Wiki/Report of Meeting 2025-01-02

Report of Meeting 2025-01-02

Present: Ed Gottsman and Bob Therriault

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

1) Happy New Year. The meeting kicked off with Bob showing the Category Table https://code.jsoftware.com/wiki/Wiki/Category_Tree_Template and reiterating that the Add-ons now link directly to anchor points on the Add-ons page. https://code.jsoftware.com/wiki/Vocabulary/Libraries#dev This skips the Add-ons category page but flattens the tree for quicker access. If there is sufficient information added for each add-on then the category pages could be brought back into the loop. The same was done for the Database category page https://code.jsoftware.com/wiki/Category:Databases_R.2.2 with direct links to the individual database pages eg. Jd https://code.jsoftware.com/wiki/Jd/Index For Labs the process was to group the pages into primary index links that bring you into a more curated experience of the labs. https://code.jsoftware.com/wiki/Category:Labs_R.2.3 The same process was taken with Open GL, Plot, Publish, Window Driver, Files, Regex, and Sockets. Essays may take a different approach.

2) Ed wondered how much of this organization is a result of fighting with MediaWiki. Bob felt that early on there was more exploration, but the process has been refined to reduce the friction of creating the information display. Ed wondered how much is shaping the information that is displayed. Bob felt that he was curating so as a result there is no doubt he is influencing the information organization. Ed wonders if MediaWiki has changed the curation process. Bob says that part of the goal is to make the wiki easier to change with the current format. Earlier Bob had made a more complicated process, but simplifying should result in a lower barrier of participation. Ed spoke of his frustrations of the web generally being ossified into a mainframe presentation when a more dynamic approach would modernize it and make information more accessible. Ed mentioned Smalltalk as an example of such an alternative. Bob mentioned the Viewer that Oleg had created the resembled a Smalltalk approach. https://code.jsoftware.com/wiki/Addons/gui/cobrowser

3) Ed mentioned that the the J Viewer is not as accessible for those who chose to use the console as their IDE. Bob pointed out that nothing stops the jconsole user from opening up the J Viewer and running it in parallel. Ed said this was enough of a barrier that it was not being done. Bob there were also barriers that existed with lab creation and the fix of Lab Author was not enough to overcome the challenges of creating content which is the bigger challenge.

4) Bob spoke about the approach to the essays section may be to contrast his curation with the work that had been done with Oleg. https://code.jsoftware.com/wiki/User:Oleg_Kobchenko/Essays Taking that information and constructing a scaffold that may be able to be filled in for the future by users as interests dictate.

5) Other business includes the potential of the J Q and A project and not much has moved on that front.

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

Okay, yeah, the agenda for the meeting was essentially, I was gonna show off, let's see some of the stuff that I had done, I'd continued to do in the category table.

And I'll just share here.

There we go, I'll share that one.

So before I had shown what I'd done with frameworks and these different links now don't go to the categories, they go directly to the add-ons page and they're indexed down for the headings.

And we had a long discussion about why that was and whether things should be expanded or not.

I haven't actually changed anything on that.

I think I'll let them get into use before I make some of those decisions.

The advantage of having the tables expanded out is that if you come into it this way, you're going to see, well, I guess you'd see the, if they were all expanded, you'd go to the right area.

If you were coming in and not through the table and you came in through some other way, you would see this long list of expanded things all the way down.

And unfortunately, the way this page is set up, I guess we could change the page, is that the add-ons only have these three areas.

Installation, main and official add-ons, these are hardly useful headings.

When you go down to this level, it no longer is, it's no longer interpreting them as headings, I'm actually putting anchors in to allow you to get to different spots.

So if all these were expanded, you quickly fill up a page with a lot of stuff and you're going to be doing a lot of searching.

So it comes down to how are you going to get into the page? - Right. - And so I'm sort of waiting on seeing how that goes.

It might be worth restructuring the page.

But again, I'm not going to make too many changes to that, but I did use that different approach to the add-ons.

I continued on, I did databases, did essentially the same thing.

So, oh, that goes straight to the guides.

That's interesting, I didn't think I did that.

Or did I?

Oh yeah, so I did do the same thing.

I took you straight to the guides from there.

Databases is where if I go to this page, which is the overall page, I give the links into the different databases. - Right. - But when I get to the individual databases, I'm going straight to the page.

At this level with labs, I did do the same thing I had done with similar things at that level.

I essentially gave them headings and then we'll take you into the primary.

You can see, so labs since J806, you've got all these labs pages down here, and I'm basically grouping them up here in this area.

So when you go to labs since then, you're going into the lab index for 806.

