Merken

The Challenges and Benefits of Automating NLM-to-ePub3 File Conversion

Zitierlink des Filmsegments
Embed Code

Automatisierte Medienanalyse

Beta
Erkannte Entitäten
Sprachtranskript
good morning everyone I keep thinking now with the microphones that if there are any unruly questions I can cut you off like a cable news show or something like that then just mentioned I'm from CFA Institute if you haven't heard reversed CFA Institute is a global association of investment professionals I always felt the the so the we we publish a wide variety of
research and educational public and here this morning to talk about my recent work with the problems Over the past year I've been working on setting up an XSLT
workflow to create the problems from our Enel and the 1st I'd like to talk about why we set up are poem workflow enhanced and also take a look at the workflow itself then I will talk about some of the problems I came across along the way and finally we'll look at some of the benefits of using XSLT to create the post if I get to form I'd like to take a quick look at the format the covers is an open
standard electronic book format it's reflowable and it's widely accepted by many distribution channels all conclusion books and look and indirectly candle the input file itself is essentially a zip package of content image metadata and navigation files the content exist as X HTML In this formatting using CSS even the navigation house for the problem is simply a html found that links to sections and shows the hierarchy the now that we've take a quick look at the pub I could talk about why we created this works last year CFA Institute released in the course of study in the interest of making this new study material flexible In his widely available as possible we we could be limited by a prince 1st workflow and to print layout based evil if was the obvious candidate to replace the PDF you will it's an open standard it's accepted by many distribution channels and it can contain rich media because the content is marked up in xhtml it's possible to create all the content files with exercising for versions there are a few reasons we decided to go with the pubs 3 equal to inventory has several advantages including the ability to include audio videos and interactive April 3 has also been endorsed by many trees and standards organizations the most recent text book series is a dynamic approachable cost study and it's going means to support that the the study material includes video and audio at some point in the future will include interactivity it doesn't support these elements and given that the countries at least theoretically backwards compatible fallbacks are included the country was the obvious choice now for a quick look at the workflow itself because we use the element 3 . 0 book DTD for content and maximal for our equations much of the infrastructure for the success of the workflow was already in place setting up the workflow mainly a a matter of setting up an XSL transform for the existing data this workflow is based almost entirely on XSLT metadata information is pulled from the XML divine XML files are split into chapters and transformed into x t x h t of here is a step by step of the workflow itself we start with a compiled volume file that contains all of the data for the book 1st you run a few XSL transforms that create the metadata files and so that the volume in 2 separate chapters 2nd we use another transform to convert those chapters into HTML 3rd re-export all the images to change using an Adobe Illustrator script then the files are package manually into music file and finally the EPA was validated using the ID validated tool and then this is through our PC group this overall this process is pretty quickly usually only takes about 15 minutes to create a new world from XML and a
lot of the time is just waiting for all of those images text for this workflow now works
reliably well but there were a few challenges along the way it turns out that writing the actual XSLT the transform was the easy part the more difficult problems that arose had to do with semantics and EPO greater fragmentation which causes display inconsistencies across the reading systems figure out the correct semantic tagging was a little confusing sometimes and display inconsistencies cost and continue to cause problems the introduction
of improved tagging in HTML 5 makes it possible to create semantically rich people files div tags that were used for sectioning can be replaced by such sometimes there is an important type actually that can be used to find section types and there a new tag specifically for figures and their captions creating semantically rich the books makes it easier for users to discover our content and can also allow reading systems to create better visual displays 1 example of that is footnotes if a footnote is tagged correctly iBooks can display it as a part of I didn't some confusion while implementing a few these and tags and attributes 1 example is the figure 10 the HTML 5 suspects states that the figure taking the
