Add to Watchlist

Conference recordings complement scholarly research communication in traditional conference proceedings


Citation of segment
Embed Code
Purchasing a DVD Cite video
here much so can you hear me so thanks for the introduction and I know
after this long afternoon session about software I could have called my talk as well and now for something completely different and I'm going
to talk about a conference recordings and as you all know by now that the conference here is also recorded as the as days last year so have been recorded as well and so
only until now we talked about the fundaments all the basis of what the research arm when we talk about data all software and how to treat software and data because that's if you think of the research process chain normally starts in the beginning but as a mark Wopat says signs isn't finished until it's communicated so what I'm now going to talk about is the at the other end of the of the chain so how do we communicate science and traditionally there are 2 main some ways to
communicate about science and the 1 is the John article and was a bit the
the topic I was talking about last year only and how you
can and hands the Johno articles with this videos like do abstracts of adjust supplements few interested in that you may find the talk of last year here in and out of that very fruitful cooperation with Tallmadge kabuki evolved and as you already mentioned in his talk we wrote a paper where which you can find on the the US preprint and in the optical and quantum electronics which was published just last month only yes with so that today I'm going to talk about something different from similar in in another way namely
conferences so that's the other big part of communicating science and this is George Whiteside chemistry professor from Harvard and he says that so this is the serious about the future of from scientific publication and the part of the Publication system on you tube which is pretty interesting and in there he says I increasingly find that the stuff that we do that has probably the greatest impact in some metrics is done in video format not in paper form at and what he's talking about is formats like for example this 3 minute thesis which is the last 1 is a formant where a PhD students I have to give a presentation about the thesis in just 3 minutes and this is quite popular and another very popular formant these text talks where you have lots of talks that have been viewed more than a million times also and just by these numbers you yeah you realize that you wouldn't have such a big audience with with the normal and conference talk all with the with a normal scientific paper and as I mentioned already last year we we did the recordings of the states and so this is the most of you to talk of last year and it it has been viewed more than 300 times in the last 11 months I mean it's not more than a million obviously but still if you if you think of here like 50 people sitting and now you have about 300 views in in 1 year which is I think pretty good only so but now to them
about the problems so most most people that almost conferences where and that that are recorded and it if you looking for for the videos if you're looking for a talk you think OK it there I know this young colleague of mine has talked about some topic I'm interested in and you're looking for the video that problem is normally you 1st of all you don't know where to look and even if you know most of the time they don't provide problem meter data so for example I've been a couple of years ago in Canada in bands at the Banff International Research Station the appears and there they always the recall to the videos and there the probably most some of you know that and that are very very good talks and very very good videos they don't provide any metadata and you can't search for that just a few months ago I wanted to look up a talk I I've seen there but I can't really read the book but couldn't remembered at that time what year was it was a 2012 for the to 6 thousand 13 I couldn't remember the exact name of this of this conference it was something about stalker state Serre systems in biology by I couldn't remember and I didn't know the name of the also so at head took me a long long time to find to find the video although I knew what I was looking for and that's called because you you you don't have to go to the right metadata even if you find the video in many cases if it's on you tube if it's on on the Conference page how do you cite that you that you can just side by the you are end then we've been in contact with a lot of conference organizers and there there are many many examples of videos that just go on and go missing after 1 year 2 years so really after a very short period of time and even the all this doesn't happen then in most cases you don't have some information on the license and so it if you want to reuse your this video and some other context you you just don't know what you what you're allowed to what not a lot to do with the video and
so these are the main problems that you know and that you have with using the sound conference videos and so what
what we did we talked to
to conference hosts and conference organizers and also to some service providers that to do the audiovisual a recording so that do the filming of the of the of the conference and but that was done the last year and we had interviews with 30 hosts and organizers and noun and also the service providers and none of them uses the persistent identifier and basically all of them published on the conference website or the on that side and about half of them on you tube or some other platform so this is the 3rd of wave that the videos are treated right now only and then we ask
them so what the main or their most important aspects concerning the the material itself in the publication process and what are the biggest problems you you encountered during this process and if you think about by the way that the videos I published it's quite surprising that long-term accessibility and sustainability is 1 of them most often named aspects by about 70 per cent so they did this work qualitative interview so you shouldn't to treat the numbers like and that these are really quantitative numbers but it's just that this should give you some idea of at what people looking for and so last year I already introduced the AV portal and so all of you that watched videos from last year already know they the portal but I I introduce
it once again just for those who don't know so videos and they the portal they on the goal of some serious all farms automatic analysis that the 1st step is whenever automatic scene recognition that you have this the video segmented from which you can see here after that we have a speech recognition and text recognition so a speech transcript and a text transcript is produced and that image recognition so for example the the system would how would recognize that this system this is a conference talk and afterwords them we have a named entering at the recognition on the speech and the text transcript to get to get the key words as an
example you can see here 1st of all the individual Islam segmented and and use in this sort of visual of a table of contents and you get the speech transcript with time stamps so that you know in each segment what what was the what was said by the ground individual and then the algorithm finds some key words like your software and extract stem and now you can search within the spoken word and within the text that is shown in the video and if you have a of video that goes for that is part of a 1 hour and the word you were looking for this edge 40 minutes you can jump right than it would be highlighted here and you could jump right in the video and see that that's what the what you're looking for or not
and then another important thing the convent the preservation of the videos so it's in the the portal every video gets a
you by from data side and the videos are an digitally preserved to with the Exley Police what was that a system as you all
know do you I you can use for 1st a stable citation and if you want for example to cite the opening session of last year's MMS days you can use this to you I in this opening session there were 4 people talking and if you just
want to side the talk by Mateus Kleiner the president of the lab it's association then you just add this has stick T you just cite the 2 minutes 33 to 40 minutes 41 and you just have to talk by how much is Kleiner so you you know that this probably from U-tube where you also can jump right in in the video but here that the special thing is the combination with the GUI so this still remains remains a stable link Of course you cannot only cited the whole talk but you can also have from just a single sentence like this about the success of the MMS network than just site a site these 12 seconds or down to just 1 2nd
so let's come back to this results of our of our sort of the another important thing the visibility 1st of all if you all the videos that are on a free license you can Of course download and publish everywhere else or distribute whenever you want of apart
from that the videos of last year's as days with in the 1st 11 months that they had more than 1 thousand views so that's about a hundred views a month which is for a conference like the states I think you them a very good a very good value so that and that's about the visibility and
it's now what we thought that the last things especially the speed of the simplicity and the financial technical and personal resources and very crucial problems and just this quote
of 1 conference of organizer that chart the they tried to do the recordings themselves and they said the video component of the conference was by far the most complicated and difficult part of our experience and that is certainly part of our delay and posting the videos because they're in that you may have they that was like half a year after the conference and they still hadn't were were not able to publish the videos because it just was to complicated if you want to do the whole process on your own and
so these are the 5 more or less 5 steps to that you need if you have to have the speakers signed an agreement and signed the license you have to do the the recording itself including the post-processing you have to add metadata you have to publish and then you have to have some preservation and most people just do step 2 and step 3 and leave out the other 3 steps but
this conference organize I just well cited they he said that he listed these problems and he said for example the videos must be post processed and edited and then posted where they can be accessed we grossly underestimated how much time and attention this takes the do yourself approach was very time consuming and extremely stressful and so we thought OK so far we do this time we how we talk about the well we care about that the licenses and the agreements and read care about the metadata we do the publishing and the preservation why not also do the 2nd step so that we all that we have all the do 5 steps out of
1 hand and can do though the whole thing and now you don't have to the have the friction between these different steps because the people that do the recordings normally don't know about their license and the people that do the license don't know about the metadata and so on and we we have experts of all these fields and so we can do all of these and just what we did and so here's the the page we set up so if you're interested in this just ask us we're here or visit this site and if Europe and if you're planning on organizing a hosting the conference and he wants to have its recorded speak to us we are happy to help we're happy to do the recording and like with
the jaundice and the video abstracts and supplements we tried to link the videos 2 conference proceedings if they are pra conference proceedings if you plan to produce conference proceedings day in your conference and want to publish them open access we also have a platform where you can publish the the conference proceedings open access and we also have an agreement with now with the EDP which is a French publisher where we can published proceedings at the PMAccess in the park at this publisher for very cheap like something like 30 uracil per article or something and of course you can also add additional leave material so like the slides off the talk or software that that is related to the talk or a data our other articles and this so you can have this permanent link between article or the related materials and the video and with that I come to an
end and thank you for your attention


