Re: Exporting data into a file, OpenWrite
- To: mathgroup at smc.vnet.net
- Subject: [mg96763] Re: Exporting data into a file, OpenWrite
- From: Bill Rowe <readnews at sbcglobal.net>
- Date: Tue, 24 Feb 2009 05:45:50 -0500 (EST)
On 2/23/09 at 5:04 AM, siegman at stanford.edu (AES) wrote: >In article <gnq6s2$87n$1 at smc.vnet.net>, >Bill Rowe <readnews at sbcglobal.net> wrote: >>When you use Export, the file will be created in the current >>working directory. That is the directory that is returned by >>Directory[] >If a slightly expanded version of your paragraph above had been >included in the Export[] Help file, for example in the "More >Information" section, it would have told the OP where his data would >be sent to (which you'll note was actually the key question in his >opening sentence). >But more than this, this simple addition to the Export[] >documentation could gently help educate novice Export[] users to the >concept of a "directory" and of Directory[] type commands -- right >at a time when they need this education, and are ready to absorb it. >Is this less than optimum result for the Export[] help command just >one particularly unfortunate example, that leaps out from the page? >-- or a classic illustration of a general weakness in Wolfram's >approach to user documentation? It's certainly the former -- my >view is, it's also very much the latter. There are several trades to be made when documenting anything as complex as Mathematica. True, it would not be difficult to add details of where files are created in the documentation for Export. But doing *just* that leads to an inconsistency in the documentation since the same information is applicable to any operation that creates files. That inconsistency can be removed by adding the same information to *all* commands it is pertinent to. Now once you do this consider the impact on maintaining the documentation in a consistent state. Changes to where files are kept now need to be made in several places instead of just one. As a consequence, more effort is needed to maintain the documentation and there is far more opportunity for error. This argues strongly for not repeating the same information in multiple places. There is no right/wrong answer as to where to document details of where files are created. No matter how Wolfram documents this information, there will always be someone to argue it should have been done differently.