| 07:16 |
gregorynyssa |
Spengler taught that will-to-power was not, in fact, invented by Nietzsche, but is rather the very essence of Germanic thought since ancient times. |
| 07:18 |
gregorynyssa |
In his view, Nietzsche and Goethe were both expositors of the fundamental Western (that is, Germanic) spirit. |
| 07:20 |
gregorynyssa |
asciilifeform: signpost: awt: Was there any discussion of the possibility of an "sbase" written in Ada? |
| |
↖ ↖ |
| 07:23 |
gregorynyssa |
awt: How was your parsing routine for IRC packets designed? |
| 07:24 |
gregorynyssa |
My strategy (using ad-hoc parsing instead of parser-generators) is first to determine whether or not the initial "source" field is present. |
| 07:25 |
gregorynyssa |
Then, find the body/payload which is marked by the first colon which is preceded by a space. |
| 07:26 |
gregorynyssa |
After the body/payload is chopped off, I take the shortcut of using java.util.StringTokenizer to get the command (such as PRIVMSG) and options. |
| |
~ 20 minutes ~ |
| 07:46 |
* |
gregorynyssa will be back later. |
| |
~ 1 hours 44 minutes ~ |
| 09:30 |
awt |
gregorynyssa: may I ask why you are parsing IRC packets? |
| |
~ 29 minutes ~ |
| 10:00 |
asciilifeform |
http://logs.bitdash.io/pest/2023-07-30#1029234 << asciilifeform is baking a wx-based cpp pestron, where worx on all major os (but, sadly, at 'geological' speed, have been at it for almost a yr nao) |
| 10:00 |
bitbot |
Logged on 2023-07-30 14:30:42 gregorynyssa[PeterL|signpost|deedbot]: The program is still operable on macOS but the frame-rate of the GUI drops to 1-2. |
| 10:01 |
asciilifeform |
http://logs.bitdash.io/pest/2023-07-30#1029241 << nope so far |
| 10:01 |
bitbot |
Logged on 2023-07-30 15:21:22 gregorynyssa[deedbot|PeterL|signpost]: asciilifeform: If I might ask, did the advent of GPT 3.5 and 4 exceed your expectations regarding what neural networks are capable of? |
| 10:01 |
dulapbot |
Logged on 2023-02-07 17:16:43 asciilifeform: signpost: thing behaves precisely like asciilifeform imagined in 1990s 'wat if i had over9000 cpu & disk to run my shannonizer' |
| 10:01 |
asciilifeform |
http://logs.bitdash.io/pest/2023-07-30#1029250 << plox to v genesis |
| 10:01 |
bitbot |
Logged on 2023-07-30 16:29:23 gregorynyssa[deedbot|PeterL]: billymg: asciilifeform: I wanted to mention that I would be interested in sharing the source-code of my client. |
| 10:02 |
asciilifeform |
http://logs.bitdash.io/pest/2023-07-31#1029286 << watsa 'sbase' ? |
| 10:02 |
bitbot |
Logged on 2023-07-31 07:20:51 gregorynyssa[signpost|PeterL|deedbot]: asciilifeform: signpost: awt: Was there any discussion of the possibility of an "sbase" written in Ada? |
| |
~ 26 minutes ~ |
| 10:29 |
gregorynyssa |
http://logs.nosuchlabs.com/log/pest/2023-07-31#1029796 << My Java implementation of Pest includes the IRC bridge. |
| 10:29 |
dulapbot |
Logged on 2023-07-31 09:28:58 awt[asciilifeform]: gregorynyssa: may I ask why you are parsing IRC packets? |
| 10:31 |
gregorynyssa |
I saw your old messages regarding the writing of the IRC parser, and was wondering whether you wrote out a grammar for the IRC packets. |
| 10:35 |
gregorynyssa |
http://logs.nosuchlabs.com/log/pest/2023-07-31#1029804 << "sbase" is a project by Suckless (the creator of the dwm(1) window-manager) to re-write "coreutils" so that the code is safer and more concise. |
| 10:35 |
dulapbot |
Logged on 2023-07-31 10:00:51 asciilifeform: http://logs.bitdash.io/pest/2023-07-31#1029286 << watsa 'sbase' ? |
| 10:35 |
bitbot |
Logged on 2023-07-31 07:20:51 gregorynyssa[signpost|PeterL|deedbot]: asciilifeform: signpost: awt: Was there any discussion of the possibility of an "sbase" written in Ada? |
| |
~ 34 minutes ~ |
| 11:09 |
shinohai |
I believe `sbase` the suckless version of base unix tools iirc. |
| 11:10 |
shinohai |
GM pestizens! |
| 11:13 |
signpost |
I pulled in busybox in this place |
| 11:13 |
signpost |
but not after a rigorous comparison with sbase |
| 11:19 |
* |
shinohai loves busybox for some reason ...... |
| |
~ 16 minutes ~ |
| 11:36 |
awt |
gregorynyssa: I used a pre-existing python implementation of an IRC server. I don't recall exactly how it worked but I believe it's in client.py in the blatta source. |
| |
~ 2 hours 22 minutes ~ |
| 13:58 |
awt |
gregorynyssa: what is the purpose of the IRC bridge? |
| |
~ 1 hours 52 minutes ~ |
| 15:50 |
gregorynyssa |
http://logs.nosuchlabs.com/log/pest/2023-07-31#1029818 << I am referring to the code which connects Pest to IRC. |
| 15:50 |
dulapbot |
Logged on 2023-07-31 13:56:26 awt[asciilifeform]: gregorynyssa: what is the purpose of the IRC bridge? |
| |
~ 1 hours 41 minutes ~ |
| 17:31 |
awt |
gregorynyssa: Why are you connecting pest to IRC? Are you writing a bridge bot? |
| 17:45 |
gregorynyssa |
http://logs.nosuchlabs.com/log/pest/2023-07-31#1029821 << I simply wrote my program according to the specification. |
| 17:45 |
