I have two main comments on this.

1. Back in the summer I produced my own modifications of the  
jmolCheckbox() function, as indicated by the comments in my modified  
version of Jmol.js for 10.2:

        The jmolCheckbox() function has been cloned and modified to give:
        jmolCheckboxMaster() and jmolCheckboxSlave(), which have their own
        counters and generate ids of the type jmolChkboxMaster0 and
        jmolChkboxSlave0, ..1, ...2. These generate checkboxes with one  
additional
        function for onClick(). In the case of the 'master' the updateSlaves()
        funtion sets all the 'slave' checkboxes to the same state as the master
        after the master has been clicked. In the case of the 'slaves' the
        notifyMaster() function determines whether the change of state of  
the slave
        has resulted in a situation in which all slaves are now checked  
(when the
        master must be likewise updated) or a situation in which not all  
slaves are
        checked (in which case the master should be unchecked).

It is in use on my web app at http://doolittle.ibls.gla.ac.uk/motif  
and notified the developers group about it (unfortunately I have not  
the email to hand) and was told that it would be incorporated in the  
next release. It is not in 11.0.RC3. I think it would be a useful  
addition, but if it is not going to be included, I need to know so  
that I can mount a branch on my own website and reference it in  
publications.

2. The SourceForge page at http://jmol.sourceforge.net/jslibrary/  
needs the following modifications to make it intelligible to non- 
cognoscenti. In making these modifications you should treat users  
like adults and tell them what the alternative are.

a. The first section entitled "Why should you use Jmol.js?" should be  
replaced by a section entitled "What is Jmol.js?" explaining what  
Jmol.js provides. As I understand it, it provides two rather distinct  
things, which should be clearly differentiated and spelled out. In  
brief:
(i) Jmol.js provides a way of providing user controls to the Jmol  
applet on the web page. You can set up a page to load a molecule of  
interest and then provide buttons and checkboxes to allow the user to  
change specific features of the appearance of the molecule, its  
background, cause it to rotate etc. (This is possible through a  
technology called LiveConnect that allows web pages to interact with  
applets.) The alternative is to leave the user to manipulate the  
molecule through the inbuilt applet controls or console commands,  
which are only really suitable for users familiar with Jmol.
(ii) Jmol.js also provides functions to control the html and css  
layout of the applet on the web page. Even if you prefer to write  
your own css, the basic javascript for displaying the applet is  
highly recommended. If you prefer to manage your own html and css,  
you can leave the css parameters blank in the functions that accept  
them.

b.  General Recommendations
REMOVE THE CHILDISH AND MISLEADING NONSENSE ABOUT ABSOLUTE REFERENCES  
BEING EVIL!
And that's shouting. But I've raised this point before.
The sentence: "Absolute URLs which start with http:// are evil and  
should never be used."
should be replaced by:
"Absolute URLs (those starting with http:/)" are generally considered  
bad practice and should be replaced by references relative to the  
calling page wherever possible. It is, however, sometimes impossible  
to avoid them (e.g. when pages are generated on the fly from servlets  
etc.)" or something of the sort.
I have to use absolute references for my servlets, and they work, and  
I resent being presented with dogmatic statements with no explanation  
involving severe misuse of English. Computer geeks may know that when  
they use the word 'evil' they don't mean 'evil' in the biblical  
sense, but who the hell do you think you are writing these pages for?  
The pages are meant to help people us Jmol.js. This you do by  
explaining the reason for recommendations, not by blanket statement  
to try to enforce your own html prejudices on people. Honi soit qui  
mal y pense.

c. Link (but there isn't one)
How about a statement that Jmol.js is part of the Jmol download  
package, and a link to where you can download the Jmol.js. After all,  
isn't that what one would expect? And let's make it accurate, unlike  
the link on the present download page which states that the current  
official release is version 10.2, and then leads you to a page in  
which 10.2 is nowhere to be found, and all the versions available  
seem to developmental versions.

David Leader




-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys - and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
_______________________________________________
Jmol-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/jmol-users

Reply via email to