Talk:Data exchange

This article is rated Stub-class on Wikipedia's content assessment scale.
It is of interest to the following WikiProjects:

Computing Low‑importance

	This article is within the scope of WikiProject Computing, a collaborative effort to improve the coverage of computers, computing, and information technology on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.ComputingWikipedia:WikiProject ComputingTemplate:WikiProject ComputingComputing articles
Low	This article has been rated as Low-importance on the project's importance scale.
	This article has been automatically rated by a bot or other tool as Stub-class because it uses a stub template. Please ensure the assessment is correct before removing the `\|auto=` parameter.

Data exchange language was nominated for deletion. The discussion was closed on 27 September 2011 with a consensus to merge. Its contents were merged into Data exchange. The original page is now a redirect to this page. For the contribution history and old versions of the redirected article, please see its history; for its talk page, see here.

I would like to add a page on "data advertising exchange" such as BlueKai. But, it would not appear that this would be the place for it. Suggestions? Ipsofacto123 (talk) 14:28, 6 January 2009 (UTC)[reply]

data exchange is introduced within Demand_Side_Platform page. it would be good to include more details related to DSP context. Thanks.

jccking (talk) 10:19, 6 August 2011

Missing languages

RSS, RDF (OWL etc.) come to my mind. I bet there are more. --178.2.56.172 (talk) 05:38, 10 October 2011 (UTC)[reply]

Possible inaccuracies in the article

XML

I marked the "Therefore, it can hardly called a data exchange language." sentence as original research, since, while using seemingly "logical" reasons, it completely overlooks the facts:

XML is being used for data exchange
as a data exchange language, XML is very popular

In my opinion, the sentence should be left out (being marked as original research long enough already)— Preceding unsigned comment added by Ladislav Mecir (talk • contribs) 22:49, 14 October 2011

Remove that sentence, it is biased. --188.104.108.229 (talk) 10:38, 15 October 2011 (UTC)[reply]

Also, regarding the "Synonyms and homonyms" column contents: I would say, that tags from different namespaces actually may be homonyms.Ladislav Mecir (talk) 21:02, 23 October 2011 (UTC)[reply]

Dialecting

One definition of dialecting used in the article is this: "For example XML is a markup language that was designed to enable the creation of dialects (the definition of domain-specific sublanguages)" This definition agrees with the definition used in the "Dialecting" article.

The other definition used is: "Whether the language definition is available in multiple natural languages or dialects."

These definitions contradict each other, since they actually don't have the same meaning. For the clarity sake, as well as to obtain a non self-contradictory article, we should stick to just one definition, and I suggest the one compatible with the "Dialecting" article.

If we do want to speak about the possibility to use words of different natural languages, we should use a different notion, not dialecting.Ladislav Mecir (talk) 22:53, 14 October 2011 (UTC)[reply]

I wouldn't tie it to natural languages. One can easily argue that "Visual Basic" is a "dialect" of "Basic" (which doesn't make Basic a language designed for dialecting, though). This is consistent with terms such as "domain-specific sublanguages", but has nothing to do with "multiple natural languages". --188.104.108.229 (talk) 10:38, 15 October 2011 (UTC)[reply]

Definition

"Data exchange is the process of taking data structured under a source schema and actually transforming it into data structured under a target schema, so that the target data is an accurate representation of the source data." - that looks to me like the description of transformation, rather than data exchange.

For me, data exchange is rather a communication between the source of the data (the transmitting party) and the target of the data (the receiving party). The source as well as the target can be e.g. a computer program, some kind of data storage system, or even a human being (if the data are human readable, which is of advantage).Ladislav Mecir (talk) 22:48, 14 October 2011 (UTC)[reply]

But then it's just "communication". Transformation is an important aspect of data exchange as opposed to just storing and transferring data. Otherwise, JPEG is also a data exchange. The key point about "data exchange" and "data exchange languages" is that not all components must be perfectly "compatible", but still are able to exchange data IMHO. --188.104.108.229 (talk) 10:38, 15 October 2011 (UTC)[reply]

Anyway, what kind of transformation do you expect when handing some data (written in a data exchange language) to a human being?Ladislav Mecir (talk) 18:32, 15 October 2011 (UTC)[reply]

Moreover, JPEG clearly is a file format used for data exchange. The only trouble is, that it is not a universal format, being designed to exchange picture data. Ladislav Mecir (talk) 18:37, 15 October 2011 (UTC)[reply]

After searching the Internet, I found many (even in Wikipedia) articles, where "Data exchange" really meant a special kind of communication exchanging data, i.e. informations suitable for use in computers. One example: [1] — Preceding unsigned comment added by Ladislav Mecir (talk • contribs) 20:37, 23 October 2011 (UTC)[reply]

In my opinion, it is only true data exchange when: (1) the applications involved are entirely independent and do not know about each other (except for maybe a data exchange protocol, say using WSDL interfaces or something like this) (2) the format is not a native storage format. Otherwise it is just accessing the same stored data (this is a very weak definition, as you will find applications that store data in XML for themselves...) (3) the format is flexible/generic/extensible and takes into account that the applications may have a different interpretation and/or do not even know what the data semantics are (say, allows for transformation with generic tools such as an XSLT processor does for XML, or a webbrowser does for converting JSON into its internal JavaScript object model)

