Users of my former free-ware text-morphing product, McPoet, will recognize that a JanusNode is extremely similar to its immediate ancestor, McPoet 5.1, which it renders obselete. Many of the differences are cosmetic and/or narrative. JanusNodes are framed within a more complex narrative in order to underscore the participatory structure of Janus. Janus gets more and more complex and wothwhile (more and more real) as Hir JanusNodes gain access to more and more resources. While I continue to insist upon maintaining my text-morphing software as non-obligating freeware, I want to make my invitation to participate in its continued evolution more explicit and clear. Janus needs you. Please do contribute any JanusNode resources that you develop.
Aside from the cosmetic changes, JanusNodes have a few changes that users of McPoet should be aware of. JanusNodes will run all your old McPoet rules if you make a few simple changes to reflect some of the new terminology. However, please read the documentation, especially about the new functions. JanusNodes have much new and powerful functionality which did not exist in McPoet 5.1.
'Rules' are now 'TextDNA'
McPoet's 'rules' are now referred to as 'TextDNA'. You need to replace calls to functions which had the word 'rule' in them with calls to the new functions which use the word 'TextDNA' instead. If you do a global search-and-replace on all your old rule files of 'rules' and then 'rule' by 'TextDNA', that should do the trick.
Note that JanusNodes ship with all the old McPoet rules (plus many more) so you don't need to translate any of those old rules yourself. Moreover, if your old rules didn't make function calls to functions with the string 'rule' in their name, then they should run without further changes.
'Macros' are now 'TextDemons'
McPoet's 'macros' are now called 'TextDemons'. If you want to use old macros (other than the ones which shipped with McPoet, which also ship with your JanusNode in their new translated format) you need to do a global search-and-replace on your macro field, renaming all the 'macro' prefixes to 'TextDemon'.
The 'SmutLevel' Function is gone.
McPoet had a function called 'SmutLevel' which allowed users to control the level of profanity which appeared in output. This function became obselete with the introduction of TextDNA sets and TextDNA subjects, which give much finer-grained and general control over the nature of your productions. The old SmutLevel function is no longer documented (except insofar as the JanusNode documentation mentions that it is no longer documented, and then comments indirectly on the self-contradiction inherent in doing so) or supported. Make different rule sets if you want to have some private rules for your own pleasure.
The 'Scan Lines' function is gone.
The 'Scan Lines' function has been removed. It may be re-introduced later if it can be sufficiently imrpoved.
JanusNodes have no hard-wired Markov chain files
When the Markov-chaining function was enhanced with the release of McPoet 5.x, all the previously built-in Markov files were removed. You can now store as many Markov files as you like in the 'Markov files' folder, so you can re-create any of the old built-in files that you want to.
The 'TellFortune' function is now open
The 'Tellfortune' function is now handled by a 'tellfortune' folder, rather than being on a 'Subjects' menu. Fortune-telling rules are now treated like any other rules, and can be selected using the new 'textDNA:' pop-up menu.
eecummingsfication is now user-configurable.
The eecummingsfication feature is now user-configureable (although it works just as it used to without any further configuration): see the documentation for details.
TextDNA files beginning with a '0' are now invisible to the 'Subjects' menu
When loading files, JanusNodes will not take into account any TextDNA files whose name begins with a '0' (zero) although such files can be loaded by function calls. This is to allow users to make only the 'entrance file' to a series of chained TextDNA files visible. It will only cause a problem for McPoet users if they happened to name an old rule file beginning with a '0' and can't understand why they can't see it.
Subjects ending with a '0' are now invisible to the 'Subject' menu
JanusNodes will not put any TextDNA subjects whose name ends with a '0' (zero) into the 'Subject' menu. This is for the same reason as above: to make it obvious to users to operate complex, chained sets of TextDNA. If you want a user to run a textDNA set using a particular starting line, make sure it is the only line classified under a subject that does not end with '0'.
'ing', 'ed', and 's' are now keywords
The strings 'ing' and 'ed' are now handled intelligently within textDNA- your JanusNode will attempt to add them correctly to the preceding word, even if doing so requires a little backwards editing or the addition of new letters. This is only a problem if you happened to include 'ing' or 'ed' within an old rule in some idiosyncratic way. If you merely intended that they be added to the end of the previous word, then your JanusNode will do a better job at that task than McPoet used to.
The 'assign' function is smarter.
The second argument of the 'assign' function need no longer necessarily be the name of a BrainFood field. If it is a quoted string of words separated by commas (but containing no spaces), then 'assign' will randomly assign one of the words in that string to the global variable. If you've done much rule-writing, you'll know this is a very useful change.