Fog Creek Software
Discussion Board

How Do You Like Your Book Reviews?

I've got a couple of programming book reviews on my web site, written in different styles.

The first is in traditional magazine-style, designed to be read from beginning to end like an article:

The second is in more of a Jakob-Neilsen-style, with different aspects of the book collected into sections and key points highlighted to help with skimming:

Which style would you, as a programmer looking for a book to help you with some aspect of your work, prefer?

Is the second one more skimmable? Does it give you just the info you need more quickly, easily, efficiently, or effectively? Would you have time to read the first one if you were trying to decide on a purchase?

Thanks very much for any light you can shed on this. I want to get the format right (or at least useful!) before I post more reviews.

Darren Collins
Wednesday, July 24, 2002

Ho ho ho, the Refactoring review is much better than the C++ one.  In fact, I couldn't read the latter because of all the BOLD staring out at me.  I felt it was selling something to me.  Does Nielsen really do this?

I think the magazine style reviews are bad, because why should anyone buy a book because a book review is skimmable?  The time investment of a book outweighs any time saved by a skimmable review.  The Slashdot book reviews are bad for the same reason.

But I liked the other one.  BTW, does Citydesk have a spellchecker?  You wrote "heirarchy," not that it's really important.

Wednesday, July 24, 2002

I like the chunky reivew style better than the magazine style review except for the bolding.

I compared a Nielsen alert box as-is to the same article without the bolding.  Without the bolding the article has no punch, with bolding it works.

The "Non-Designers Design Book" says there are rules, but the good designers know how to break the rules.

Wednesday, July 24, 2002

Bold text in a paragraph should not be necessary if you already have the sections clearly marked. IMHO it is too much of a clutter. Improve the section head if you feel it doesn't pull the weight of identifying the content that follows.

Skimming only works well if what you skim has relatively few landmarks standing out. And if they instantly tell you what is going on. You now have text like 'only a page or so,' or 'a few minutes' marked with bold, but if I skim and encounter such a text, I have no reference as to what it means. So I have to stop and decide if I investigate or not, making me read part of the surrounding text. Before long I'll have read the entire text, but now in fragments that don't always hold together.

My advise: stick with short, clear markers that have recognisable meaning in the context of a book review - such as what you have now as section markers - and leave the rest of the text unmarked.
If you feel the text contains something noteworthy, rewrite and put it in its own section.

Just my two eurocents.

Erik van Linstee
Thursday, July 25, 2002

Thanks all for your comments so far. And please, keep them coming!

Darren Collins
Thursday, July 25, 2002

There was a recently usability study undertaken that pertains to your question:

The article found the following:

<i>"Overall, there were no statistical differences in search time across the three presentation types. However, the Summary condition was perceived most positively in terms of ease of finding information, being visually pleasing, promoting comprehension, participants' satisfaction with the site, and looking professional. "</i>

So, people may perceive scannable text more positively, but there may be no actual performance differences.


Thursday, July 25, 2002

I prefer the traditional, magazine-style review.  It's more pleasant to read.

Friday, July 26, 2002

(sorry, I meant "Nielsen-style" reviews are bad, not the Refactoring article's magazine type.  But I think everyone noticed I confused the two here.)

Friday, July 26, 2002

*  Recent Topics

*  Fog Creek Home