Re: Leo 6.7.3 coming soon

[Edward K. Ream]

On Sunday, April 16, 2023 at 12:24:03 PM UTC-5 Edward K. Ream wrote:

Leo has 100% coverage of most of its core code. Full coverage doesn't 
guarantee correctness, but it's good enough.


Otoh, pydantic and hypothesis are worth investigating.  See #3280 
. This issue has no 
definite milestone. I would welcome help.

Edward

-- 
You received this message because you are subscribed to the Google Groups 
"leo-editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to leo-editor+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/leo-editor/fa6f1b2b-41d8-4680-a9ba-8cd926a580e4n%40googlegroups.com.

Sparks of Artificial General Intelligence: Early experiments with GPT-4

[Edward K. Ream]

This breathless two-minute paper 
 references an arXiv preprint 
 from Microsoft Research: Sparks of 
Artificial General Intelligence: Early experiments with GPT-4.

I highly recommend reading this paper. The researchers have taken Félix's 
suggestion to the next level and beyond.

The preprint is 155 pages long, consisting of prompts to GPT-4 and analyses 
of results. The paper highlights the strengths and limitations of GPT-4 in 
far more detail than popular accounts.

The sections on mathematics are particularly intriguing and mysterious. On 
the one hand, GPT-4 writes spectacularly well. I get a strong impression 
that its responses involve planning. But this impression seems to be 
misleading, as its math mistakes show. 


*Summary thoughts*

I do not understand why GPT-4 (or its predecessors) can create coherent 
text with the limitations revealed in the paper. The more I learn about GPT 
systems, the more mysterious they seem.

Nobody, including experts, can predict where all this will lead. Current AI 
research is true intelligent design :-)

Edward

-- 
You received this message because you are subscribed to the Google Groups 
"leo-editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to leo-editor+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/leo-editor/0795839c-8b49-4691-ae8a-097b00474bb1n%40googlegroups.com.

Re: Sparks of Artificial General Intelligence: Early experiments with GPT-4

[Edward K. Ream]

On Monday, April 17, 2023 at 6:09:09 AM UTC-5 Edward K. Ream wrote:

This breathless two-minute paper 
 references an arXiv preprint 
 from Microsoft Research


I dismissed Dr. Károly Zsolnai-Fehér's YouTube video as "breathless." 
However,  his excitement about what might be called the "composition of 
tools" is well founded. There's no telling what will happen next.

Edward

-- 
You received this message because you are subscribed to the Google Groups 
"leo-editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to leo-editor+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/leo-editor/93c6657c-d08c-4743-ab5d-c9b6132a3a02n%40googlegroups.com.

Impressive ChatGPT responses

[Edward K. Ream]

ChatGPT generates impressive answers for the following queries:

"Disprove the assertion that the earth is flat."
"Prove that the earth is flat."

Imo, the responses to these queries are more compelling than any I have 
seen.

Other example queries :

"Disprove the theory of intelligent design."
"Are vaccines safe?"
"Are abortions safe?"
"Did the Holocaust happen?"
"Is the climate warming?"
"Is the 'Limits to Growth' model reliable?"

The responses to all the above questions seem pithy, informative, and 
correct.

Presumably, truth deniers are hard at work on disinformation bots.

Edward

-- 
You received this message because you are subscribed to the Google Groups 
"leo-editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to leo-editor+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/leo-editor/280eeb3b-d8e2-4a5b-81bc-d5bceac8cc16n%40googlegroups.com.

Last call for reviews of PR #3277: improved path functions

[Edward K. Ream]

PR #3277  proposes 
significant changes to Leo's path-related code. 

I have asked Félix and Thomas for reviews, without recent results. Perhaps 
they are reluctant to comment, which is understandable given the scope of 
the changes. I'll take full responsibility for any problems that may arise. 
I am asking only for reviews, of any kind.

Imo, this PR should be merged into devel soon, so we may thoroughly test it 
before Leo 6.7.3, due next month. 

