freenode/lisp - IRC Chatlog
Search
14:39:47
warweasle
As a bonus, this system could be used as an editing tool...like the heart of a distributed text editor.
14:40:43
aindilis
warweasle: git is insufficient? replicating something as complex as git would be difficult I would think
14:41:18
aindilis
it would be cool to have an index of who has what projects, and to see where people are aligned and could team up
14:41:26
warweasle
Git is just diff and patch with hashing. This would allow you to create documents within documents.
14:42:57
aindilis
I worked on a now defunct project called POSI to coordinate what people were working on by having them list their goals, and coordinate to achieve them: https://frdcsa.org/~andrewdo/writings/flourish-2009.odp
14:43:36
aindilis
the main innovation was using RTE (recognizing textual entailment) / NLI (natural language inference) to figure out which goals were identifical
14:44:14
warweasle
My end idea was to use "safe-lisp", this sexp-database/revision mangager interface and some gui tools to create new applications based on your data. As you need them.
14:49:06
warweasle
LdBeth: Similar to basic javascript. It can't interact with anything outside itself. Unless you plug in something like a DOM or some other connection.
14:49:56
warweasle
Anyway, it was just a recursive diff function. But I couldn't figure out how to work with comments.
14:50:51
warweasle
I could apply a "diff" command which was just a list of what to change in a tree.
14:51:05
aindilis
I keep a large Knowledge Base in Cyc of people on IRC and what they're working on...
14:51:50
aindilis
beach: do you think better organization is critical to the expansion or at least mantainence of the community?
14:52:09
warweasle
End the end I wanted to create a single application which could replace all others.
14:52:17
LdBeth
warweasle: I had similar plans on diff, but I choose a “flat” representation instead of trees(or linked list), and comments can be preserved well even they’re multiline
14:53:18
beach
aindilis: I think the community is doing quite well, and it appears to be adding a few people regularly. What I think we ought to do, but it is hard to obtain, is to factor more work so as to avoid duplicate effort.
14:53:39
LdBeth
Although diff is not my first concern, the intention is to be compact and share as much as possible
14:53:50
warweasle
The basic idea is there are only a few "basic" interaction types. Text, rich text, vector graphics, etc. But embedding them into complex documents, you don't need new apps anymore. The app will form around your data.
14:56:59
aindilis
I'm interested in learning how to persuade others both for the sake of lisp and other similar ecosystems I rely on but mostly since my project for the last 20 years has had no new members
14:58:45
beach
aindilis: Given what I now know about the psychological forces behind this issue, I think you can basically forget about persuading others to change their ways.
15:00:23
aindilis
I have two printouts on my wall: "Slow and steady wins the race", and "Show people something that works"
15:01:57
aindilis
:) is there any shared hosting that I can use to host some lisp related infrastructure work? I lost 9 hard drives last year and that's my main deterent from working more on these things?
15:03:11
beach
aindilis, LdBeth: Most people who need convincing have a very long-term investment in something significantly different. They are often experts in their group, and enjoy a lot of prestige.
15:03:16
beach
When they are confronted with the possibility that their investment was the wrong one, they react by denying that possibility. They may even use ridicule and even stronger tactics to avoid that anyone in their group may want to test the better thing.
15:03:17
beach
If not, they may risk losing the prestige. This is not a conscious reaction. They basically can't help it. They firmly believe that they are right and that the person trying to persuade them is totally wrong.
15:03:21
warweasle
Xach: Are you speaking in French again? I don't like it when the voices in my head do it either.
15:09:22
aindilis
one thing that could be done is iterate over ghtorrent extracting the urls of projects that list Lisp as their primary language
15:10:48
aindilis
and then you could use NLP text classification to classify the projects into preexisting folksonomies like debtags or sourceforge's classification
15:17:59
beach
aindilis: For example, I just saw a recent paper by Adams on reading and writing floating-point numbers. There have been work in the past, but this one claims to be much better. Now every Common Lisp implementation needs those routines.
15:18:01
beach
But you can be sure that most implementations use their own code, perhaps their own technique as well, and possibly even a technique that is not as good as the published ones.
15:18:57
beach
So taking those articles, writing portable Common Lisp code with the techniques, comparing them and making the code available, would make it possible for many Common Lisp implementations to use this code and avoid duplicate maintenance.
15:21:20
aindilis
ah, now I understand, with the number of different versions of CL out there, you have this problem. I come from Perl originally so we didn't have that problem
15:22:41
aindilis
for instance, I heard it can be hard to get things running on SWI-Prolog to run on other Prolog, another language which has this kind of fractionalization
15:22:57
beach
They all have their specificities, but something like reading and writing floating-point is typically done in portable Common Lisp code.
15:24:32
beach
CLISP is written in C and uses a bytecode interpreter which makes it very portable but not that fast. ECL compiles to C and interoperates with C very easily. SBCL generates fast native code, CCL has a fast compiler. Clasp interoperates with C++. etc etc.
15:27:29
beach
The only way I see to fix that problem is to make sure that the new thing is faster, more portable, better documented, better tested.
15:46:42
pjb
I have an idea, let's write what we work on in a ~/.plan file, and let's the finger command fetch and dump this file…
15:49:08
ck_
hold on not so fast I need to copy this down -- "let's ... the finger ... command fetch and dump the file" ok got it
15:49:33
ck_
will proceed with this plan but I will make a feint to the north-east first I'll surround them
15:49:59
pjb
aindilis: there's the notion of conformance in CL. This is stronger than portable. Portable, is only a potential. conforming is definitive.
15:50:34
pjb
aindilis: If you write conforming code, then it is guaranteed that it will run the same in all conforming implementations. No work to do, no porting of portABLE code, or whatever.
16:13:06
Xach
My cron job failed last night, but not to worry, it's because a lightning storm knocked out power. time to get another ups...
16:20:45
ck_
Xach: I just read your conversation with a scammer. Wonder why they're all called Solomon.
16:31:36
ck_
ah, good news -- githubstatus.com says 'error rates have dropped'. I guess the levees are holding.
18:13:35
decent-username
Hi, does someone of you know how I could disable the quoting behaviour of paredit in Emacs? Right now when I type a backslash character it will quote the next key stroke, but I just want a simple backslash character.
18:17:03
ck_
decent-username: \ is probably bound to paredit-backslash. Bind it to self-insert-command or remove the binding to get the behaviour you're looking for
18:23:48
Xach
C-q <anything> will insert that thing rather than going through fanciness, it's a generic emacs feature.
18:24:09
decent-username
I'm not an elisp hacker, so I have no clue how to rebind keys. I just started writing some function for the paredit mode hook.
20:57:51
Guest5703
I'm wondering where it comes from, there seem to be references outside emacs lisp.
21:02:51
zhlyg
Gah, wrong terminal again.. did obarray provide a mechanism for symbol specialization?
21:08:05
housel
REMOB corresponds to http://www.lispworks.com/documentation/HyperSpec/Body/f_uninte.htm
21:20:02
zhlyg
housel: yes of-course, the symbols value vs its reachability/existence. Anyhow, thanks for the link. Those pre-CL lisps really seems to be the origin of the obarray concept.
21:22:32
zhlyg
Interesting style, is it to get temporary reachability of the special symbol X? (LET ((X ...)) ... (MAKUNBOUND 'X) ...) ;bad style
21:23:25
housel
https://www.dreamsongs.com/Files/HOPL2-Uncut.pdf page 12 says that before 1971 it was a list (and thus OBLIST)
21:28:37
zhlyg
housel: wow! thanks for digging that up! So the name is most probably a short for object-array.