Merken

Tracking Changes to JATS XML in an Online Proofing System

Zitierlink des Filmsegments
Embed Code

Automatisierte Medienanalyse

Beta
Erkannte Entitäten
Sprachtranskript
the the Director Technology Strategy at damageable services I I'm here today along with my colleagues to preserve the paper that was written and submitted to Jax and regarding a web-based axonal editing environment that we developed over the last several years and and the key aspect of the paper is as was that indicated is focused around change tracking and so I think she rightfully refer to it as part the Holy Grail and for
us it's been something we've been trying to attain for for quite some time and have put quite a bit of effort money and time in developing the system and so between the 3 of us presenting the giver the back on our company in a bit of background on the
project to a kind of frame the discussion and then hopefully by the time we get the presentation of will go to the technical approach in Charleston and it'll take you through some the technical aspects of the change tracking and so as you can see on the slides and so that the General Services if you don't already know who we are where basic content and services company serving STM publishing market and damage all services is located in Waterbury Vermont were about 150 employees and it's not a company that was founded in 2002 to really serve the printing services side of our business so were part of a larger company called the Sheraton group you may know it's often industry just the Sheraton and and and so several of the companies in our group operating companies serving journals market the book market and the magazines market but when damage all services was started in 2002 was specifically focused on content services for STM publishers and it was actually the merging of several different divisions in our other companies that all came together and now located in our water of Vermont location only about 150 employees and at at Johnson's we're you know intensively focussed on XML based workflows and but yet we got presentation chocolate and then made its losses this the you variability that of so this
is our error a building in in whatever remarks we have when worse than that which shows up but basically what we
do is we we don't peer-review we don't do online hosting we do everything in between and so editorial axonal 1st workflows almost all over workflows for the publishers we work for XML 1st of course to all kind of composition services and importantly is where for electronic deliverables this is the actual based deliveries to the online host but just a high out of bonds are the only holds the world the cost of medical mid-central except for In addition as I mentioned we also blueprint product a sister companies we also in the last few years have started developing mobile apps as well for our customers but when I
when I mention our customers it's important to do state four-dimensional services that lower part of the shared a group company and you may have heard of some the other companies without technology lab and Hunt Valley Maryland as in Vermont and then 3 other printing sites overall the company focuses on the general segment on all STM publishers including large commercial publishers however services we are really exclusively focus on medium to large-sized associations societies university presses and so we tend to be really focus on customers who need a high degree of support and development and customers something that have the staff and how to do a lot of this development so we've been focusing on growing our technology to increase significantly over the last 5 years or so so beyond just kind of a big company background about because the 5
years ago we started asking ourselves although DGS was in a significant growth period we started asking ourselves where is this going were looking at the side of the business that is right for us and as most of you would know the imprint is is declining not rapidly in our business we've been able to sustain the prince declining we start to look at the front and services and said is the composition services which is
typically how we used to refer to ourselves as a composition services company but but you know I that earlier refer to this is really a content services company we've looked at composition services as being something that will in the future common kind of under fire if you will not unlike print has and it may begin to decline were the willingness of publishers to put money into the composing of articles may be under pressure to reduce their costs and so we started asking ourselves the question of all the online journal the journal record then how come all production costs are are competitors except for a are so focused on the 2 dimensional PDF proves the frankly just print and so we started to say the online journal the journal record the PDF can incorporate all the things that are going to be more common in scientific articles but without significant cost is not to say that P can be more robust more interactive and but it's not a pretty significant cost to do that so things that multimedia data linking Semantic Enrichment supplemental material all those things can be