unit of content that is self-contained In this typically referenced as a single unit from the mean of the document that definition itself isn't too complicated but it wasn't immediately clear what else could take advantage of that In content blocks text fits that description this self contained and they don't necessarily need to be read in line most of our tables do as well but not all of them and there was some confusion regarding which would work in which would mean in the UK ultimately decided that if a table is labeled and called out in the text it could go on of you take it otherwise it just gets left behind another issue is 110 a bold and italic formatting the B and I tags and rugmaking H T M L 5 with semantic meaning he's now exist in addition to that mn strong tags having these for text presented a couple of questions 1st figuring out which tag is appropriate can be difficult and
subjective different authors and editors who have different ideas as to what's call that is important and what should only be different styles 2nd annual LaMarca only has bold and italic pain so it's not possible to transform to 2 different tags without more information in in the XML so we wind up just using B and I throughout another the issue is the use of type actually this attribute is used to identify the type of content in a section or other elements within these identified chapters footnotes and other content types the idea here maintains an extensive vocabulary of the problem types there are a few things that we couldn't find for content are textbooks have learning outcomes statements had used sections and problems and solutions I couldn't find appropriate texts for these items in the vocabulary so sideline of using the more generic Introduction and that matters section those tender a little confusing but there
are a few things that I couldn't find specific paying for some of our content is optional it should help the reader better understand the material but isn't directly testable in print these are marked by an icon in the margins since there aren't really margins any probe by simply place the icon minus style text and that's what we have examples of on the slide they should be hard to miss in the hominid reader can decide if they wanna leave that section not I also wasn't sure what to do with labels for equations in the XML these are label tag inside the display formula tag In operant template they just appear on the same line as the equation of a line on the edge of the page and is strictly layout would work very well and on the what would happen on a small screen or if there was a rendering with the label look like it was part of the equation itself or something like that instead of risking this sort of confusion I simply moved the equation label to appear before the equation and clearly spelled out that it was in equation this may not be absolutely necessary to have semantic waiting for some of these things and I kind of feel obsessive sometimes we even looking at you know how common optional content is but you know maybe a reading system can make use of this maybe it could hide optional content or give the user the option to high-dimensional confident if they're in areas of the library after the
semantics are figured out 1 of the biggest challenges to creating problems is getting the book to appear consistently and across reading systems CSS rendering differs across training systems and there's a very low level of support from the of suspect somebody problems that subsumes somebody pub platforms handled market well some dumb Our goal is to have only 1 epoch of file per book we don't have the time or desire to create and manage and multiple different problems for a single book it's lot to keep track of we work on multiple different curricula and that means I can barely keep track of what year it is so I left and simple this
shouldn't be a problem because technically heap of 3 is backwards compatible with the pub to it 3 features can be wrapped in in the province which element which allows for all x to be included for example it's possible to include a maximal equation inside which element if the rating system can render the mathematical it should display the image instead as a fallback some of the reading systems no don't handle this which correctly and it will display a mathematical as an unformatted string of characters with an image of the equation immediately following this example on the slide is a fairly simple equation but some of them went up looking this makes it difficult to create a universally backwards and have a file and becomes a judgment call as to what's more important supporting as many people of platforms as possible or including the features In our case if the public is being distributed through our website we lean towards making it With the toward supporting as many platforms as possible if it's being distributed through a single proprietary platform we can use the new features like mad the figure 10 is another new tearing
