112 von 121 Kunden fanden die folgende Rezension hilfreich
Let's Compare Options Preptorial
- Veröffentlicht auf Amazon.com
The review trends of Yau's last book have already started with this edition: "too basic." Maybe we could graph the stats of those reviews, then look at the very topmost band of readers to find the "perfect" audience, vs. the large body of outliers who will trash this as oversimplistic. So, get into alpha, and visualize a bell curve, with "perfect for me" on Y and age/experience on x:
DON'T BUY IF:
--You're in the heavily skewed, lightly shaded, experienced right side of the curve, with even good basic experience in data presentation. I'd include any mid level manager who has decent powerpoints in this group. The colorful pictures are gorgeous, as in Visualize This: The FlowingData Guide to Design, Visualization, and Statistics, and if you have a LOT of disposable income, you "could" buy it just for the pictorial ideas (paper is coated matte, images are 4 color, very high quality book production wise). If you're a post undergrad freshman, you might find the advice too basic. There also are a lot of discussions of data "types" but very little about psych. For example, starting a presentation with the statment "My purpose here is to INFORM" often gets audience hackles down if they're resistant to being sold or convinced-- not much of that is covered here.
--You're a graphic artist or graphic pro, unless, again, you're just looking for pictorial and presentation ideas, and not advice (the illustrations, as in the last edition, are stunning).
--You're very new to data presentation and aren't even sure whether red goes with green or tables are better than scatters in a given situation.
--You're, again, looking for VISUAL ideas to supercharge your presentations, NOT programming tips or even English advice on details. ONE EXCEPTION to this volume compared to Yau's last book: there ARE a good number of example visuals by artists other than Yau (although his are still astonishing), and in THOSE CASES, the author does give the website. In some cases, these are just bigger online pictures of the graphics, in others, there actually is an explanation of the techniques.
Now, for the good stuff. If you KNOW that this book is NOT for pros, you won't buy it, then downstar it because you're disappointed. JUST DON'T WASTE YOUR MONEY if you are looking for comparisons between R and visual basic, steps on translating LaTex and PostScript to .jpg, etc. The level of technical advice amounts to: "R is being used by more and more researchers and statisticians" (and that not until p. 283 of 290). There ARE a number of examples of open source and other software like indiemapper, GeoCommons, ArcGIS, Gephi, Imageplot, Treemap, Tilemill, etc. but the author only mentions them, and leaves you the autodidactic task of figuring out, for example, which do and don't work with Python, RSS, PHP, HTML5 and other pertinent questions pros would ask. But think about this: if you ARE very new to presentation, these tips WILL be eye openers and of great value, as you could surf for hours and not be able to compare or value what's worth it and not. At least beginners get a head start on what this very experienced statistician and author USED throughout the book.
The biggest problem I saw with previous reviews is that the purchasers seemed to expect detailed explanations of how the author created the stunning graphics. This is NOT that book. The software is still not always mentioned with each visual, and steps are really NEVER given that detail "how to" get that effect, let alone scripting, code, or even pseudocode. The book is truly more of a coffee table text showing best practices, as an artist would, but not giving a tutorial on techniques. I know you've watched some tutorials on YouTube that are really "show off" steps by the programmer, with no real intention to show you how to do it. This isn't that bad, as it does have many important "rules of thumb," especially on mistakes to avoid if you're a novice.
So, people who say this is a must buy, or people who say this is a waste of money are both wrong. The solution to that axis of opinion is an intesecting plane visual-- if you're relatively new, don't expect technical detail, and love to get visual ideas and inspiration, you won't go wrong with this volume. If you're expecting to learn tricks and tips in R vs. Excel, get dashboard and data texts on those specific programs instead, and you'll be much happier. Expect a lot of beauty, but not how to get there!!!
Library Picks reviews only for the benefit of Amazon shoppers and has nothing to do with Amazon, the authors, manufacturers or publishers of the items we review. We always buy the items we review for the sake of objectivity, and although we search for gems, are not shy about trashing an item if it's a waste of time or money for Amazon shoppers. If the reviewer identifies herself, her job or her field, it is only as a point of reference to help you gauge the background and any biases.