composed into an article but it's pretty difficult to do in an automated fashion therefore becomes costly and so we were heavily focused on you know although this was in our specific area of expertise in terms of hosting the regional content we really recognize that's the future of scientific articles even if the PDF is the record of publications that I say that because most of our customers in any publishers newest will probably support this you know you have full taxation online the PDF tends to be far more downloaded in the HTML full-text regional rednecks that's changing a little bit and time we think there's technologies that are coming about things like regional 5 CSS 3 and other things that will come after that that will drive the it's not really to the forefront so we after cells back in 2009 2010 are we going to be ready for when that costs and if were just focus on composition services were not going to be ready and and even though axonal 1st is really the focus for a composition process we want to do something different and so what we did was
we look at the current workflow written much knows this PDF goes through a cycle corrections we produce a PDF for crime as the primary product and the online journalist post-secondary deposit the downstream deliveries the delivery of content to the online host metadata crossref content Department except for and so we decided to organize countless small
talented team to focus on this problem of where service is going to be 5 to 10 years from now and what we need to develop in order to respond to the needs of publishers and so among team to others few Charles O'Connor and Anthony non-Protestant but they were key original players in the team low myself only recently added another protein of Fleischer who's in the audience here and has joined us Israeli air quality assurance person doing our use case testing etc. and several the team members have experiences production others which is usually
helpful 1 of which has experience of being an offer as well and so i'm as some of the edges and why we don't take ourselves too seriously but we we have a bit of fun doing this project but it's been a long long process and what were about to release in May
and SSP this year is the new workflow that we refer to as of article expressed it's a combination of a web-based editing environment that really is the main interface for users along with an automated composition so and that's all axonal driven and the focus of this paper in the discussion was really around the track changes component of the Web-based having learned because that proved to be a 1 of the greatest challenge to give you a better sense around this time you
know this is just our focus was you know make that a smell presentation really the central focus for the corrections PDF will still be there was stability PDFs but make the PDF the secondary content that's produced make the action the primary content driver Nacional presentation so authors have the benefit of reviewing in HTML presentation at the correction stages rather than only being able to review PDF presentation and so on over the course
of about 4 4 years of investigating the idea of building the concept and coming to where we are today which is really in the last frame of our development we did a number of things that were really crucial to the success of the project 1 of which was we broke built a prototype early 2011 this is a screenshot of just that original prototype system as we took it on a road show with our
customers some prospects mostly customers we basically went sent out and talk about the concept and were trying to build and that this was something that only we could build something that customers would be interested in you wanna build something that is new technology for technology's sake and so we got agreement from customers that this was the approach for the future in something they strongly felt was something we should proceed with a and then importantly we sat
down with a group of postdocs and graduate students and actually the focus group using the prototype system we walked focus group and basically through a script and said here's a system no training it's going to be in year 0 learning curve so we give the script and said his a publishing workflows you're an corresponding the receipt approved go through the system to make all these changes that are required and let us know whether you know you have feedback at the end to the mall that an hour or so to do their processing the script was rather lengthy but the number 1 feedback that came out of that
interface with the authors was the need to have a collaborative environment from ability for the corresponding wanted to share that proof what their co-authors and In the PDF workflow we typically would not really be aware of how much the PDF is shared corresponding office so that up their colleagues they give feedback they merge into 1 pdf only have the interaction with the pdf and so here collaboration became a critical component and the other component was now you can have multiple users coming to the system how we possibly can track the changes of each independent user and be able to that those changes in the case of a production accept reject or modify the changes that authors have made no who made it and when and so that's really what led us to this the prototype system that we built for attracting just about 9 months to develop so our overall development period but about 3 years but 9 months let alone was
