[open-linguistics] GOLD not reachable

Wed Jan 28 12:59:52 UTC 2015

Hi everybody,

GOLD was a project funded until 2010, I believe. We maintain its website (Malgorzata, Lwin and I) and have it integrated in many tools. In summer 2014 we relocated from Michigan to Indiana. At Indiana University we are rebuilding the team for our research group and new institute (for 6 months now), and setting up all servers and technology from scratch. Not only were we forced to acquire new hardware and set up a new IT-infrastructure, integrate everything into the new environment of the hosting institution, most systems had also to be recoded and brought up to date wrt. the underlying technology (programming language version, database support etc.). This has consumed our time (three of us coding and maintaining), completely for the last 7 or more months.

GOLD will be up again in the next weeks, this winter. Priority was given to funded projects. Nevertheless, all systems will be up again soon.

We have not only completely replaced all LINGUIST List code, this happened to OLAC too (which you maybe want to look into as well, or are using?), to MultiTree, and now this happens to LL-Map. GOLD is less a problem, and it has been integrated in its last version in ISOCat and can be browsed there.

Since you mention ODIN, there is a site where Will Lewis and others make the ODIN datasets available for download. There was a recent LREC paper on that. We set up the only online site for ODIN again at IU, but the author and former PI of this grant is Will Lewis, and he would have to repair some issues that emerged when switching from some ancient Perl version to the current one on our new server, to enable full browsing capabilities. We for ourself, we are working on a new form of something similar to ODIN, with a new interface etc. It should pop up soon, maybe even with a link from LINGUIST List. We can talk about that using a private channel.

Getting back to a response by Steve Moran, Malgorzata and I are the moderators of LINGUIST List, Lwin is the tech manager, but we should treat the projects and linguistic resources as independent of that operation. LINGUIST is the independent community funded and supported service. The one thing that might be interesting to discuss in this context is, these projects that create resources and tools for linguists and others, often have been project results related to private research interests, or funded by some science foundation or other sponsor, and the funds are/were always limited in two dimensions, amount and time. The resources that were created in the past by Helen Aristar Dry, Anthony Aristar and their team, were cyber-resources and online applications. While also static data packages require a storage somewhere and an access channel, which comes with real costs, it is even worse with real applications that require hardware, bandwidth, an OS and a software environment, and in addition real human administrators and developers to maintain the operation. While we are taking care to get heritage systems up and going, it is a problem not really solvable easily in the most common scenarios. Running all the servers costs us now real money, and it is hard to find sponsors for that. Even harder to find volunteering admins... :-) I don't know how you all solve this problem. It would be interesting to hear about some strategies and solutions.

Thanks for asking and for your interests in the resources and technology of the former Institute for Language Information and Technology in Michigan, and the new "institute without a name yet" at IU, and maybe you will allows us to remind you:

- support LINGUIST List on the next fund-drive, this is not only good for it, but for all of us, and

- you might want to consider volunteering or coming as an intern to LL and IU next summer to help us set up the resources and infrastructure much faster and work on new ones: http://linguistlist.org/getinvolved.cfm

Thanks!

Damir

> Hi Bettina, 
> 
> I'm quite sue there hasn't been much development after 2010. Partly, this is because people involved have considered redesigning it (Damir Cavar, pers comm at LREC 2012), but AFAIK, there hasn't been any movement in this direction. I include Damir Cavar (http://cavar.me/damir/) in my reply, he should be able to give us an overview over the current state of affairs and plans for GOLD.
> 
> Best,
> Christian
> 
> 2015-01-28 9:10 GMT+01:00 Bettina Klimek <klimek at informatik.uni-leipzig.de>:
>> Hi everybody,
>> 
>> I was about to have a closer look at the GOLD ontology but unfortunately it seems, that something is broken. At least I cannot reach it via http://lov.okfn.org/dataset/lov/vocabs/gold. There is, however, this file https://ckannet-storage.commondatastorage.googleapis.com/2014-10-19T09:35:54.447Z/gold-2010.owl. But it I am not sure if this is the last version and if it is complete.
>> 
>> Does anyone know, who I could contact to get some information what is going on and if this problem will be fixed or does anyone happen to have a more recent version of the OWL file (if one exists).
>> 
>> Similiarly, I have trouble finding data in ODIN (http://odin.linguistlist.org/). Do we provide meta data such as "dataset not maintained any more" to separate up to date data sets from those which still have data files but are not maintained any more. It seems that quite some data sets only persist as long as the project under which it was compiled.
>> 
>> How do we deal with such cases? Do we leave this issue to the data providers?
>> 
>> Regards,
>> Bettina
>> 
>> 
>> _______________________________________________
>> open-linguistics mailing list
>> open-linguistics at lists.okfn.org
>> https://lists.okfn.org/mailman/listinfo/open-linguistics
>> Unsubscribe: https://lists.okfn.org/mailman/options/open-linguistics
> 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.okfn.org/pipermail/open-linguistics/attachments/20150128/c8d405f7/attachment-0003.html>