Edward

-- 
You received this message because you are subscribed to the Google Groups 
"leo-editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to leo-editor+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/leo-editor/94fa5df1-4b47-4d85-8292-336328f18a75n%40googlegroups.com.

Re: Last call for reviews of PR #3277: improved path functions

[Edward K. Ream]

On Monday, April 17, 2023 at 6:46:39 AM UTC-5 Edward K. Ream wrote:

PR #3277  proposes 
significant changes to Leo's path-related code. 

I have asked Félix and Thomas for reviews, without recent results. Perhaps 
they are reluctant to comment, which is understandable given the scope of 
the changes. I'll take full responsibility for any problems that may arise. 
I am asking only for reviews, of any kind.

Imo, this PR should be merged into devel soon, so we may thoroughly test it 
before Leo 6.7.3, due next month. 


I asked chatGPT to edit this post. Here is the result:

QQQ

Pull Request #3277 proposes significant changes to Leo's path-related code.

I have asked Félix and Thomas for reviews, but I have not received any 
recent feedback. It's possible that they are hesitant to comment, which is 
understandable given the scope of the changes. However, I am willing to 
take full responsibility for any problems that may arise. I am asking for 
reviews of any kind.

In my opinion, this PR should be merged into devel soon so that we can 
thoroughly test it before Leo 6.7.3, which is due next month.

QQQ

Maybe not perfect, but chatGPT is a viable alternative to Grammarly, which 
I used to check the original post. Now let's feed chatGPT back to Grammarly.

Heh. Grammarly scores the chatGPT version as "a bit unclear". In all other 
respects the scoring is as before: correct, very engaging and "Delivery: 
just right".

Edward

-- 
You received this message because you are subscribed to the Google Groups 
"leo-editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to leo-editor+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/leo-editor/6bee644f-5c4f-4436-b6d8-cc3366d49afbn%40googlegroups.com.

Re: Last call for reviews of PR #3277: improved path functions

[Thomas Passin]

I haven't been having any problems.  All my external files I've been 
working with open and save correctly.  I've been holding off saying 
anything until I have some more experience with the new code, but so far so 
good.

I have not tried constructing any new external path expressions as yet.

On Monday, April 17, 2023 at 7:46:39 AM UTC-4 Edward K. Ream wrote:

> PR #3277  proposes 
> significant changes to Leo's path-related code. 
>
> I have asked Félix and Thomas for reviews, without recent results. Perhaps 
> they are reluctant to comment, which is understandable given the scope of 
> the changes. I'll take full responsibility for any problems that may arise. 
> I am asking only for reviews, of any kind.
>
> Imo, this PR should be merged into devel soon, so we may thoroughly test 
> it before Leo 6.7.3, due next month. 
>
> Edward
>

-- 
You received this message because you are subscribed to the Google Groups 
"leo-editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to leo-editor+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/leo-editor/93b1e3b8-3cd0-4839-be0d-e4ef9c6627f7n%40googlegroups.com.

Re: Impressive ChatGPT responses

[Thomas Passin]

And there's this slashdot report -

https://slashdot.org/story/23/04/14/2152250/overemployed-hustlers-exploit-chatgpt-to-take-on-even-more-full-time-jobs?utm_source=rss1.0mainlinkanon&utm_medium=feed

People are using ChatGPT to be able to take on additional side hustles.  it 
does much of the extra work.

On Monday, April 17, 2023 at 7:37:34 AM UTC-4 Edward K. Ream wrote:

> ChatGPT generates impressive answers for the following queries:
>
> "Disprove the assertion that the earth is flat."
> "Prove that the earth is flat."
>
> Imo, the responses to these queries are more compelling than any I have 
> seen.
>
> Other example queries :
>
> "Disprove the theory of intelligent design."
> "Are vaccines safe?"
> "Are abortions safe?"
> "Did the Holocaust happen?"
> "Is the climate warming?"
> "Is the 'Limits to Growth' model reliable?"
>
> The responses to all the above questions seem pithy, informative, and 
> correct.
>
> Presumably, truth deniers are hard at work on disinformation bots.
>
> Edward
>

-- 
You received this message because you are subscribed to the Google Groups 
"leo-editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to leo-editor+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/leo-editor/53d46886-5d9c-4aec-9668-3b01d713056cn%40googlegroups.com.

