hpr-knowledge-base/hpr_transcripts/hpr1816.txt

Episode: 1816
Title: HPR1816: Visualising HPR tags
Source: https://hub.hackerpublicradio.org/ccdn.php?filename=/eps/hpr1816/hpr1816.mp3
Transcribed: 2025-10-18 09:39:21

---

This in HPR episode 1,816 entitled Visualizing HPR Tab.
It is hosted by Dave Morris and in about 9 minutes long.
The summary is using Bluff Vista Visualizing Tab on HPR episodes.
This episode of HPR is brought to you by an Honesthost.com.
Get 15% discount on all shared hosting with the offer code HPR15.
That's HPR15.
Better web hosting that's Honest and Fair at An Honesthost.com.
Hello everyone, this is Dave Morris.
Today I'm going to do a different type of show from the way I normally do things.
Content might be a bit different, certainly the way I do it is a bit different.
I'm just trying to do this in a pretty much off the cuffway.
I thought I'd share with you a thing I've been playing around with.
I came up with the idea yesterday and put together some software to make my idea real.
The idea is based around the HPR database.
In the database there is an entry per show, not too surprisingly.
And in this entry, in this row in the database,
one of the things that's stored there is a list of tags.
You'll be aware hopefully that you are asked to provide some tags when you submit a show these days.
I think this is a relatively recent invention. Ken would would remind me but I haven't asked him
within the past year or so the tags came along.
We're not doing much with them at the moment but we're thinking of
using them to give more information about shows in general.
Anyway, there's this list of tags and I was looking at them.
We really need to populate them and that means we could do with crowd sourcing
this amongst the HPR community asking people to look at, select a show, look at it,
or listen to it more at the point and come back with a set of tags which we can then add to the
database. One of the things about adding tags though is quite useful to know what tags exist.
We need to do some sort of standardization anyway.
So I thought what I would do is try and analyse what we have at the moment with the view to just
get my head around what we could do in the future.
So what I thought I'd do would be to make a graphical representation of the tags
in the shows that have tags at the moment which is not very many.
Back in the day when I was at work I used to mess around with the package called
GraphViz. There'll be a link in the show notes to this. It's still going strong.
In fact it's developing, it has developed quite a lot from the way I knew it.
GraphViz is a package which lets you define items, whatever they are,
nodes in the parlance which are words or structures or whatever.
And then join them together in various ways. So you define
here's a node, here's its name, here's what I'd like you to label it.
Various things about its colour and how you want it drawn and things like that.
And then you then say I want to create an edge.
An edge is the line between one node and another.
So the way I decided to do this was to write a purl script. Of course in purl there's
an interface to this GraphViz package. And what I'm actually doing is I'm going through
the database, finding all records which contain tags.
And then I'm returning the show number, the title of the show and the list of tags.
Then in my script, the script will be available in the show notes.
In the script I'm parsing out the tag list, separating them out by commas.
And I'm building a structure where it's indexed by tag and attached to each tag,
the list of all the shows that have that tag. So having built this structure which is meant
to represent the sort of picture I want and then using GraphViz to simply walk through this
thing through the tags in a sorted order. I don't think it actually matters because GraphViz has
a bit of a mind of its own. I'm walking through the tags anyway and for each tag I create a node
with a label of the tag. So you'll see this in the notes as to how this turns out. And then I'm
walking through the list of IDs associated with that tag and making nodes from them. Each
ID I'm turning it into the string HPR followed by a four digit number, then the title of the show
on a new line. And it's in a box with a green border. And then I'm doing an ad edge,
is the call you need to do to make a link between the tag box, it's actually an oval in the current
iteration, and the show number, show name, show title box. And once that's all been done and there's
something like, check my list, we've got 727 tags and 303 shows with those tags. So that's pretty
small number in comparison to the number of shows, which is in the 1800s now as you probably know.
Then I run the graphical, the GraphViz application to make it produce the output. What it does
is it's creating an SPG file which you can then look at in a browser. And I'm putting this in my show
notes so that you can gaze at this. It's pretty big and it's not anything very beautiful,
but it is interesting, very least anyway, I find it interesting. You might too as well. So
just looking at it at the moment and how it's laid out, I've made it do the tags in red ovals
on the left and the show details on the right in green boxes as I said. And then there are arrows.
So for example, at the top of the display at the moment, I've got teachers as a tag and tutorials
as a tag. And they're pointing at show 1735, which is in type of free tutorials for teachers.
So that's actually quite simple. It's only got two tags. If you scroll further down and it's a long
scroll to get to the end of it, you start to see things like, well, if you look for the word
Linux, and there's no way of searching in this, but it's more of an exercise in showing what's
in there, I think. I'm not sure this would ever be a practical thing. This isn't the way to
produce tag clouds and I don't think, but I thought it was an amazing thing. I thought you might
be interested in it. Let's find one. See, I've got HPR Live as a tag and it's pointing at shows 1411
and 1413, which are Omrup Live shows that need ODE, if I remember rightly. So I hope that if you
looked at this, and I look at this, this one, I'm going to do a show, this one, particularly,
there'll be a bit more of a clue as to what sort of tags are out there and what people are doing
and doing with them. And I hoped it might generate a bit of discussion about using tags and get people
enthusiastic about adding something to shows. So as you scroll way down this list and you start to
find that things get pretty tangled. And I'm not quite sure how it's all ordered these. It's
graph is made its own decision about how this is to be done, but you start to find that the
plate of spaghetti problem starts to manifest big time. And so, for example, if you find
Android, I think there's only about three links from Android to stuff, but the way the thing is
optimised it, you can't actually follow the lines from that at all. And if you find Linux,
a lot of shows been tagged with Linux, then it's just ridiculous. It just goes into a revolting
tangle. Now, if I understood graph is better, then there might be ways in which I could optimise
this. I've been fiddling around with the attributes and there's a new enormous number of them.
Not too any great success to be honest with you, but it's something I'm going to mess around with
a bit longer, just from my own amusement more than anything else. So, right. So the show notes
will contain the script, which is not very big. It's only 200 lines long and it will contain
a bunch of links to things to do with graph is and it's going to, you should be able to find the
the latest SVG image from my experimentation. So, let me find that. It's interesting. Okay, bye.
You've been listening to Hecker Public Radio at Hecker Public Radio. We are a community podcast
network that releases shows every weekday Monday through Friday. Today's show, like all our shows,
was contributed by an HPR listener like yourself. If you ever thought of recording a podcast,
then click on our contribute link to find out how easy it really is. Hecker Public Radio was
founded by the digital dog pound and the infonomicum computer club and it's part of the binary
revolution at binrev.com. If you have comments on today's show, please email the host directly,
leave a comment on the website or record a follow-up episode yourself. Unless otherwise status,
today's show is released under Creative Commons, Attribution, ShareLite, 3.0 license.