Production of good chemical structure depictions is something of an art form. There will likely never be any way to say that a specific representation is "right" and that all others are "wrong". These guidelines don't try to do that. Rather, they are designed to produce drawings that are likely to be interpreted the same way by most chemists.
Throughout these guidelines, various structure drawing styles are labeled with the description "avoid". That word was chosen carefully. It does not say "don't ever consider this". It says, simply, that a particular style might not be understood universally, and that it might be worth considering a different style.
The most important guideline here, as in any style guide, is to know your audience. In general, the more specific the audience for a structure, the less important it is for that structure to follow the guidelines discussed here. A structure drawn on the back of a napkin isn't going to be draw with the same accuracy and precision as one that will appear in a printed journal. There is nothing wrong with that. A napkin drawing has an audience of one, your colleague on the next stool, where a printed journal has a much broader audience. Similarly, the types of structures that are appropiate for the Journal of Very Specific Chemistry might not be appropriate to Chemical & Engineering News or Science or Nature.
The opposite, however, is not true. Structures drawn for a general audience can be understood just fine by a more specific one. Your colleague on the next stool can surely understand a nicely printed diagram if he can also understand your scribble.
Accordingly, these guidelines suggest that you "avoid" unusual, archaic, and ambiguous drawing styles, and instead recommend that you use those styles that are most likely to be understood by everyone.
It is necessary to consider not just your audience, but also your presentation medium. For the most part, these guidelines are written as if you will be using a "perfect" presentation medium, where nothing will detract from the chemical structures themselves. In reality, that may not be the case. If you are preparing diagrams for a low-resolution format such as the WWW, for example, you may want to make your diagrams slightly larger or use a larger font than you normally would, so that they can be more clearly read on the computer screen. Presentations in printed journals, on the other hand, have an absolute maximum width determined by the page size of that journal, and structures have to be sized an dpositioned accordingly. It is certainly reasonable (and altogether proper) to consider how your structures will eventually be presented and processed, and there is no problem in deviating from these guidelines whenever necessary.
The ever-growing use of computers in chemical research provides some special problems. Compared to the number of human chemists, there are very few computer applications designed to process (display, store, search, analyze, etc.) chemical structure diagrams. When creating chemical structures that are likely to be interpreted by computer, you can consider yourself has having an extremely limited audience, and a fairly stupid one as well. Even the best computer programs available today are quite sensitive about the way that structures are drawn. Although these programs will surely become more intelligent over time, they will not be rivaling human intelligence in the near future. Structures drawn in the recommended styles are much more likely to be interpreted correctly by computers than those that are not (and they are also likely be interpreted correctly by humans).
In some cases, no computer software currently available will be able to interpret a depiction that is otherwise competely reasonable and even preferred. We have tried to indicate those cases clearly, and we hope that software will evolve over time. If you need to produce structures that must be interpreted by computers now -- for example, for entry in a chemical registration system or for searching of an electronic chemical database -- it is particularly important to understand the strengths and limitations of the software you are currently using. Again, structure drawings that follow these guidelines are more likely to be interpreted correctly than those that do not.
Throughout these guidelines, you will see two recurring themes: reduction of
ambiguity and proper use of context. With no context, the symbol
might represent 4
tungstens, 8 vanadiums, 17 connected carbons, a wiggly bond, or a diagrammatic
fracture. A simple line might represent a single bond, half a double bond, an
iodine atom, or a negative charge. On occasion, it might even represent nothing
more than a simple line itself. Context is key. The end of one bond should not
touch the end of another unless they truly are both bound to the same atom. Text
should not be placed near the end of a bond unless it is intended to provide an
atom label, or is so visually different from other labels (in font, size, style,
color, or some combination of those) that it couldn't possibly be confused to be
an atom label. If you create diagrams that are difficult to interpret, you
should not be surprised if people have problems interpreting them.
The same is true when creating diagrams that need to be interpreted by
computer. In many ways, computers today are much more demanding than human
chemists. Few programs will interpret a block of text as being an atom label, no
matter how close it is to the end of a bond -- unless the software is told,
specifically, "Hey, that's an atom label". Fortunately, most software makes it
easy to do so. On the other hand, software programs may let you assign specific
meanings to objects that otherwise look identical, so that the symbol
could be made to mean
17 connected carbons without any ambiguity at all.
Whatever your audience, keep it in mind as you create your structural diagrams.