just spent on building the change tracking system and so for a bit more the technical details and I handed over to my colleague Charles O'Connor but take you through what some of those technical challenges were related to track changes and and then unlawful build wrap subtree guys thanks I have chosen primarily as myself and and it's going to go through some of the the the challenges that we face I think I might give a good story about you know where we went through the whole project for a retraction use it was more about cyclical the issue is the process we would go out of blind alley and then Anthony and I would lock cells in a room yell at each other for about 4 to 6 hours and then come up with a new way of of of looking at it and progress in the problem so that the technical challenge to track changes or and
and and this is this is key is that you know would we can have as points in multiple users interacting with this with the system and even the same workflow steps the we have to have the ability to act upon the changes that this is a this is a alive document in a production system so not only do we want the production editor to say of what the author did we also want to give them the ability to act on those changes but and importantly if you think about XML for a while you'll you'll realize that if you allow people to act upon the changes to accept and reject if you don't enforce some way of of of ordering the acceptance
rejection out you can break the the XML and so those with the major problems we have to solve but now are underlying
environment for for articles press is we use so I think it's called like content create from STL now I used to be called the open and it has a lot of good features it's it's online it's with the with its a continuously validating XML environment so it it works well for all of those but there are a couple
limitations and the 1st limitation that we had to overcome there's really no way for all authors on on this 0 learning curve at work trying to achieve to add more complex XML elements so we use a variety of form based pulls up as well as for the references we use at effects from i nearer to to to pass references in and put them into the document but really in an editorial workflow without track changes you have you have nothing authors could could you know change the data they could make new claims they could do all sorts of things and if you don't know what's happening but then then you don't have an editorial workflow and in now
we 1st thought that we would be able to get and through a workflow through sequential editing 1 person does 1 thing the the next person doesn't act and it would go on like that if necessary we compare you know version 1 to version 2 version 1 to version 3 or were any of the versions that with and that would satisfy the the basic idea of tracking the changes however has like pointed out meaning
we went to a focus group we realized that that authors to share the document and so we thought well we should give them to have a way to share the document gives a a bunch of people copies of it and we thought well maybe I could do it in in in parallel each of them gets a copy and then we would have to merge them that this would be it actually incredibly difficult can't we read but really thought that we could get away with this uh and you know we go on to the rest of the workflow but is this has a lot of difficulties that especially since what we were looking at that time was the in x amount difference in approach now there are
limitations 2 different saying that are critical to to the way we want to put this together and the the the major 1 and it seems like a subtle difference but it's very important is that 2 different you get an accurate representations of the difference between 1 version of the document and the other
version of the document but we don't necessarily see what was done and to to illustrate that so I have a very example and if you if you start out with
ICG is to you and your changes to make it as a whole old cleans means to you they that a straight differencing estates free text differencing will give you that and that doesn't that doesn't really make editorial sense to anyone you see it accurate it shows what was taken out of that but it really doesn't help and editors say all I want to accept this changeover were rejected that it opens up a lot of the possibility for air now you can be more sophisticated than that and have a real exon of difference engine water or but if you attempt to overcome the problem you thereby and clear parameter grouping changes etc. you run into other sorts of difficulties 1 of which is you you lose the granularity of of changes and but if you could changes within changes aren't necessarily marked as such so today to show the loss of of granularity if you start out with a low world and you and then you you add the so we unintelligible text within that and then make the entire paragraph I tell it what you'll get from XML difference in is the entire string as a change it's not discrete between this is a formatting change here and the text content change here and the reason that's important is that if we had ability were given the ability to a production editor to accept or reject changes and they reject the I talent then they could inadvertently be rejecting that in addition of content so we started on our our