that can sometimes cause issues most reading systems can display but if you have issues with as you can see in the slide digital editions is and rendering this figure correctly the image is below the node and the node is duplicated again the question is what's more important new features or supporting as many patterns as possible I tend to think that if users see a major issue like this their thought process will be that CFA Institute the books don't work very well rather than maybe my reading system doesn't support speaking
of major issues tables and boxes text can also be problematic tables it box text are even more money inference are boxed text is set off from the rest of the text with the shaded background in heading this visual display can be recreated in permanent works fairly well across some readers this image here is an example of box text in 1 of our recent the pubs the problem here is that this doesn't always fall back well especially on heating devices the box text is readable but it can be difficult to tell where the box ends I'm still struggling with this 1 then make it fall back as well as possible as the table like don't attempts
to make them always fit on the screen we have a lot of large tingling could easily wind up with a table that's so narrow that only 1 character would fit in each column instead the tables are wrapped in a div element and through the assessed the set to scroll an overflow automatically this allows the tables to be readable on-screen even when the inside of the box also these elements do cause
problems fortunately it is not too difficult to get a lot of the book formatting CSS to work well across platforms here are a few examples of our chapter openers they look pretty much the same in each example I could actually show you a lot more examples of our CSS that works well but it would probably start to feel like I was making your offered through my vacation pictures of mind that generally have tried to keep our CSS simple which helps the probes display more reliable across platforms for example I let the reading system decide how to invent lists initially tried to customize the list in then but the slight visual change wasn't worth the time it would take the set up on the added complexity in the CSS table sizing and line breaks inside tables are also up to the reading system immediately this can produce some results that are not ideal if your stakeholders are used to seeing them perfect line-breaks that you can get in print then this can be hard to accept however I'm not really sure that that level of perfection is possible In a reflowable document was an automated process at the very least I'm pretty sure trying to make it definitely for the quadratic decreasing after all of
this wrangling with technology and tagging 1 problem I didn't really anticipate with problems is user acceptance of a new format via handed myself I sometimes forget that most people just want technology to work and don't enjoy trying to make it work we recently produced in the media rich people and distributed through our website this was an epoch only publications there was no print version we very quickly starting getting the West the PDF questions the 1st issue people had was actually with the file size the embedded videos put on combat over a hundred men this is more of an issue with rich media in general rather than in the pub specific issue but we are currently trying to figure out how to get the file size down maybe a lower resolution on the videos would help maybe
straining would work but both of those have their drawbacks the 2nd issue that people have is that there used to the ease of using a PDF everyone knows what it is you just double click on it and it opened the networks the public doesn't really work this way for most operating systems hopefully something in the future but not yet we've tried to avoid try to avoid endorsing the specific heat of readers in the past but choose your favorite recovery here doesn't really work for most people are familiar with it it's easy enough to load any kind of into iBooks but it's a lot more challenging to find a capable reader for the Android we're still researching this see if we can figure out if we can find a reader or 2 that would work well for all our users and I really would like to make it as easy as
