One entry was yours, @mckinley. The Ladybird note in https://mckinley.cc/notes/atom.xml links to the Tor Browser note. Anyways, interesting project you're following there. I reckon that metric isn't too bad to find bloated HTML and CSS.
One entry was yours, @mckinley. The Ladybird note in https://mckinley.cc/notes/atom.xml links to the Tor Browser note. Anyways, interesting project you're following there. I reckon that metric isn't too bad to find bloated HTML and CSS.


> Once they are made public, [archive feeds] are supposed to be left alone and won’t receive further updates. Deletion or editing is still allowed, but feed authors should not expect clients to retrieve archived feeds on a regular basis (or at all).
Looking at your main feed URL
https://twtxt.net/user/prologic/twtxt.txt
, the prev
points to https://twtxt.net/user/prologic/twtxt.txt/1
and that in turn links to https://twtxt.net/user/prologic/twtxt.txt/2
. Here the chain ends. That means whenever a new archive feed is generated, the URLs change. Or viewed differently, the URL suddenly provides a differnt content. That probably causes clients to not update their caches properly, because they always see the same …/twtxt.txt/1
in the main feed's prev
and think that nothing changed, which contradicts the promise of the spec.

And yesterday:

Closing wih something more substantial. This morning I didn't think I would go out without an umbrella. But the weather god ment well for us, rain stopped for our afternoon hike. Autumn is just beautiful, we never tire of looking at the colored leaves.

Next time I come across a GPX file, I'll check out GPXSee, never heard of that. When I used OsmAnd myself years ago, I also made use of the offline maps feature. But I'm sure I did not have to pay anything. I thought it was even open source software, could be wrong though. Up until now I didn't even know that there are multiple applications out there with the same name. Hmm.
curl https://txt.sour.is/twt/u4bs34q -H 'Accept: application/json'
That exfiltrated the original twt from @xuu's yarnd instance, which I cast into a test case: https://git.mills.io/lyse/go-lextwt/commit/66d8e56141233af66026cfb72ebd74b02cc02201

Now that I updated my feed with this twt, all yarnd caches should hopefully receive a new twt with the correct hash, once they refetch my feed.
Did you have a power or network outage?

2021-06-12T08.03.22Z
is an illegal timestamp format. The periods must be colons.


What do other clients see here? Which hash does that linked twt produce on your implementations?
A few years ago, we clocked mountainbiking vs. hiking. By bicycle (real ones, no fancy electric motors) we weren't any faster than by foot. Just on the way home, of course, the bike won by a large margin. The fastest ride home took only 12 min. But that was totally insane. Just plain stupid. If anybody had come our way, it would have ended really, really badly. So we only did that once.
Depending on the picked route, 1:30 or even 2:00 hours one-way to the top is easily possible at our walking pace. So your guess of a bunch of hours is not that wrong. Just depends on our mood and path choices. :-)

Welcome to Lyse's evening of pessimistic views.
As for twtxt in particular, I see growth more as a bad thing, I don't want to deal with all these idiots out there, not even indirectly. If the network grows, other folks, best not to tangle with, are naturally attracted, too. It's a normal thing, there's no way around it. I'd like to rather keep my filter bubble small and familiar. I know, a lot of, if not most, people here disagree with me.
Also, I even don't know whom of my friends I would want to create their own twtxt feed, since I'm in touch with them in other ways, anyways. The ones I told about twtxt in the past were very hesitant to join. So was I in the beginning. For very good reasons. Writing shit publicly on the internet is not everybody's cup of tea. And by now people have already established ways of publishing truly interesting things. So, there's simply no need for alternatives to them. My best mates have low opinions on social media, so do I. I try to convince my brain not to consider twtxt as social media. :-) I'm just here because I initially was intrigued by the simplicity of twtxt and I like the people I follow.
Lastly, there are days I hardly can keep up with the amount of new messages, so a larger community would make that job even harder.
Quite a pessimistic and selfish view, some might say. And they were probably not too far off. It's very hard to explain. And then even not in mother tongue.
Don't get fooled with jenny and mutt. ;-D Twtxt is way much more like a website. Well, in fact, it basically is.
And I try not to talk people into twtxt. It's just something I try to avoid in general.

