Re: [Freemarker-devel] miscellaneous

[Jonathan R. <jre...@te...>]

Travis Low wrote:
> 
> Since I'm not currently using FM, I try not to pollute the list very
> much.  But I do plan to use, and possibly develop, FM in the future,
> so I feel compelled to add my $0.02 to the recent discussions.
> 
> Featuritis: Adding extra features is never an issue in itself.  It's
> only a problem if it negatively impacts code size, stability,
> usability, or backwards compatability.  

I think we basically agree. Another way of looking at this is in terms
of   cost-benefit analysis. Any feature you add imposes an ongoing cost
in terms of both code maintenance and documentation -- and possibly
answering newbie questions about the feature on the list. If the feature
is of such marginal interest that the benefits do not outweigh the
aforementioned costs, the feature should not be introduced.

So there are very marginal features that should not be introduced for
that reason. However, it does not seem to me that having a standard way
to get the upper case of a string could possibly be an example of that
kind of "marginal" feature.

> I would suggest, that before
> adding major new features directly to CVS, you write a brief design
> note for the list and then invite comments.  Also, you might consider
> providing a clear fault line between the features that most people
> will use, and the features that only advanced users will use.  

Well, it never occurred to me that getting the upper case of a string
was a special-interest feature for advanced users. 

Now I have not been perfect in terms of letting consensus form on
things. The problem there is that I do not have a full set of gears. I
only have neutral and overdrive. There are times where it has been an
order of magnitude less time-consuming just to implement something than
it would have been to propose it and wait for a consensus. And if I am
in overdrive mode, I don't have the patience and just blast ahead
sometimes.

Still, on many occasions, I have tried to float various ideas before
implementing them, particularly where I thought there was a potential
for some controversy. 

There is a sense of mounting tension here, I think, but I do not think
the basis of it is really technical. I think there are
psychological/sociological/emotional reasons for it. Some people may
feel that I've just sort of come in and taken over. And that is not
entirely incorrect. Also, I do have an abrasive personality, which does
not help matters. 

But as I said, I don't think that any tension you perceive is based on
genuine differences on technical issues. It's as if your wife gets
absolutely enraged because you leave the lid of the toilet up. It may be
that there are much deeper issues of much longer standing and the
supposed cause of the argument (leaving the toilet seat up) is a red
herring.

> One way
> to do this is an alternative syntax for those features.  Another way
> is to maintain a "quick reference" of some sort.
> 
> Syntax: I am unable to keep up with the huge number of postings
> (whew!), but after skimming them it seems that most of the extensions
> are of the form ${list@operator}.  How about $#{list} for size of a
> list, $#{string} for the length of a string?  As for "\lt" & co., I
> find it butt-ugly.  What's wrong with using parenthesis for "math
> mode"?  I find it hard to believe that <if (x > y)> could be
> considered more awkward and cluttered than <if x \lt y>.  Besides,
> everyone already knows about parentheses and greater/less than signs.
>   They don't know about "\lt".  One of the cardinal rules of usability
> is that you capitalize on the existing knowledge of users, not turn
> that knowledge into a liability.
> 
> More syntax: $^{string} for uppercase, $_{string} for lowercase, and
> $~{string} for title case.  Just off-the-cuff suggestions.
> 
> Dipshits, etc.: They, too, are part of the divine plan.  :-)  

If you have not had the time to read through all the stuff, I should
make very clear that I did not call anybody on this list a dipshit. I
simply referred to the presence of such people in companies. Since these
people -- unlike vampires, ghosts, and leprechauns -- actually exist,
maybe we should have an official terminology. There is "jerk" and
"asshole" and of course the British people on the list would propose
"wanker". This is a subtle matter because there is a whole taxonomy
there, different gradations. There is also a whole range of various
unary operators that can be used to adjust the intensity level, as in:

#!$%& jerks!!!

> Let's
> try to be polite, and not take the non-technical stuff too seriously.
>   We all want pretty much the same thing.
> 
> FreeMarker/Freemarker: Did someone make a command decision on this?

Yes. It look like it's FreeMarker. Ben Geer said that was his original
intent. That is also the spelling that appears on our  sexy new logo,
courtesy of Mr. Sridhar and the Zerocode folks.

(I'll answer your point on the syntax of the comparisons separately,
because I've actually implemented a new pass on that.)

> 
> Thanks for listening.

Thank you for taking the time,

Jonathan

> 
> cheers,
> 
> Travis
> 
> --
> Travis Low
> <mailto:tr...@da...>
> <http://dawnstar.org/travis>
> 
> _______________________________________________
> Freemarker-devel mailing list
> Fre...@li...
> https://lists.sourceforge.net/lists/listinfo/freemarker-devel