Hacker News new | past | comments | ask | show | jobs | submit login

The advice doesn't seem friendly, and ignores valid points. I agree with the above poster. The article is long and meandering.

Let's raise some specific complaints:

>So if you see "2 days 18 hours" that means "66 hours total time spent on this activity".

Completely superfluous. He goes on for 3 sentences explaining that a day contains 24 hours. WOW! AND he couldn't figure out how to keep his units, or make it unitless.

Chart 2:

>Here's how my work week tends to break down:

We break down the day into 4 time frames. Why? When does each timeframe begin and end? If the time is for a "work week" then why the hell does "monday" contain almost 200 hours? That's for 6 months, you say? Why?! It should be an average for a work week! No one intuitively knows how many hours occur on mondays over a 6 month period. Is that a lot of hours, or a few? You would have to do some mental math to even interpret this chart.

Chart 3:

>stacked bar graph

Enough said. It's impossible to comparatively interpret stacked bar graphs over time. You want to use a line graph when you want to compare things over time. Tell me, did he spend more time podcasting on 7/21, or the week after that? "Obvious!" you say, "he spent more time! It's clear!" Look again, however! The next week is 7/28, but the next label is 8/18. What? Where is 7/28? The labels are seemingly randomly distributed, with no rhyme or reason! Eventually you find 7/28 and lo-and-behold... they look about the same, but had to mouse over it to check, didn't you? Because they are separated vertically by nearly the entire chart.

And we didn't even get around to Happiness 3.0 Make a Change App Activity, which contains 41 distinct labels, and you can only read 3 of them at a time, and has all the same problems as the above... Or the Instagram chart that seems to contain the entire message as the label? Does an instagram post take 4 hours to make? What does this chart even mean?

Let's not sit here and suggest that it's "friendly advice" that a person isn't being helpful when they don't defend every statement they make. Anyone who tried to read this, actually read this, would immediately see that these charts are garbage and nearly meaningless. They probably make sense to the person who made them, but as a method of conveying information in a blog post to others, we can criticize without writing a dissertation about the mistakes here.

^ unfriendly advice




curiously, that third graph for me has order 7/7 7/14 8/18 8/25 7/21 8/4

If you didn't mistate the dates, then it might even be changing its random distribution :-)


Hah, in this case, I think the labels were so confusing I forgot what data I was even looking at. Yes you're right, it was 7/14 and 7/21.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: