intro.tex

% vim:ts=1:et:nospell:spelllang=en_gb:ft=tex

 \chapter{Introduction}

  For over 25 years, \ppt* has been the market leader in digital prsentations.
  Admittedly, it was a revolutionary software package when it was first
  introduced, and its ease-of-use combined with its supreme graphical
  capabilities --- at least compared to other software in the same era --
  quickly made it one of the most popular software packages in history. 25
  years later, \ppt* can claim over 90\% market share in presentation software,
  and on average 30 million \ppt presentations are created every day
  \citep{parker-1, drucker-1, bajaj-1}.

  % TODO Currently, at the time of writing, up to now ?
  In this time, \ppt* has gotten many new features, and certainly improved and
  grew with every new version, but it never really changed its core approach.
  It started out mimicking the then-popular and widespread use of dia and
  overhead projection slides. This technique where digital media imitates
  physical objects and actions is known as skeuomorphism, and it is often used
  as a way to make people who are new to digital media feel more familiar with
  the interface, the content, and the behaviour of software. However, when
  many people of the current generation have never seen the physical object or
  action because the digital version has almost completely replaced it, this
  skeuomorphism no longer makes sense (e.g. the `save' button in many
  applications still showing a floppy disk).

  % TODO ask is this ok
  In \ppt, this was at the time a good way to convince people of its merits and
  purpose, allowing them to feel comfortable with a familiar format instead of
  alienating potential customers with a new and potentially confusing
  interface.  However, this interface is quite restricting, and in recent years
  different approaches have seen the light of day. The zoomable user interface
  of Prezi is probably the most well-known, but apart from abandoning the
  traditional slide format it does little to improve or extend the concept of
  presenting information to an audience.

  This is where \mxp comes in. Its extensible plug-in system allows anyone with
  some knowledge of programming to create new functionality to use in
  presentations. Examples are interactivity with the audiencer through various
  means, controlling the presentation from another device --- or several! ---
  and (re)modelling data while presenting it, based on feedback from the
  audience.

  While this is obviously a big improvement on the traditional presentation
  model of \ppt* and related tools, it remains hard to convince the general
  public of its merits. People are generally afraid of change, and it is
  important to make the transition as smooth as possible. On top of that,
  people are often worried that the work they did in the past may be lost ---
  or worse, irrelevant --- after switching to a new tool. This alone may be
  a huge factor in deciding wether or not to start using new software, or to
  stick with what they know.

  That is where the subject of this thesis comes in. We aim to provide a way
  for people to convert their existing \ppt presentations into \mxp
  presentations, allowing them to take their previous work with them in their
  switch to \mxp. This way, we lower the treshold for them to make the decision
  to start using \mxp as their presentation software of choice. Once all their
  existing \ppt content is available, usable and editable in \mxp, it should be
  obvious to anyone why \mxp is the better option for their presentations.

  Another common problem with \ppt presentations is the way they look. This is
  not necessarily the fault of the software; most people just are not trained
  in graphical design, and as such they know very little about proper layout,
  color choices, or slide content limits. Everyone has probably encountered
  slides with full paragraphs of text, too small to read and / or too much to
  process in the short time the slide is visible --- (too) many people have
  made those slides themselves.

  When we say this is not the fault of the software, that is mostly true, as
  the creators of these slides obviously made a conscious choice to make their
  content appear like that. It could be said however that \ppt* and other
  presentation tools are guilty through inaction. We believe it is possible to
  have software either warn its users against these choices and practices, or
  --- even better --- have the software fix these problems automatically.

  One of the primary purposes of \mxp is to provide automated layout, much like
  \latex does, ensuring that the content creator only has to worry about the
  actual content, while the software takes care of layout. In practice, both
  \latex and \mxp currently use template-based layouts, where the contents'
  position is predefined in the template and not related to or based on its
  size, shape or nature. In the end, everyone who has ever used \latex knows
  that sooner or later you will struggle to get a certain image incorporated in
  the text correctly, ending up doing the layout yourself anyway, because the
  predefined template just does not work properly for your specific content.

  Our goal is to eradicate those situations. Automated layout should
  dynamically adjust to any content it is given, no matter the size or aspect
  ratio. This may seem hard, if you consider the limits of slides and the fact
  that you can only fit so much content on them before they are full. This is
  where another important aspect of \mxp comes into play: we are not
  necessarily bound to the limits of slides. If we do not have to consider the
  boundaries of traditional slides, we can fit content together in an
  aesthetically pleasing way much easier, without having to scale anything.

  As such, the second part of this thesis focuses on implementing true
  automated layout in \mxp. Again with the goal to convince \ppt* users to
  switch, showing that their presentations actually could look better in \mxp,
  but at the same time we also provide new functionality to other \mxp users.

  We believe this functionality will improve the aesthetic aspect as well as
  the effectiveness of presentations. If content is not scaled down to fit the
  articifial confines of a slide, but can instead be shown and studied in
  detail, this should clearly increase the flow of information towards the
  audience. Providing an overview of the information in a presentation becomes
  easier and more effective too: where traditional presentations rely on a
  boring table of contents, in which the presenter announces the subjects
  they will be talking about one by one while the audience forgets the first
  thing in the list by the time they get to the last, \mxp allows the presenter
  to just show all of the presentation's content at once just by zooming out.
  Here automated layout can help as well: content can be arranged in such a way
  that an overview effectively hilights the important subjects, different parts
  or keywords of a presentation.

  Last but not least, we hope this functionality improves the experience of
  creating a presentation. Everyone who ever created a presentation knows, and
  research has shown \citep{lok-1} that often more time is spent on creating
  and fine-tuning the layout than actually putting in the content. Most
  presenters however have not had any significant training in creating
  effective layout, which means this time is often wasted on a layout that ends
  up not actually benefiting the presentation as a whole. We want to eliminate
  this problem by taking control over the layout away from the presenter and
  instead providing them with a programatically generated layout that presents
  the information provided by the presenter in the best, clearest way possible.

  In summary, we determined how to take the\ .ppt file format apart to harvest
  a \ppt presentation's content, in order to transform the components into a
  \mxp presentation. Then we compared different layout algorithms and
  approaches, and finally we created a new \mxp plug-in to apply this knowledge
  to the content of \mxp presentations. In this manner, we attenuated the user
  migration process from \ppt to \mxp and tempered the pitfalls of content
  layout in presentations.