Distributed LINQ is an architecture for distributed computing, objects and collections, with LINQ available for both SIMD and MIMD scenarios on clusters, grids and peer-to-peer. Distributed LINQ is available at http://distributed.codeplex.com.
The history of science includes continuous changes to the meanings of concepts, often fundamental to our understanding of the universe, over time. By representing and storing both contemporary and historical scientific knowledge, mechanical reasoners can better infer the very processes of science. With regard to mechanical reasoners:
A repository of contemporary and historical scientific knowledge, in a machine-readable format, can enhance both science education and the development of such mechanical reasoning artifacts.
Philosophy of science topics include: confirmation holism, coherentism, contextualism, conventionalism, deductive-nomological model, determinism, empiricism, fallibilism, foundationalism, hypothetico-deductive model, infinitism, instrumentalism, positivism, pragmatism, rationalism, received view of theories, reductionism, semantic view of theories, scientific realism, scientism, scientific anti-realism, skepticism, uniformitarianism, vitalism, metaphysics.
The semantic view of theories indicates models as relating to theories. Each concept’s or theory’s definition includes, beyond semantics, logic and mathematics, its inclusion in a set of models. Each model contains a set of interoperating concepts and theories.
Semantic holism describes that the meaning of concepts may arise from their relationships within models. Incommensurability of meaning can arise when argument participants use language from different models.
As models can be represented by means of distributed field- or topic-specific files or resources, such files or resources can be composed and combined algebraically. Combinations of such files and resources can describe schools of thought applicable to regions at instants. Individual scientists can be described as having been utilizing various models or combinations of models at various points in time.
Concepts within and between models can be related as can entire models, with such relationships resulting from scientific processes including investigation and argumentation. Machine-utilizable resources including the details of such historical scientific processes will be tremendously advantageous.
The matter can be phrased as seeking a record of the repairs made to the ship of scientific knowledge, using Neurath‘s simile indicating that ”we are like sailors who must rebuild their ship on the open sea, without ever being able to dismantle it in dry dock and reconstruct it from the best components.”
Argumentation theory, which can be phrased as an alternative to both absolutism and relativism, can explain the piecemeal processes over the course of the history of science. With participants arguing about a main topic or matter, while making use of data resulting from scientific investigation, the substantiation of each claim, and otherwise communication, makes use of other concepts. The concepts used thusly, that are not objected to by participants, can be phrased as having either been accepted, useful or convenient to the participants of argumentation, at least for the purposes of argumentation, at that instant, while the argumentation focused on addressing the main topic or matter.
Argumentation theory and scientific reasoning interrelate in other interesting ways. For example, in the context of the hypothetico-deductive model, there is the problem of induction, and, in argumentation theory, there is the topic of the accrual of arguments.
A robust means of representing argumentation, including scientific argumentation, should include a means of indicating which models were in use by which participants, perhaps to the granularity of utterances, so as to enhance the accurate interpretation of sentences. Many formalisms of argumentation take it for granted that the semantics of arguments are available; knowledge representation formats that include natural language can include features to enhance the correct interpretation of sentences.
Representing scientific argumentation builds upon the previous topic of argumentation and mathematics while including more linguistic premises and a dynamic set of ontologies with each perhaps composed of multiple model components.
Robust resources for contemporary and historical scientific knowledge, resulting from described processes of investigation and argumentation, can provide numerous advantages including to science education, while enhancing the development of advanced scientific instrumentation.
With regard to argumentation, some representation formats include the Argument Interchange Format (AIF), Argument Markup Language (AML), Legal Knowledge Interchange Format (LKIF) and the Proof Markup Language (PML).
Argument mapping and visualization software include Araucaria, Argumentative, Athena, bCisive, Carneades, Collam, and Rationale. Web applications include Argumentum, bCisive Online, Cohere, Debategraph, truthmapping.com, Compendium, Argunet, MIT’s Deliberatorium, debatewise.com and CoPe_it.
Additionally, many computational mathematics software include their own proof formats such as HOL, Mizar, PVS, Coq, Otter/Ivy, Isabelle/Isar, Alfa/Agda, ACL2, PhoX, IMPS, Metamath, Theorema, Lego, Nuprl, Ωmega, B method, and Minlog.
At the 23rd International Conference on Automated Deduction, the first workshop on proof exchange for theorem proving (PxTP) occurred with goals that included facilitating the exchange of knowledge, such as machine-checkable proofs, between automated reasoning systems. Furthermore, the workshop indicated that “emerging work from several groups proposes standard meta-languages or parametrized formats to achieve flexibility while retaining a universal proof language.” Proof formats, meta-languages and logical frameworks for proofs were discussed.
