XSF Discussion - 2018-03-12


  1. jonasw

    moin

  2. Zash

    Hm, no MUCs in clients.json

  3. Ge0rG

    What?

  4. jonasw

    support MUCs, I assume

  5. Zash

    Yes

  6. Ge0rG

    Do we have a field for that?

  7. jonasw

    Dave Cridland, I heard you had some (internal? non-public?) client which does reactions. Would you care to give feedback to this? https://github.com/jabbercat/jabbercat/issues/80

  8. jonasw

    Ge0rG, no

  9. Tobias

    https://news.ycombinator.com/item?id=16567009

  10. Ge0rG

    Tobias: HN is probably downgrading / shadowkilling upvotes from direct links, it's better to go from the main page.

  11. Tobias

    HN users probably know it and can go back to the mainpage to upvote

  12. Ge0rG

    Maybe.

  13. daniel

    Nice good job

  14. daniel

    Don't even care about the content but having xmpp on the front page of HN is always an achievement

  15. Zash

    Dave Cridland: Did I see you saying something along the lines of email stagnating around the time everyone switched to gmail?

  16. pep.

    "Markdown-like syntax" in that article. This is going to come back to bite us at some point

  17. Ge0rG

    pep.: that also came on the ML recently, with the suggestion to call 393 and 394 "markdown" and "markup" respectively

  18. pep.

    Yeah I've seen that one and your comment :)

  19. Zash cries in groff

  20. Ge0rG

    I had to resist the urge to counter-propose "markleft" and "markright"

  21. Ge0rG

    I think that 0394 qualifies for markleft, because it's based on tags starting with <, and 0393 is markright because it specifies > as a quoting character.

  22. Ge0rG

    jonasw, SamWhited: ^

  23. jonasw

    I’ll slap you

  24. pep.

    Ge0rG: :D

  25. Ge0rG

    jonasw: luckily I didn't attend CLT

  26. jonasw

    yeah

  27. jonasw

    we still need that slapping-into-face-over-plain-TCP/IP

  28. jonasw

    ;)

  29. Ge0rG

    jonasw: over XMPP

  30. jonasw

    <slap/>

  31. SamWhited

    *fry*

  32. SamWhited

    wait, that doesn't work on here…

  33. SamWhited

    http://www.sydneyjanebaily.com/wp-content/uploads/2014/12/0tsVo2k.gif

  34. Ge0rG

    Damn, with Sam on the Council I will probably not get this through Council against the authors wishes.

  35. Zash

    -xep poke

  36. Bunneh

    Zash: Presence Obtained via Kinesthetic Excitation (POKE) (Humorous, Active, 2004-04-01) See: https://xmpp.org/extensions/xep-0132.html

  37. Ge0rG

    SamWhited: I'm actually glad you didn't try to fry me :D

  38. Ge0rG

    > How do you use emoji reactions in a mission-critical way? 🤣

  39. Zash

    Who are you calling a rectangle?!

  40. jonasw

    fry: not sure if demonstration on how to do that or quote and laughter.

  41. flow

    jonasw, what's your stance on swapping the names of xep39(3|4)?

  42. jonasw

    flow, I said that I’m going to change the name of 394 anyways, so I don’t really care

  43. jonasw

    I’m not sure either styling or markup really applies

  44. flow

    jonasw, what shall be the new name of xep394?

  45. jonasw

    swapping the names seems like a bad idea though

  46. jonasw

    not sure

  47. flow

    XEP-0394: Not Sure

  48. jonasw

    seems legit

  49. flow

    That appears to be confusing

  50. Ge0rG

    jonasw: "markleft"

  51. jonasw

    Ge0rG, no.

  52. flow

    Text Style Annotations

  53. Zash

    You are thinking too 2D

  54. jonasw

    maybe Markup Annotation or something which carries the notion that it’s not directly inside the message

  55. Zash

    Markforward!