roots of of solving the problem and of the 1 of the 1st things we realize that we needed we're worried about custom elements to be added to the jets DTD and we use that the the event handlers in like content created to inserted these elements as changes that were made when we realize that we needed a lot more than we we originally thought because edits can be made in many different ways if you just think in terms of the different ways you can delete something you can hit backspace character by character or delete character by character where you can select a bunch of text and hit backspace or delete where you can select a bunch of text and just start typing over it and those are all deletions but they all happened in different ways and for some applications of it is necessary to treat them differently so here it's just that much work a list of the sum of many of our in addition to b jets DTD and some of them we have some of them might strike you as well as all other than such as and both now I wish I could show this to you a little bit better as well as switch over to oxygen to show this to you but this is a an unlabeled node here and what what we show here is is because you know bold and another for members of the cobbles and so you go back and forth but we have to keep a history so we have a history of who made bold words was originally bold who made it unbowled all the different actors and so you could have you know 10 different people having done this it will show up in our track changes which it as 1 change but as you can see just this once a change it requires a lot of xml under the hood so now we
we we also realized that the the idea of of having parallel editing wasn't really going to work and so instead we have what we call out random axis is sequential editing and yes we made that term up and what that means is that we give the corresponding author the file and everybody else that he wants to share with works on the same file and if there is somebody is working on it them and then other people see and read only version but in this way we can put together all the different the changes that they make and now it was the important to realize so it is that we now now have a document that has the changes of many different people image and those all have been kept streets and so for a little bit about how the history alternative so you and I and and the units of In this knowledge in the context of
his finding out from the idea that the tho that the 2 things that we need here are their religion that
protects the structure of the XML so it governs the order of acceptance and rejection and for that rule engine
to work that what we realized that we needed to do was the normalized the XML which means where you have yeah like a nice clean neat XML with the minimal number of tags necessary while we messed up again can you tell if you so the ruling doing
products that changes is like a group of references and the the JavaScript library which controls user acceptance actually how we use that so is build
upon excessive the and delisted late so so this is how our review more no proof explosive received the left side right so is their pricing
reject it to show how the use of and not surprised we don't actually so can go intuitive changes and to do that this is the point where the place where the team just listening and economics of the region similarly you can do model changes by switching back and come back to this in since the beginning of the world except that the data go back a a little more and combat idea was to make a change in sense they did cases that he has made it probably gonna be something called in sick and come back actually and similarly the changes of the group that was the most is the common on the innermost is the less common sense later and this should be taken up and the most the recognition in innermost actually so so is that
the light not something and thinking processes group because we want to make the excellent non-derogable below it they image should not be intensely if someone is meaning edition and in the most they make an album was based on the test that's on the phoneme our following this case is actually so most of it is that decision is made an outermost then a division is made in in the most and that the basic instinct and deletion is always group that delta most and the the sample so in this case the
user comes to review more so that a paragraph and then you can say that additional takes to the power of the needs to make that 1st edition of incident and the apparent because it that background if you think that that's which is inside the which is inside the it's actually so that's a basic common things that most of the rules which is like if you should see was mean being them all but we cannot but if you see there is a real text so users can and integration actually so if there is a based it did not that all the most of the topics that will be the innermost actually and the next 1 which we need
to which is the main thing in the the process of writing this is the number of connections so the normalization economy is alternatingly optimize the use of advanced speaking grouping we have anything of that that's so that the business rule of accidental that works as well as on the axis being used let's work Blake vixenish maybe being like broken out something which so as this is a ligand simple example of and that like the order of the names of the nominees so in the 1st thing that and the time has the concessions they aren't out who later but live in exciting board then the harmonization function place to spilled the border into groups and next thing station as the time for the board so when you come back to that act this year so in this the
the painting this will be grouped under concession
1st then the board so that the vision of inserting them that I have beginning faster than the actually so this is a simple and he had like but
it would be more being Dallas as it is in the last time I
