Running AntConc as a backend. Any ideas?

65 views
Skip to first unread message

Yannis Panagis

unread,
Sep 28, 2023, 11:20:01 AM9/28/23
to ant...@googlegroups.com
Dear all,

I am thinking of possible solutions for sharing AntConc with colleagues, without eg. them to have the need to install, configure, and load corpora in their own computers. I was therefore thinking of a web front-end, also inspired by the link from AntConc's homepage, on "AntConc emulation". 

The webpage hosting the emulation is not available right now, and AntConc does not make it's API available, but I just wanted to ask if someone could share any good ideas on how something similar can be done.

Screen scraping could be an option, I guess, but any ideas for less brute force solutions are more than welcome.

Best, 
Yannis Panagis 

Laurence Anthony

unread,
Sep 29, 2023, 2:21:06 AM9/29/23
to ant...@googlegroups.com
Hi Yannis,

A few people have asked me about an API for AntConc, but if they are already familiar with programming, it would seem just as easy to use something like the Python NLTK library. So, I haven't really pursued it. Can you explain what advantages you would get from an AntConc API over NLTK?

What you really seem to want is a web-based version of AntConc. This is something that I've wanted to create for a long time. The main problem with this is allowing users to upload their own corpora and tracking these. It would necessitate authentication with accounts/passwords and would also require a server with a massive amount of available space for thousands of people to upload their own data. I could apply for funding to develop AntConc in this way, but I would worry about what would happen if the funding ran out. At the moment, my wonderful Patreon supporters help me to maintain the current infrastructure, but a web-based tool would incur many more costs.

For Windows environments, have you tried simply putting the portable version of AntConc on a network drive with launch permissions there? That should work for many use cases.

I hope that helps!

Laurence.

###############################################################
Laurence ANTHONY, Ph.D.
Professor of Applied Linguistics
Faculty of Science and Engineering
Waseda University
3-4-1 Okubo, Shinjuku-ku, Tokyo 169-8555, Japan
E-mail: antho...@gmail.com
WWW: http://www.laurenceanthony.net/
###############################################################


--
You received this message because you are subscribed to the Google Groups "AntConc-Discussion" group.
To unsubscribe from this group and stop receiving emails from it, send an email to antconc+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/antconc/CAJ0cCiTYU4VzuyHVSUuzVVf4seeX1hhQ7znPXCH85phFvRPERg%40mail.gmail.com.

Yannis Panagis

unread,
Oct 3, 2023, 8:35:01 AM10/3/23
to ant...@googlegroups.com
HI Laurence,

Thank you for your reply. Your answer in fact made me realise that sometime ago I wanted to post the same question and then I didn't because I thought of the same argument, ie. use NLTK or another NLP platform in the backend.

My main motivation to ask this question is that AntConc seems to have an easy(er) way to extract concordances and collocations, and show them to a front end, whereas a library could probably require more work to yield a similar result. Furthermore, since there is in fact something, using an older version of AntConc as a backend, there seems to be someone who has more or less done the same, in the past.

In any case, thank you for your answer and I might use a network share at the end, as you also suggested.

Best,
Yannis  

Laurence Anthony

unread,
Oct 3, 2023, 10:08:18 AM10/3/23
to ant...@googlegroups.com
Hi again,

Perhaps you can post a question in that old thread asking how they used AntConc as a backend. It seemed like a very interesting solution. I would say that with the new corpus manager, the ability to save settings, and an easy way to distribute single file corpora, even the standard AntConc is fairly easy to distribute to users. Basically, you just need to guide users to the website to download the software, and then send them a link to the corpora that you want them to use.

Laurence.

###############################################################
Laurence ANTHONY, Ph.D.
Professor of Applied Linguistics
Faculty of Science and Engineering
Waseda University
3-4-1 Okubo, Shinjuku-ku, Tokyo 169-8555, Japan
E-mail: antho...@gmail.com
WWW: http://www.laurenceanthony.net/
###############################################################

Reply all
Reply to author
Forward
0 new messages