Oh, and if you are only interested in attention values, that is "a lot easier". The attention bank already wraps all calls to change the attention values; and adding instrumentation there, to track what is going on, "make sense". Its .. easy, even, so, yes, please do.
--linas On Mon, Apr 15, 2019 at 6:14 PM Linas Vepstas <[email protected]> wrote: > > > On Wed, Apr 10, 2019 at 1:58 AM Juffin Hally <[email protected]> > wrote: > >> Perhaps writing such a visualiser can be attempted? Is there any actual >> need for it (apart from me thinking it'd be helpful, I mean)? >> >> Another point would be documenting the attempt; there might be someone >> working on it right now, but in secret, so far as I can tell from other >> public resources such as the wiki and the slack chat rooms. >> > > I don't think anyone is intentionally working "in secret", although there > is a tendency to work quietly, and not talk very much. It would be cool if > people talked about their work more often. Anyway, I'm 99% sure no one is > working on this now. You are welcome to try it. (create a github issue to > track progress, and post to mailing list if your actually getting > somewhere.) > > ... but!! A few comments about the design of such a thing. The atomspace > does currently have a callback-notification mechanism for when atoms are > added/removed. Its .. not used much (or at all??). I kind of very much want > to rip it out, because its in the critical performance path. Currently, > you can add-remove atoms at about 500K/second and anything that slows that > down would be bad. The current notification system is in that direct path, > and, misused, can quite easily slow down that path by factors of 10x or > 100x .. even by a factor of 2x even if you were super-careful and clean > and elegant. I don't want to pay a 5% penalty, nevermind a 2x penalty. > > So it would be cool if you (or a mythical "someone"?) figured out how to > monitor the atomspace from a different thread, (i.e. outside of the > critical performance path) so that if users don't care about notifications, > they would not have to pay any penalty. Exactly how to make this fast & > efficient, I don't know -- its a challenging, hard problem. I just want to > avoid the problem of paying a performance penalty for a "gee whiz" feature. > > --linas > > >> On Monday, April 8, 2019 at 11:10:04 PM UTC+3, Ben Goertzel wrote: >>> >>> We had some dynamic visualizations of certain quantities... emotion >>> levels within OpenPsi, some parameters of ECAN etc. .. .but never a >>> dynamic visualization of Atomspace itself... >>> >>> On Mon, Apr 8, 2019 at 6:34 PM Juffin Hally <[email protected]> >>> wrote: >>> > >>> > As far as I'm aware, there're ways to visualize atoms in a static >>> atomspace -- the AtomSpaceExplorer can be used for that. It's also possible >>> to post/get atoms to/from ana atomspace and see the results in an Explorer >>> as well. >>> > But I couldn't find any correct & still working way of using the >>> publisher/subscriber. So, if I create a new atom and link it to some other >>> atom, I'd currently have to refresh the explorer page manually. And if I >>> want to see some value change in real time (say, an attention value of some >>> atom), I don't really know how to. >>> > >>> > So, are there any examples of dynamics visualisation? >>> > >>> > The best I could find was the workbench repo, which apparently was >>> created 5 years ago and doesn't seem to be maintained. >>> > >>> > -- >>> > You received this message because you are subscribed to the Google >>> Groups "opencog" group. >>> > To unsubscribe from this group and stop receiving emails from it, send >>> an email to [email protected]. >>> > To post to this group, send email to [email protected]. >>> > Visit this group at https://groups.google.com/group/opencog. >>> > To view this discussion on the web visit >>> https://groups.google.com/d/msgid/opencog/5904ef06-7075-411f-85bc-f18cfa3e3569%40googlegroups.com. >>> >>> > For more options, visit https://groups.google.com/d/optout. >>> >>> >>> >>> -- >>> Ben Goertzel, PhD >>> http://goertzel.org >>> >>> "Listen: This world is the lunatic's sphere, / Don't always agree >>> it's real. / Even with my feet upon it / And the postman knowing my >>> door / My address is somewhere else." -- Hafiz >>> >> -- >> You received this message because you are subscribed to the Google Groups >> "opencog" group. >> To unsubscribe from this group and stop receiving emails from it, send an >> email to [email protected]. >> To post to this group, send email to [email protected]. >> Visit this group at https://groups.google.com/group/opencog. >> To view this discussion on the web visit >> https://groups.google.com/d/msgid/opencog/715c7720-d569-4cc8-a3b1-b609dbc3dafb%40googlegroups.com >> <https://groups.google.com/d/msgid/opencog/715c7720-d569-4cc8-a3b1-b609dbc3dafb%40googlegroups.com?utm_medium=email&utm_source=footer> >> . >> For more options, visit https://groups.google.com/d/optout. >> > > > -- > cassette tapes - analog TV - film cameras - you > -- cassette tapes - analog TV - film cameras - you -- You received this message because you are subscribed to the Google Groups "opencog" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send email to [email protected]. Visit this group at https://groups.google.com/group/opencog. To view this discussion on the web visit https://groups.google.com/d/msgid/opencog/CAHrUA34CBQwzBc3%2B66GWJrmDTFAuq%3De_Pmv3y2svBiZzq9-MYg%40mail.gmail.com. For more options, visit https://groups.google.com/d/optout.