Interestingly, argumentation formats can include mathematical proofs in their expressiveness. Knowledge representation formats can be devised capable of simultaneously representing conversational, mathematical, scientific, legal and political forms of argumentation and debate.
Recently, in a blog article, Citations, References and Footnotes in HTML Documents, I discussed extending or redefining the <cite/> element. While a SOAP-like @ref/@id pattern was presented, @cite can also be utilized, as presented in this article, resembling other existing HTML elements, as per <cite cite=”…”>.
With a <reference id=”ref1″ … comprehensive metadata attributes for referenced resource …/> in a document, a <q cite=”#ref1″>text</q> could, configurably with CSS3, render as: “text” (Smith 2011). The rendering could automatically add parenthetical references to hypertext. <cite/> and the HTML elements with the @cite attribute could include more attributes, including some or all of the comprehensive metadata attributes on <reference/>. That would allow rendering “text” (Smith 2011, p. 123), for example, from <q cite=”#ref1″ frompage=”123″/>. A <cite/> element could render as the parenthetical reference, as per CSS3, that is <cite cite=”#ref1″/> could render as (Smith 2011).
Footnotes differ from references. While putting footnotes in a document structure, such as a footer, is possible, a desired rendering might include footnotes per page or per document column. In any event, with something like <footnote id=”footnote1″>footnote text</footnote>, a <q cite=”#footnote1″>text</q> could render as “text”1 and a <cite cite=”#footnote1″/> would place the 1.
The <cite/> element could be useful for portions of a document that cite other materials. With document text such as
Lorem ipsum dolor sit amet, consectetur adipisicing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo consequat. Duis aute irure dolor in reprehenderit in voluptate velit esse cillum dolore eu fugiat nulla pariatur. Excepteur sint occaecat cupidatat non proident, sunt in culpa qui officia deserunt mollit anim id est laborum (Smith 2011).
it is unclear, in a simple mechanical sense, whether the parenthetical reference is intended as for the previous sentence, sentences, paragraph or paragraphs. However, the following markup Lorem ipsum dolor sit amet, consectetur adipisicing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. <cite cite=”ref1″>Ut enim ad minim veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo consequat. Duis aute irure dolor in reprehenderit in voluptate velit esse cillum dolore eu fugiat nulla pariatur. Excepteur sint occaecat cupidatat non proident, sunt in culpa qui officia deserunt mollit anim id est laborum.</cite> clarifies that and, furthermore, with such an approach, nesting is possible as per the markup Lorem ipsum dolor sit amet, consectetur adipisicing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. <cite cite=”ref1″><cite cite=”ref2″>Ut enim ad minim veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo consequat. Duis aute irure dolor in reprehenderit in voluptate velit esse cillum dolore eu fugiat nulla pariatur. Excepteur sint occaecat cupidatat non proident, sunt in culpa qui officia deserunt mollit anim id est laborum.</cite></cite>. The semantics of such a <cite/> tag could include that the contained hypertext content is the author’s while specifically indicating one or more references and/or footnotes.
CSS3 could be of use for rendering the <reference/> elements, envisioned as rendering as per references in a references section of a document, and for rendering the <footnote/> elements, envisioned as rendering per document column, per document page or in document sections.
With HTML5 video, video tracks and JavaScript, video documents are possible. A video document is a video that includes either interactive applets, charts, diagrams, equations, figures, graphs, illustrations, pictures, tables, videos or other document-related objects. A video document primarily uses video, instead of text, while making use of the same interactive document objects that modern text- and hypertext-based documents can have.
Video documents can be authored by writing hypertext or office-style documents and then making use of upcoming software to utilize the computer as a teleprompter while recording video. The written documents can be processed into video tracks. Video post-production software can also be of use when generating video documents.
In addition to business-related applications, video documents can enhance scholarly, scientific and educational forms of communication. Scientific and educational videos have often included the use of 3D computer graphics for video diagrams and scientific visualization. With computers and consoles, client-side 3D rendering is available for interactive scientific visualization.
XML video tracks are possible, in addition to WebVTT video tracks, with regard to HTML5 video. With a minimal set of XML attributes, start, end and id, as per:
<… t:start=”00:00:00″ t:end=”00:00:30″ t:id=”part_1″/>
XML files can be processed as video tracks. A video track XMLNS can contain additional markup attributes for use with the various kinds of video tracks or additional XMLNS can add to the functionality of a base video track XMLNS. Video track XML is envisioned as attributes-based, so that it can extend other markup formats. RDFa may be of use to such XML tracks.
When computers are utilized as teleprompters, authors’ compositions, even hypertext compositions, can be of use to generating such tracks. Post-production software can additionally be of use to generating such tracks.
Such XML video track files can advantage end users both directly, as downloadable files for use with software, and indirectly, by enhancing information search and retrieval and content discovery.