24 von 25 Kunden fanden die folgende Rezension hilfreich
- Veröffentlicht auf Amazon.com
Many people drawn to this book will find it entertaining and informative. Yau covers a wide range of data graphics from unconventional data art to more conventional statistical graphics. As you would expect, there are numerous examples, often excellent and mostly in color. The tone is informal and enthusiastic. If you want something very light as an overview of or introduction to visualization, this is quite a good choice. If you want a more serious but not strongly technical book on what to do with your data, Naomi Robbins' "Creating more effective graphs" is my best recommendation. At a deeper or wider level, nothing has superseded the works of Edward Tufte, William S. Cleveland, and Leland Wilkinson on statistical graphics.
Yau is broad-minded on many of the debates that now bedevil graphics. He stands aside from the mutual incomprehension or even hostility often seen between groups dedicated to "data art", information graphics or "infographics", and statistical or scientific graphics. He gives some examples of data art, but is not especially articulate on what we are expected to see or on how it should be evaluated. He is not dogmatic on many smaller points, such as whether pie charts should just be avoided. I do not see, however, how citing research that shows that people are poor at comparing angles is consistent with his indulgence. The overall attitude is liberal, even anarchist: anybody's "rules" are just suggestions, not absolute rules.
There is a marked downside to the informality. I found Yau's book to be rambling and unstructured and to lack a clear roadmap. Bluntly put, the graphics are of higher quality than the commentary, which is usually sensible but often banal. You might fairly read the book by looking through the graphics and reading the accompanying text only when interested. A more detailed and informative list of contents, better chapter summaries, bullet lists of key tips, and a better index would have helped browsers, and indeed all readers.
This is a non-technical book: there is no code and the one equation comes with an apology. That is as advertised, but Yau does on occasion assume some knowledge beyond very elementary statistics or mathematics: "sine wave" is mentioned without explanation on p.177; density, meaning probability density, is also not explained on p.195. How far will the non-technical focus suit those readers taken to want to follow up by learning how to program their graphics? The software choices surveyed do, however, start with MS Excel.
Yau's writing is breezy, often colloquial. Antsy, heck, kind of, "like" as a universal conjunction, lulz, nuke, okay, psyched, scrobbles, snotty all feature. "A lot", "a lot of", "a ton of" appear again and again as obtrusive verbal tics, sometimes several times on the same page or in the same paragraph. There are some lengthy personal digressions: photographs from Yau's wedding, how he briefly played a lot of blackjack in his youth, and how he learned to cook from his mother. The wedding example is engaging, but only for the usual reasons, not because the point being made about graphics is especially cogent. The other personal stories come across as labored digressions which obscure what they are supposed to illuminate.
Copy editing and proof-reading were both poorly done. If you regard data as a plural noun, or "compared with" as standard rather than "compared to" or "compared against", or prefer "different from", you will not like the choices here. Those are all debatable points, and my comments go beyond them. The text would have benefited from much more care. "About the author" starts "Nathan Yau has written and created graphics for FlowingData since 2007, a site on visualization, statistics, and design and believes that visualization is a medium that can be used as both a tool and a way to express data". What an awkward sentence! Even if you insert the missing comma, it still needs rewriting. Similar problems recur throughout. Common lapses include unmatched singulars and plurals, misplaced conjunctions and prepositions, missing punctuation, and sentence fragments that appear inept, not acceptable stylistic variation. Trivial typos are also common, even on the graphs: "Feburary" (p.28), "targetedd" (p.150), "Montanta" and "in habitants" (p.236), "SUMBIT" for "SUBMIT" (p.266). Unedited code fragments are visible on graphs at pp.268 and 270.
On more technical points, the author is generally sound, but I spotted a few errors. William Playfair's "charts were handmade on paper, of course" (p.45); not quite so, as printing of Playfair's books depended on copper engraving as well. "Sample population" (p.118) is a statistical solecism, and Yau is a statistician. His use of "dot plot" adds another meaning to an already confused literature, but quite what it includes is not clear. "exponential" is used as if it just meant "big".
References to books and papers are typically either missing or incomplete. Edward Tufte is quoted on p.xiii, for example, but there are no details of which book (or other material) is being referred to (let alone a page number for the quotation). Perhaps the attitude is that you always can Google that. Perhaps the attitude in this example is that you already know in detail about Tufte's work, but that is likely to be wrong for the readership that seems most intended. Short reference lists harm no readers and help many.