possible 4 are used to load and exercise books I think a reflowable in book for our content is generally a better reading experience then a PDF especially across different devices and especially if that PDF was initially intended for so enough about problems let's move on to the benefits of using the XSLT workflow to create pubs the 1st benefit is consistency the epochs created using this workflow don't need any manual changes and they displayed consistently across our publications this also helps with our functionality testing because the output is consistent and reliable we can do the functionality testing upfront not on a per book basis the 2nd benefit is reliability after quite a bit of testing this workflow has proven to work reliably and it produces value pubs every time In the past we've tried to create the post from our print layout program but we often run into problems with the stability of the program it would crash for seemingly inexplicable reasons on export we haven't encountered any mystery areas like that in our actions our flow the 3rd benefit is speed it's taking a value XML file and turning it into many by using this workflow is pretty fast it only takes around 15 minutes to create a complete valid file and a lot of that time is waiting for images to export and me trying to remember what all those in the folder and winning their manually for the flexibility because this workflow was developed in house and is maintained by CFA
Institute employees any changes or improvements can be made and tested quickly this flexibility allows us to make improvements in our pubs and experiment with new features
easily there are actually a few things I'm working on it to improve clubs 1 thing I'd like to do is to include high resolution images forest using images the normal 72 dpi images can look pixelated on screens with higher resolutions or higher pixel densities I think including STG would be ideal but in my testing run into some occasional text rendering issues about mostly on digital editions based readers it appears that this can be solved by converting the text in the STG to outline but that removes the life text from this G the other option I've experimented with is to simply include a high resolution image and then restrict the image to the width of the screen using the CSS this also works really well and it looks better on the higher resolution screen but it also has some drawbacks on a small screen an image that's fit to the width of the screen could be too small to read some older EPA readers don't downsample the image very well and they can look kind of choppy and I think the worst issue is it can be difficult to science in image equation match the size of the surrounding text so if you have to use in imagery equations instead of methanol in different resolutions can cause problems when they're in line has set up are transformed to easily switch between P. ingenious cell whenever I figure out which 1 of these works best I can make that change is have also attempted to include some JavaScript widgets in our problems 1 of our recent publications has the potential to include some interactive exercises for example there are some tables in the book that the user was meant to fill in and then calculate a store based on those answers I think it would be cool if they could be done within the book itself and to have the school or pop up when it was still there getting the JavaScript for that to work was fairly easy but the default caged mode in some deeper readers breaks the javascript functionality what essentially happens is that the 1st part of the exercise appears and works by the exercise breaks if it
continues to another page so when you flip to the next page in the book that the exercise quits working I wasn't able to overcome that before publishing deadline what I hope to keep working on it on the next section and I know you are starting to think twice he talking about a bunch of stuff doesn't work but you know hopefully someday I can make it work and there's actually 1 more thing that I'm working on it has been fairly successful and that is including some responsive design principles in r equals most of the content in our books works well in a linear form but some of our country groups like to include pull quotes inside bars and they would like them to appear like that in the book I was able to include pull quotes in a recent book we did and they work fairly well the PO box is set to appear in line by default for smaller screens but CSS media children Media query checks the screen size and calls the box off to the side like if the the screen is large enough this gives us pull quotes on larger screens but they're still readable most so In conclusion despite the challenges of greater fragmentation occasional semantic tagging confusion and other issues setting up this problem workflow has been worth the investment for in the course of study we need in the book format that is small and can contain rich media and interactive the eve of absurdity format provides pretty much exactly what we need and this workflow allows us to create the profiles quickly initiation thank you increased the 1 book is independent
