Menu ⇓

  9. Bad Gauge Design

    Compilation of some sub-optimal gauge design I’ve come across recently. First up is BBC iPlayer’s speed tester+comparison tool:

    A circular gauge indicates line speed, speed test results are shown in a table above a graphic of a pyramid, a tube and a small stick figure. Below a table is shown comparing your speed to various reference speeds including HD TV, TV, TV, TV, and radio.

    Ignoring the questionable pyramid chartjunk, there are a bunch of issues here — the use of arc length and exceedingly inconsistent scale on the skeuomorphic gauge, the strange choice of colouring on the comparison bars (is green good and red bad? or better/worse than my connection?) and the totally useless “status” column add up to a confusing graphic.

    A series of semicircular yellow gauges show a hash rate of 18 kilohashes per second, a pool hashrate of 211 megahashes per second, a share rate of 0.02 shares per second and net hashrate of 12.60 gigahashes per second

    The next is this set of dials from pool.dogechain.info — again the use of arc length is unnecessary, and very few comparisons are possible. At first, the scales look fairly sensible — perhaps the maximums are pool maximums, or averages or something?

    An identical series of semicircular gauges is shown with different numbers but all still half-full

    Uh, apparently not.

    These dials are always half full. They use up hundreds of pixels of colour and drop shadow, taunting you into thinking you’re getting some sort of useful comparison, when actually they’re informing you that two times twenty three is forty six. Astonishing.

    Simply erasing the colour creates a much cleaner graphic

    which can be further improved by clearing some of the resultant whitespace

    The addition of time-series data might make a genuinely useful contribution, as could some comparisons or proportions allowing your rate to be compared to the pool average or total.

    Another welcome, and obvious addition would be “how much do I make per second/minute”, “how much have I made so far”, “how does that compare to other miners” and “how much is that in real money” — these are the questions that I as a miner actually want answered when looking at a dashboard, but the few relevant statistics which are shown are relegated to a small corner.

  11. Found this interesting piece of from my GCSE Geography project whilst digging through site archives whilst trying to fix some dead URLs:

    Apart from the obvious flaw of hard-to-read text, and the more subtle distortion of the results due to them being overlaid onto a contoured 3D landscape, it’s actually not that bad.