Beefy Boxes and Bandwidth Generously Provided by pair Networks DiBona
more useful options
 
PerlMonks  

Re: bivnn.cgi -- an alternate interface to newest nodes (batch older nodes by parent)

by grinder (Bishop)
on Sep 30, 2001 at 19:37 UTC ( #115735=note: print w/ replies, xml ) Need Help??


in reply to bivnn.cgi -- an alternate interface to newest nodes

You are right when you say xNN is difficult to hack on. It takes a while to get used to epoptai's style. I keep telling myself I should sit down and refactor it one day.

One of the nice features xNN has, that bivnn yet lacks, is displaying the response nodes in their threaded order underneath a primary node. It's a really effective way of letting you see how busy a node is, long vs. deep, bushy vs. thin and so forth.

Secondly, for new nodes that are replies to older nodes that themselves have fallen of the end of the newest nodes, they get listed in a separate section, with a link to the parent node as well. This lets you jump straight to the parent node, so that you can read the response in context.

I think the default xNN behaviour in this section was to sort the titles alphabetically. Whatever, I didn't find it very useful. I hacked it to display by sorting the parent nodes ascendingly, which means you see the reponses to the oldest nodes first, and multiple new nodes to an old node are grouped together.

The final feature that xNN has is, put it this way, your HTML design skills are about on par with mine (that is, essentially non-existent). epoptai is a grandmaster paramount HTML designer who moved from designing web pages to writing Perl code without any formal training (which goes a long way in explaining why the code is as it is, and why the results look as good as they do).

--
g r i n d e r


Comment on Re: bivnn.cgi -- an alternate interface to newest nodes (batch older nodes by parent)
Re: Re: bivnn.cgi -- an alternate interface to newest nodes (batch older nodes by parent)
by blakem (Monsignor) on Sep 30, 2001 at 19:55 UTC
    You're right, I wasn't going for a fancy HTML interface. I also avoided the threaded look, even though that functionality is built into the PerlMonks Modules. My primary goal was simply to generate the shortest list of links needed to catch up on my perlmonks reading. Since the threads aren't usually all that deep (atleast not like USENET) I didn't think the theaded functionality would be particularly helpful to me. Although, if someone wants to take the code and run with it..... ;-)

    -Blake

(ichimunki) re x 2: bivnn.cgi
by ichimunki (Priest) on Oct 01, 2001 at 16:33 UTC
    Hmmm. I think I prefer blakem's plain text approach. On my browser, epoptai's page produces a font size that requires a magnifying glass to view. Thankfully my browser includes one, but pressing it increases all the text on the page and I had to press it a couple of times to get back to my default "regular" text size-- which is the size I expect most of the text on a page to be in. That's why that size is the default.Then he's gone and used form inputs as display elements. I suppose form elements can be seen as graphic widgets, but I have a tendency to expect them to do something.

    So maybe our opinions differ about what is good web design, but blakem's code is solid Perl and that gives it a decent foundation upon which to build and add embellishments and prettifications. But as I said, I like it the way it is. :)

      Odd that you should say that. The output from xNN and reputer is no smaller than the output of Perlmonks itself, plus, you have the source to make it bigger, which is not as easily accomplished here. (But you're right, I too dislike pages that use a blanket <font size="-1"></font>).

      Could you elaborate on form inputs used as display elements? All the form elements I can get my hands on have a function, but I could be missing something.

      --
      g r i n d e r
        (I swear I responded to this, but that response seems to have gone missing-- maybe I previewed without submit.)

        My bad. I was looking at the reputer demo on Konq 2.2.1. Under IE 5.5 (here at work) it looks a lot better. The xNN demo also looks quite nice on IE 5.5, but I suspect it is also going to have problems on Konq (I'll have to check it when I get home). But the fact that the pages change fairly dramatically between clients is exactly why blakem's plain old CGI-generated HTML is so great. It looked the same on both IE5.5 and Konq 2.2.1-- a big win in my book.

        The form elements was talking about really do serve a purpose. They just don't do it on the demo page, so my apologies to epoptai for that unfair criticism.

Log In?
Username:
Password:

What's my password?
Create A New User
Node Status?
node history
Node Type: note [id://115735]
help
Chatterbox?
and the web crawler heard nothing...

How do I use this? | Other CB clients
Other Users?
Others having an uproarious good time at the Monastery: (5)
As of 2014-04-19 00:07 GMT
Sections?
Information?
Find Nodes?
Leftovers?
    Voting Booth?

    April first is:







    Results (473 votes), past polls