If you go to labs before that, you go to the labs before that.

So it's actually doing a grouping for you.

So you don't have to go down through all the pages.

It's grouped to the pages and you get to the, basically the control point for the different areas.

And then in these cases, these were labs in text form.

Oop, taste of jail, I'll have to fix that up.

Yeah, and these were some tutorials that sort of have the flavor of labs.

And I thought that was just kind of useful information to put at that level, but that's sort of the curation I'm doing at that level with these headings.

I'm trying to structure the information that would be fully available down here.

If you knew to go to lab 805 index, you'll go to the same thing as the labs prior to 806.

If you know to go to the labs index here, it will be post 806, but you have to know that.

So that's the added form that I'm putting in here.

And so I've done that as well with OpenGL, same thing, but lots of pages and I've grouped them so that you sort of get a quick access into those different areas.

And I think I'm gonna continue to do that.

Well, I have done it with OpenGL plot, publish window driver files, regex and sockets.

And so it's just a way to, if you were interested in say in the window driver, you're gonna go here and you go to the overall information on the window driver, specifically for animation, they've got a whole section on animation I hadn't actually seen before.

I thought that was interesting to put in.

And then a reference, which is also part of this one, but it's just a quick way to get into the reference, command reference if you wanna see what the commands are. - Right. - As opposed to having to search through these to try and figure out what's going on in this grouping. - Right.

Bob, how much of your curation is just fighting with media wiki, if you had to guess? - Fighting with media wiki? - Yeah, you wanna achieve a particular effect and media wiki wasn't designed to do it.

So you come up with workarounds. - Initially, yeah, that was pretty common because I was learning it.

But now I think like, for instance, if I go into edit this page, I gotta go log in.

Let's get logged in.

Okay, if I go into edit this page, I guess I'm gonna just read source.

The format I'm using is pretty easy.

I'm just putting a heading and it happens to also sometimes contain a link.

So you go straight to it and then a description and a line underneath.

So I'm not fighting with it very much at all.

At this point, that's my pattern that I'm building there. - Yeah.

I guess I would return to a comment you made earlier about chat GPT, which is that it's gonna tend to guide you towards certain solutions rather than maybe the best solution. - Yeah. - I wonder to what degree that's happening to you.

And there's no good, there's no way to answer that.

I just can't help but wonder. - I'll tell you, it's certainly happening because how could it not happen?

Because I'm the same, in that sense, I'm playing the same thing as chat GPT without the vast corpus of information behind me. - Yeah. - So I'm having to make choices about, and that goes back to the initial categorization because that's what fueled what pages here show up, which in turn fuels how I break these headings up.

So you've got two levels of it happening.

So there's no doubt.

And I guess the first level of it, which is a little wider because it involves more people, was the initial information that was put into a Wiki to start with. - Sure, of course. - Because the interest of people will go to certain areas. - I wasn't so worried about the curational aspects, how you group stuff and how you come up with headings.

I was interested in all the non-curational work that you have to do in order to get the effect that you're searching for.

And if I understand you right, now that you've accommodated yourself to MediaWiki's capabilities and limitations, you're not fighting anymore, but you're not fighting because you don't bother trying things that you know MediaWiki can't do or can't do well. - Yes. - See what I mean? - Yeah, I do, I do.

That's a good point.

I think, as I was saying, when I started out, when I was learning it, I was pushing those boundaries all the time. - Yeah. - And what I've come around to, as I've got deeper into this, is the challenge of pushing those boundaries is that there's a tremendous amount of work to have anybody else come in and do things.

And that reduces the effectiveness of the Wiki. - No, that's a good point. - That's kind of why I've gone to the simplified approach is because I wanna make it easily accessible to people to go in and change things.

It's one of the reservations I have about the category system, although I think there's power in the category tree that it pays off.

But one of the reservations I have is how easy is it for people to go in and add categories or how easy is it for people to understand how things would fit in a category?

And that's a whole other level.

And I think as the Wiki develops, those questions will have to be answered.

But those are questions that I see are the difficult challenges now.

But it's not MediaWiki that influences that. - No, no.

I have the root of all my problems with all of this is that I have become extremely annoyed with most of the web because it's definitely there a mainframe, green screen, page-oriented presentation experience for the most part.

You can find parts of the web that aren't, that are much more dynamic, that don't feel like a mainframe terminal.

But MediaWiki is dead on a mainframe terminal. - Yeah, it is. - And the '70s were an unattractive era in many ways.

And one of them is the mainframe.

And also in the '70s, we had Smalltalk, which is what I think of as the JMEGANAV.

