in reply to Re^9: Update the GUI (What we want?)
in thread Update the GUI

thanks, just some remarks (more to come)

> There is no REST interface; pretty sure.

Some ppl wrote plenty of xml tickers, some of those even needing authorization.

Adding a new vote-interface shouldn't be harder, unless I'm missing something.

See What XML generators are currently available on PerlMonks?

> Perl6 is today, and for the foreseeable future, ...

Sorry not my point, it was meant as analogy. There is no Meta-Perlmonks available today.

We didn't stop continuing Perl5 development 15 years ago because of the prospect of a better tool to come, which is 10% better. (Well some did)

We (except you ;) should continue developing PM too.

> Metacpan is just a better site. "Design" is not all that different. Information layout is, which some might call design.

I agree from the perspective of the experienced Perl user.

Most visitors probably aren't and will judge their "Perl xyx" search hits after 500 ms by the colors and fonts.

My priority ATM is to improve this impression, not to compete with SO (yet)

Cheers Rolf
(addicted to the Perl Programming Language and ☆☆☆☆ :)
Je suis Charlie!

Replies are listed 'Best First'.
Re^11: Update the GUI (What we want?)
by Your Mother (Archbishop) on Jul 23, 2016 at 21:22 UTC

    The reason why I think it's a terrible way to go about it is it completely violates good web app design concerns. It uses the View and the Model and half of the Controller as a TERRIBLE API to construct a new view, a new hybrid model which consumes the view and the model and turns it into a new model, and new controllers which have to account for all this mess to create a new, more reasonable, interface. It's possible, sure, and if someone smarter than I am can show how to do it well, I'd rejoice and sing hosannas. I've been doing this awhile though and CSS improvements, like what tobyink did, and perhaps some Ajax voting are about as much as could be expected and it wouldn't be super easy and would (probably) still require user action to add the CSS and JS manually. Therefore this would be of zero use to newcomers and casual visitors... If the Gods determine the scripts/CSS are safe enough to implement in the actual site delivery, maybe... I would never advise anyone to build out technical debt though. I wouldn't try to stop them to strongly either. :P

    The core of what this site is/does is quite basic and if just a couple of monks with the chops committed to it, it would be done in 6 months. Less if bikeshedding could be avoided.

[reply]
      > It uses the View and the Model and half of the Controller as a TERRIBLE API

      I agree.

      But what are the alternatives?

      (Closing PM down and "defecting" to SO? ;-)

      Since you are rejecting the DB scheme, even building an new alternative MVC application on top of the data seems out of discussion, right?

      And if you did this new MVC front end, it might take so long that new requirements would arise, not talking about all the additional legacy plugins which might get obsolete in the meantime.

      I have a strategical point of view on the things, improving the design will bring the best and fastest return of investment ... politically, technically and from a marketing perspective.

      And afterwards a long term strategy can improve the other issues you listed

      > If the Gods determine the scripts/CSS are safe enough to implement in the actual site delivery, maybe

      That's my plan.

      For me Toby's approach is "only" a prove of concept.

      I plan to invest a week to improve it (in my own "dogfood" interest ;) after YAPC::EU, build a show case and try to convince the gods of a long term strategy.

      Cheers Rolf
      (addicted to the Perl Programming Language and ☆☆☆☆ :)
      Je suis Charlie!

[reply]
[d/l]

        I look forward to your stabs, as it were. Who knows, maybe you will decynicize me into participating.

[reply]
Re^11: Update the GUI (What we want?)
by Anonymous Monk on Jul 23, 2016 at 21:50 UTC
[reply]

      *It's.

      Are you using it? This is how it interacts when used from the free nodelet–

      The XSS Auditor refused to execute a script in 'http://perlmonks.org/?' because its source code was found within the request. The auditor was enabled as the server sent neither an 'X-XSS-Protection' nor 'Content-Security-Policy' header.

[reply]

        And then what happened?

[reply]
      From what I understand this approach is wasting resources.

      I don't see the point to return several kb of html with the full node just to let JS parse the votes out. (Except maybe for a prove of concept of DOM manipulation)

      update

      as I already said

      > > > > Adding a new vote-interface shouldn't be harder

      Cheers Rolf
      (addicted to the Perl Programming Language and ☆☆☆☆ :)
      Je suis Charlie!

[reply]

        From what I understand this approach is wasting resources.

        :) Not that many extra ones that aren't wasted already

        Its very much like hitting talk or refresh in chatterbox

        It is a common viewing pattern to view a node, vote on it, then scroll down, to see a reply you like, vote on it, and so on

        Also common to open a node in a new tab and do the same

        So resource wasting? Not really

        But it doesn't have to do that, code could pick a different node that returns less byes, like 162 bytes

        Or it could simply "cue" votes and cast them every 30 seconds at a cost of under 500 bytes

        as I already said

        as I already agreed :p

[reply]

In Section Perl Monks Discussion