the benefits of ELAN

Fri May 2 13:12:08 UTC 2008

How you indicate or annotate depends on what it is you want to look at. In
our case, one thing we are looking at how often indicating verbs (agreement
verbs) do modify spatially. So on one tier (or actually one for each hand)
we use an English gloss. We are able to be totally consistent because Trevor
Johnston has created the Auslan Lexical Database so each sign has an
assigned ID gloss. Doesn¹t have to be the best translation in that context,
as translations occur on another tier. But it means every time we search
for, say FAMILY2 we know we are getting that particular variant for family.
On another tier we have a controlled vocabulary of grammatical class,
including whether a verb is indicating or not. And on another tier we say
whether it was, indeed, modified spatially to show semantic roles.

We have well documented annotation guidelines if you are interested that
show how we code on each tier. But this means we can search for the most
frequent verbs and how often they modify, or how often verbs modify if they
occur before or after a point sign etc. The possibilities are endless. And
if later we want to get different information, we just add another tier to
the template we use and then go through the data again looking for that
additional feature.

On 2/5/08 10:46 PM, "Gerard Meijssen" <gerard.meijssen at gmail.com> wrote:

> Hoi,
> I have been doing some reading.. I try to understand this.
> ELIAN is open source, the software is GPL licensed - I think I understand how
> it would work. What I wonder about it is how you annotate a signing session.
> Mark it in such a way that you can indeed search for things. You still need a
> method to indicate that a particular sign or part of a sign is used in a given
> time frame. Is this where SingWriting / HamNoSys come in by associating it
> with a time slot in ELIAN ?
> 
> Amanda Brown wrote on a separate mail thread about "Comparison of multimodal
> annotation tools: A workshop report. Gesprächsforschung, 7:99-123." would this
> be available in a digital format for me ?
> Thanks,
>     Gerard
> 
> On Fri, May 2, 2008 at 2:19 PM, Lou de B <luisitadb at optusnet.com.au> wrote:
>> Sign writing, as Sonja has said, is a transcription system or orthography, as
>> is HamNoSys etc. I don't know anything about ANVIL, but have been using ELAN
>> daily for close to 2 years and it is a fantastic program that allows you an
>> infinite number of tiers to annotate what you want (such as individual signs,
>> or whole clauses) however you want (eg English glosses, or some sort of more
>> phonetic transcription). Unless you are looking at the phonology, generally
>> an indepth phonetic transcription is not needed as the annotations are time
>> aligned with the video. That means you always have access to the raw data in
>> front of you. The great thing about using a program like ELAN is that you are
>> then able to search your data for (just a couple of egs, the uses are
>> multiple) frequency of something, length of time of something in a text (eg
>> overlaps), or co-occurrence of one thing with another. This means instead of
>> having an archive of data, you have a machine-readable corpus. Absolutely
>> invaluable for going back to the data again and again.
>> 
>> ELAN is free, constantly updated, help is available by email or to the ELAN
>> list. Additionally, you can give feedback to the guys working on it so it
>> evolves to meet your needs. Basically, I have found very little that ELAN is
>> unable to do in our use of it in Australia (with Trevor Johnston and Adam
>> Schembri in London). That's my rant and rave for it. I'm a serious convert.
>> 
>> Louise de Beuzeville
>> 
>> 
>> On 2/5/08 8:06 PM, "Sonja Erlenkamp" <sonja.erlenkamp at hist.no> wrote:
>> 
>>> I wouldn't compare a writing- (or transcription)system with a dataprogramm
>>> (as Elan is), which is used to link annotations to a videofile. They have
>>> different functions in the analysis of the data. I'd rather combine them. :)
>>>  
>>>  
>>> Best regards
>>>  
>>> Sonja Erlenkamp
>>> 
>>> 
>>> Fra: slling-l-bounces at majordomo.valenciacc.edu på vegne av Ingvild Roald
>>> Sendt: fr 02.05.2008 11:16
>>> Til: A list for linguists interested in signed languages
>>> Kopi: A list for linguists interested in signed languages
>>> Emne: Re: [SLLING-L] ANVIL versus ELAN
>>> 
>>> For transcription and analysis of dialogues, I don't know about ANVIL or
>>> ELAN, but it could be done in SignWriting (hhtp://www.signwriting.org
>>> <http://www.signwriting.org> ). Examples of communications between two and
>>> three persons have recently been published on the sw-list. SignWriting can
>>> be exported as XML
>>> 
>>> Ingvild Roald, Norway (not a linguist)
>>> 
>>> 
>>> _______________________________________________
>>> SLLING-L mailing list
>>> SLLING-L at majordomo.valenciacc.edu
>>> http://majordomo.valenciacc.edu/mailman/listinfo/slling-l
>> 
>> 
>> _______________________________________________
>> SLLING-L mailing list
>> SLLING-L at majordomo.valenciacc.edu
>> http://majordomo.valenciacc.edu/mailman/listinfo/slling-l
>> 
> 
> 
> 
> _______________________________________________
> SLLING-L mailing list
> SLLING-L at majordomo.valenciacc.edu
> http://majordomo.valenciacc.edu/mailman/listinfo/slling-l

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.linguistlist.org/pipermail/slling-l/attachments/20080502/b9bc115e/attachment.htm>
-------------- next part --------------
_______________________________________________
SLLING-L mailing list
SLLING-L at majordomo.valenciacc.edu
http://majordomo.valenciacc.edu/mailman/listinfo/slling-l