can display day that to be constraints and at the last link failing them all but that you we would like to show you
that no but my computer doesn't get along with the projected so I guess the best thing for
us to do would be to the questions you but just in you don't
have to know that the I will make full comments on the velocity questions I don't think that if would do that was the 1st comment is that I really like the solution is a great solution the not many people know and understand the difference between track changes and comparison as a set of quot subtle differences and you've you've understood that such as the uh and implemented various solution which which does change tracking relevant comparison and uncertainty from the for point instance the 2nd point I think or made is that the example of that comparison showed comes from our tools and at then this slide I just want to make a point that actually we we have understood the issues that you will face it's time we do have solutions for making the distinction between content changes of formatting changes and if that we just about reason you which has a new change represented representation for way can represent overlapping structures which does something very similar to the fragmentation you talking mainly in the normalization to represent these changes in a way that you would be able to say I want to accept performance and changed but I want to get the content change so that we have addressed these issues and we do not solve this problem my point is that they don't merge is really really difficult problems that has anything to them a really good job of to most most doctors together and have something sense of the end that this is not an easy task we are also what can emerge from it to be released next month so I like to chat with you about that and and just get something about some of the problems we face in and see if there's some common ground and falsely I think I just remembered and there is a w 3 c community groups are looking at change tracking all change representation in XML as an XML standard and I think your experience in developing this would be really useful on group so that if you're involved already all refusing integral but but look it up in them and get involved with that global thank you and and I appreciate that I think we talked you guys originally when we 1st started looking and I think that you were starting to work on that the formatting and overlapping at that time but it was just a matter of of development schedules and so we had to move but but there about the future applications this is the want so it is not I don't know if we did not go with energy and has 1 if it was already clear that we we didn't both emerging the merging we recognize as a huge challenge to deal that is I think your data we can when a different way so we're getting essentially the effect merging without doing parallel and I but but we work about optimal reliance on axonal different was the place we started and we spoken so this work as the online content create now they were the ones that actually initially put us in touch with the filtered delta acting on really recommended axonal diffing as the initial solution we should approach and I think was really what is the matter of the couple are learnings in time to find the subtle differences between different and really what we need for track changes in and you know we as Charles describe went down a path after we recognize OK we can do this that's another thing we went down a path that actually ran as into a number of benefits until we can come come around to what a solution would look like that would actually work but will definitely be interesting topic model to culture crossref but it seems that you're at the normalization process was predicated on the recognition all of all some basic concepts like format change content but so my question is uh what other kinds of changes are possible in the document that can be recognized by the normalization process for example structure changes do you can handle those for the former changing content changes that really because the 2 golden rules that you start from what the stability is our our more complex and and we try to treat them as though they are insertions and deletions of this you know it when you when you have enough about a large change like that you do have treat it more as a discreet unit as opposed to treating it as a broken down unit as you see here and so those those changes if you have no in the replacing larger more complex elements if it is all at once so see for example semantics of bulleted list and changes it to a number of places and the reviewer wants to change of back to the forum list or something like that that's structure change tags that change right in the back case we would treat that as a deletion of all bolted listen agents and the insertion of of the entire numbered lists that somebody could act by markets relevant to the checks questions on a test set of management our reviewers a part of this cycle because the substantive content changes well it's just this way if publishers expressed concern that we reach a level of substantive change possibility that might render the article not quite completely peer-reviewed short reviewers part of this is this is this is at the proving stage and services Beyond peer review at at this point we do talk to to customers and say 1 sort of oddly the disadvantage of of the system is that it is easy to support offered to make changes about Our copyeditors there's as well are are quite adept at spotting what