This page is not created by, affiliated with, or supported by Slack Technologies, Inc.
- # announcements (8)
- # beginners (22)
- # calva (42)
- # cider (2)
- # clj-kondo (1)
- # cljdoc (3)
- # clojure (63)
- # clojure-chicago (1)
- # clojure-uk (29)
- # clojurescript (16)
- # clojureverse-ops (2)
- # core-matrix (6)
- # cursive (1)
- # datomic (23)
- # emacs (1)
- # heroku (2)
- # luminus (1)
- # off-topic (47)
- # pathom (1)
- # planck (3)
- # re-frame (4)
- # reitit (1)
- # rewrite-clj (5)
- # shadow-cljs (47)
My Sunday is just starting (I haven't gone to bed yet on Saturday) and I've spent nearly all day working on tests and documentation for
next.jdbc... Alpha builds are up on http://clojars.org but I haven't announced it yet... trying to get http://cljdoc.org to play nice with it first.
I keep pushing updates. It's 1:25 am. I need to go to bed (but I'm also chatting online to my wife in China right now).
It's amazing how much you are able to push forward and chip away at this. Do you have a particular process?
I hate documentation so this part is painful. I hate writing tests too -- but a lot of the tests committed today come from REPL history logs.
I'm just really motivated to draw a line in the sand after
clojure.java.jdbc and move forward to what I believe the API should be. And I know that adoption means tests and documentation 😐
My wife just sent me some pretty scary pics of how bad the air quality is in Shanghai right now -- big storm coming in for the next few days (she flies home tomorrow night -- T-storms predicted for three days there!).
next.jdbc, I've been working on it slowly since early January. Just over 100 commits so far. And there was a lot of hammock time before that due to nearly eight years of using
clojure.java.jdbc so... ¯\(ツ)/¯
Like most of my projects, it started off as a series of scribbles on a paper napkin, quite literally...
When I originally wrote what turned into one of the most popular MVC frameworks ever for CFML (ColdFusion), it all started with a set of bullet points on a paper napkin. I blogged about that http://framework-one.github.io/blog/2010/02/06/fw1-the-napkin-spec/
If you want another insight into my approach to projects http://corfield.org/blog/2014/06/03/getting-started/ 🙂
@dominicm We switched everything to JIRA/Confluence/BitBucket several years ago. We needed more flexible ticket workflows and better integration with other tools. We all loved Unfuddle tho'...
@seancorfield Tangentially related, did you play much with Frege? Anything you'd use it for that you wouldn't do in Clojure?
@dominicm In the end, no, I didn't get very deep with Frege. I was trying to help port some of the Haskell libraries over to Frege but it was a pretty painful experience. I just don't think the project was mature enough and evolving fast enough to be a serious contender for any production-near work. I think Eta is a safer bet if you want to go that direction: https://eta-lang.org/
I've had a love/hate relationship with Haskell for close to 30 years now. Back in the 80's, I was heavily into ML-family FP languages (and my PhD was around the design and implementation of such languages), so I was very excited about the potential for Haskell -- it was meant to be the best-in-breed of all the things all of us university researchers were working on -- so, for a while, I was convinced it would take the world by storm because it was just so obviously The Best Language(tm).
But the community that grew up around Haskell pretty much sabotaged any likelihood that industry would take it seriously and so, very soon, AT&T released Cfront E to education and C++ took over the world and we all had to learn OOP to be employed. It was quite depressing.
So, every few years, I go back to Haskell, with a view to "learning it properly" (again!) and coming to terms with whatever state the ecosystem has gotten itself into... and every time I end up getting frustrated by so many of the same old sharp edges, weirdness, and a lot of the "attitude" from (some) Haskellers...