An illustrative example involves the corpus of argumentation and debate videos. Some formats for argumentation and debate include Argument Interchange Format (AIF), Argument Markup Language (AML) and Legal Knowledge Interchange Format (LKIF). It occurs that each existing knowledge representation format can either already encompass the corpus of videos or can version to encompass the corpus of videos. In any event, the entire corpus can be enhanced by means of XML tracks describing the structured content of the videos.
In HTML5, multiple sources for multimedia objects can be indicated. Extending upon the multipart MIME concepts, as illustrated in Multifile Protocols, Multipart MIME and Multimedia, other HTML scenarios include indicating multiple resource locations for hyperlinks. The following example illustrates indicating two resource locations that differ by protocol:
<a>text
<link rel=”alternate” href=”http://example.com/file.ext”/>
<link rel=”alternate” href=”ftp://example.com/file.ext”/>
</a>
Such expressiveness is possible with both multipart MIME and Metalink. Illustrated, herein, are approaches to expressing multiple resource locations for HTML hyperlinks. The following example illustrates combining multiple resource locations with a multifile protocol and multipart MIME:
<a>text
<link rel=”alternate” type=”multipart/mixed” href=”torrent+http://example.com/file1.torrent”>
<link rel=”part” type=”video/avi”/>
<link rel=”part” type=”image/png”/>
</link>
<link rel=”alternate” type=”multipart/mixed” href=”torrent+http://example.com/file2.torrent”>
<link rel=”part” type=”video/mpeg”/>
<link rel=”part” type=”image/png”/>
</link>
<link rel=”alternate” type=”multipart/mixed” href=”torrent+http://example.com/file3.torrent”>
<link rel=”part” type=”video/ogg”/>
<link rel=”part” type=”image/png”/>
</link>
</a>
An example illustrating multiple resource locations and the URI fragment syntax for referring to files inside of a torrent:
<a>text
<link rel=”alternate” type=”video/avi” href=”torrent+http://example.com/file4.torrent#file=videofile.avi”/>
<link rel=”alternate” type=”video/mpeg” href=”torrent+http://example.com/file4.torrent#file=videofile.mpg”/>
<link rel=”alternate” type=”video/ogg” href=”torrent+http://example.com/file4.torrent#file=videofile.ogg”/>
</a>
In a previous article, Segmented Downloading and Web Browsers, a project was initiated to robustly integrate segmented downloading and the web. Torrent was described as possible to implement as a networking protocol.
The multifile protocol ”torrent+http://” was arrived at. The protocol is named for its use of HTTP to obtain a torrent file for use in segmented downloading. As the “torrent+http://” multifile protocol encapsulates the use of the torrent file, the MIME type for “torrent+http://” is not “application/x-bittorrent”.
A multifile protocol can reference one or more files simultaneously. When referencing multiple files by means of a multifile protocol, the envisioned MIME type is “multipart/mixed” as per
<a type=”multipart/mixed” href=”torrent+http://example.com/example.torrent”/>text</a>
One possible syntax to detail multipart MIME structure in HTML is:
<a type=”multipart/mixed” href=”torrent+http://example.com/example.torrent”/>text
<link rel=”part” type=”video/ogg”/>
<link rel=”part” type=”video/mpeg”/>
<link rel=”part” type=”text/vtt”/>
<link rel=”part” type=”text/vtt”/>
<link rel=”part” type=”text/vtt”/>
</a>
The use of XML, in that regard, is advantageous in that data per resource item, such as filenames and hashes, can be indicated with XML attributes, as per Metalink.
Indicating files inside of a torrent is possible with a URI fragment identifier syntax:
<a type=”video/ogg” href=”torrent+http://example.com/example.torrent#file=videofile.ogg”/>text</a>
Such a URI fragment identifier syntax is interoperable with media fragments. The following HTML5 video example illustrates a video source of an interval or a clip of a video file inside of a multifile torrent:
<video>
<source type=”video/ogg” src=”torrent+http://example.com/example.torrent#file=videofile.ogg&t=10,20″/>
</video>
A more detailed HTML5 video example illustrates that the same torrent can contain multiple versions of a video resource as well as video tracks:
<video>
<source type=”video/ogg” src=”torrent+http://example.com/example.torrent#file=videofile.ogg”/>
<source type=”video/mpeg” src=”torrent+http://example.com/example.torrent#file=videofile.mpg”/>
<track type=”text/vtt” kind=”subtitles” srclang=”en” src=”torrent+http://example.com/example.torrent#file=subtitles-en.vtt”/>
<track type=”text/vtt” kind=”subtitles” srclang=”es” src=”torrent+http://example.com/example.torrent#file=subtitles-es.vtt”/>
<track type=”text/vtt” kind=”subtitles” srclang=”fr” src=”torrent+http://example.com/example.torrent#file=subtitles-fr.vtt”/>
</video>
Previously, the ways for indicating citations, references and footnotes in HTML have been numerous with combinations of text strings and HTML formatting. Mechanically obtaining the interconnectivity between web documents, for example for information search and retrieval or content discovery purposes, as seen at the arXiv and CiteSeerX projects, has required complex document and text processing to reobtain the structure from the indicated permutations of text and formatting. By introducing a <reference/> element, possibly <footnote/> element, and extending the attributes on the <cite/> element, a robust structural system for citations, references and footnotes can exist for HTML documents.