is the substantive change instance like the system as a whole and actually so Holman correction has been made so it's crosses the level below thing invented was so that the said like 3 304 paragraphs it is not in the group it's important and that's about my but it's important that we got about initially every single class from the rotor said that same question if you made it so easy problems to make these changes how we manage that they don't make substantial changes content and the reality is all the editing directly in the system and or having the yes everything still going to be reviewed by the reduction vectors job basically identify whether the changes are stylistic with their stances and if they are all were concerned the rates that reflect back culture so there is that the loop gets close by leveraging the reviewing all of the changes and hence that the great importance of the tracking ability itself the time this has to be like I said it wasn't any fuel publishers have been expressed an interest that using this in the peer review process so in obtaining accurate and it yeah during the course of the road show we've got a number of questions from publishers about you know where else we could use this in although Oracle Express combines both the web editing environments was automatic composition system it's the web editing environment that people are most kind excited about the track changes as part of it and we are actually considering not right at the moment but for the future and licensing of the web editing environment and so we are kind of initially framing how will handle that and so were not ready to go remarkable license private but we do expect that there will be interest from book publishers use in book publishing CME content referred from pharmaceutical companies as well as period companies as well you might have an interest in using the system you know using point double wide accepted to manage comments in and edit the content in the purity streams at the great opportunity for to to move upstream literally a moral licensing by Mike Freeman from American neurological society you have to use the focus group to help guide you along this this hat and and and it looks great I'm just wondering if you following up with a focus group in terms of usability yet or if you plan to do that and so you know what the that you've already gotten where some things that they may have identified to improve their it's a good grade point is that as like stage we use a focused very early on and this is now a a couple years ago to get kind of initial feedback and while we show the system and its prototype stage to a number of customers and frankly are best feedback came from the end users directly and so we have talked about doing a 2nd follow a larger focus group when the system is complete 1st step is where this development in about a month and a half ready go through what we call you you experience user fees user experience period actually use our own staff and to kind test the system out make sure the balance of the right points prompt the user through the system because it doesn't need to be frankly 0 or near-zero learning curve and we have talked about going back and doing a larger focus group to really that whether the system is meeting expectations users before we go on of our 1st phase customer which will be in the 3rd quarter of this year the
Metropolitan area network
Objektverfolgung
Geschlossenes System
Bit
Dienst <Informatik>
Geschlossenes System
Web-Applikation
Mereologie
Mathematisierung
Vorlesung/Konferenz
Ideal <Mathematik>
Computeranimation
Einfügungsdämpfung
Rahmenproblem
Wasserdampftafel
Mathematisierung
Hochdruck
Gruppenkeim
Dienst <Informatik>
Kombinatorische Gruppentheorie
Division
Computeranimation
Rechenschieber
Metropolitan area network
Dienst <Informatik>
Mereologie
Vorlesung/Konferenz
Projektive Ebene
Inhalt <Mathematik>
URL
Fehlermeldung
Assoziativgesetz
Addition
App <Programm>
Web Site
Stab
Hochdruck
Gruppenkeim
Applet
Ausnahmebehandlung
Dienst <Informatik>
Biprodukt
Computeranimation
Rastertunnelmikroskop
Dienst <Informatik>
Minimalgrad
Gruppenkeim
Mereologie
Rastertunnelmikroskop
Softwareentwickler
Dimension 4
Grundraum
Data Mining
Aggregatzustand
Multimedia
Dichte <Stochastik>
Hochdruck
Ausnahmebehandlung
Biprodukt
Online-Medien
Frequenz
Computeranimation
Formale Semantik
Dienst <Informatik>
Datensatz
Druckverlauf
Vorlesung/Konferenz
Inhalt <Mathematik>
PRINCE2
Subtraktion
Bit
Prozess <Physik>
Element <Mathematik>
Multimedia
Zellularer Automat
Term
Computeranimation
Metadaten
Hypermedia
Datensatz
Inhalt <Mathematik>
Umwandlungsenthalpie
Dichte <Stochastik>
Browser
Dichte <Stochastik>
Ausnahmebehandlung
Online-Medien
Biprodukt
Fokalpunkt
Dialekt
Kombinatorische Gruppentheorie
Dienst <Informatik>
Flächeninhalt
Dreiecksfreier Graph
Softwaretest
Bit
Dienst <Informatik>
Prozess <Physik>
Anwendungsspezifischer Prozessor
Vorlesung/Konferenz
Projektive Ebene
Biprodukt
Ordnung <Mathematik>
Computeranimation
Autorisierung
Geschlossenes System
Stabilitätstheorie <Logik>
Landau-Theorie
Gruppenoperation
Web-Applikation
Schaltnetz
Mathematisierung
Dreiecksfreier Graph
Dichte <Stochastik>
Kombinatorische Gruppentheorie
Fokalpunkt
Computeranimation
Druckertreiber
Geschlossenes System
Komponente <Software>
Inhalt <Mathematik>
Schnittstelle
Geschlossenes System
Rahmenproblem
Landau-Theorie
Gebäude <Mathematik>
Zahlenbereich
Gebäude <Mathematik>
Projektive Ebene
Softwareentwickler
Computeranimation
Prototyping
Rückkopplung
Wellenpaket
Mathematisierung
Gruppenkeim