  56. Ge0rG

    jonasw: you don't really care, and "markleft" and "markright" are great memory hooks

  57. jonasw

    Ge0rG, they’re also bad for people with right/left weakness :P

  58. Zash

    Mark Here

  59. flow

    Mark Tawin

  60. jonasw

    intended typo?

  61. flow

    I'm just going to say 'yes'

  62. jonasw

    I don’t see the joke then :)

  63. Ge0rG

    jonasw: don't get all ableism on me

  64. Ge0rG

    What about Mark Zuckerberg?

  65. flow

    I've heard he is pretty rich

  66. Ge0rG

    rich text formatting with zuck-up!

  67. moparisthebest

    haha the irony is excellent, that slack-bait-and-switch hn thread a guy 'I'm writing an open source client...' -> follows link -> 'The app will be open-sourced at some point, so you don't have to worry about it being abandoned.'

  68. jonasw

    lolwat

  69. moparisthebest

    I fell for that once with minecraft, never again

  70. jonasw

    i don’t like that "we’ll open source that at some point"

  71. Zash

    Minetest!

  72. moparisthebest

    yea minetest is excellent :)

  73. jonasw

    people who don’t open source right away usually won’t lose their reasons not to later on

  74. jonasw

    Syndace might be an exception, still wary on that one ;-) (jk)

  75. moparisthebest

    ah I can link directly to the comment https://news.ycombinator.com/item?id=16567285

  76. Syndace

    jonasw, Haha nice one :D

  77. jonasw

    multi-protocol client. suuure

  78. jonasw

    > There will also be an affordable premium plan for things like having more than 5 accounts. The base functionality will always be free.

  79. jonasw

    > Open-Source

  80. jonasw

    nope

  81. moparisthebest

    but he's fixing the walled garden problem jonasw

  82. jonasw

    deak

  83. jonasw

    *yeah

  84. moparisthebest

    he even says so...

  85. moparisthebest

    :)

  86. jonasw

    multi-protocol messengers have always been the best

  87. jonasw

    Ceterum Censeo Pidgin Delendam Esse.

  88. moparisthebest

    especially ones written in pure C

  89. moparisthebest

    they never have security issues

  90. jonasw

    moparisthebest, you should renick to "multiprotocolisthebest"

  91. jonasw

    IN PURE C

  92. jonasw

    holy smokes

  93. rion

    =)

  94. jonasw

    I thought we were past this stage

  95. Zash

    C is nice. In small doses.

  96. moparisthebest

    I dislike 100mb javascript monstrosities too, but the solution to me doesn't seem to be 'rewrite everything in "pure C"'

  97. jonasw

    yeah

  98. jonasw

    using something inherently memory-unsafe in 2018 seems like a super stupid thing to do.

  99. jonasw

    especially if network interfaced

  100. jonasw

    I’d go along with C++. that has at least the option of writing mostly compile-time checked memory-safe code.

  101. jonasw

    > eul is a registered company, and all binaries are signed. Your data is safe.

  102. jonasw

    I’ll... just ... stop reading that site now

  103. moparisthebest

    mainly the irony of being linked from that thread and 'solving the walled garden problem' was the best part :)

  104. Ge0rG

    he was plugging eul in earlier threads as well.

  105. Kev

    Slack doesn't give me much grief using the Slack client, although it's annoying, but Discord...someone makes a nice open source Discord transport and they will be my friend.

  106. Ge0rG

    someone make a nice XMPP client.

  107. edhelas

    Kev wow you need a lots of effort to be your friend

  108. Kev

    Ge0rG: I'm working on that.

  109. Link Mauve

    https://securelist.com/somebodys-watching-when-cameras-are-more-than-just-smart/84309/ Botnet control by Samsung using XMPP~

  110. jonasw

    Ge0rG, I’m working on that.

