I’m going to try and finish the Clojure Koans: https://github.com/functional-koans/clojure-koans/.
I have a sneaking suspicion this wasn’t supposed to be published for another 48 hours or so…
Joke or no, this is the systemd endgame. Whether systemd merges Linux, or Linux merges systemd, it’s going to happen. It’s bad for Linux as a hobby, but good for Linux as an operating system for businesses and the computing community at large. Companies smaller than Google and Amazon pay for RHEL or SLES for a reason. The “year of Linux on the desktop,” if it ever comes, will be after that merge. Until then, Linux is only viable for people who have the time to deal with the fragmented community, or pay for someone else to do it for them.
Whether all this is truly good or bad remains to be seen.
Why do you think it’s bad for Linux as a hobby?
Good question. Part of the fun is putting together the puzzle that is Linux. I don’t think it will destroy the hobby, but it will detract from it somewhat, bring it closer to Windows as a hobby. For example, Gnome 3 already depends on systemd, removing some freedom of choice. There are shims but that’s beside the point.
Yet gnome3 even runs on openbsd example, which IIRC doesn’t even have PAM, not speaking about systemd.
With stuff missing, hence this project. And this Gnome 3 status post also notes the same problems that the systemd shim aims to solve, namely timedated, localed, hostnamed, and logind.
I also use octosplit to get side-by-side code diffs.
What a surprise to find that Go ever supported Windows 2000
Imagine my surprise when there were users on the mailing list that requested support to continue.
If anyone is looking for a good book that covers topics like this in detail, I can wholeheartedly recommend “The Linux Programming Interface” by Michael Kerrisk.
“Advanced Programming in the UNIX Environment” by Richard Stevens is also quite good, and less specific to Linux - it covers standards, and notes variations in BSD, OSX, Linux, and commercial Unices.
This statement has been a while coming but I’m glad it’s (finally) been made.
I’ve known for some time (from an IRC discussion) that there are currently no engineers at Mozilla working on Persona full time. We’ll continue to use it on forums across our platform (e.g. http://forum.espruino.com/) for the foreseeable future though.
I hope very much that community ownership is viable long-term. Open source projects don’t (just) die from lack of contributors or users, they die from lack of direction too. But there is a healthly list of contributors, so I hope this won’t be the case.
On a more practical note, the fact that the server is node.js is a blocker to me contributing at the present moment. Regardless of the merits of the technology, it’s one that I don’t know and don’t have time to learn. I think it would have been better to write it using a more established and widely known technology. There is a far larger pool of Java/Python/Ruby server programmers.
I wish I could be more constructive, but (one of) our original considerations for using Persona was to free up some of our development time for other features. I contribute to various open source projects but I don’t have time to do that for everything.
I personally came to lobsters to get away from the Silicon Valley/startup-tabloid type of submissions that get a lot of upvotes on Hacker News. News about an acquisition is not really news about technology.
After reading the post you submitted, I agree with the downvoters because there isn’t any technology news in there (FWIW, I didn’t downvote you).
News about an acquisition is not really news about technology.
There are also numerous places to read startup news. Harder to find are technical communities with an ethos like lobsters.
I would prefer startup news to be tagged specifically as such so it can be filtered easily.
I am working on an implementation of Raft in Ocaml. The rough outline can be found here
It doesn’t do anything yet but the idea is that this repo is purely the state machine for Raft. And then another repo will implement the I/O layer above it, which is what users would actually develop against. The interesting thing about this repo is it’s an attempt to encode the state machine into the type system, in the hopes it will make using it incorrectly really hard.
For those that don’t know, Raft is a consensus algorithm, similar in idea to Paxos.
For the curious, here’s an excellent intro to Raft from Ricon West 2013: http://www.youtube.com/watch?v=06cTPhi-3_8
The original paper is very readable too, linked here with some other resources: http://raftconsensus.github.io/
I’m going to be working on our site owner signup process (where users create their own forum) and improving the code quality of our web application.
The latter was written in a hurry and it shows, but I’m pleased with the usage we’re seeing and the bugs aren’t too serious or numerous.
This is nifty tool. Out of curiosity, does this work in a similar way to how templates are embedded in the godoc binary?