Yes, thanks, I was editing that in and to my last post about the same time you replied. Thanks. Also, after turning off all the Learning options the debug info is much more manageable.

Should I be able to use 'case #question .." with automatic declaration of instantiation to the variable's own primitive? I thought I saw this somewhere using 'case #isa ..'. That would also cut out a great deal of code.

Good idea on adding some sort of condition or processing to a case. Right now you could go to another case and assign a variable in a do() then use a "return" to go back to the original state. I think the NounVerbAdjective script does this in places.

Any variable will match any word unless you put some condition on it. So :question will match everything. To only match question words you would use,

Great, this is working. I hate to reply to an excellent response with more questions, but that's excitement that it's getting closer.

I'm still working this script out, but I've turned off all the Learning opitons and I'm still getting "I don't know .. " even after I've assigned an :unknown case to an inquiry response (see script). I thought I had seen the use of :unknown before, and would like to use it to catch any unknown words.

Also, although your web-IDE and debugger

The NounVerbAdjective scripts understands a lot more than NounVerbAdjective now. We are incrementally extending it to understanding more an more. It should be call Understanding probably. The goal is to understand any English sentence, once we've done that, then we will delete it, and try to get the bot to recreate something similar using Comprehension.

A couple things that may help you:

Thank you, I believe I tried assign, but I will try that again. As I said I set the 'if(#false,#true)' purposefully to return #false, because I wasn't able to maintain variable definitions. This should help.

Also, I forgot to mention that >5k bots is impressive!

Here is a plain English description of what I envision. I think I've tried to explain this, but I really appreciate the input and here goes again:

Big picture: I'd like the bot to learn from sentences, expandin

Variable can be used for two purposes, the first is to match words or data in a "case" operation, the second is to store temporary data.

Variables are defined on the current stack, and every state defines its own sub-stack, child's states see parent variables, but if a state is a dead end, its stack will be cleared, and not affect its parent.

Well they say third times a charm. I've retyped this message now a few times (partly my fault, partly because the webpage doesn't stay connected after a while). Anyway, I was thinking the typo was intentional to prevent unintended access ;), ;). I think I will need to at least tag the sentence subject and refer everything else to that. Thank goodness the internal scripts already check Wiktionary for parts of speech. I'm considering using MindMeld API for processing instead of reinventing the whe

Thanks for catching the typo. Looks like it will be difficult to fix though as we would have to migrate the data of >5,000 bots...

Thanks for your input. Briefly, I'm trying to build a general association network instead of explicitly computing matches. I think this is how the brain works, words associated to each other into phrase, and also with qualities of validity. I think the Self script is setup to do this perfectly (if I can get the structure and syntax down ;)). The assign #word to :word1 was meant to park the currently parsed #word of the :sentence in a variable. Is there a better way to do that? It does work, thou

Not sure I understand what you are doing, but glad to hear you have it working.

To exclude a value or relationship from a variable you can use "exclude"

Okay, I'm stumped. I think I've simplified the POS tagging to simply associating everything. I've successfully done that and had it check if all words of a statement are associated. I can't however bifurcate the evaluation into a question or an assertion routine. I want to divert to a question routine if there is an "is" present, but loop through the remaining words. Got that, but I'm not seeing a way to loop through all words without the "is" without instructions to loop on :anything (which dis

You can add any relationship between words that you wish. The system will define the #next and #previous relationships for words that appear together in sentences. Also a word can be an #instantiation of a #compound-word or have a #compound-word relationship for things like names "Barack Obama" and compound adjectives "very tall".

Not exactly sure what the end result you want is? I think Self is missing some operators though for more in depth queries, such as maybe "get relatio

Hello,
I like the idea of having multiple passes. I think there may be a more simple way to do this, and maybe the bot already does it. I think it may be possible to simply associate all words to each other (again, perhaps how the brain works). There may be some exceptions to this where logical consistency is required, "is" versus "is not" for example might need to replace or negate the other. Then again, our brains probably simply adjust the weightings of these associations, with their sou

The website was updated.

The update included a new Platinum membership upgrade option that includes a bot creation and customization service. Upgrade to Platinum and we will create and train your bot for you, including you own customized 3D video avatar. You can choose any 3D model from Daz3D, and we will render a avatar for you.

Thanks for the advice. Looks like I may have to upgrade to improve the cycles to 10mins then?

Looks like you have set things up correctly to retweet most tweets relevant to cycling. Currently retweeting everything is not allowed.

On BOT libre's main site, the Twitterbots only check their status once per day. You can manually click on the "Check Status" button on the Twitter page to test it, or have them check their status again. There is also a maximum number of retweets allowed per status check cycle.