  111. Kev

    It's C++ and it's Qt and I hope this could be the year of XMPP on the desktop ;)

  112. Ge0rG

    > the entire Hanwha smart camera cloud is a Jabber server. It has so-called rooms, with cameras of one type in each room. An attacker could register an arbitrary account on the Jabber server and gain access to all rooms on that server. Sound security design. I'm on the look for a "web cam" style camera I can point to the great view I have in the new home

  113. jonasw

    lol

  114. Zash

    Kev: Haven't you heard? The desktop is obsolete!

  115. Seve/SouL

    Better someone build an XMPP Discord-kind-of-client

  116. jonasw

    should probably take a look at discord at some point

  117. Kev

    Much like Slack.

  118. Kev

    Except you can't have multiple accounts

  119. jonasw

    I never used slack

  120. Kev

    Much like Discord.

  121. jonasw

    ELOOP

  122. Zash

    Someone wanna take one for the team and look at all these things and tell us what their point is?

  123. Kev

    I use both extensively.

  124. Zash goes and sits in the "IRC and XMPP are fine, why would you want all that garbage?" corner.

  125. jonasw

    Kev, wanna agree on how to handle Emoji Reactions and do a field test?

  126. Kev

    Zash: They both do things that are useful and XMPP doesn't do, to my eyes.

  127. Kev

    jonasw: Reference.

  128. Link Mauve

    Last argument I heard for Discord was “it’s possible to do an audio conference at any moment, so we should migrate that in the eventuality we may want to do one someday”.

  129. jonasw

    but there are many details

  130. Seve/SouL

    Kev, it would be nice to know which kind of things we are missing then

  131. jonasw

    Kev, https://github.com/jabbercat/jabbercat/issues/80

  132. Kev

    That's the easy answer for live messages. For non-live messages we need the server archive to understand references.

  133. Link Mauve

    I then proceeded to close this biboumi tab to never come back there.

  134. Zash

    Kev: Things.

  135. jonasw

    Kev, I argue that Message Attaching is more well-suited for emoji reactions, but feel free to weigh in if you disagree. I consider that one of the more irrelevant details though.

  136. Zash

    Kev: All I hear are repeats of my Skype hating days. "XMPP just can't replace Skype" - said by peolpe who use it exactly as you would IRC

  137. Kev

    jonasw: I *think* that References simply supersedes Message Attaching, once it's fleshed out.

  138. jonasw

    Kev, it does too much

  139. Kev

    Zash: I think you can give me a little mileage in having an understanding of XMPP if I think some of the things Slack does that I use are hard to currently do in XMPP :)

  140. jonasw

    but I argue that in the linked thing already

  141. Kev

    And Reactions that Jonas is talking about is one useful example.

  142. Kev

    Other things are what I'm hoping to build on top of MIX once we've got MIX usefully out there.

  143. Ge0rG

    I've abandoned MIX by now.

  144. Ge0rG

    I'm also in favor of Message Attaching over References for reactions, because what jonasw said

  145. Ge0rG

    I could live with a reference containing the message ID and a type='reaction', though

  146. jonasw

    Kev, also, my beef with references would be that it uses codepoint ranges into body but doesn’t handle multiple languages. that’s not relevant to reactions, but I expect some breakage originating from that.

  147. Ge0rG

    however, neither XEP provides an element to put the actual reaction into

  148. jonasw

    (and thus namespace bumps for things irrelevant to reactions)

  149. jonasw

    Ge0rG, I’d just put that as a simpling into the <message/>

  150. jonasw

    Ge0rG, I’d just put that as a sibling into the <message/>

  151. Ge0rG

    jonasw: my point is: we need a separate reactions XEP anyway

  152. jonasw

    yeah, that’s true

  153. jonasw

    hm, actually, I might post some aggregated feedback on references, I am finding quite a few things I dislike about it when reading it

  154. Kev

    jonasw: Give it a week or so, Edwin's sending out a major update imminently.

  155. jonasw

    Kev, oh-kay

  156. jonasw

    Kev, no wait, I’m going to send out that feedback anyways. From how I see things working usually, the major update will be delayed for some reason and then my feedback is lost, because I’ll forget about it

  157. Kev

    Or send it anyway and Edwin can integrate it, either way.

