summaryrefslogtreecommitdiff
path: root/community/weblogs/ArneBab/how-i-write-a-qoth.mdwn
blob: 87b1f07dd8aeee48a263fef94a4982159b8f5ed0 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
I just read on the hurd IRC channel (chat: #hurd at irc.freenode.net), that people consider my work valuable (I knew that, and I think that myself, but it is still nice to hear), so I want to dispell any possible myth about it :)

What I do is not hard - at least not anymore, since I created a simple structure for it (But it still takes time).

First I open up the relevant mailing lists for the quarter. I get them from [[contributing/web_pages/news/writing_the_qoth]]. Normally I just use the following:

* <http://lists.gnu.org/archive/html/bug-hurd/YYYY-MM/threads.html>
* <http://lists.debian.org/debian-hurd/YYYY/MM/>

Then I copy them 3 times and use M-x replace-string (in emacs) to adjust them to the correct months.

Additionally I open the Arch Hurd news:

* <http://www.archhurd.org/news.php>
* <http://planet.archhurd.org/>

Having all those news at hand, I read every thread-starter and every news-item. For each of them I first check if I understand them (no use trying to explain something I don’t get myself) and if they provide a way for people to test what they improved (however complex that might be), then I 

* note the name of the main contributor(-s), 
* write a line of text what it does (often partly copied from the news-item),
* add a link to the news-item, a code-repo or a patch and
* a note how that new development helps achieve the goals_of_the_Hurd (see [[contributing/web_pages/news/writing_the_qoth]] for details).

With that list of short news I go into [[contributing/web_pages/news/qoth_next]]. 

Now I identify 2 to 4 main news items by some kind of “helps the Hurd most when more people know it”, “biggest change” and similar fudgery :)

Finally I sort all the news items by intuition, crude logic I develop on-the-fly writing and the goal of making the qoth read somewhat like nice prose. 

On the way to that I commit every little to medium step. I never know when I have to abort due to an interruption (I’m sure tschwinge loves my super-non-atomic horrible-to-review commits :-) - but better that than losing work == time, and I try to prefix the commit-messages with “news:” so he knows that it’s useless to review them as in-flight-patches…).

Having finished the text (usually after 3 to 6 hours of overall work), I send it by mail to bug-hurd: <http://lists.gnu.org/archive/html/bug-hurd/>

After about a week I incorporate the comments from there and publish the qoth as described in [[contributing/web_pages/news/writing_the_qoth]].

Then tschwinge reviews it, does some last-minute changes and pushes it from the staging wiki to the website.

And that’s it.

I hope this small insight was interesting to you. Happy hacking and have fun with the Hurd!

-- Arne Babenhauserheide

PS: Writing this blog entry took about 20 minutes. The raw text is longer than a qoth, but it is much faster to write, because it avoids the main time-eater: Gathering the info with the necessary references to make sure that people can test what’s in here.