An example is provided to illustrate the differences in the presentation layer rendering for the same structural element, a referenced material:
APA style
Reference. (2011, August 22). In Wikipedia, The Free Encyclopedia. Retrieved 18:11, August 25, 2011, from http://en.wikipedia.org/w/index.php?title=Reference&oldid=446196654
MLA style
Wikipedia contributors. “Reference.” Wikipedia, The Free Encyclopedia. Wikipedia, The Free Encyclopedia, 22 Aug. 2011. Web. 25 Aug. 2011.
MHRA style
Wikipedia contributors, ‘Reference’, Wikipedia, The Free Encyclopedia, 22 August 2011, 19:17 UTC, <http://en.wikipedia.org/w/index.php?title=Reference&oldid=446196654> [accessed 25 August 2011]
Chicago style
Wikipedia contributors, “Reference,” Wikipedia, The Free Encyclopedia, http://en.wikipedia.org/w/index.php?title=Reference&oldid=446196654 (accessed August 25, 2011).
CBE/CSE style
Wikipedia contributors. Reference [Internet]. Wikipedia, The Free Encyclopedia; 2011 Aug 22, 19:17 UTC [cited 2011 Aug 25]. Available from: http://en.wikipedia.org/w/index.php?title=Reference&oldid=446196654.
Bluebook style
Reference, http://en.wikipedia.org/w/index.php?title=Reference&oldid=446196654 (last visited Aug. 25, 2011).
Bluebook: Harvard JOLT style
Wikipedia, Reference, http://en.wikipedia.org/wiki/Reference (optional description here) (as of Aug. 25, 2011, 18:11 GMT).
AMA style
Wikipedia contributors. Reference. Wikipedia, The Free Encyclopedia. August 22, 2011, 19:17 UTC. Available at: http://en.wikipedia.org/w/index.php?title=Reference&oldid=446196654. Accessed August 25, 2011.
We can see from the BibTeX representation that the different notations have a common structure:
BibTeX entry
@misc{ wiki:xxx,
author = “Wikipedia”,
title = “Reference — Wikipedia{,} The Free Encyclopedia”,
year = “2011″,
url = “http://en.wikipedia.org/w/index.php?title=Reference&oldid=446196654“,
note = “[Online; accessed 25-August-2011]“
}
A robust model for materials, a metadata model, can be devised such that its structural elements would be the substructure of or attributes upon a <reference/> markup element with a corresponding JavaScript interface. By means of CSS3, such a <reference/> element could render to at least any of the existing styles.
Furthermore, with such a <reference/> element, a <cite/> elements could connect to <reference/> elements. A <cite/> element can include an attribute, @ref, with a text string value identical to the @id attribute value of its corresponding <reference/> element. Some or all of the attributes on <reference/>, from the aforementioned metadata model, can be overridden upon <cite/> which allows for indications, such as page ranges, or even to paragraphs or sentences on pages, in referenced materials. The rendering of the <cite/> elements could, additionally, be described by CSS3.
Furthermore, with a <cite/> element, as described, a <cite/> element could refer, additionally, to <footnote/> elements. The typography, layout and formatting styles for citations, references and footnotes, which often occur in scientific, scholarly and encyclopedic documents, can be facilitated by the <cite/>, <reference/> and <footnote/> (or <note/>) elements, utilizing CSS3, as described.
In searching with regard to technology and archives and library systems, I found an interesting technology resembling multi-metadata format syndication with additional synchronization capabilities. It contains the capability for set-based catagorizations and allows for paginated result sets in the delivery of larger response sets. The Open Archives Initiative Protocol for Metadata Harvesting (OAI-PMH) allows one to harvest and synchronize from a repository of metadata, for example from websites including arXiv and CiteSeerX.
It might be interesting to create federated desktop search services that connect to systems implementing OAI-PMH, OAI-ORE, and/or SRU. Presently, desktop search appears to have implementations utilizing OpenSearch.
Regarding the space of search and archives, WorldCat has an interesting feature where users can make and search lists of items.
Another interesting idea is the use of segmented downloading with archives, harvesting, synchronization and search.