  158. jonasw

    also, maybe it can be incorporated into that update :)

  159. jonasw

    yeah

  160. Seve/SouL

    Are you just considering Reactions emoji-only things?

  161. Zash

    <reaction>meh</reaction>

  162. Ge0rG

    Seve/SouL: no, but emoji will be the recommended way to react.

  163. Ge0rG

    Seve/SouL: the challenge is to determine how many codepoints correspond to one emoji

  164. Zash

    A bunch

  165. Zash

    <{reaction}meh/>

  166. Kev

    Why do you need to count the codepoints in an emoji? They're the reaction, not the source

  167. Zash

    Did someone suggest shoving semantics into <body>? Otherwise there shouldn't be any need for codepoint counting.

  168. jonasw

    I did not.

  169. jonasw

    I only want to shove fallback into <body/> :)

  170. Ge0rG

    Kev: in case somebody wants to limit reactions to "at most one emoji"

  171. Ge0rG

    so you can't write a novel in there.

  172. Kev

    For this one thing, I think not having a fallback is actually sensible Jonas.

  173. flow

    where is +notify specified? Is it only that single sentence in xep163?

  174. Zash

    flow: might be a bit in 60 too

  175. Kev

    9.2, 60.

  176. jonasw

    Kev, why?

  177. Zash

    Spam?

  178. flow

    hmm then xep163 is misleading, it tells me that appending +notify will make me receive all notifications related to the protocol that precedes the suffix

  179. jonasw

    Zash, if in response to the Reactions thing: people are doing that already anyways

  180. Zash

    Isn't half the point of these to compress eg a ton of 👍 replies?

  181. flow

    whereas xep60 tells that that I will only receive notifications about the node whose NodeID is the prefix

  182. Zash

    No the obvious solution here is to write a message with only 'm', which triggers a bot to say "$name says +1". Now you got twice the spam.

  183. Zash

    (This what actually went on in those Skype channels)

  184. jonasw

    (not to mention that if we don’t have a body on the message, it will be omitted from many archives.)

  185. jonasw

    (or may not be carbon-copied or something)

  186. Kev

    flow: How are the nodes named in 163?

  187. jonasw

    Zash, I feel you’ve got quite a trauma from skype.

  188. Kev

    jonasw: We need to update how archiving works in the world of references. And it's by archiving bare-JID.

  189. Zash

    jonasw: Mmmmmmmmyeah

  190. flow

    Kev, xep163 doesn't really name nodes, its protocols like tune who do

  191. jonasw

    Zash, so, I’ve seen people in MUCs already doing the "name\n> message\nemoji" thing.

  192. flow

    but what if such protocols define multiple nodes? xep163 gives the impression that I'll get notifications for all of them by a single +notify

  193. Zash

    jonasw: right

  194. moparisthebest

    that article is interesting "One of the main problems associated with the cloud architecture is that it is based on the XMPP protocol."

  195. MattJ

    moparisthebest, which article?

  196. Ge0rG

    MattJ: https://securelist.com/somebodys-watching-when-cameras-are-more-than-just-smart/84309/

  197. flow

    xml:lang='*'

  198. flow

    from the articel's "Message sent over XMPP using a test account created for research purposes" picture

  199. ralphm

    Kev: the idea of splitting resources, as in the examples in MIX, have those ever been written up?

  200. Kev

    I don't remember if I've put anything about it into bind2 or not.

  201. ralphm

    I didn't find it in XEP-0386. I did notice it being expired.

  202. ralphm

    Kev: in any case, the paragraph confused us here

  203. Ge0rG

    I really hope that those split resources are going to die soon.

  204. moparisthebest

    sorry I got cut-off mid-type there it says "One of the main problems associated with the cloud architecture is that it is based on the XMPP protocol." but then proceeds to say the *real* problem

  205. moparisthebest

    "An attacker could register an arbitrary account on the Jabber server and gain access to all rooms on that server."

