Wiki/Report of Meeting 2024-11-14

Report of Meeting 2024-11-14

Present: Ed Gottsman and Bob Therriault

Full transcripts of this meeting are now available below on the this wiki page.

1) Bob plans to finish up the landing pages for the category by providing a consistent set of links to sub pages https://code.jsoftware.com/wiki/Category:Promoting_J_C.8. Next would be to include the search bar and access to NuVoc below the search bar. https://code.jsoftware.com/wiki/J_Code_Search Also to include the category table at the bottom of each page. https://code.jsoftware.com/wiki/Wiki/Category_Tree_Template Raul may be instrumental in that process and it may require a temporary test wiki to ensure that there are not issues.

2) Bob feels that the next steps will be to fill in some of the gaps in NuVoc such as the foreign conjunctions https://code.jsoftware.com/wiki/Vocabulary/bangco which are better described in the old Vocabulary pages https://code.jsoftware.com/wiki/Vocabulary/Foreigns. Bob also received an email from Eric that may impact on future wiki directions, but that meeting had not happened at the time of this meeting.

3) Ed wonders whether there is a potential to move the J Wiki Viewer https://code.jsoftware.com/wiki/J_Viewer onto the web and make it more accessible to the wiki. Perhaps the J Viewer working on the virtual machine and linking it to the wiki. This is in early stages but Ed regards it as possibility and would be interested in working on that if it is deemed useful to the community. Ed thinks that Eric may be interested in this and Bob wonders if this might have a use with the JHS platform in the cloud that Eric has been working on. JHS does have a subset of HTML that is coded in J for the presentation of JHS. Ed would be happy to migrate the J Viewer on the JQt platform to the web if there is an appetite for it. Bob said he would ask him about the possibilities.

4) Bob has the feeling that Eric is hoping that Bob will move on to other projects from the wiki.

For access to previous meeting reports https://code.jsoftware.com/wiki/Wiki_Development If you would like to participate in the development of the J wiki please contact us on the J forum and we will get you an invitation to the next J wiki meeting held on Thursdays at 23:59 (UTC)

Transcript

I'm recording now.

It's official.

What my plan is to do over the next couple of weeks, months, is to finish up all of the pages that I've, the landing pages for the categories.

So essentially, when you come into a category, there'll be an explanation about the category you're on, and there'll be links to the things that are most pertinent.

And then there'll be the category page underneath that.

You know, the category table will be underneath that.

And then the next step after that is to move the search.

This is Raul, I think, who's going to end up doing this.

Move the search up onto every page so Chris's search doesn't have to be a sidebar.

And then really easy thing to do, change the sidebar so all the major categories will be in the sidebar on top of having the category table down below.

And at that point, I mean, there's all sorts of other things we can do with the wiki.

But at that point, as far as the organization of the wiki goes, that's pretty much where I think we're going to sit with that.

And that's kind of where I am with that, thinking about it.

And I guess the only other change could be that if Jan comes up with a really good, and I haven't got back to him, I was in Italy when he sent out that email and stuff.

So I should get back to him now and just say, "Oh, by the way, I'm back now."

So you can take a look at that.

But if he comes up with a real bomber way to do it, I might move some categories around.

But that won't change the content of the pages.

All that's going to do is change the structure of what's linking to what.

And I can do that too.

So anyway, that's where I am now.

I got an interesting email from Eric while I was away, and I haven't talked to him yet.

But he said, "We should set up a time to talk."

And he said, "I want to talk to you about some other stuff, not the wiki."

So we'll see what's happening with that.

That builds into the wiki.

In terms of the wiki, the next stuff I'm seeing, I'm thinking probably if I was going to aim somewhere, I would fill out NUVOC.

There's still holes in NUVOC, like all the foreign conjunctions.

That information is really partial.

Like if you really want to know what's going on, you have to go back to the old vocabulary.

That's where all the information is.

Yeah.

And I mean, if we're trumpeting NUVOC as this thing that's developing, it never got transferred across and that should be there.

Because Henry now is making a point of updating NUVOC to every new change that comes in, which is great.

But there are old things in that are still not coming across.

So I think that would be the next thing I would look at is to bring NUVOC up a level so it's at least consistent.

And then after that, I don't know.

I realized how big a difference the JViewer made in all that.

Well, it's the way you look at the information now.

Like the wiki's there and that's great because it's got this expansive wiki, but the JViewer is such a good window on the information.

I just don't know why anybody would use anything else.

I wonder, Bob, I wonder what it would take.

The problem with the JViewer is that it's this big thing.

And I wonder what it would take to put the interesting bits of the viewer into the wiki.

And here I'm thinking, I have not thought this through in any way, but it strikes me that we could dramatically expand the audience or the viewer style of experience if we could get it into the wiki rather than insisting that you load this rather inconvenient add-on.

And so I don't, as I said, I'm not sure what I mean by that.

I think probably the main use of the viewer, and because I didn't collect any usage information, I have no idea whether this is true.

I think the main use of the viewer is the search mechanism, the full text, full code search mechanism.

Absolutely.

I've started to wonder whether you couldn't do that up on the web.

You know, it would be work, and I'm not quite sure how you'd go about doing it.

Would you still use SQLite?

Would you have, luckily the traffic isn't huge.

You don't have a massive performance problem that you have to address.

So it might be SQLite running in a virtual machine that the web page talks to, the virtual machine running on whatever it is you guys use, Amazon Web Services, I don't even know.

Yeah, I think that's what, I think Eric's using AWS, I think for the JHS in the cloud.

Right, but where's the wiki?