t-Test
Zahlenbereich
Interaktives Fernsehen
Gebäude <Mathematik>
Computeranimation
Geschlossenes System
Skript <Programm>
Kurvenanpassung
Softwareentwickler
Prototyping
Schnittstelle
Autorisierung
Teilnehmerrechensystem
Geschlossenes System
Dichte <Stochastik>
Biprodukt
Fokalpunkt
Frequenz
Office-Paket
Kollaboration <Informatik>
Komponente <Software>
Beweistheorie
Autorisierung
Webforum
Teilnehmerrechensystem
Geschlossenes System
Bit
Prozess <Physik>
Punkt
Landau-Theorie
Mathematisierung
Mathematisierung
Zellularer Automat
Biprodukt
Computeranimation
Teilmenge
Texteditor
Arithmetische Folge
Geschlossenes System
Vorlesung/Konferenz
Projektive Ebene
Geschlossenes System
Geschlossenes System
Güte der Anpassung
Vorlesung/Konferenz
Computeranimation
Folge <Mathematik>
Autorisierung
Soundverarbeitung
Geschlossenes System
Folge <Mathematik>
Mathematisierung
Versionsverwaltung
Mathematisierung
Element <Mathematik>
Quick-Sort
Computeranimation
Arithmetisches Mittel
Bildschirmmaske
Datenstruktur
Inverser Limes
Kurvenanpassung
Varietät <Mathematik>
MUD
Autorisierung
Subtraktion
Subtraktion
Selbstrepräsentation
Versionsverwaltung
Gruppenkeim
Inverser Limes
Versionsverwaltung
Fokalpunkt
Computeranimation
Parametersystem
Addition
Subtraktion
Einfügungsdämpfung
Wasserdampftafel
Mathematisierung
Versionsverwaltung
Applet
Biprodukt
Quick-Sort
Computeranimation
Texteditor
Reelle Zahl
Subtraktion
Parametersystem
Vorlesung/Konferenz
Inhalt <Mathematik>
Hilfesystem
Zeichenkette
Folge <Mathematik>
Autorisierung
Addition
Subtraktion
Bit
Gravitation
Element <Mathematik>
Landau-Theorie
Mathematisierung
DTD
Kartesische Koordinaten
Mailing-Liste
Element <Mathematik>
Information
Elektronische Publikation
Term
Ereignishorizont
Computeranimation
Knotenmenge
Attributierte Grammatik
Äußere Algebra eines Moduls
Wort <Informatik>
Ereignishorizont
Parallele Schnittstelle
Bildgebendes Verfahren
Objektverfolgung
Extremwert
Element <Mathematik>
Zahlenbereich
Mathematisierung
Schlussregel
Computeranimation
RFID
Teilmenge
Modallogik
Datenstruktur
Vorlesung/Konferenz
Datenstruktur
Ordnung <Mathematik>
Formale Grammatik
Punkt
Mathematisierung
Gruppenkeim
Ausnahmebehandlung
Biprodukt
Mustererkennung
Computeranimation
Teilmenge
Rechter Winkel
Beweistheorie
ATM
Programmbibliothek
Vorlesung/Konferenz
Modelltheorie
Softwaretest
Geschlossenes System
Prozess <Physik>
Prozess <Informatik>
Kontrollstruktur
Gruppenkeim
Schlussregel
Inzidenzalgebra
Division
Computeranimation
Entscheidungstheorie
Integral
Stichprobenumfang
Ordnung <Mathematik>
Bildgebendes Verfahren
Leistung <Physik>
Einfach zusammenhängender Raum
Prozess <Physik>
Gruppenkeim
Zahlenbereich
Sprachsynthese
Kartesische Koordinaten
Schlussregel
Whiteboard
Computeranimation
Arbeitsplatzcomputer
Harmonische Funktion
Binäre Relation
Ordnung <Mathematik>
Ordnung <Mathematik>
Normalvektor
Geschlossenes System
Kontrollstruktur
Prozess <Informatik>
ATM
Vorlesung/Konferenz
Ordnung <Mathematik>
Binder <Informatik>
Maschinelles Sehen
Whiteboard
Computeranimation
Einfügungsdämpfung
Punkt
Prozess <Physik>
Selbstrepräsentation
Gruppenkeim
Kartesische Koordinaten
Element <Mathematik>
Computeranimation
Gradient
Übergang
Formale Semantik
Eins
Streaming <Kommunikationstechnik>
Einheit <Mathematik>
Geschlossenes System
Prozess <Informatik>
Vorlesung/Konferenz
Kurvenanpassung
Parallele Schnittstelle
Phasenumwandlung
Prototyping
Softwaretest
Benutzerfreundlichkeit
Güte der Anpassung
Mustererkennung
Bitrate
Frequenz
Rechenschieber
Scheduling
Dienst <Informatik>
Menge
Digitale Videotechnik
Dateiformat
Instantiierung
Standardabweichung
Geschwindigkeit
Rückkopplung
Maschinenschreiben
Subtraktion
Stabilitätstheorie <Logik>
Stab
Klasse <Mathematik>
Mathematisierung
Zahlenbereich
Term
Task
Loop
Erwartungswert
Benutzerbeteiligung
Webforum
Delisches Problem
Inhalt <Mathematik>
Modelltheorie
Softwareentwickler
Datenstruktur
Demo <Programm>
Soundverarbeitung
Medizinische Informatik
Mailing-Liste
Peer-to-Peer-Netz
Vektorraum
Paarvergleich
Fokalpunkt
Quick-Sort
Ordnungsreduktion
Summengleichung
Energiedichte
Identitätsverwaltung
Mereologie
Dreiecksfreier Graph
Normalvektor

Metadaten

Formale Metadaten

Titel Tracking Changes to JATS XML in an Online Proofing System
Serientitel JATS-Con 2013
Teil 04
Anzahl der Teile 16
Autor O’Connor, Charles
Gnanapiragasam, Antony
Hepp, Michael
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 10.5446/21792
Herausgeber River Valley TV
Erscheinungsjahr 2016
Sprache Englisch
Produktionsort Washington, D.C.

Inhaltliche Metadaten

Fachgebiet Informatik
Abstract When Dartmouth Journal Services began building ProofExpress, an online, XML-based proofing and editing system for STM journals, we knew that the most difficult challenge would be creating an accurate change-tracking mechanism. Change tracking is an essential feature, both to ensure that author corrections conform to journal style and to catch any changes to data or claims. The system must not only track each insertion, deletion, and formatting change, it must also give production editors the ability to accept or reject changes without breaking the XML. ProofExpress is built on SDL LiveContent Create (formerly Xopus). We use its extensive API to add custom elements and attributes to mark changes in the XML. The XML is then transformed through XSLT to group and nest changes so that they can be acted upon by the production editor. To prevent breaking the XML during this process, a rule engine enforces the order of acceptance and rejection of changes.

Zugehöriges Material

Ähnliche Filme

Loading...