consultant so thanks for the update on it looks like a really catalysts work on my question has to do do with the people of the Republic development community and in particular on their awareness of what organization that you're still doing out on the boundaries because it seems to me like you have tremendously valuable on inside and test data for them and things that they need to be paying attention to in my question is only doing so if not what can be done to make use of a I would have to say that they are probably not aware a lot of the work that I have done has been through you know research and right now at the pub 3 best best practices book has been very helpful but I haven't really been in communication with yeah well I mean I would I would suggest to me and I know it's not your day job and not necessarily something that you have cycles to do about part of the island the real potential here is that in in a system like the 1 that Europe will stand up and this goes for anybody who's doing a jad XSLT based workflow you're able to isolate the issues in ways that that many other experimental pressure not because we simply you're not do generating a lot of invalid content and the can effects of all this is valid right so you're able to appear to vary but is particularly exposed where's the assistance working where the job scripts in working with long page model all this very very specific nitty-gritty kinds of things which number 1 article today to differentiate between the products out there because you know which words reduce work which would stop and number 2 are able to give them actually tested on and the purpose of really gonna go forward unless there's feedback loops can be completed in the and because otherwise you can be a lot of noise and mayhem and confusion so I'm wondering what there's no as publishing community so we can do to help that process forward that's great idea we have a good idea which we do that but yeah but honestly I don't really know what's happened my head and how we would go about doing that I suppose I could complete into more people than just my co-workers about all the things that I mean I don't know if the variance to testing system doing this but but maybe you can find somebody in the input space who it was who would who would like to thank and say look we have this great of and works on in this this and this always look and call now you know maybe it makes people uncomfortable and give an incentive to fix that sounds like a good idea and no 1 of the issues has always been that we don't make covariance is 40 . 3 because no 1 uses it but we don't keep of 3 because there are no resources for a while you know now we have of 3 cell maybe we can get a it will display good idea thank you and 1 of I have a question on do you have any information they could give us on the demographics of your users and have you done any research about which years they prefer if any I would really love to have that information for the things that are distributed through our website all we have is download can actually it's page view counts so it's not even that count so we don't have that information I suppose that we do have some things that are distributed through apple but I don't know if we have access to enable the access any information like that but I would love to have that information we did a more global member survey that said something like half of our users have I had the I kind of think that they would be using that for a book and it is 1 of the more capable ones so I'm hoping they're using that but yeah I would love to have more information like that because you know I don't really know how many people actually use digital emissions but it's kind of a quick easy thing for PC group to use so we do use that for checking thank you do so it looks through iTunes there's 1 publication minded it does go through right some of our stuff cannot what how to sell the other stuff but well we're actually a member organization so most of our content is free it's either but is either free through our website or its the course you sign up for to secure though even tho it is for your members of the textual courses are secure you have to be registered to our platform the with the or they just over 2 that there is I don't put any DRM but are distributed who I believe absent in the format of the the incredibly 