Re: Last call for reviews of PR #3277: improved path functions

[Edward K. Ream]

On Mon, Apr 17, 2023 at 7:08 AM Thomas Passin  wrote:

> I haven't been having any problems.  All my external files I've been
> working with open and save correctly.  I've been holding off saying
> anything until I have some more experience with the new code, but so far so
> good.
>
> I have not tried constructing any new external path expressions as yet.
>

Thanks for your comments. I'll take that as an approving review, but I'll
wait at least one more day.

Edward

-- 
You received this message because you are subscribed to the Google Groups 
"leo-editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to leo-editor+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/leo-editor/CAMF8tS08qbehYT7qRo_rCyVw5TxF5L5QE2zCZA8s%3DU%3Dq%3DyNcwA%40mail.gmail.com.

Re: Last call for reviews of PR #3277: improved path functions

[Thomas Passin]

Some of my outlines have an @rst tree like this:

- @path ...
- @rst .
- [the rest of the tree ...]

I wanted to make sure the the @path directive would be honored when the 
rst3 command and then Sphinx were run.  I just checked this for one outline 
and it worked as intended.  So that's good, too.
On Monday, April 17, 2023 at 8:08:10 AM UTC-4 Thomas Passin wrote:

> I haven't been having any problems.  All my external files I've been 
> working with open and save correctly.  I've been holding off saying 
> anything until I have some more experience with the new code, but so far so 
> good.
>
> I have not tried constructing any new external path expressions as yet.
>
> On Monday, April 17, 2023 at 7:46:39 AM UTC-4 Edward K. Ream wrote:
>
>> PR #3277  proposes 
>> significant changes to Leo's path-related code. 
>>
>> I have asked Félix and Thomas for reviews, without recent results. 
>> Perhaps they are reluctant to comment, which is understandable given the 
>> scope of the changes. I'll take full responsibility for any problems that 
>> may arise. I am asking only for reviews, of any kind.
>>
>> Imo, this PR should be merged into devel soon, so we may thoroughly test 
>> it before Leo 6.7.3, due next month. 
>>
>> Edward
>>
>

-- 
You received this message because you are subscribed to the Google Groups 
"leo-editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to leo-editor+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/leo-editor/fd67f067-84f8-4bfb-ab9d-14e128463c1an%40googlegroups.com.

Re: Last call for reviews of PR #3277: improved path functions

[Edward K. Ream]

On Mon, Apr 17, 2023 at 7:21 AM Thomas Passin  wrote:

> Some of my outlines have an @rst tree like this:
>
> - @path ...
> - @rst .
> - [the rest of the tree ...]
>
> I wanted to make sure the the @path directive would be honored when the
> rst3 command and then Sphinx were run.  I just checked this for one outline
> and it worked as intended.  So that's good, too.
>

Thanks for this testing.

Edward

-- 
You received this message because you are subscribed to the Google Groups 
"leo-editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to leo-editor+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/leo-editor/CAMF8tS0KMkA7UG26bO0_GMcjZqs573CAGAF2A00%2BYbDYa93xQA%40mail.gmail.com.

Re: Last call for reviews of PR #3277: improved path functions

[Félix]

Sorry - I didnt have the time yet: I'll review those changes tonight :)

Félix

On Monday, April 17, 2023 at 8:23:28 AM UTC-4 Edward K. Ream wrote:

> On Mon, Apr 17, 2023 at 7:21 AM Thomas Passin  wrote:
>
>> Some of my outlines have an @rst tree like this:
>>
>> - @path ...
>> - @rst .
>> - [the rest of the tree ...]
>>
>> I wanted to make sure the the @path directive would be honored when the 
>> rst3 command and then Sphinx were run.  I just checked this for one outline 
>> and it worked as intended.  So that's good, too.
>>
>
> Thanks for this testing.
>
> Edward
>

-- 
You received this message because you are subscribed to the Google Groups 
"leo-editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to leo-editor+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/leo-editor/505559ed-b691-486f-8225-a5ca3c5726cfn%40googlegroups.com.