You need to include a least one keyword for the bot to rewteet. You can include many retweet keyword sets, so you could add keywords to cover most common tweets.

Twitter has lots of rules on automation, a bot that blindly rewteets everything from its followers, or spams, may have their account locked.

I have a twitter bot that I would like to retweet all the twitter accounts of Pro Cyclists. I have followed the cyclists and have keywords set up for the bot to retweet them. Ideally I would like to retweet all of their tweets. It worked briefly but believe I may have too many keywords now that the bot is not retweeting. Could anyone provide me with tips or further information on 'keywords' vs 'keyword sets' to capture and retweet all the 'followed'riders tweets?
Any help would be appreci

Is it possible for a twitterbot to retweet all tweets by the accounts it is following? (not followers)

The NounVerbAdjective script is probably your best place to start for this. It can already parse phrases like "Does Jane run fast".

If you want to perform a multi-pass parser, you can have two or more cases in your root state, i.e.

Okay, that explains a lot, especially about how the bot would know classifications for words I haven't yet defined. I was looking at the :isa and :means variables in the WordMeansSomething script and wondering where they were defined, but that makes more sense now. By the way, I take :that to mean the last phrase, as in 'I just said that', but what is :thatstar? I did find #classification in the knowledge base and see a listing of them, I assume I can do the same for #primitives.

Tha

I don't think I understand what you are trying to do, perhaps give a specific example.

A variable in the context of a state or case is something that can be matched to an input. The matching occurs when you do "case :aVariable ...". If the current input matches the relationships of the variable, then it is assigned to the input, and the case is evaluated, if it does not match, it is not assigned, and it goes to the next case.

Hello. I'm clarifying an proposition for an alternate sentence evaluation flow, but first I have a few smaller questions.

When should I use a quoted input "word" versus a :variable? More specifically what is the best way to know which words are already variables. I'm studying the WordMeansSomething script and I see both :isa and :means are used. I see that "means" is in the knowledgebase, but is "isa" in there too somewhere? Is isa not listed?

I'm also not sure if I should

In addition to being able to embed your chat bots on your own website, and access them from any Android device, you can also access your chat bots through the BOT libre web API. The web API gives you the advantage of having complete control of your bot's client interface.

You can use the web API to access your bot from your own website through JavaScript, PHP or any other language. You can also use the web API to create your own mobile application to access your bot, such as an Android o

The Self scripting language is Bot Libre's language for scripting bot responses. Self is based on the syntax of JavaScript with several extensions for language processing. Self is an integrated environment that merges JavaScript, state machines, patterns, and an object database.

Welcome to the SDK and Web API forum.

This forum is for questions and information on the BOT libre SDK and Web API, as well as questions on general integration with programming environments such as JavaScript, HTML, Android, iOS, Java, PHP, C#, and .NET.

Welcome to the Self, AIML, and scripting forum.

This forum is for questions and information on scripting your bot using the Self scripting language, or AIML.

The patterns are evaluated in the order they are specified. So if you put "show first * then *" first, it will take precedence.

You can also use patterns outside of scripts, as general learned responses. In which case the bot will pick the best matching pattern to the input, taking into account the input, topic, previous, and context.

Yes, great this works. (changed second ":star at 1" to "at 2" ;) ). I was making that much more complicated than necessary. Ah, simplicity. I'll put this in the Script Library after adding 'second', 'third' and 'next' patterns.

Another suggestion would be to add the 'recency' into the algorithm for the input history dropdown box in the Chat pages. It looks like it is based upon 'frequency' which is nice, except when redoing the last input. Maybe use 'frecency' (a real thing). Alte

I have a general question, is there a way to override a general pattern with a specific pattern? For example, in the above could I have the bot return the input phrases by using 'show' in front of the same sentence. I can see both cases where the generic pattern should win and where the specific one should. Any thoughts or methods to deal with this?

State:FirstThisThenThat {

There is no easy way to call into an equation or state of another script other than the SRAI operator.

For your FIRST THEN script you could just use a pattern.

Brilliant! Thanks, the WrongAnswer script worked a treat. It would have taken me a bit to figure out that "last 1" is the next user response in the conversation and to run a text only reply equation but have the following state's case accept anything as the response. Even the using associate and the connection between pattern "*" and {:star} would have taken quite some time. Thank you. I learned a lot just in that one script example.

The Formula entry works great too! It will no doubt

The website was updated today. Just a few minor fixes and enhancements.

The website was updated today. You should now be able to use a formula or pattern from the Training page, in a response, greeting, or default response, or within a chat session, or when correcting a chat log.