3 saying from 10 from the work we've been doing with I GPs liaison liking said the report I consider reports that's interoperability I also analytics measuring his got which reader and how many people read books and this kind of issue in a way that respects the privacy this is the issues that are being workflow I have a specific question for you which is about how you build a final the about so you from XSLT you've created your xhtml files in a folder I've been experimenting with having XSLT write a shell scripts that run which covers all the images in place runs in a conversion necessary and runs it with all the correct options to put things in the right order but I wondering if what I'll be doing that sounds cool I think that so yes I've always wanted to automate the actual packaging of this file but it's kind of I guess the lower priority because I can you know just kind of cobble everything together and I think when more people than just me start working on these the books it would be mn would be very helpful to have something like that but that was just assisting automatically and even even I forget things and I throw at the validator and it's like you don't have any anything assessment of you know that's probably kind of important thanks yeah and because my grandmother's from multiple systems and it appears that 1 of the our limitations or the obstacles user acceptance of European but uh files is the size the it seems to me that if you had created 11 people find the rather than the entire book level your size would have been significantly lower and it would even borrowed the way pdfs so usually provided that its optimal level and optionally 1 can have the book wide pdf as well that's true that's a good thought yeah so this 1 work and thinking of there we did the they splayed out somehow so like many videos are in chapter 3 and they're kind of so here and there throughout the rest of the book but that would that would be make it easier for people to download the file it's good our hobbies USA has less security the other things so since most of our content is free we we don't have any DRM minority books for a textbooks our distributed there handled badly rapid and there container I think in the it is secure following up on that this the code available this I would say you know you but it's not now I suppose we could discuss that yeah we would have to give our you know division by spread is division directors and that sort of thing involved but here we could discuss that anyone actually Blair each publications I was wondering if you're CSS in customized my hand on sort of yes I have 1 CSS file that handles you know and I know what the percentage anymore maybe like 80 or 90 % of the book for mining and then different publications have their own separate CSS file handles brand colors and different things like that so a lot of it is common but there are some differences in each publication yes linear wind up with 3 was 1 of the other things being worked on is links between the books if you go further down the path of a separate the book 3 chapter you then have the problem that a lot of books might have references between chapters and you can't have links between the books right just well In many intractable Wessel and I is working on has worked on respect for doing that but I don't think anyone implements it yet region has started until the what happens the right now include linking between chapters of problems that sounds familiar where called the I remember reading about it and thinking that sounds really complicated and then I did you quite units I think next chapter but yeah and they have this this really complicated way of referring to other chapters within a publication the the involving going to the get going to the index file and stepping down through XML elements but I think the linking between books as something separate OK 1 of the I haven't followed the details about so I don't know if you noticed I just wanted to mention it because the ones that have separate chapters all of this is the problem with unfortunately in ways pointing that out of you know about it when I have separate set yesterday I might say a prayer from pedophile ring you through your quality control hold you draw the line on look these have the devices we looked at some of these have the capabilities look 1 the speed or not kind with some insight into this this discussion yeah we tend to right now and we tend to favor against the newer devices but we pretty much don't even look at like the older candles and stuff there are other devices that I think may be a more common like that and that we're trying to support more versions of those so it's kind of wait when we get into the spot were like this just actually isn't gonna work on this all e-book reader