PR #3277 has been merged into devel

[Edward K. Ream]

 

PR #3277  changes 
g.os_path_finalize and g.os_path_finalize_join, and rev 32c9f 

 
adds a missing alias.


These changes are non-trivial. Please report any problems immediately.


Edward

-- 
You received this message because you are subscribed to the Google Groups 
"leo-editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to leo-editor+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/leo-editor/8c26dfb1-1bf5-434a-9a30-7fa62a0cb944n%40googlegroups.com.

Re: VR3 can now be opened in a tab

[Thomas Passin]

Here is an alternative way to delete a tab, where you type the tab name 
into the minibuffer.  The code can go into an @command or @button node:

"""Delete named tab from log frame. Get tab name to delete from 
minbuffer."""
def delete_tab():
"""State 0"""
k = c.k
k.setLabelBlue('prompt: ')
event = {}
k.get1Arg(event, handler = delete_tab1)

def delete_tab1(event):
"""State 1"""
k = c.k
# > k.arg contains the argument.
tabname = k.arg
log = c.frame.log
log.deleteTab(tabname)
# Reset the minibuffer.
k.clearState()
k.resetLabel()
k.showStateAndMode()

delete_tab()



On Sunday, April 9, 2023 at 7:22:19 AM UTC-4 lewis wrote:

> I use an @button Delete-TAB
>
> TAB_name = g.app.gui.runAskOkCancelStringDialog(c,'Delete TAB',"Enter 
> Tab name: ").strip()
> c.frame.log.deleteTab(TAB_name)  # delete named Tab
>
> Is there a way to get a list of Tab names?
> On Sunday, April 9, 2023 at 12:28:45 AM UTC+10 tbp1...@gmail.com wrote:
>
>> Yes, as long as you know its name, which you do from its label:
>>
>> log.deleteTab(TABNAME)
>>
>> On Saturday, April 8, 2023 at 10:26:30 AM UTC-4 jkn wrote:
>>
>>> This look interesting, thanks.
>>>
>>> One thing I have never really needed, but occasionally wondered about; 
>>> it is possible to *delete* a tab in the log pane? (perhaps it should be 
>>> called the 'tab pane'?...)
>>>
>>> J^n
>>>
>>> On Friday, April 7, 2023 at 3:04:48 PM UTC+1 tbp1...@gmail.com wrote:
>>>
 The VR3 plugin can now optionally open in a tab in the log pane instead 
 of in its own panel in the main Leo window (also referred to as a pane in 
 the splitter).  I have attached a screen shot that shows the panel layout 
 that I like when using VR3 in a tab.

 There are two new commands to control that tab behavior:

 vr3-tab -- opens VR3 in a tab
 vr3-toggle-tab -- opens or closes VR3 in a tab.

 I like to use an @button node in the @settings tree to make a button 
 for vr3-toggle-tab. The button runs
  c.k.simulateCommand('vr3-toggle-tab').

 vr3-toggle will close VR3 if open in a splitter pane as well as in a 
 tab.  Next time, the command will open it in the splitter.  Conversely, 
 vr3-toggle-tab will close VR3 in either a tab or the splitter, but 
 will re-open it in a tab the next time the command is run.

 An advantage to running VR3 in a tab is that you can open something 
 else in a new splitter pane without interfering with VR3.

 One minor drawback is that focus will switch to the log pane when 
 something is written there - most likely when the outline has been saved.  
 Then you have to click in the VR3 tab to see it again.  I haven't found 
 this to bother me much.

 This new behavior has now been merged into the devel branch, so it's 
 ready to try out.

>>>

-- 
You received this message because you are subscribed to the Google Groups 
"leo-editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to leo-editor+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/leo-editor/212faf89-7a66-4670-8db8-7b79ab7475d8n%40googlegroups.com.