Yesterday, I forgot my SD card at home, so the camera was totally useless. We had a very crazy, fiery glowing, blood-red sunset from what I could see through and above the trees. But I was about 15 minutes too late. Only saw the last red stripes over the horizon when I finally made it to the lookout outside of the forest. Todays sunset was quite boring.
<updated>
over <published>
. Easy as that. As a bonus it did that forever. No idea, why the generated feeds suddenly caused all the trouble. Basically all feeds were affected. Luckily, no Newsboat bug.I just filed a question for Newsboat about the unsupported
<author>
on <feed>
level: https://github.com/newsboat/newsboat/issues/2256 Is it a feature or a bug?
<published>
timestamp was used as <updated>
in an <entry>
. And now it's the original feed's <updated>
timestamp. I never used <published>
elements in <entry>
s. I have absolutely no clue why now the updated instead of the published timestamps are used. I definitely should just use both fields properly and not just one. In the Newsboat source code <published>
take precedence over <updated>
.But I guess this explains the weird behavior I've seen in Newsboat, that suddenly, read entries were marked unread. They were older than a year and I've set
keep-articles-days 356
, so they were removed from the cache. With updated timestamps Newsboat just recreated new articles in its cache, which of course are unread.
/usr/bin/ld: /lib/x86_64-linux-gnu/libm.so.6: unknown type [0x13] section `.relr.dyn'
What the heck!?
And it also appears that I'm not really able to reproduce this unread bug. It only kind of works a single time. And it has something to do with my config. Not sure what it is yet. I also noticed that the
<updated>
timestamps in the entries somehow shifted between the old and new feed. Da fuq!?
/usr/bin/ld: /lib/x86_64-linux-gnu/libm.so.6: unknown type [0x13] section
.relr.dyn'
What the heck!?
And it also appears that I'm not really able to reproduce this unread bug. It only kind of works a single time. And it has something to do with my config. Not sure what it is yet. I also noticed that the
timestamps in the entries somehow shifted between the old and new feed. Da fuq!?
author
column in the rss_item
table of the _cache.db_ is now empty and also Newsboat does not render an "Author" header in the article view anymore. The feed's author is ignored.Looking at the code, I noticed the
<contributor>
is not supported at all. Gotta file a bug report or even better directly submit a fix MR (to some of the issues) tomorrow. Gotta test my C++ "skills". Duck and cover everybody!In my enqueue script I now just fall back to the "Feed" field, so at least that's working again.
Alright, multipass is basically for administrating multiple systems at once. Good thing, I'm not an admin. :-) But now I remember a good mate doing this, too. If I'm not mistaken, he used tmux's builtin functionality to update all his, I don't know how many, machines at the same time. Pretty cool actually. This also reminds me that I really should finally take a deeper look at tmux some day. Never did that before. I just open a new terminal and my tiling window manager takes care of most things, well the layout. Except that I would have to explicitly ssh into the other system if I were working some remotely.
Yes, I'm on i3 and really not in a mood to fiddle around and waste my time with broken or unfinished stuff if I can easily avoid it. I just want my system to work. :-)
What's your use case to send keyboard events to an asortment of windows? I never did that and can't see when that would be useful. But I'm sure there are lots of applications out there for that.
If I want a fucked up clipboard, I can also simply use a VM… Oh dear, this is a total killer. So Wayland is absolutely worthless to me.
I never tried Wayland myself. Nor did I do any research on that matter. From all you said and I heard from other mates, let me answer your question why you want to switch: You don't. :-) This model really sounds totally insane. It feels like it's heavily contradicting KISS at all possible levels. Complicating everthing at basically no gain. Security is all good and nice, but if you can't do basic things anymore, then that's the opposite of progress or even a working system. I fear it's going to be the same with systemd and System V Init. Eventually everybody is forced to switch over. But not now. Hopefully, by then some things are sorted out and even simplified. Hope dies last, but it dies.
Very nice quote, I completely agree! :-D
<author>
from <entry>
s to <feed>
, Newsboat marked all old affected articles as unread. IDs were untouched, of course. Need to investigate that. Had something similar happen with another feed change I did some time ago. Can't remember what that was, though.
<link>
would be treated as <id>
if there's no explicit <id>
. But I guess, it's too late now.Interesting, tag URIs are a first for me, too.

@nick@hostname
yarnd syntax was a mistake. Long live proper @<nick url>
twtxt syntax, only.
xml:base
attribute, just like @mckinley said. Never heard of <link rel="self" … />
being used for that. Maybe some last resort fallback in some feed readers, though. But that would be very fragile, too.Unfortunately, the reasoning behind
rel="self"
remains a mystery.