Top Tips for Using Skimcast

  1. Skim between 1% and 99%. You can set the skim length to anything between 1% and 99% of the length of the original document. The longer the skim, the fuller your understanding of the document will likely be.
  2. 30% is a good place to start. We find a 30% skim is often a good place to start. The document reads well, and you can save some good time. If you feel you're missing too much, raise the percentage. If you need to get the gist of things quicker or can't focus any longer, lower the percentage.
  3. Better or worse than reading the whole thing? Reading a skimmed text is not the same as reading the original text. In some ways it may be better, and in some ways it is surely worse. For many people and with many texts, it may be a quicker route to a fuller understanding of the text. On the other hand, when you skim, you miss some stuff. We find that the algorithm does a really good job of catching and highlighting the key themes, but we can't claim perfection.
  4. So how can I be sure I'm not missing something? It's a skim, a summary, so you can't. A good, attentive reader reading the full text will have a better understanding (but still might miss something). For the rest of us, Skimcast is often better. SO, try it out first by skimming a document you've already read and know well. See how much of the key stuff you get at a 10% skim, a 20% skim, a 50% skim. You will quickly develop a sense of at what percentage you feel comfortable that you're "getting it."
  5. View in Context – like a security blanket. Try this mode. We love it. The summary appears as an overlay on top of the full text of the document. It lets you keep reading out of the skim if you want to.
  6. Try View in Context mode with fiction. See number 7, below, but the former high school English teacher on our team feels fiction reads well while in the View in Context mode. Often the themes list will look a bit like a character list, but that can help when studying or writing a paper.
  7. Works better with non-fiction. Generally, Skimcast will do better with non-fiction than with fiction. This is because fiction is less fact and information oriented. Picking up key facts and themes is Skimcast's strength. Also, a lot of what makes a work of fiction great has to do with dialogue, metaphor, imagery. Skimcast won't focus on these key attributes of literature.
  8. Use Skimcast with study guides. That said, if you're using something like SparkNotes, Skimcast can be a good, time-efficient way to get a sense of the sound of the language of a particular piece of literature.
  9. Use Skimcast with study guides – part 2 Also, if you find you're running short on time and/or attention while using one of the commercially available study guides, you could skim the study guide to really hone in on what is likely the core of the core. Again, not as good as taking the time to read the study guide fully, but stuff happens.
  10. Noise. Some of the themes or highlighted terms will just be noise. It should be pretty obvious. The algorithm is pulling them in for some reason, but if a term or theme seems totally irrelevant, don't stress out about it. Again, we're not perfect.
  11. Better than the average human. On the other hand, the Skimcast skims will probably be better than the average human skim. Why? Well, first of all, the algorithm doesn't zone out or day dream. It doesn't miss a thing. Second, we all have our own personal biases of what's important and what's not. We've all had the experience of "Oh, I didn't realize that was important." Skimcast operates by the rules of language, so while it's not perfect, it's also not susceptible to our many individual biases about what's important and what isn't.
  12. Greatest thing ever? Is Skimcast the greatest thing since sliced bread? Since the advent of the auto-mobile? Possibly. Try it out. Form an opinion. Give us feedback. It would be nice if together we could make Skimcast something that really improves the quality of life and quality of understanding for millions of people. Help us out. Thanks.