  206. moparisthebest

    which is addressed in the base spec, not even an extension, in fact arbitrary account registration could just be off...

  207. Ge0rG

    moparisthebest: I think you are preaching to the choir

  208. winfried

    When designing a protocol, you have to anticipate on user and implementer stupidity, but this a level of stupidity I would not be able to anticipate....

  209. Zash

    What's the saying?

  210. moparisthebest

    well sure, just pointing out how dumb it is, it's only seperated by 2 sentences, don't know why an otherwise smart reviewer would do that

  211. Zash

    If you design something idiot-proof, the universe will design a better idiot?

  212. Guus

    hargh. Did some kind of DST change happen anywhere in the world, recently?

  213. pep.

    Lots

  214. pep.

    It's the week of hell

  215. moparisthebest

    yea EST did sunday morning

  216. moparisthebest

    or EDT, whatever, flipped between them one way or the other :)

  217. pep.

    Its changed in the US but not in Europe aiui

  218. pep.

    It's

  219. jonasw

    when’s the switch in Europe btw?

  220. Zash

    Still +1 here

  221. pep.

    Next week I think

  222. Guus

    https://depositphotos.com/33062973/stock-illustration-comic-curses-speech-bubble.html

  223. Zash

    My feeling is "last weekend in mars"

  224. jonasw

    this time the last weekend is split between march and april, so I’m unsure

  225. jonasw

    right, march 25th in DE

  226. jonasw

    so probably the same in europe

  227. moparisthebest

    whenever I think about how terrible timezones and DST are now, I just imagine how much more terrible it'll be when people live on other planets :)

  228. Zash

    jonasw: Yay EU-standardization

  229. jonasw

    moparisthebest, star time 1520872849

  230. Guus

    I think all board members are in the same DST-change period, but council might not be. SamWhited, you're probably the only one this week with a different time. Don't be like Guus and miss a meeting.

  231. SamWhited

    ack; I forgot about that, thanks.

  232. Guus

    (afaik, both board and council fixed the time to London time, thus using its DST switchover date, whenever that might be).

  233. moparisthebest

    jonasw, maybe but it won't be like that in the beginning, if people go to mars in the next 50 years or whatever it'll just be yet-another-far-harder-to-calculate-timezone, presumably

  234. Guus

    Well, you probably would've been an hour early, and not missed it by an hour, like I just did. :)

  235. ralphm

    Zash: it is actually 'the last Sunday in March'.

  236. jonasw

    that makes sense

  237. Ge0rG

    It's really exciting what you can do with timezones (this is from 10 years ago) https://op-co.de/1984.html

  238. moparisthebest

    am I the only one to find the opkode.com vs op-co.de distinction super confusing

  239. MattJ

    Nope

  240. Ge0rG

    I find it too.

  241. moparisthebest

    I read about half of the opkode.com blogpost earlier thinking odd Ge0rG updated his website until I got to the 'author of converse.js' part

  242. Ge0rG

    I don't know how to resolve this problem, besides of undeservedly bitching on JC

  243. moparisthebest

    if you weren't both involved in XMPP it wouldn't be so bad :)

  244. Zash

    #OPCODEBOWL

  245. moparisthebest

    we clearly need a dual to the death

  246. moparisthebest

    duel even

  247. Zash

    Cleganebowl can go throw itself into the wall! Opcodebowl hype!!!

  248. jonasw

    duel to death sounds like a bad idea

  249. jonasw

    really bad

  250. jonasw

    Ge0rG will lose

  251. jonasw

    jc knows how to wield a sword.

  252. jonasw

    I’ve seen things

  253. moparisthebest

    but what a way to go out

  254. Ge0rG puts on his wizard hat

  255. Ge0rG

    https://github.com/davisonio/awesome-irc - we need the same for XMPP!