Formal Metadata

Title Conference recordings complement scholarly research communication in traditional conference proceedings
Title of Series Leibniz MMS Days 2018
Author Drees, Bastian
Contributors Leibniz-Institut für Oberflächenmodifizierung e.V. (IOP)
Leibniz-Institut für Troposphärenforschung (TROPOS)
License CC Attribution 3.0 Germany:
You are free to use, adapt and copy, distribute and transmit the work or content in adapted or unchanged form for any legal purpose as long as the work is attributed to the author in the manner specified by the author or licensor.
DOI 10.5446/35349
Publisher Weierstraß-Institut für Angewandte Analysis und Stochastik (WIAS), Technische Informationsbibliothek (TIB)
Release Date 2018
Language English
Production Place Leipzig

Content Metadata

Subject Area Information technology, Mathematics
Abstract Traditionally, research results presented at scientific conferences are published in conference proceedings. Additionally, an increasing number of conferences are recorded and the videos are subsequently published. In most cases these videos are published by the organisers on commercial platforms or directly on the conference website. Therefore, a systematic search for conference recordings is difficult. Moreover, the videos are often lost after a short time e.g. because the URL changes or external links lead to nowhere. Usually, conference websites are not maintained on a long-term basis and commercial platforms may remove videos or change the conditions for access for a variety of reasons. Finally, to ensure an unproblematic re-use of the material questions of licencing, technical quality and formats have to be solved in advance. In order to prevent the loss of conference recordings the German National Library of Science and Technology (TIB) has developed the AV-Portal. The AV-Portal provides the ideal infrastructure to host, find and reuse scientific videos. It's a single access point for videos from different conferences and years. All videos are assigned a Digital Object Identifier (DOI). These persistent identifiers allow for reliable referencing both as stable online links and as correct citations in scholarly work. In this talk we describe how sharing scientific results via audio-visual media has become an important part of scientific communication and how conference recordings may complement the classical conference proceedings and add new value to the presented research. By discussing the whole process from pre-conference organisation to publishing and archiving the recordings we show pitfalls, best practices and opportunities of recordings for both conference organisers and participants.

Related Material

The following resource is accompanying material for the video


  737 ms - page object


AV-Portal 3.7.0 (943df4b4639bec127ddc6b93adb0c7d8d995f77c)