Re: [xsl] Q: to Jeni Tennison regarding your APress titles...

["M. David Peterson" <m.david@xxxxxxxxxx>]

Jeni Tennison wrote:

> Hi David,
>
>  
Hey Jeni :)

> Right: "Beginning XSLT 2.0" won't be out for a while -- I want the
> book to be as accurate as possible, and the specs are still changing
> (albeit not by much and ever...so...slowly...). My original plan was
> to wait until the end of the Last Call period before finishing off the
> book, though Apress is keen for me to get it done sooner.
>  
I am looking forward to it!

> Meanwhile, "Beginning XSLT" (1.0) has just been reprinted for a second
> time, I guess due to customer demand.
>  
Nice!  They're demanding a good title for sure!

>  
>
> > What would your advice be on using your Beginning XSLT title as an
> > instrument to sharpen and fine tune my skill sets as such that, in the
> > end, I will be a better XSLT programmer because of the effort put forth?
> >    
>
> If you don't want to read the book from start to finish, 
Would be a good approach but for now I have about an hour a day for 
studies related to all languages and technologies I use... possibly 
first part of next year would be a good time to simply sit down and run 
through from beginning to end over maybe a 2-3 day session so I can get 
a better feel for the order and flow you chose for this title.... But 
for now...

> one approach
> would be to look at each chapter's Review Questions and try to answer
> them. 
Ah!  Very good idea...

> I don't give "correct" answers, but the purpose of the questions
> is to make sure you understand the material in the chapter fairly
> thoroughly, so if you find a question is making you think, you could
> take it as a hint to go back and read the relevant section of the
> chapter.
>  
This is a very good idea for an approach.  Will work quite well with 
everything else combined in with it.

>  
>
> >        - Are there specifics areas in which you feel a lot of so-called
> > "experts" tend to miss the boat in how they understand something or 
> > approach particular problems?
> >    
>
> I know that it took me a long time to understand the difference
> between XSLT patterns (which match) and XPath expressions (which
> select), but once you understand this particular distinction many
> things in XSLT become a lot clearer. It's also an area where the
> language used in some books that I've read can be pretty fuzzy.
>  
Agreed.  Although this is an area I feel pretty comforable in there are 
a few that, although I understand them, still trip me up from time to 
time.  Probably the biggest is namespaces and as such I plan to spend a 
good deal of time taking my level of understanding to a much higher 
level in this particular area. 

> Whitespace and namespace handling are other things that it took me a
> long time to get my head around, but they are more rarely important.
>  
I should have read forward one more sentence :)  It seems namespace 
handling is a zinger for a lot of us. :)

> Constructing a good algorithm for performing a calculation using
> recursion is a major skill in XSLT, particularly balancing the
> conflicting requirements of writing readable, maintainable code,
> writing efficient code, and writing reusable code.
>  
Absolutely!  I dont think anybody (except maybe Knuth of course :) could 
claim to know all there is to know about building solid and efficient 
alogorithms for whatever the task may be or even the language being 
used.  They always seems to be a better or faster way of performing a 
task.  It seems to me that with a greater understanding of all the 
little details it will be a much easier task to fine tune and rework 
code such that it takes full advantage of what the language has to 
offer.  Going through this process with your book is something I feel is 
going to help me locate and internalize all these little details that I 
know are out there, I just dont know exactly what they are (obviously if 
I did I wouldnt have the need to go through this process :)  A good 
example of one that came up on the list last week is the proper useage 
of the number() function.  I have always used the number function when 
stripping out non-numeric characters from a string to then compare as a 
number.  Of course, through the help of David Carlisle, I discovered 
that this was an unnecessary step as XSLT is designed to cast from 
string to number (if possible) when doing a comparison.  Something else 
I learned was a fantastic little trick that I had never seen used 
before: using the number function to dynamically determine what element 
to use based on document position (e.g. foo(number(@bar)) would allow 
you to select the element in the position that was contained in 
foo/@bar.  These are the types of little details that I am hoping to 
possibly re-discover during this process.

>  
>
> >        - Are there portions of XSLT 1.0 that are rarely used in 
> > practice (and as such not well understood) that, if implemented 
> > correctly, could make my XSLT code cleaner, leaner, more efficient, 
> > easier to maintain, or easier to fine tune and/or debug?
> >    
>
> Keys are the only thing that comes to mind, though perhaps
> <xsl:strip-space> and <xsl:preserve-space> are contenders. XSLT 1.0 is
> pretty lean: there aren't many features there, so people tend to use
> most of them.
>  
Yeah, even as much as I use them I still find myself having to 
methodically step through my code everytime to make sure that I using 
proper context and such.  This is definitely an error that can easily 
trip up even the best of developers... definitely something that you can 
never spend enough time practicing :)

>  
>
> > - And finally, leading up to your 2.0 release due out in a few
> > months what areas of this title would you suggest as areas that will
> > help me better understand (and therefore implement) the content you
> > set forth in this upcoming title? Or in other words, from your
> > standpoint, what portion(s) of XSLT 1.0 is and will remain as the
> > absolute core of XSLT from now until the end of the foreseeable XSLT
> > future?
> >    
>
> "Beginning XSLT 2.0" will be approximately 75% the same as "Beginning
> XSLT". The essence of the basic processing model of processing nodes
> in order to get results doesn't change: I'd view that as the core of
> XSLT.
>  
Ah, thats a good number to remember.  Definitely will help in both 
learning and teaching XSLT 2.0.

> My other book, "XSLT and XPath On The Edge" is written for more
> advanced XSLT programmers, so perhaps you'd find that more helpful
> (though I think "Beginning XSLT" is better written).
>  
I didnt see that one there but if its not there when I do my daily "fix" 
at B&N I will have them order it for me.  Again, the idea of even doing 
this excercise in the first place is to simply rework my brain tissue in 
such a way as to locate and internalize all the little stuff that seems 
to get forgotten or never realized when you first begin to use a new 
development language.  Its been quite some time since I frst started 
developing with XSLT so this process I think is well overdue.  
None-the-less, its never too late to learn and I'm confident that going 
through this process will be just the learning process I need to kick my 
expertise level up one more notch.

In fact (I'm going to advertise this in a seperate email that is short 
and to the point so that more people become aware of it) I am going to 
document this process using a blog format at the domain 
"UnderstandingXSLT.com" and encourage people to go out and buy a copy of 
your book and follow along with me, discussing through comments the 
experience each person may be having, details they have discovered that 
may interest others, and questions they might have that anybody reading 
might be able to answer.  It would be great if you stopped by to chime 
in on things now and again as I am sure there are many people out there, 
myself included, who would love to hear what you have to say about 
certain areas contained withing your book.

BTW... is there a preferred vendor or URL that you would like me to list 
for people to visit to buy a copy of this title?

Thanks for everything Jeni!  I am looking forward to getting the process 
started and to possibly helping to create a community momentum that will 
help us all understand XSLT just a little bit better than we did before.

Cheers!

<M:D/>

> Cheers,
>
> Jeni
>
> ---
> Jeni Tennison
> http://www.jenitennison.com/