Wednesday 21 December 2011

Good paper... must... reject...

A colleague and I recently received the review outcome for one of our conference papers. This is an excerpt of the feedback from one of the reviewers:
*** Reasons to accept: Please give the main reasons for accepting the paper.
(this was actually blank)

*** Reasons to reject: Please give the main reasons for rejecting the paper.
This article is very well written. The contribution is interesting and the results are encouraging.
The final outcome was "accept" :D

Guesticide!

I was browsing through a recent issue of the Times Higher Education today and I stumbled across a very fascinating article. It is entitled "Research Intelligence - Phantom and parasitical menaces" and is dated November the 10th 2011.

The article is a report on a survey published by the British Medical Journal, on the phenomena of guest and ghost authors on academic publications. The survey's title is "Honorary and ghost authorship in high impact biomedical journals: a cross sectional survey". The original survey (BMJ 2011;343:d6128) was aired in October 2011, with a subsequent correction (BMJ 2011;343:d7677) published in November 2011.

This is a link to the publication: http://www.bmj.com/content/343/bmj.d6128.full
The correction can be found here: http://www.bmj.com/content/343/bmj.d7677

In my humble opinion, the interesting part is not the discussion on the phenomena themselves. Rather, it is the discussion on methods to decrease their occurrence. THE's article lists various recommendations, from various sources. Among them are ghost/guest explicit prohibition; signed contribution statements from authors; publications bans; more rigid institutional monitoring; legal action. Some may seem a tad bureaucratic or unrealistic but, to me, this is an indication that the community acknowledges the situation as a problem that needs solved.

Being outside the biomedical field myself, I will refrain from commenting on the ghost part. When it comes to guests though, I shall point you to the first reader comment (the one on November the 10th) below THE's article. Read the third paragraph, the one starting with "On reason for this...".

Having read this comment, I have nothing further to say...

Thursday 15 December 2011

Contiki for cc2530 with uIPv6/RPL support

I recently finished porting the Contiki Embedded Operating System for Texas Instruments cc2530 devices.

The port can be found on github:
https://github.com/g-oikonomou/contiki-sensinode/wiki

As part of my work for Loughborough University, I have also ported Contiki for Sensinode/cc2430 devices. Due to hardware similarities, it made sense to co-host the two ports on the same repo.

Both ports use the Small Device C  Compiler and have full banking support, which means that we can build working images up to 256 KB (for cc2530-F256) or up to 128KB for Sensinode/cc2430.

Still a couple of drivers to implement (SmartRF LCD, joystick), but the interesting functionality is all there: IPv6/6LoWPAN support with RPL, UDP and ICMPv6. TCP is currently disabled but I am hoping to enable and test it as soon as I get a chance.

The cc2530 porting effort would have been impossible without the direct support from TI, who kindly donated development/testing kit. Many thanks to them!