  256. moparisthebest

    ah yea not bad, I like it

  257. Tobias

    https://www.unicode.org/mail-arch/unicode-ml/y2018-m03/0075.html

  258. moparisthebest

    so the same as https://github.com/moparisthebest/freespeech if the wordlist just consisted of 1024 emojis

  259. jjrh

    Ge0rG, we already have it https://github.com/bluszcz/awesome-xmpp

  260. MattJ

    and it lists 'Pidgin' under 'Clients'

  261. jjrh

    I'm not saying it's perfect

  262. Zash

    awesome used to mean something

  263. jjrh

    It's also missing a lot of stuff. Gajim isn't even listed :p

  264. SamWhited

    If it's about "awesome" clients that's going to be a short list…

  265. Zash

    Now it seems to just be some curated list of stuff.

  266. jjrh

    not sure if you guys are being facetious or not - the 'awesome' thing is just a github thing for a repo that is just a big list of resources/information relating to something. It doesn't really quantify quality

  267. Zash

    It used to be a window manager!

  268. SamWhited

    I think some of my coworkers still use it as their window manager

  269. SamWhited

    But maybe that means it's just a list of window managers, most of which aren't actually "awesome"?

  270. MattJ

    Zash, it still is!

  271. jjrh

    It's reasonably popular still I think. I was never able to adapt to tiling first workflow

  272. MattJ

    My window manager can do XMPP!

  273. SamWhited

    I use i3, but keep meaning to try awesome. No idea what the difference is.

  274. MattJ

    => awesome XMPP

  275. Zash

    i3 represent!

  276. MattJ

    The downside is that I only got it working in blocking mode, so my desktop was unusable

  277. SamWhited

    *highfive*

  278. jjrh

    I should try i3 again - but I'm guessing even if it's great i'll never manage to get used to it coming from openbox

  279. Zash

    MattJ: Cool story bro

  280. Zash

    I'm happy with i3, in that I finally found a window manager that let switch virtual desktops separately on each monitor

  281. Zash

    And does tabs

  282. SamWhited

    I used to use a mix of open box and xfce-wm, but stopped when I switched to i3 full time.

  283. Seve/SouL

    I use KDE Plasma

  284. SamWhited

    What Zash said; that was the big draw for me as well.

  285. jjrh

    Does it play nice with floating windows? What I always wanted was tiling for one monitor/virtual desktop and floating for others.

  286. Zash

    I've been trying to switch to LuaKit for browsing, so that i3 can do the window management instead of the browser

  287. Zash

    buuuut now I'm back with a ton of firefox tabs :(

  288. moparisthebest

    I still use KDE since 2006 , been meaning to try i3 though

  289. jjrh

    I got a coworker still using twm

  290. Zash

    jjrh: duno what qualifies as nice, but it lets you have floating windows if you want to

  291. Seve/SouL

    moparisthebest, noo, please don't leave me alone on this

  292. jjrh

    I guess can I get the best of both worlds? Because some stuff like coding I want tiled but a lot of other stuff is better floating.

  293. jjrh

    Because I tried one tiling wm - umm a haskell one forget the name - and it was like "yeah you can do floating but uh don't"

  294. moparisthebest

    Seve/SouL: I've tried various things and never liked them, lxdm, gnome, xfce etc, i3 is much more different though, we'll see

  295. SamWhited

    i3 will respect whatever the free desktop attribute thing is that windows can set to say "I should be floating", so most tiny windows that should be floating already are in my experience

  296. Zash

    and you can write rules in the config

  297. jjrh

    Can you set a workspace to be floating by default? Basically can I use i3 like I use openbox but have a few workspaces that are tiled?

  298. Zash

    Duno, floating is set per window usually

  299. Zash

    Probably doable with some config

  300. jjrh

    I guess I should give it a go at some point.

  301. jjrh

    With the workspace per monitor - will it just make double the workspaces when you lose a display (aka undock your laptop) because that would be super rad.

  302. Zash

    I'd recommend trying dwm too

  303. Zash

    Used it way back before laptops, it was very nice

  304. Zash

    but on laptops it's always ended with abrupt power failure due to no battery monitoring

  305. Zash

    might be better now, it's probably managed by systemd these days ..