do we really need to make it work I like I don't know if anyone uses a generation 1 candle to review anything this complicated anymore so it kind of happens on I guess per-device bases that we run into problems when Michael I was wondering if you could talk a little bit about your quality assurance process if I QC manager is watching you know he he made me like that's not what we do at all but from my understanding of of and they have checklists that they go by come and take you know specific things that tend to break or things that I might forget and they view it in they view in like if it's a proprietary distribution system the viewing that reader if it's something that's going through our website they will view it I think in digital editions this is the and in the and I want I was intrigued to hear that term you know it's only about 15 minutes to basically created people thought you know 5 out of the XML and tell our current then that takes us far more than that so I was just interested in having a little more color and insulation better then so when we start actually the volume has generally been through a round of print QC already so the creation of the epoch then pretty quick from that existing XML and then to see us to look at it again in the pub version thank you but the accident was the worst mutual has not been the UAW and suppose that we make the text million
Assoziativgesetz
Reverse Engineering
Güte der Anpassung
Vorlesung/Konferenz
Umsetzung <Informatik>
Dean-Zahl
Elektronische Publikation
Dateiformat
Vorlesung/Konferenz
Umsetzung <Informatik>
Computeranimation
Überlagerung <Mathematik>
Distributionstheorie
Offene Menge
Punkt
Prozess <Physik>
Extrempunkt
Gruppenkeim
Versionsverwaltung
Gleichungssystem
Element <Mathematik>
Computeranimation
Videokonferenz
Netzwerktopologie
Spezialrechner
Metadaten
Standardabweichung
Adobe Illustrator
Skript <Programm>
Vorlesung/Konferenz
Auswahlaxiom
Prinzip der gleichmäßigen Beschränktheit
Dean-Zahl
Reihe
Dichte <Stochastik>
Ein-Ausgabe
Dateiformat
Arithmetisches Mittel
Adobe Illustrator
Dateiformat
Garbentheorie
Information
Standardabweichung
Metadaten
Selbst organisierendes System
Vektorraum
DTD
Hierarchische Struktur
Interaktives Fernsehen
Transformation <Mathematik>
Audiodatei
Spezifisches Volumen
Elektronisches Buch
Hypermedia
Skript <Programm>
Inhalt <Mathematik>
Spezifisches Volumen
Inklusion <Mathematik>
Bildgebendes Verfahren
Beobachtungsstudie
Diskretes System
DTD
Telekommunikation
Elektronische Publikation
Binder <Informatik>
Umsetzung <Informatik>
Hochdruck
Hypermedia
PRINCE2
Beobachtungsstudie
Formale Semantik
Datensichtgerät
Content <Internet>
Physikalisches System
Umsetzung <Informatik>
Elektronische Publikation
Computeranimation
Formale Semantik
Mereologie
Datentyp
Attributierte Grammatik
Garbentheorie
Inhalt <Mathematik>
Figurierte Zahl
Widerspruchsfreiheit
Attributierte Grammatik
Lesen <Datenverarbeitung>
Aggregatzustand
Autorisierung
Addition
Formale Semantik
Befehl <Informatik>
Subtraktion
Content <Internet>
p-Block
Element <Mathematik>
Umsetzung <Informatik>
Computeranimation
Formale Semantik
Arithmetisches Mittel
Generizität
Texteditor
Deskriptive Statistik
Einheit <Mathematik>
Datentyp
Attributierte Grammatik
Vorlesung/Konferenz
Garbentheorie
Inhalt <Mathematik>
Gerade
Attributierte Grammatik
Tabelle <Informatik>
Randverteilung
Formale Semantik
Wellenpaket
Datensichtgerät
Hochdruck
Content <Internet>
Gleichungssystem
Systemplattform
Computeranimation
Formale Semantik
Homepage
Ausdruck <Logik>
Übergang
Weg <Topologie>
Total <Mathematik>
Programmbibliothek
Inhalt <Mathematik>
Gleichungssystem
Gerade
Inklusion <Mathematik>
Prinzip der gleichmäßigen Beschränktheit
Nichtlinearer Operator
Template
Systemplattform
Einfache Genauigkeit
Physikalisches System
Umsetzung <Informatik>
Elektronische Publikation
Bildschirmsymbol
Hochdruck
Quick-Sort
Konfiguration <Informatik>
Rechenschieber
Flächeninhalt
TVD-Verfahren
Mereologie
Garbentheorie
Lesen <Datenverarbeitung>
Prinzip der gleichmäßigen Beschränktheit
Web Site
Prozess <Physik>
Systemaufruf
Sprachsynthese
Gleichungssystem
Element <Mathematik>
Physikalisches System
Umsetzung <Informatik>
Elektronische Publikation
Bitrate
Rendering
Systemplattform
Mechanismus-Design-Theorie
Computeranimation
Neue Medien
Rechenschieber
Knotenmenge
Digitalsignal
Speicherverwaltung
Strom <Mathematik>
Figurierte Zahl
Bildgebendes Verfahren
Lesen <Datenverarbeitung>
Zeichenkette
Tabelle <Informatik>
Prinzip der gleichmäßigen Beschränktheit
Prozess <Physik>
Quader
Inferenz <Künstliche Intelligenz>
Datensichtgerät
Element <Mathematik>
Umsetzung <Informatik>
Division
Computeranimation
Inverser Limes
Menge
Pufferüberlauf
Ruhmasse
Bildgebendes Verfahren
Modallogik
Tabelle <Informatik>
Touchscreen
Resultante
Web Site
Prozess <Physik>
Mathematisierung