So, this should appeal to most of its likely readership. Everyone else, technical people, picky people, and especially picky technical people, may sniff and snort throughout. You have been warned.
14 von 14 Kunden fanden die folgende Rezension hilfreich
M. L Lamendola
- Veröffentlicht auf Amazon.com
Having suffered through agonizing PowerPoint "presentations" that make me yearn for being waterboarded just so I can get out of there, I'm very familiar with being subjected to data visualizations that confuse and annoy. I've also seen this in books, magazines, and online articles.
Many years ago, I worked as a magazine editor. I wrote and edited many articles. In the course of this, I submitted illustration ideas. Some were pretty good. But some were, to be kind about it, inept. I just didn't have that graphical eye to make the intended point. The magazine's graphic artist, who is still there, would ask me a few questions and come back with something that totally nailed the idea I was trying to convey. Sometimes, he would just strip away semi-relevant detail. Other times, he'd add a flourish that made sense of it all. And other times, he'd come up with something quite different from what I submitted.
One time when I told him how much I appreciated his "magic," he said, "There's no magic. Think of your graphics the way you edit your articles." What insight! I told the story efficiently and clearly in text, but didn't carry that thought process over to the graphics.
This concept is one of the major themes that Mr. Yau is relating in this book. And it's one that is typically not understood.
While Yau's intended target reader is the quant who needs to communicate ideas from the data, everyone else can benefit as well. In this modern age of spin, lying, and misrepresentation, especially from those frauds occupying positions of power in the District of Corruption, it helps to understand the difference between an accurate visualization and one that is produced with intent to deceive. My personal estimate, for what it's worth, is 9 out of 10 people are regularly deceived by dishonest visualization techniques. Odds are, you're in that group. Read this book, and you won't be.
His tour of various charting types is extensive, and his explanations are insightful. Most people who create charts think in terms of bar charts and pie graphs, with little or nothing else in their repertoire. But the available chart types go far beyond that. Understanding what your options are can be quite liberating for you.
Selecting the right option can make the difference between a yawn or a wow on the part of your audience. On our Crystalkeen site, my company sells data graphics tools for Crystal Reports. Hop over there, even if you don't use this software, and look at the various chart types. That'll be an eye-opener, and it will also help you see another way a book like this adds hugely to the literature.
This book is lightly technical in nature. Non-quants might not understand some of the references. Statisticians, professional report designers, professional presenters, and others who present "from the data" via visualizations might find this book too basic for their tastes. But I caution such people: I have seen your work, and you are found wanting! Well, most of you anyhow. Your takeaway will be "simplify, simplify, simplify." That, and probably some related tips such as don't add marginally relevant information and don't mislead by "zooming in" on scales.
I think too often people get lost in the wiz-bang, see what I can do with the software mentality and forget the reason they are producing the report or presentation aid in the first place. My impression of this book is its primary purpose is to help the reader correct that deficiency and consequently be able to convey the intended message with clarity, accuracy, and precision. Yau understands this about graphics. He needs to apply it to his writing, however.
My advice to him would be, "There's no magic. Think of your text the way you edit your graphics." He does the reader a disservice with his many grammar gaffes and syntax errors. His word choices are sometimes inappropriate to the text. An example is his frequent misuse of "a lot." This phrase originally referred to the space that physical objects occupied. They took up an entire lot, so you had a lot of them. It would be better, when discussing the non-material, to say "many" or something similar.
I also don't like it when people use a Latin plural noun as a singular noun in a sentence. Data is the plural of datum. I'm willing to overlook this here, partly because Yau knows the material and partly because a common (mis)usage of "data" is to denote a body of data. It's increasingly becoming shorthand for "group of data."
Despite the writing weaknesses, Yau left me feeling that my time reading this book was well-spent. Sometimes, a good subject matter expert can stumble with the language but still help the reader or listener raise his or her game. I feel Yau achieved what he set out to do, and I am the better for it.
This book consists of 7 chapters occupying 290 pages. It's graciously endowed with ample color photographs and color illustrations.