Following up on our discussions below concerning how to write references. I've written a proposal here:

http://www.cfd-online.com/Wiki/index...nts#References

I think that it covers most of the things we discussed. Note that I changed the way we should write author names to Lastname, Firstname instead of Firstname Lastname. Although the I think the later is more easy to read the former is more common and makes the more important Lastname more visible. Hope you agree.

One thing which I haven't covered is the possibility to upload publications to the Wiki and link to them internally instead. This is something we should definitely think about but it has a much wider scope than just agreeing on how to write references - it has potential copyright and owner problems and we need to specify naming and format standards.

I like it. Is it possible to hyperlink directly to a reference? I know we can link to the "Reference" section, but I don't know if one can set it up so that you jump right to the particular link.

As for uploading of publications: I don't see how that would work without explicit permission from the author and/or publisher. This is a form of publication (albeit less formal), and the standard rules would probably apply. Especially if lawyers get involved.

Jason

I have a suggestion about quoting references. If there are several references in a page, then I suggest that we refer them in the text by the author name and year (in case there are multiple references by the same author)

Batchelor (1965)

instead of using some reference number, since reference numbers could change as we add more references and arrange them in alphabetical order.

I am not happy with numbering of equations since there is no automatic way of doing it. If I want to add an extra equation in between and number it then it will be a pain to change the number of all the succeeding equations. I prefer not to number the equations. I think we can manage to do without this if we keep each page length small which is also a good idea in terms of organizing the content. In case you think it is not possible to do without equation numbering see the book "Introduction to Numerical Linear Algebra and Optimisation" by Ciarlet, which has loads of equations and not a single equation is numbered !!!

Good suggestions.

I think that it is better to quote references using author name and year as you suggested - avoids having to renumber them later on as new references are added. And if a reader is already familiar with the paper he can recognize it directly in the text from the author and avoid having to check the reference list. So lets adopt that style.

Also, if there are more than one publication by the same author on the same year we should distiguish between them by adding "a", "b", "c" to the year both in the quote and in the reference list.

We can also have two ways of writing references, either as a part of the text or separate from the text in square brackets. Perhaps both should be allowed, although it might be a bit confusing to have two styles. Or what do you think? Here is an example to illustrate the two ways of writing a reference:

"Following Hinze (1954b) the stress tensor can be decomposed into two parts"

"The stress tensor can be decomposed into two parts [Hinze (1954b)]"

In this case I think the former is easier to read. But the later style makes it more clear that you are indeed quoting a reference and it is more easy to see where in the text references are used if they are within square brackets. Also, if you want to quote a few different references the later style is probably better. Here is an example:

"As shown by several authors [Launder (1991a), Wilcox (1998), Menter (2001b)] the k-epsilon model suffers from..."

Any opinions on this?

About numbering equations. I agree with you that it is preferable if we can avoid this since there is no buildt in support for it and it makes a text more difficult to maintain and change. But still I think there might be cituations when numbering equations might be really benefitial and make it more easy to follow the reasoning. So I'm a bit hesitant to completely ban numbered equations. Perhaps it would be sufficient with a strong recommendation to avoid numbered equations unless it is really necessary? Any opinions? I'm afraid I haven't got access to the Ciarlet book here but I'll try to remember to check it out next time I'm in the library.

It might be possible to hot-link references so that the browser jumps to them directly. Probably needs some html coding though. I'll investigate it later.