And, of course, you can always construct things that essentially violate any of this. Say, I can put a JPEG image base64-encoded into a single-tag XML file. That isn't really XML anymore. And I can convert an XML file into an image, transfer it, OCR it, decode the XML. That doesn't make the image format a data exchange format either IMHO. So we must talk about what the formats were designed for. JPEG was designed for compressed storage of pictures, ergo it is a storage format. --178.2.60.180 (talk) 21:04, 24 October 2011 (UTC)[reply]

Schemas

"Whether the language definition is available in a computer interpretable form." - I think, that e.g. DTD "works" as a schema language for XML, but, the greatest disadvantage of DTD is, that it actually isn't XML. This problem has been solved by defining XML Schema, which is itself a dialect of XML. These two cases should be discerned in the article, informing the reader, when the respective option is available only "externally". Ladislav Mecir (talk) 23:17, 14 October 2011 (UTC)[reply]

Well, DTD itself also doesn't define XML itself, but dialects. I don't see an explicit requirement for the dialect definition language to be the same as the data language. --188.104.108.229 (talk) 10:38, 15 October 2011 (UTC)[reply]

" I don't see an explicit requirement..." - What you do or don't see is not very important. That requirement was actually the reason why, even though DTD existed, XML Schema was specified. See the relevant sources. Ladislav Mecir (talk) 18:21, 15 October 2011 (UTC)[reply]

The decision to add an XML-based Schema language was a voluntary choice by the XML people themselves, wasn't it? So who says this is a requirement? It's just another feature, that you can now use regular XML tools to generate XML schemas, isn't it? (without going through hacks such as the text output module of XSLT, that is) --178.2.60.180 (talk) 20:54, 24 October 2011 (UTC)[reply]

Flexible

"Flexible - Whether the language enables extension of the semantic expression capabilities without modifying the schema." - this definition is quite complicated, and, in my opinion, it is not understandable. — Preceding unsigned comment added by Ladislav Mecir (talk • contribs) 23:22, 14 October 2011 (UTC)[reply]

I don't get it either. --188.104.108.229 (talk) 10:38, 15 October 2011 (UTC)[reply]

Semantic verification

Why is this not possible with XML? Especially with schemas (XMLSchema, Schematron etc.) that for me clearly says "semantic verification". --188.104.108.229 (talk) 10:38, 15 October 2011 (UTC)[reply]

Well, this is an iteresting question. I guess, that for XHTML, semantic verification is possible. But, OTOH, I would say, that it is not possible for XSLT.Ladislav Mecir (talk) 18:26, 15 October 2011 (UTC)[reply]

Moreover, for some types of data (semantically rich), the semantic verification is impossible. Languages always allowing semantic verification are not suitable for such data. Thus, a data exchange language which does not offer semantic verification in all cases may be more flexible than a language always offering such verification.Ladislav Mecir (talk) 00:58, 16 October 2011 (UTC)[reply]

The question is, what "semantic verification" actually means. REBOL is also a programming language. Full semantic verification probably involves solving the halting problem. So what is it really? Schema validation? (What good are schemas if they cannot be validated - redundant to column 1) That clearly is available for XML as well. Using Schematron or some weird XSLT hacks, you can probably even require lists to be sorted in order to be valid; I'm not sure which of the other formats has such a rich choice in verification options. --178.2.60.180 (talk) 20:50, 24 October 2011 (UTC)[reply]

Bias

I have the impression that many of the definitions here are strongly biased towards the promotion of REBOL and Gellish, which both don't yet have widespread adoption. The article shouldn't just be another "why XML sucks" article. --188.104.108.229 (talk) 10:38, 15 October 2011 (UTC)[reply]

That is certainly not the purpose here. What I did was to try to remove the original research "Therefore..." part as mentioned above. Moreover, I tried to put in verifiable informations about REBOL, but that does not mean that I think the article is already in a perfect form. Quite the opposite is true, see my criticisms above. If you think that there are some positive properties of XML missing that should be mentioned, then go ahead, and add them to the article, please. Or, otherwise, if you think any information mentioning properties of XML is not fair, then, again, correct, it, please. As mentioned above by me, even the definitions of the table columns look unclear and contradictory (either to the rest of the article, or to some other Wikipedia article), so, lots of corrections are needed. That does not mean I like the type of comments like this one. Please, try to be more constructive, and correct some of the mistakes, if you can.Ladislav Mecir (talk) 18:15, 15 October 2011 (UTC)[reply]

Sorry, I did not mean to offend you. It's just that I got this impression from the article. I'm not an expert on XML; I don't feel qualified to help with the table. But some things just sound like they are worded in a way that XML barely misses the mark, and the other system gets the points. And I mean, when you look at the REBOL article, it claims that

view layout [text "Hello world!" button "Quit" [quit]]

is easy to use and human readable. Next we'll see S-expressions as data exchange format? And on one hand, Gellish comes with a Dictionary-Taxonomy, on the other hand it claims to be lightweight? Somehow these don't agree for me. Ah, I see: this has been changed from an earlier version to "lightweight version available" ... sorry, I still say this is biased. --178.2.60.180 (talk) 20:45, 24 October 2011 (UTC)[reply]

Ion

Could somebody please add a mention of Ion to the page. Ion is a binary and text interchangable, typed JSON-superset that has recently been open-sourced by Amazon [2]. Bonomont (talk) 08:05, 23 April 2016 (UTC)[reply]