Uni Göttingen
Institute for Informatics
Databases and Information Systems

dbis

Semistructured Data and XML
Summer 2024

Prof. Dr. Wolfgang May

Date and Time:

  • Monday 14-16 ct, IFI SR 2.101
  • Wednesday 10-12 ct, IFI SR 2.101
  • This year, DBIS will use mainly non-live teaching by pre-recordings. There will be some live online meetings with BigBlueButton provided by GWDG; the rooms/meetings can be entered via StudIP. Maybe, there are also optional meetings (Monday afternoon slots) in presence at the IFI.
  • Materials for self-studying (in english) will be linked below weekwise:
    • revised videos taken from summer term 2020 (as the "original" dates in the filenames indicate)
    • PDF slides
  • Please also read the general and technical information about DBIS virtual teaching.

Lecture and Exercises mixed (see announcements on this page). There will be non-mandatory exercise sheets whose solutions will be discussed as parts of the lecture.
All materials and announcements can be found HERE on the "blue DBIS pages".

Module M.Inf.1141, 4 SWS, 6 ECTS.
The module's home is the MSc studies in Applied CS. It can also be credited in the BSc studies in Applied CS (as "Vertiefung Softwaresysteme und Daten", B.inf.1706),
and in several other studies:
BSc/MSc Wirtschaftsinformatik, Mathematik (BSc/MSc), Digital Humanities, Teaching/2-Fach-Bachelor, PhD GAUSS, ...

Course Description

One of the most important facts that lead to the overall success of XML is that the "XML world" combines a lot of already known concepts in an optimal way for coping with a broad spectrum of requirements. The course will first review some of these preceding (partially even historic) concepts (network database model, relational databases, object-oriented databases) and the integration of data and metadata (SchemaSQL). Then, the idea of "semistructured data" is introduced by showing early representatives that helped to shape the XML world (F-Logic, OEM).

In the main part, XML is presented as a data model and a markup-meta-language, and the current languages of the concepts of the XML world are systematically investigated and applied: DTD, XPath, XQuery, XSLT, XLink, XML Schema, and SQL/XML.

The lecture uses the geographical sample database "Mondial" in its XML version for illustrations.

For practical exercises, the XML software is installed in the IFI CIP Pool. The software playground page can be found here; the XPath/XQuery/XSLT Web interface is available here.
The sample code fragments can be found in the CIP pool under /afs/informatik.uni-goettingen.de/course/xml-lecture/ .

Dates & Topics

  • Mo 8.4.: 14ct-16 live meeting via StudIP->SSD-SS24->Meetings->SSD-2024-04-08-intro-meeting
    Administrativa, Overview, ...

    The lecture is intended to have three dimensions:

    • A typical technology course (XML and its languages) with much practical contents. Perfect for self-learning and experimenting.
    • XML is a perfect example of many computer science concepts (practically: around data, data exchange and interoperability, theoretical: trees, language design)
      By now, most of you have already practical experience with many of them: ant, maven (XML), JSON (a form of a much newer "lightweight" reduction of XML).
      Learning: Always be aware and analytical of the underlying concepts and structures.
    • "History": how did (do, and will do) computer science and IT concepts evolve?
      The "XML world" (and many other things, like most basic Web technologies) were developed as a response to new requirements in the mid/late 1990s. A very interesting time, where many already existing ideas and experiments were improved and combined.
      Learning: see how to analyze and question requirements and existing things/software.

Part 0: Preface

Materials for self-studying:

Part I: History, evolution, and comparison of data models until 1995 and requirements for XML

  • This part is not to be seen as a "technical lecture" to learn details of some languages, but to show how ideas and concepts, in this case, data management and data models (and the concept of high-level declarative query languages) evolved, and how new requirements (Web, data integration, data interoperability, handling documents+data and metadata) lead to XML in the mid/late 1990s.
  • 10.4., 15./17.4., 22./24.4.: Self-studying, no pre-scheduled live meetings.
  • Overview:
  • Data Models: about structuring data and the development of query languages.
    Slides: Review of the Relational Model
    Recording: General concepts of data models, the relational model (with its querying concepts) as an example data model.
  • The Requirements for "semistructured data" in the mid 90s, history of data models: appropriateness of the data model for modeling data and other requirements of the time.
    Slides: data models
    Recording: the network database model (1960s, pre-declarative querying) and the object-oriented database model (late 1980s)
    Recording: the object-oriented database model (late 1980s, OQL, OIF, Corba)
  • Some references to read about database history (optional):
  • "History" continued - requirements and academic prototypes of the early 1990s:
    SchemaSQL (an extremely powerful, yet syntactically minimal "opening" of SQL to metadata) and early semistructured data models (Tsimmis/OEM and F-Logic):
    Slides: early semistructured data models
    Recording: SchemaSQL
    Recording:Tsimmis/OEM (part 1)
    Recording: Tsimmis (part 2), F-Logic and the situation pre-XML
  • (Exercise: if you have knowledge about JSON (JavaScript Object Notation), compare JSON with the concepts discussed above in Part I)

