Problems with dive planning module.

Dirk Hohndel dirk at hohndel.org
Thu Oct 17 07:39:53 UTC 2013


On Thu, 2013-10-17 at 11:18 +0200, Willem Ferguson wrote:
> Having spent a lot of time playing around with the Dive planning module, 
> trying to do anything (even stupid) that a user might attempt. There are 
> several issues:
> 
> BUGS
> 1) Setting the 'Last stop at 6m' setting has no effect at all on the 
> final dive plan.

That's a bug :-)

> 2) Scaling:  The scaling on the Time axis of the plan, using the '+' and 
> 'v' buttons: If you magnify along the time axis using the 'v' button, 
> even after full scale has been reached, it keeps record of the number of 
> clicks. So if you click 'v' 5 times after full scale has already been 
> reached and then you then want to reduce the size of the profile along 
> the time axis, one needs to to click the '+' button 6 times before the 
> image is reduced. It looks like a pointer is updated or increased every 
> time one hits the 'v' button, and decreased every time one hits the '+' 
> button. But this happens irrespective of whether the image of the 
> profile has actually changed. I have managed to hang Subsurface several 
> times now (having had to force-kill the process) by hitting the 'v' 
> button several times after full magnification and then the '+' button, 
> trying to reduce the size of the image. It seems that the 'v' button 
> should only lead to any action (including background management) if the 
> image is not at maximum size. Maybe the '+' button also needs attention 
> with respect to the status of the graphics display. I hope this is 
> understandable!

OK - you completely lost me. What are you doing?
These are NOT scaling buttons. You add maximum time / maximum depth
displayed. Yes, in some way that is related to scaling, but that is not
what the buttons are supposed to do.

> 3) Several dive plans in dive list: When a dive plan is saved, it gets 
> listed in the dive list, together with all one's other real-world logged 
> dives. Now, if one creates the same dive plan more than twice, the dive 
> plan actually looks and behaves very differently. You do not get the 
> same dive plan. An example:

Yes, well understood because of the whole repeated diving concept. What
is needed is the ability to edit the start time. The whole point is that
I want to plan the next TWO dives and have the second dive and its deco
take into account my dives so far and the first dive I have planned.

> USER INTERFACE
> 1) Scaling buttons: Please use '+' for magnification and '-' for 
> reducing the image. The two present buttons '+'=reduce, 'v'=magnify are 
> totally counter-intuitive.

Those are NOT scaling buttons.

> 2) When entering the durations of dive segments, please allow entering 
> the durations that applies to a specific segment only. Look at this 
> screendump from VM-Plan that I sometimes look at: 
> https://www.dropbox.com/s/27q6watntoxayg0/VM-PLan1.png
> The interface is easy to work with because one does not have to do the 
> mental calculation of the total time duration. VM_Plan calculates that 
> automatically.

We are not trying to re-implement VM Plan. Anyone who has VM Plan would
be NUTS to try and use Subsurface for planning. This is NOT a design
goal at all. Ever.

> 3) Allow for a larger range of gas mixes, specifically with respect to 
> EANx and trimix.

Yes, known limitation and intended to be addressed

> 4) The completed dive plans are stored with the rest of the regular 
> dives. I have not managed to discover a way to modify an existing dive 
> plan (e.g. alter bottom duration, change gas mixture). There is a strong 
> need for a more interactive and iterative modeling environment at that 
> level, in the same way that it is now highly interactive in setting up 
> an individual dive plan before saving it.

Yes, editing dives is on the list of features. Not sure if we'll have
that in time for 4.0.

> Having said this, I can only complement the developers who have gotten 
> the planning interface so far. It knocks the hell out of any other open 
> source software that I have seen and it reflects the amount of time and 
> care that has gone into implementing this feature. This is making 
> Subsurface so much more than just a simple dive log program and I think 
> it is so far beyond what the original expectations of the software were.

Yes. And 99.9% of our users will never use this and don't care. And the
remaining 0.1% of our users tell us all the way in which VM Plan is
better and gets things right that we get wrong.

If there is ONE feature that I regret having added, it's dive planning.
This has sucked an incredible amount of my time (for the Gtk version)
and now Tomaz' time with zero benefit to the vast majority of people.

I am still tempted to keep the interactive input and simply drop the
planner as it seems absolutely not worth the pain it is causing. Not to
mention the risk if some bozo decides to dive this plan, gets bent for
whatever reason and then decides that this couldn't possibly have been
their fault.

/D



More information about the subsurface mailing list