  306. jjrh

    My pain is I have 2 monitors and I undock my laptop and not only is my laptop not 1080p so that messes stuff up - but all the windows from the two screens get crammed into one virtual desktop

  307. jjrh

    so if i3 treats each monitor as a workspace then it should play nice with switching from 2 to 1 monitors.

  308. Zash

    if your video drivers are sane enough, you'd just get all the workspaces moved onto the other monitor

  309. SamWhited

    That seems to be what mine does; I unplug my monitor and windows remain exactly the same on whatever workspace they were on (except maybe resized a bit)

  310. Zash

    The nicest thing is being able to move entire workspaces between monitors

  311. Zash

    IIRC no keybinding for it by default, but one can be set

  312. SamWhited

    Although I don't actually use a multi-monitor setup anymore; got an ultrawide and haven't looked back.

  313. Zash

    I've mostly got an xclock on the laptops monitor :)

  314. Zash

    and an xterm to mke the clock be the right shape

  315. Zash

    https://www.zash.se/upload/za9lYCrKESH1.png

  316. jjrh

    Zash, that's the problem - I want my wm to be smart enough to instead put them on a virtual desktop so everything isn't crammed together.

  317. jjrh

    So if I3 treats each monitor as a virtual desktop/workspace it should do the right thing.

  318. Zash

    Hm?

  319. jjrh

    Like when I undock my laptop (which docked has 2 monitors) all the windows between two monitors go to one - which makes a big mess.

  320. Zash

    Yeah, not like most others where it's treated as a single large workspace that covers all monitors

  321. jjrh

    Yeah exactly - so that's awesome if that's what i3 does :)

  322. Zash

    It is and it is

  323. Ge0rG

    I'm using fluxbox for something like 14 years now, and recently switched from VMware to VirtualBox. The latter seems to get something wrong, so I never see the full screen overlay. 😐

  324. jjrh

    I'd suggest you switch to openbox - I started on fluxbox but eventually switched.

  325. jjrh

    depending how complex your config is it should be pretty painless.

  326. jjrh

    I don't think anyone has been working on fluxbox for a long time.

  327. Ge0rG

    Pretty complex.

  328. Ge0rG

    Funny thing. I've met the fluxbox maintainer in real life, long before realizing that he's the maintainer, on a retro irc party, when those were a thing

  329. jjrh

    Yeah - might take a bit of time then. Everything is XML config in openbox and I think they renamed a few things so it's not a 1-1 mapping unfortunately.

  330. Ge0rG

    I shouldn't be complaining about xml too loudly in this channel

  331. jjrh

    not complaining - just pointing out it's radically different than fluxbox configs :)

  332. jjrh

    Ouch - looks like last release for openbox was 2010 :(

  333. Zash

    XML for config? Mmmmhm.

  334. jjrh

    If I remember correctly the fluxbox configs were really brittle, one wrong space and everything breaks.

  335. moparisthebest

    Ge0rG: might go ahead and and skip virtualbox and go directly to virt-manager/kvm

  336. moparisthebest

    I wish I had

  337. Ge0rG

    moparisthebest: Hm. Does it run macOS and Windows on a Linux host?

  338. moparisthebest

    Ge0rG: windows for sure

  339. moparisthebest

    Probably Mac? Haven't tried

  340. Holger

    jjrh: I used Notion (successor of Ion3) in the past, that one has full-blown support for floating vs. tiling workspaces and handles multiple displays like i3 does.

  341. jjrh

    Interesting - thanks for that :)

  342. Holger

    jjrh: I also ended up with i3 though. It's the younger and easier project and way more popular these days. Compared to that, Notion feels a bit clumsy, with a somewhat complex Lua configuration; i3 has a straightforward key=value config file.

  343. Zash

    I see no `=` in `bindsym $mod+Tab move workspace to output right`

  344. Holger

    Well it's straightforward at least :-)

  345. Holger

    It's actually commands that take arguments, yes.