Oh, the wiki?

I think that is AWS as well.

I think, I don't know.

And then the other possibility would be AWS has native full text search, so we could just use the same trick I use to translate J symbols, J glyphs, and just instead of feeding it to SQLite, feed it to whatever AWS uses by default.

And then there's nothing to stop us from continuing to supply results that are not only in the wiki, but in all these other places where the J ecosystem manifests itself.

So Rosetta code and so on.

As I said, I haven't thought it through, but if somebody is looking for an interesting project to do, it strikes me that this might be one.

You were concerned at one point about the latency with the web, not being able to, the connections and stuff.

Yeah, yeah.

You know, we'd have to see about that.

I don't think it would be as good in that regard, but I think it still might be a significant improvement both in quality of results and presentation and scope of results if we could do it on the web.

I mean, Chris did a nice job, but it's an older implementation.

I think we can improve on that using a different approach.

Well, and Chris was really just using regular expressions, wasn't he, to do that?

That is my understanding.

Yeah.

And there's nothing wrong with that.

No, no.

We got a lot done with that.

Yeah, I think the viewer provides better results.

If only for a couple of reasons, I guess.

One, just a much wider scope of results.

But secondly, if we could do the underlining the places where the matches have occurred, that would be really slick.

If we could actually duplicate that display, that would be really quite slick.

I don't know what would be involved in doing that either.

I mentioned this only, you said that you're going to be talking to Eric.

To Eric, yeah.

So just for fun, you might mention this possibility to him.

I don't know if he's a JViewer user, but just throw it into the mix as a possible project.

It's something I'd be happy to work on, but I'd want to have some faith that there was appetite for it.

It's not something I'd want to go off and do by myself.

I'm wondering how well that would fit with what he's already...

And I don't know whether this is what he wants to talk to me about, but the whole J in the cloud thing is moving along.

And whether it would make sense to put that information up with J in the cloud.

Well, I mean, once you've got a web service that can provide that kind of functionality, you could invoke it from the wiki, you could invoke it from J in the cloud, it wouldn't really matter.

Yeah.

I mean, actually, I'll qualify that in a moment.

The hard part is setting up the web service.

The user interface, and you're going to qualify it, the user interface to that web service is also hard if you're trying to do something that looks like the JViewer's results set.

That's like work in HTML.

It wasn't that hard to do using the...

JQT.

JQT library.

Yeah.

But HTML, I think you'd probably find yourself working a little harder.

And I don't know how Eric would want to present the results in J in the cloud, assuming he wanted to use it at all.

I don't have an intuition about that.

Well, Eric already has sort of the subset of HTML sitting in his JHS in J.

Like he's using that to format what you see.

Yeah.

And it's got...

It can do all the stuff that...

Well, I shouldn't say it can do all the stuff, because it's dependent on the particular browser you're using.

But there isn't very much that I've hit that JQT can do that Eric can't do on JHS.

And in some ways, Eric can do it better, because you're looking at a web page.

So when you look at an SVG or latex or anything like that, it's embedded.

It doesn't have to be in a separate window.

It's actually embedded on your page.

Right.

So you can see it right there.

Yeah.

No, I'm sure it could be done.

The only question is, how do you want to present it?

And I don't know what the answer to that is.

Yeah.

Well, ask him about that for sure.

I wouldn't mind retiring the JViewer in favor of...

I mean, the functionality really belongs in the wiki.

It's sort of a technical accident, and me being bloody minded, that it wound up being a separate application.

But you could do it on the web, and the wiki is really where all that functionality belongs.

Yeah.

Well, and you got the model for it now, too.

I mean, you got a working prototype, basically, if you wanted to port it over to the web.

That is certainly the case.

Yeah.

It would be a rewrite.

You'd really have to start over from scratch.

I did it all in J.

You couldn't do it in J, really.

Although, I guess maybe...

I think you could do it in J, because that's what Eric's done with the JHS.

Eric is doing.

Yeah.

That's interesting.

Maybe it would just be a port, just.

Maybe it would be a port.

Well, look, good idea, if you'd be so kind.

Yeah.

Because I would be happy to tackle it.

The only question is, is there any appetite for it amongst the powers that be?

Yeah.

The J in the cloud and the JHS seems to be Eric's baby.

But that alone, I mean, the fact that JHS is web-based, that brings you right back into that area, right?

Yeah.

Because that's how it's done.

That's how their whole interface is just your static web pages, interactive.

Yeah.

I'll ask him about it.

As I said, I'm just finishing up with this call now.

I'll probably try and get in touch with him next week and see if I can...

We can set up a meeting and just have a chat and see what's going on.

I'll let you know what happens out of that.

Yeah.

I would appreciate that.

Thank you.

Yeah.

I think that's it.

Not that I can think of.

Other than that's my roadmap over the next while.

And I think he's trying to drag me off the wiki a bit.

I just get the sense that he's saying, "You've been working at this for four years now."

Perhaps you deserve a break.

New vistas to regard new...

Yeah.

I was wondering at first, he mentioned this about a year ago.

I was wondering at first a year ago whether he was saying, "I think you've worked at this long enough.

We'd like to see you on something else."

But that wasn't the vibe I got from him.

It was more just, "Are you okay?"

Oh, that's nice.

Yeah.

And I said, "No, I'd like to see this through."

And then this most recent email I got, it was more like, "You know, you..."

I said, "Okay.

Well, I do have a plan.

I will like to wrap it up and that's how I'd like to do it."

He says, "Yeah, we should talk when you get back from Italy."

So, we'll do that.