formatting the response as JSON, or using specifically the IFRAME element, or whatever. Just because Joe Shmoe made a blog post or website about their discovery that they just had and coined a word for does not make them "new technologies" and worthy of a whole article. A line needs to be drawn somewhere or the words and articles will never cease. All of these articles are of minimal content already, and together, they have redundant content and just interlink amongst one another creating this endless cycle that seems to say there should be a main article somewhere. --Quilokos (talk) 06:28, 1 August 2009 (UTC)[reply]

Merged and redirected. Quilokos is right on and there is no need to wait 3 more years. ~ Don4of4 ^[Talk] 04:40, 19 August 2011 (UTC)[reply]

Definition[edit]

Please note that Reverse Ajax is not the same as either Ajax or Comet. Reverse Ajax is any way to get asynchronous data from the server to the browser, and includes polling, comet and piggyback. For more information, see http://directwebremoting.org/dwr/reverse-ajax. JoeWalker (talk) 20:09, 1 July 2008 (UTC)[reply]

Yes it is, those methods are included in Comet as well (see Comet (programming)). — ^Fatal Error 19:28, 5 July 2008 (UTC)[reply]

I don't think it is. Reverse Ajax has a clear distinction to other technologies, Comet on the other hand is an umbrella term and more vague, it claims to include other technologies. As far as I can see there is no consensus among web developers and Wikipedia editors for one or the other to be the more dominant term. Btw, the term piggyback that JoeWalker mentioned is not part of the (current) Comet article. - 83.254.214.192 (talk) 14:49, 14 September 2008 (UTC)[reply]

Comet is long lived HTTP connections. Polling and Piggyback are not comet. Reverse ajax may use comet or polling or piggyback. Reverse Ajax != Comet. JoeWalker (talk) 22:17, 15 September 2008 (UTC)[reply]

Whether polling can be considered Comet is an open question. It is, either way, covered in the article about Comet, which is what matters as far as Wikipedia is concerned. “Piggyback” is a word used by precisely one source. If it’s important/relevant, it also can be discussed on the Comet page. The fact remains, 90% of the information for proper explanations of “Comet” and “Reverse Ajax” is overlapping. They are terms which mean, to the extent anyone reading a Wikipedia article to learn about the subject cares, the same thing. Having two articles about the same subject under two different names is pointless and a good way to let errors and contradictions sneak in. –jacobolus (t) 23:04, 31 March 2010 (UTC)[reply]

Merge please[edit]

Merge the cited information into the other two articles and make this article a redirect to AJAX. Heroeswithmetaphors (talk) 08:19, 10 February 2010 (UTC)[reply]

Oppose Reverse Ajax is in some ways the opposite of Ajax. Current technology is inefficient because it requires polling. But that will not always be case. Conceptually, Comet is just an inefficient mechanism for Server Pushing (Reverse Ajax), using polling (inefficient) or a long-time TCP connection (which violates the philosophy of the Web and increases global latency and load). Reverse Ajax is the parent concept; it potentially includes an implementation in which the server creates a temporary connection back to the client to notify it of a result or of other server-generated data (this isn't WP:OR, it's just not available yet). The only merging that makes sense to me is to change the Comet article to a section in this article. Someday soon, when companies and standards organizations create a Reverse Ajax object (or extend the Ajax object to have this design pattern and functionality), WP will be ready with a logical article structure. Since neither Ajax nor Reverse Ajax overlap in their basic functionality, neither should their articles. David Spector (talk) 19:08, 31 March 2010 (UTC)[reply]

There are two relevant “objects” created by “companies and standards organizations”. The first is called “Server-side events” (Opera has an implementation), and the second is called “Web Sockets” (Chrome has an implementation). One is part of HTML5 and the other is being discussed through the IETF. No one is planning to make a “Reverse Ajax” object. (For what it’s worth, the Comet article already discusses both of these – in other words, the “logical article structure” already exists. That coverage was highly detailed before the article was eviscerated; sadly, adequately describing the subject seems to be impossible while the user responsible for said evisceration insistently watches over the relevant articles.) –jacobolus (t) 23:11, 31 March 2010 (UTC)[reply]

“Reverse Ajax” is a term used by almost no one. “Comet” (for better or worse) is widely used and generally understood in the field. It really doesn’t matter all that much where the article finally sits, but the two articles should definitely be merged together, and one title redirected to the other. If someone wants to contact everyone who has expressed an opinion and have a discussion somewhere, that sounds like a good idea to me. –jacobolus (t) 23:08, 31 March 2010 (UTC)[reply]

DO NOT MERGE[edit]

Even if it is original research, it is notable. The only thing missing is example code, which I'm sure SOMEONE will provide. —Preceding unsigned comment added by 71.220.218.164 (talk) 04:11, 20 June 2010 (UTC)[reply]

Merging doesn’t remove content. Instead, both terms and their relationship can be defined and contextualized, and we can avoid a bunch of redundant and misleading material (as 100% of the contents of this article currently is). After a merge, this title would redirect to the other article, and readers would be able to read a more complete and useful description. –jacobolus (t) 18:07, 21 June 2010 (UTC)[reply]

Example[edit]

The example in the article is unclear to me. Aren't the "hi"'s supposed to be "hello"'s? --Beforez (talk) 11:39, 27 March 2011 (UTC)[reply]

It appears so. However, the example is more problematic because it directly conflicts with the stated definition. According to the definition (top section), reverse Ajax can be done via either polling or long connection. However, the example then goes on to claim that the polling model is traditional Ajax. So either the definition is wrong, or the example is wrong. (I'm inclined to lean toward the definition being wrong, but I'm not a reverse Ajax expert - I was coming here to learn more about it in the first place... GalacticCowboy (talk) 12:55, 13 June 2011 (UTC)[reply]

Expert help[edit]

Could we get an expert on computing to resolve this merge issue? It's been there for three years. Let's either do the merge or delete the tags. D O N D E groovily Talk to me 05:16, 7 August 2011 (UTC)[reply]