JMEGANAV is a Smalltalk-style interface.

It's not the opposite of the mainframe, but it's very, very, very, very different from the mainframe.

And I like it much better myself.

And I'm really annoyed that order of the way through the 21st century, we're still building mainframe applications, MediaWiki being a prime example of that particular sin. - Yeah, I agree with you 'cause there's a lot of things I like about the Smalltalk approach. - Oh, yeah. - Have you seen, I'm trying to think of what it's called.

Is it Object Viewer that Oleg put together?

Like in 2008?

It's essentially an object view of your variables in J.

And it looks, it's patterned on a Smalltalk window with the various columns, those kinds of things.

It's within the wiki, but it's quite neat.

Again, it never caught on.

The same way you were talking about your viewer being bound by JQT, I know Devin doesn't use JQT.

He uses the console, but using the console doesn't preclude you from running the viewer in JQT and having it beside you. - It's just not, it's too much of a barrier for I think probably a lot of people, for whatever reason. - Yeah, I don't know.

Yeah, I think the same way when I was talking to Eric about labs and we talked about the barriers to making labs and the one that they attacked was how easy is it for you to construct the lab as a framework?

So we'll build the framework.

You can put this all through.

You wanna put this here and now you've got the framework.

And then what they found is nobody was making labs because the framework isn't the hard part.

The hard part is figuring out how to put content into the framework.

But when you've got to do both, the framework feels like it's getting in your way. - Right, right. - But that's not what keeps you from making labs because the combination does, but the hard part is the content. - Yeah, surprise. - Yeah, but I think it's a good example of what we're talking about.

Put together, it's really hard.

Take the easy part out, it's still really hard.

So you have to know, yeah, you still have to know what part is the hard part. - Right. - Yeah.

Anyway, so I've gone forward with, as I said, these different areas and done some of that curation of the pages that would show up below, like down here.

Essays is really interesting because Oleg is actually, he did, I found out where I actually pulled these headings from.

I pulled them from him, but he's done, and I've got to go through this.

He's put different essays in different areas given these headings than I did.

But that's great because now I'm gonna get a chance to go back and actually see how he did it and what he put.

And I basically have a dialectic, a dialogue with somebody else about you put this here, I put it there, and I can think in my head, well, I know he's right here, I shouldn't have it, or maybe we should expand it or whatever.

I've got a different point of view, a separate curation to work against. - Right. - And that's one of my challenges with essays.

So honestly, I think what I'm gonna do as I develop this whole table, I'm gonna skip through essays and go right to phrases, articles in these areas, finish this off, and come back to essays and probably initially have something that looks more like this with the list of essays that are under a group and not do as much of this.

I'm not gonna spend all the time figuring out what an essay means.

I just maybe put links to each of the essays and have the user go in and find them, decide on the title.

And over time, then these explanations can be added by other people underneath. - Yeah, right. - And I think that's probably the way that I'll finish off this curation of this whole table.

And I'm hoping that, I'm hoping within a month or two, it's actually to the level where that's been done for essays and we can actually publish this and say, oh, by the way, you should look at our new wiki.

We've been putting the wings on the plane while it's been flying all this time.

You wondered what was going on.

We'll drop the other wings away and how do you like it?

(laughs) - Right, exactly. - Yeah.

And in terms of other developments, I think was the only other thing on the agenda.

I haven't really done very much on that.

Done a little bit of thinking, but not very much.

And I think in the next week or two, I've got lots of calls to make.

And I wanna talk to Devin, as I've said many times before, about the J-Users group, New York J-Users group, and figure out how that can all work together and maybe have a dialogue and start to building those ideas together.

Yeah, so that's about it. - All right.

Okay, well, I'll keep chugging away and I hope to be able to send you something at the end of the, or maybe during the call next week. - Okay, well, that would be great. - To try it out.

And have a great week. - You as well.

And as usual, you're breaking lots of new ground.

It's really cool to see what you're coming up with. - Well, we'll see what we can do.

I appreciate all the encouragement.

Thank you very much. (laughs) - Well, the last thing I'd wanna do is anything that discourages you right now. (laughs) 'Cause you got enough work ahead of you.

I can see a lot. - I know how much work it's just going to be.

I might not have done it.

It has turned out to be, even with chat GPT, much more effort than I expected. - Yeah, but that's why I encourage people. (laughs) It's a trick.

Get them to do stuff they wouldn't normally do.

Because otherwise it would never be done. - It's true. - Yeah, I think it's gonna be really cool.

Anyway. - We shall see. - See you next week. - Great. - Great work. - Next week.

Take care, bye. - Bye-bye.