SamAmco
commented
1 year ago Hi @KonTy I'm sorry but I disagree that this is a good standard behaviour. Looking at the graphs you posted, this graph is perfectly readable:
I personally find it quite confusing because the X-Axis seems to imply that there is a gap of almost a year between the second two points but a gap of only a few days between the first two points. Yet geometrically it's as though the two are equidistant. Still I think it's subjective.
The last graph is much less clear:
There are three points there that don't have a date under them, and the erratic nature of the X-axis does not help you infer the dates intuitively. Perhaps this app has a function where tapping the point will tell you the date. That's all very well but many people using track & graph (my self included) have graphs with hundreds, maybe thousands, of points on them. You rely on the X-axis to gain a sense of the relationship between your data and the passing of time.
Say for example I track exercise, and for a whole year i exercise every single day, but then i take a full month off before exercising again. If the graph behaved the way you are suggesting you wouldn't see that gap at all as the distance of one month between exercising would be squashed to the same scale as the day taken between every other time.
If this feature were to be implemented you might as well remove the X-axis labels almost entirely as they are mostly misleading, perhaps you would just have a start and end label. Still I haven't seen a lot of demand for this and I don't really understand the use case so I am unlikely to implement this.
KonTy
Right now T&G plots graphs based on time periods. It would be nice to have an option to skip time periods when there is no data.
Let's say I log my weight irregularly sometimes once aweek, some other time every day, other times once every 3 months. The point is irregularly. Right now, T&G will pick time periods and plot a graph during those time periods even if we don't have a data point. As a result the line looks jagged and very hard to read. See picute below:
If we had an ability to only plot point which have data and divide graph to equal points that do exist it would look something like this:
In most cases I think this is prefered behavior, because this is the default behavior in all plotting solutions that I've used (ex Excel, PowerBI etc)