Importing CCR data [was: CCR Testing]

Anton Lundin glance at acc.umu.se
Mon Nov 3 15:47:34 PST 2014


On 01 November, 2014 - Paul Sargent wrote:

> 
> On 1 Nov 2014, at 10:57, Miika Turkia <miika.turkia at gmail.com> wrote:
> 
> > Together they give enough information to know what the diver was breathing at any time.
> > 
> > Thanks, this gives me something to work on. The current Shearwater import is indeed quite limited in this regard. And of course the whole CCR concept was not there when I wrote the transformation. 
> 
> Indeed, and I think we’re going to find things rather fluid as we decide what’s useful information to have from a CCR dive. Willem’s building a road with the Poseidon work he’s doing, but it’s quite an esoteric rebreather, so I think we’ll find some concepts map well to others, and some won’t.
> 
> > The CSV import has no options for Gasses, Setpoints, OC/CC mode, ppO2 Sensors, all of which are in this file. I agree the dialog is a limiting factor here. I think there is also a problem with how these concepts map into the Subsurface XML.
> > 
> > Fields are named similarly to the XML.
> > 
> > I am thinking of 3 different options for this:
> > 1) just keep it as it is and hope that CCR divers have better import option than the CSV
> > 2) Implement the setpoints and other info under the hood with no configuration options for users
> > 3) Make a new tab for CCR import
> > 
> > Personally I would of course prefer something that does not include GUI work, but overall I kinda like the third approach. At minimum, implement another tab for the CCR import and possibly cimplify the current CSV import for recreational divers. I just wonder if tech divers will need all the fields in current CSV import when diving OC.
> > 
> > Any comments on these? Anton, what is your opinion?
> 
> CSV is nice to have as a fall back. There’s always going to be a computer / rebreather that we don’t support, and having CSV available gives a user a fighting chance. Even if it’s a limited set of info. If you’ve got a better method for a particular device (e.g. XML, or a specific format of CSV) you’re going to do that.
> 
> My opinion is that for the configurable CSV import it has to be limited, otherwise you’ll be on a never ending quest to try to build in more flexibility.
> 
> 	Time
> 	Depth
> 	Water Temp
> 	ppO2 (Single Value - CCR/SCR)
> 	CC/OC (CCR)
>         Stop Depth
> 
> …and then it’s a manual process to enter gasses / switch depths.
> 
> (I’ve dropped off NDL, CNS, TTS, & pressure from the current list. CNS is calculable. NDL & TTS aren’t that important if you’ve got the ceiling IMHO. Pressure I assume is cylinder pressure - How many do you want? Possibly just 1, but no more)
> 

I think those are quite interesting to look at, and to compare with what
Subsurface calculates and what the DC reported. I use it as a sanity
check between different DC's and Subsurface.

I think we should import as much data as possible from the different
formats. If we can find a format that contains data that we currently
can't import, I kinda enjoy hacking on such features =)

> Things like multiple sensor readings, or multiple gasses probably need to be limited to known vendors. There’s just too many ways they could be represented. 
> 
>     - Is an O2 sensor reading in bar or millivolts? 
>     - Is a breathing gas a number into a gas table, an fO2/fHe, or a ppO2/ppHe? 
>     - Is a set point column boolean Hi/Lo, or a value?
> 
> Any dialog box that deals with all of that is going to be huge. So I’d keep the options low (as present, maybe trimming a couple), and add a couple more for CCR/SCR. Then have predefined schemes which are able to do more under the hood (e.g. APD, Seabear, etc)
> 

Hmm... Sane defaults and hiding some of the gory details is probably the
right way to go, but why not just put those gory details in a "advanced"
page or something? If you open a "advanced" page and get a ridiculous
amounts of settings, nobody will be surprised.


//Anton


-- 
Anton Lundin	+46702-161604


More information about the subsurface mailing list