Acoustic tag in front, PSAT tag at the rear. Source.
Please do read this.
Prima vista, nothing spectacular.
As far as I understand it (correct me if I'm wrong), it simply states that the researchers don't have direct access to raw data but instead, they are using data that are being processed/analyzed either by the tag manufacturers or via proprietary software given to them by the manufacturers; and that some of those processed data might be faulty; and that because they are commercial entities, the manufacturers tend to be reticent about their algorithms and software specs - meaning that the researchers need to engage in constant dialogue with the manufacturers and not simply blindly rely on those processed data.
So what's the big deal?
Turns out that behind the scenes, this has led to some major kerfuffle, with some reviewers encouraging a more open dialogue about this issue but not wanting to stand in the front lines; whereas others have been positively outraged and have accused Juerg of being cavalier and incendiary, of disrespecting fellow researchers and more nonsense of this sort.
IMO this is a clear indication of the fact that a) there is a problem, b) many researchers are a tad too timorous (a euphemism), maybe because c) the tag manufacturers, being quasi-monopolists, are exerting way to much influence (another euphemism) on their customers and d) journals may be lacking fortitude, too!
IMO this is a clear indication of the fact that a) there is a problem, b) many researchers are a tad too timorous (a euphemism), maybe because c) the tag manufacturers, being quasi-monopolists, are exerting way to much influence (another euphemism) on their customers and d) journals may be lacking fortitude, too!
Long story short, I'm disappointed.
I thought that science was better than that.
Kudos to Juerg for having persevered and gotten this published in the end - the question being, will this now lead to more pointless demagoguery, or will this now be discussed publicly as should have happened all along?
Anybody taking bets?
2 comments:
Funny coincidence and very interessting as I was trying to find more
informations about the tag manufacturer some days ago and was oft stumbling
over "tags will then be send back to the manufacturer for preparing data".
As someone who is working in IT Monitoring, I was interested in "preparing
data" and was
wondering about how they will process. So Thanks for the Info on that.
Also I was wondering about the fact, that you are, as the one who payed for
the tag, are not able to get this raw data by your self. Having the Tag full
with data and then have to wait again for probably month till I get
Informations from the manufacturer which are, as we know now, are maybe not
correctly analysed. And on what are there algorithm based on? Best practice and
some references from Ritter? ;-)
I'm doing mostly Monitoring in IT and this would be like my colleagues would
only get access to all Status and Performancedata I'm collection about their
Servers/Applications if they are asking for it and maybe also inviting me
for lunch. And maybe going one step further as some of the managers asked
me that in the past. Manipulate them before putting them in a PDF-File as
they are looking then better and more rounded and nobody will question anything.
:)
Of course those tags are business and contain proprietary stuff, so I kinda understand that there must be a filter.
I guess Microsoft and Apple don't share all of their proprietary info either, right?
But if that then lead to errors, especially in the scientific conclusions, then there's certainly a problem that needs discussing.
Post a Comment