Hochdruck
Versionsverwaltung
Systemplattform
Komplex <Algebra>
Computeranimation
Übergang
Videokonferenz
Physikalisches System
Perfekte Gruppe
Visualisierung
Kontrollstruktur
Gerade
Cross-site scripting
Bildauflösung
Prinzip der gleichmäßigen Beschränktheit
Umwandlungsenthalpie
Dichte <Stochastik>
Mailing-Liste
Dichte <Stochastik>
Physikalisches System
Umsetzung <Informatik>
Elektronische Publikation
Teilmenge
Offene Menge
Hypermedia
Dateiformat
Lesen <Datenverarbeitung>
Tabelle <Informatik>
Bit
Stabilitätstheorie <Logik>
Gruppenoperation
Mathematisierung
Hochdruck
Computeranimation
Physikalisches System
Vorlesung/Konferenz
Inhalt <Mathematik>
Optimierung
Widerspruchsfreiheit
Bildgebendes Verfahren
Funktion <Mathematik>
Softwaretest
Prinzip der gleichmäßigen Beschränktheit
Lineares Funktional
Dichte <Stochastik>
Datennetz
Benutzerfreundlichkeit
Dichte <Stochastik>
Physikalisches System
Humanoider Roboter
Umsetzung <Informatik>
Elektronische Publikation
Datenfluss
Widerspruchsfreiheit
Flächeninhalt
Basisvektor
Wiederherstellung <Informatik>
Lesen <Datenverarbeitung>
Subtraktion
Mathematisierung
Zellularer Automat
Interaktives Fernsehen
Gleichungssystem
Computeranimation
Interaktives Fernsehen
Widget
Kontrollstruktur
Vorlesung/Konferenz
Speicher <Informatik>
Inklusion <Mathematik>
Default
Gerade
Bildgebendes Verfahren
Bildauflösung
Touchscreen
Prinzip der gleichmäßigen Beschränktheit
Lineares Funktional
Videospiel
ATM
Pixel
Wald <Graphentheorie>
Cliquenweite
Bildauflösung
Umsetzung <Informatik>
Endogene Variable
Dichte <Physik>
Konfiguration <Informatik>
Neue Medien
Mereologie
Fitnessfunktion
Tabelle <Informatik>
Beobachtungsstudie
Dean-Zahl
Quader
Gruppenkeim
Abfrage
Profil <Aerodynamik>
Umsetzung <Informatik>
Computeranimation
Homepage
Formale Semantik
Bildschirmmaske
Hypermedia
Endogene Variable
Dateiformat
Vorlesung/Konferenz
Garbentheorie
Inhalt <Mathematik>
Inklusion <Mathematik>
Default
Gerade
Touchscreen
Kovarianzfunktion
Bit
Vektorpotenzial
Prozess <Physik>
Nabel <Mathematik>
Hochdruck
Versionsverwaltung
Gruppenkeim
Element <Mathematik>
Sondierung
Zählen
Verteilte Programmierung
Raum-Zeit
Homepage
Eins
Videokonferenz
Übergang
Einheit <Mathematik>
Datenmanagement
Stützpunkt <Mathematik>
Notepad-Computer
Skript <Programm>
Vorlesung/Konferenz
Gerade
Umwandlungsenthalpie
Prinzip der gleichmäßigen Beschränktheit
Softwaretest
Sichtenkonzept
Computersicherheit
Güte der Anpassung
Dichte <Stochastik>
Ein-Ausgabe
Biprodukt
Kontextbezogenes System
Checkliste
Konfiguration <Informatik>
Neue Medien
Teilmenge
Randwert
Generator <Informatik>
Druckverlauf
Automatische Indexierung
Rechter Winkel
Digitalisierer
Dateiformat
Information
Ordnung <Mathematik>
Subtraktion
Web Site
Selbst organisierendes System
Digital Rights Management
E-Book-Reader
Zahlenbereich
Geräusch
EDV-Beratung
Zellularer Automat
Unrundheit
Term
Systemplattform
Code
Division
Data Mining
Loop
Informationsmodellierung
Multiplikation
Unterring
Inverser Limes
Spezifisches Volumen
Inhalt <Mathematik>
Softwareentwickler
Ganze Funktion
Varianz
Bildgebendes Verfahren
Schreib-Lese-Kopf
Soundverarbeitung
Trennungsaxiom
Datenmissbrauch
Validität
Physikalisches System
Umsetzung <Informatik>
Binder <Informatik>
Elektronische Publikation
Quick-Sort
Modallogik
Dreiecksfreier Graph
Mereologie
Gamecontroller
Wort <Informatik>
Kantenfärbung
Verkehrsinformation

Metadaten

Formale Metadaten

Titel The Challenges and Benefits of Automating NLM-to-ePub3 File Conversion
Serientitel JATS-Con 2013
Teil 03
Anzahl der Teile 16
Autor Dean, Mike
Lizenz CC-Namensnennung 3.0 Unported:
Sie dürfen das Werk bzw. den Inhalt zu jedem legalen Zweck nutzen, verändern und in unveränderter oder veränderter Form vervielfältigen, verbreiten und öffentlich zugänglich machen, sofern Sie den Namen des Autors/Rechteinhabers in der von ihm festgelegten Weise nennen.
DOI 10.5446/21804
Herausgeber River Valley TV
Erscheinungsjahr 2016
Sprache Englisch
Produktionsort Washington, D.C.

Inhaltliche Metadaten

Fachgebiet Informatik
Abstract While converting NLM book tag XML to an ePub seems like a relatively straightforward process (hey, an ePub is mostly just HTML, right?), setting up a workflow to do just that is quite challenging. It turns out writing the XSLT could be considered the "easy" part. Other problems, such as dealing with ePub display issues across ebook readers (anything from minor CSS differences to major MathML display problems), deciding what tagging makes the most sense semantically, and figuring out how to give semantic meaning to visual formatting such as table cell shading add a layer of complexity to the process. This paper discusses the challenges, rewards, and as-yet unresolved problems encountered in the process of creating an NLM to ePub3 workflow.

Zugehöriges Material

Ähnliche Filme

Loading...