|
in reply to Dancer2 developers on PerlMonks?
Hi, I've contributed a couple of Dancer2 plugins. The guys who maintain Dancer are very smart and have been very nice. It's a great project, highly focused on extensibility and compatibility, rather than an all-in-one environment as I perceive Mojo to be. Contributions are welcome, but the maintainers are all volunteers and busy. It's sometimes disheartening that your PR isn't reviewed, or if it is accepted, a new release doesn't happen for months, but that is the way of most OSS projects in general. IRC sucks, always has. There is the Master of Supercilious Tongue-lashings who sort of dominates the IRC channel, and besides him, usually very slow response there to questions/comments. I'd love to see more Dancer2 questions here.
Update: On reflection and reading your response I think I gave too negative of an impression about the Dancer project. In particular I neglected to mention that the Github issues queue for the Dancer2 repo is well-curated (by MrCromeDome), to the point that issues are tagged with Beginner suitable and Up for grabs, so that would be a good place to start looking for ways to chip in. There's also a good deal of discussion goes on among the devs in the issue comments (to echo Laurent_R's point about the benefits of threaded conversations).
The way forward always starts with a minimal test.
|
Re^2: Dancer2 developers on PerlMonks?
by nysus (Parson) on Dec 12, 2017 at 22:18 UTC
|
Yeah, one thing that impressed me was the credentials of the guys who helped found the project. And I really like the philosophy of Alexis Sukrieh but who has apparently stepped away from the project. Overall, I'm getting the impression that the project is mature and stable but could use some more promotion and new blood. My guess is, Sawyer X–who helped carry for the project for the past few years–is wrapped up on Perl 5 core development as a pumpkin. I don't have a lot of involvement in open source (made some very minor contributions to Drupal 10 years back) but I've always been fascinated in getting involved if the right project came along. This one seems to be right up my alley. I'm looking for ways to do what I can to help.
It's tough to help breathe life into a project without support, though. If existing core developers are too busy to foster the next wave of Dancer2 developers, what can be done?
| [reply] |
Re^2: Dancer2 developers on PerlMonks?
by Laurent_R (Canon) on Dec 12, 2017 at 23:04 UTC
|
IRC sucks, always has.
No, not always has. It was good and useful at some point.
But, yes, IMHO, it sucks today. To start with, it is obsolete technology. IRC was invented almost 30 years ago, this is very very old in the context of fast-moving technologies on the net. And it is backward and time-consuming. I don't understand why so many people in the Perl community are still wanting to use it. Just as an example, threaded forums or chats are much better, people know easily who is answering whom or what, you can answer a post from a couple of days ago and make sense, and so on. As for the specifics of the Dancer IRC channel, I just don't know.
| [reply] |
|
|
Well, I don't remember any time where IRC was not a flame fest, but anyway you are right that it's a worthless medium for collaboration and for offering support. The problem is that the Dancer project, like many in the Perl world, as you say, continue to state that the IRC channel is the "official" support method.
The Dancer channel is largely silent, and when newcomers ask a question there's usually a long period of silence and most often the newb has left when someone acks. On the other hand, as I mentioned in an update, like many projects nowadays, the most activity for Dancer is found around the Github pages for the repo.
The way forward always starts with a minimal test.
| [reply] |
|
|
And it is backward and time-consuming. I don't understand why so many people in the Perl community are still wanting to use it.
IRC has helped me for a lot of projects where I collaborate, and where I had questions. I admit it's only really useful if you are logged in permanently. I use it in a tmux session on one of my servers.
I also admit that platforms like slack have advantages (easier sharing of files, showing online status), but also disadvantages.
I can use IRC in textmode, the status line is always visible and it will simply highlight the "windows" that need my attention. No pop-ups required. I can do everything with my keyboard in one terminal window with currently 30 channels.
Browser apps have improved, but I still need to use the mouse a lot.
| [reply] |
Re^2: Dancer2 developers on PerlMonks?
by nysus (Parson) on Dec 13, 2017 at 00:32 UTC
|
Yes, I have gotten some response and interactions on GitHub. Even Sawyer X chimed in on an issue I posted. The project is not dormant, but it does have the feeling of a beehive in late autumn with a few bees flying slowly in and out. From my observations, which may certainly be completely wrong, the project was most active from about 2011 to 2015 and involvement has dwindled a bit.
I'm not sure what to attribute this to.
I did chime in on IRC and was pleased to see so many people there. Though you are right, response is a little slow.
So I'm curious to know if you are still using Dancer2 regularly and what you have done with it so far. Please let me know.
| [reply] |
|
|
You can't always interpret a lack of buzzing around as stagnation. The Template Toolkit hasn't had much new core development for years: that's because it's complete. Dancer2 core is not as complete as TT, but it's very full-featured and stable, and the plugin architecture means you can easily build your own if something's missing. Most of the things introduced to core started as plugins anyway.
As far as what I use Dancer and Dancer2 for : all the new webapps and APIs I build, which is most of my work: "microservices" where the bulk of the requests are for REST endpoints rather than traditional HTML page content. I also have plenty of more human-oriented websites on Dancer/2. (I just started a $job where the app framework is Mojolicious, though.)
The way forward always starts with a minimal test.
| [reply]
[d/l] |
|
|
What do you see as D2's big differentiator from Ruby Sinatra which it is based on? If you're not a Perl developer and you're looking for a new web framework to experiment with, what about D2 might make you say "D2 looks cool, I'm going to check it out."
| [reply] |
|
|
