Too Good to be True

So there were a lot of things to be done in a short time and, apart from one or two minor hiccups, everything went quite well: The paper presentation, the demo, submission to the ACL main conference, and the two workshops. In fact, one of the workshops (NLP for Less Privileged Languages) was an unexpected success, both in absolute and relative terms. The funny thing is that while this workshop was the one my heart was set on, I had to do much much more work for the other workshop (NER for South and South East Asian Languages). Believe me, the ‘much much more’ part is more than justified. Conducting a workshop (when you have to do most of the work) is not very easy. Conducting a workshop that includes a shared task is tough. And a shared task which requires new training and testing data is tougher. And it’s even more tough when you don’t have any funding etc. You get the point.

So, everything went quite well.

Not quite.

After the workshop work was finished I just happened to look at the final version of the paper which was submitted to ACL. And I found to my shock that a figure which I had added and on which I had spent around two hours on the night before the workshop (my heart was set on) was missing. The caption was there and so were the references to the figure, but the figure itself was gone.

What happened was that while I was in the middle of the workshop, one of the co-authors noticed a minor typo. But he is now doing a job and is not here. He was unable to compile the Latex files on his system. So he asked another co-author to correct the mistake. The other co-author did as he was told, but he didn’t check before submitting the revised version. Apparently some library was missing or incompatible on his system and so the figure simply disappeared from the paper. Now, I had worked all night before the workshop (my heart was set on) to complete that paper and I thought I had done a good enough job. Moreover, that paper has already taken two rejections, in spite of the fact that it has taken a lot of work from more than one person. Even as an unbiased reader (to the extent it is possible for me to be), I think that paper was worth selection in a major (‘first class’) conference.

So many things to do in so short a time and everything going almost perfectly?

It was too good to be true. Something had to go wrong. Murphy’s law.

That jinxed paper seems set for another rejection. Leading to more dejection.

Just because a library was missing (or incompatible) on a system and the version submitted had one figure missing.

I don’t know whether I should say this, but had this happened to some other author and I was the organizer, I would allow the author to submit a revised version with the figure included, provided the authors had not added anything more after the deadline.

Does that sound like a crazy idea?

Author: anileklavya

मैं सांगणिक भाषाविज्ञान (Computational Linguistics) में एक शोधकर्ता हूँ। इसके अलावा मैं पढ़ता हूँ, पढ़ता हूँ, पढ़ता हूँ, और कुछ लिखने की कोशिश भी करता हूँ। हाल ही मैं मैने ज़ेडनेट का हिन्दी संस्करण ( भी शुरू किया है। एक छोटी सी शुरुआत है। उम्मीद करता हूँ और लोग भी इसमें भाग लेंगे और ज़ेडनेट/ज़ेडमैग के सर्वोत्तम लेखों का हिन्दी (जो कि अपने दूसरे रूप उर्दू के साथ करोड़ों लोगों की भाषा है) में अनुवाद किया जा सकेगा।

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.