Part II: XML concepts and technology - aspects of XML as a data structure in Computer Science

  • Provisional Exam Date: Thursday, July 25th, 10-13h Online exam with ILIAS (as in the previous years)
    Confirmed in the Meeting on April 29
  • Monday, April 29th 14:15: Live online meeting
    to get some feedback, answer questions, and to discuss/give a roadmap for the rest of the lecture.
    From now on, the course gets "productive" and continues with XML and the languages of the XML world.
  • (Exercise: if you have knowledge about JSON, compare JSON continuously with the following aspects of XML)
  • 29.4.: XML: data model, language, DTDs etc.
    Slides: XML basics
    Recording: XML basics
  • 1.5.: Holiday
  • 6.5.: XML: DTDs etc. (cont'd)
    Recording: DTD, the xmllint tool
  • Exercise Sheet 1
    (XML basics, parsing, grammar aspects, parsing)
    If there are questions etc., the RocketChat dbis channel can be used (also participants are encouraged to answer questions from others).
  • 8.5. XML parsing ...
    Recording: XML parsing, XHTML (and parsing)
  • 13.5.: Discussion of Exercise Sheet 1:
    Solutions to Exercise Sheet 1.
    Recording:
    • You find the recording from 2022 as follows: register for "Semistructured Data Summer Term 2022", and use the recording of the Meeting "2022-05-30-exercise-sheet-1" (this was an "experimental" session in the seminar room, using the Smartboard and BBB)

Part III: Languages of the XML world: XPath, XQuery, XSLT ...

Exams

The exam will take place as a written/online exam using the ILIAS system.
Thursday, July 25th, 2024, 10-13h (time details see below)
Online-at home(or @anywhere) with ILIAS.

  • The exam is an "open-book-exam", i.e., you can use documentation whatever and whenever you want (but it is intended that you should not need it much, except maybe for looking up syntax - DBIS exams are not "learning" exams, but competence exams). Help by any other persons is not allowed.
  • Strongly recommended: prepare a "cheat sheet" (German: Spickzettel) where you put everything that you may want to lookup quickly. This preparation also helps to become aware of the material.
    Details of the syntax of the pre-XML history section are not relevant. That section is for understanding the concepts and the problems.
  • Like in a "paper exam", solutions that do maybe not work (completely) can be delivered and will be graded with appropriately partial points.

    The Exam Procedure with IDENT, ILIAS, and BBB (as 2023, technical details may change)

    • Official information about online exams: german/ english
    • We will use the IDENT feature (with photo). Enter IDENT via FlexNow (the IDENT feature opens at 10:15) - you must identify by uploading a photo of your face and your study ID card (and find an information text, basically the same as here, and the Ilias-password [not yet the password] ) before entering Ilias. (note: when entering, the IDENT system usually starts in german, so you have to switch manually to english there).
    • We meet in a StudIP-BBB-Room, via the StudIP lecture -> Meetings -> "SSD-2024-07-25-EXAM".
      • The BBB meeting officially begins at 10:30 (I will be there at about 10:10).
        Then, I will read through the whole exam, give some comments and answer first questions - like in a synchronous in-presence-exam. This is expected to take about 15 minutes, until about 10:45.
      • Also in the StudIP course, under "Learning Modules" -> "Course in Ilias" you can then log into ILIAS from 10:30 on, using the above password. It is strongly recommended not to start working in ILIAS immediately, but to follow reading the whole exam first. After reading, ILIAS will remain open for 120 minutes, ending about 12:45.
    • You should have installed xmllint (for XML-DTD validation, it has better error messages than saxon) and saxon (for XQuery and XSLT) -or whatever XML software you want to use- on your computer.

Exam Preparation

Note: this section is usually published at te end of the lecture, when it starts being useful for exam preparation.
By now, it is just to get an overview what the exam looks like.

  • you need to design a small, but useful XML instance from the given text (which will contain sample data as in the earlier exams). Note that for queries in the "all x such that for all y ..." style, it is helpful to have such x,y in the instance.
  • have a plan, "experience", how to edit an XML file quickly with copy+pasting elements. This can be much faster than in a paper exam where the chatty XML stuff must be written several times. Choose short element names and attribute names. You may also abbreviate text contents like person names to initials.

Training example exams