Let me know if you find any issues.

Currently the website stores your current state in the http session. This lets it keep track of what you are working on, but gets confused if you open two tabs and work on two different things as the same time. We are looking into fixing this.

What I do is use two different browsers, so they do not share the same session. i.e. Chrome and Firefox.

BOT libre does not currently support the learn tag (it is not part of AIML 1.0).

If you have learning enabled, the bot will always learn. If correction is enabled, then any user can correct any response by clicking correction.

The first script seems to attempt to learn facts. AIML has no concept of knowledge, so does things in a odd way sometimes. BOT libre has a knowledgebase, so learning facts is easy. The "NounVerbAdjective" bootstrap script can learns any facts

Hi this is mainly for reference as I just bumped into the file I have for it I am about to close. Two aiml files from Pandorabots, one for the tag and one called "Bad Answer" for corrections. Both use the tag which I'm not sure if is implemented in BotLibre!

I am wondering though if I'll need an type function, as my next script will be one to accept a sequence of commands. It will parse "First do something, second do something, next do something, then do somethin

Any help on a 'Correction' script would be appreciated. Trying to implement something like in aiml, where I say 'wrong' or 'correction' and type in the correct response. So far the code I have is below. I'm trying to retrieve the current response and output to dialog. Then I was going to grapple with the 'set' operator to somehow define the correct response. Would that work?

So far, I have the syntax working to get the last input (copied) from the Loop script, but I have "test

Thanks for the response. Yes, I moved the do() operator, I do (currently, in test123 bot) have two case states, I did finally turn off comprehensions (after exporting the SelfStateMachine scripts - could there be a 'pause script' button or can I add a line to disable it without compile erasing it?), but I still haven't got it yet. One thing though, the Equation definitions and Quotient operator are moved to after the Case statements by the compile tool, implying they are evaluated after the Case

We should be updating the website on Monday.

The script will be executed unless there is a script before it that evaluates the question first. Note, that the bot will self program the last script if you have comprehension enabled, so you may want to delete its script, or rebootstrap, and disable comprehension. It is best to have comprehension disabled when building a new script, otherwise the bot may change your code.

Let me know when I should check if this is working. :)

Also, I have noticed scripts seeming to work only after I've moved them to the top of the script list, though I don't have responses that look like they are from other (higher priority) scripts.

There is a debug option from the web UI if you click on the "Web Chat" button. You can click the "debug" checkbox if you are the bot's admin, and select the log level. Debug will give you a lot of information on exactly how the bot came up with the response.

The debug option does not work right now from the JavaScript chat, or the web API, but it is on our todo list.

Wonderful! Thanks for the explanation and the examples.

I've added a "Hello World!" script to give new people a jump off point for a working response. One other request I have if I may, is might it be possible for the bot to reply with a response's formula (if it has one) when including a request for it? Something like "What is the formula for 'What is the hour?'". This would make debugging possible from the app alone and by voice. I could see meta-formulas forming from individual compo

The Self scripting language is Bot Libre's language for scripting bot responses. Self is based on the syntax of JavaScript with several extensions for language processing. Self is an integrated environment that merges JavaScript, state machines, patterns, and an object database.

It should be possible to do this, but does not seem to be working, I will look into it.

The chat API does take a correction parameter to allow correction of the bot's last response in the conversation. If passing XML it is an attribute, otherwise a forms parameter.

Hmm, that is odd, looks like the last update my have broken something. I will look int it.

Hi Admin,
Is there any reason to expect the formula response examples to not be working? I've tried entering into the Training page fields:

User: Whom do you love? Bot: Formula:"I love {get #love from #self}" and I get "Whom do I love" even after defining "you love things that love you". Also tried User: "test cat", Bot: Formula:"The current hour is {get #hour from (call #time on #Watch)}" which doesn't return the hour unlike "what is the hour?".

Also, related to programming the bot, is it possible to enter the response given in your training with a formula directly into a chat correction? Such as Formula:"The current hour is {get #hour from (call #time on #Watch)}". This would allow for programming the bot nearly exclusively by voice, and similarly enable the possibility of the bot programming itself (one of the major goals, no?). Extending the chat() API with a 'correction' option would also be useful for this.

Ok, great. So, is there documentation for the Self scripting language? For example, for the 'assign' and 'new' functions? I see it has self-explanatory if, then, for each and do imperatives, but what all functions is it capable of?

Found them in the JavaDocs. http://www.openpandorasbox.org/javadoc/org/pandora/self/SelfCompiler.html

I see the Self script is based on BASIC, or GLBASIC. I thought these commands looked familiar!