dulapbot |
Logged on 2023-07-31 17:29:55 awt[asciilifeform]: gregorynyssa: Why are you connecting pest to IRC? Are you writing a bridge bot? |
| 17:46 |
gregorynyssa |
Please look at the specification, section 2.5. |
| 17:56 |
awt |
gregorynyssa: we tried using IRC clients as pest clients, but it didn't work out well. See for example: http://logs.bitdash.io/pest/2022-06-17#1007244. I'm not sure if that's your intent. |
| 17:56 |
bitbot |
Logged on 2022-06-17 02:44:44 asciilifeform: naturally impossible w/ irc frontend tho |
| 17:59 |
gregorynyssa |
http://logs.nosuchlabs.com/log/pest/2023-07-31#1029825 << I did not know that you and the others gave up on the IRC portion of Pest. |
| 17:59 |
dulapbot |
Logged on 2023-07-31 17:55:09 awt[asciilifeform]: gregorynyssa: we tried using IRC clients as pest clients, but it didn't work out well. See for example: http://logs.bitdash.io/pest/2022-06-17#1007244. I'm not sure if that's your intent. |
| 17:59 |
bitbot |
Logged on 2022-06-17 02:44:44 asciilifeform: naturally impossible w/ irc frontend tho |
| 18:00 |
gregorynyssa |
I have not kept up with the logs. |
| 18:00 |
gregorynyssa |
Thus, I wrote my Java implementation with both the IRC bridge and a GUI. |
| 18:00 |
* |
awt formatted messages |
| 18:02 |
awt |
How would you display the thousands of messages during the initial getdata sync on an IRC client, for example? |
| 18:02 |
gregorynyssa |
http://logs.nosuchlabs.com/log/pest/2023-07-31#1029832 << Have a look at this line within the logs. This is a leaky abstraction. |
| 18:02 |
dulapbot |
Logged on 2023-07-31 17:59:04 awt[asciilifeform]: formatted messages |
| 18:03 |
gregorynyssa |
http://logs.nosuchlabs.com/log/pest/2023-07-31#1029833 << By maintaining a reordering buffer, which can be queried by the operator. |
| 18:03 |
dulapbot |
Logged on 2023-07-31 18:00:39 awt[asciilifeform]: How would you display the thousands of messages during the initial getdata sync on an IRC client, for example? |
| 18:06 |
gregorynyssa |
awt: I am reading the transcript of 2022-06-17 which you linked. |
| 18:06 |
gregorynyssa |
Have a look at this line: http://logs.bitdash.io/pest/2022-06-17#1007309 |
| 18:06 |
bitbot |
Logged on 2022-06-17 16:44:01 asciilifeform: http://logs.bitdash.io/pest/2022-06-17#1007298 << afaik all attempts at gui glue are doomed to be either intimately os-specific (and straightjacketed firmly into whatever knobs $os supports) or 'hobbyist joke'; or, often enuff, ~both~ of these |
| 18:07 |
awt |
Messages come in in order of latest to earliest. If there are 10k messages, do you wait until you've received all 10k to ensure they're ordered properly before outputting to the irc client? |
| 18:07 |
gregorynyssa |
During the early years of Tk, one of the design-goals was close integration with X Window. |
| 18:07 |
gregorynyssa |
Thus Tk fell into the category of being "intimately OS-specific." |
| 18:07 |
gregorynyssa |
However, it was later adapted to other operating systems. |
| 18:09 |
gregorynyssa |
http://logs.nosuchlabs.com/log/pest/2023-07-31#1029841 << You can treat the IRC client as a command-line. |
| 18:09 |
dulapbot |
Logged on 2023-07-31 18:05:22 awt[asciilifeform]: Messages come in in order of latest to earliest. If there are 10k messages, do you wait until you've received all 10k to ensure they're ordered properly before outputting to the irc client? |
| 18:09 |
gregorynyssa |
Wait for 200 messages before sending them to the client. Let the operator query the rest later. |
| 18:15 |
awt |
I believe there is a more recent draft of the spec. I have a local copy but can't find a link. |
| 18:16 |
awt |
Where for ex. there is no longer an order buffer, and there's something like a long buffer but it's called the "Filter." |
| 18:16 |
gregorynyssa |
http://logs.nosuchlabs.com/log/pest/2023-07-31#1029848 << I have only read the 0xFB version. |
| 18:16 |
dulapbot |
Logged on 2023-07-31 18:14:10 awt[asciilifeform]: I believe there is a more recent draft of the spec. I have a local copy but can't find a link. |
| 18:19 |
gregorynyssa |
signpost: Does your implementation of Luby codes support streams of indefinite duration, or only files of known size? |
| |
↖ |
| 18:20 |
gregorynyssa |
Implementing a TCP-like protocol using UDP is a common problem in professional-level game-development. |
| 18:22 |
gregorynyssa |
signpost: asciilifeform: If I am not mistaken, the benefit of Luby codes is that we no longer need a TCP-like flow-control mechanism. |
| |
~ 45 minutes ~ |
| 19:08 |
awt |
gregorynyssa: latest draft: http://loper-os.org/pub/pest/pest_predraft_html.tgz |
| 19:11 |
gregorynyssa |
http://logs.nosuchlabs.com/log/pest/2023-07-31#1029855 << I am reading it now. I can see that a lot has been changed. |
| 19:11 |
dulapbot |
Logged on 2023-07-31 19:06:58 awt[asciilifeform]: gregorynyssa: latest draft: http://loper-os.org/pub/pest/pest_predraft_html.tgz |