#ossasepia Logs for 08 Apr 2020



April 21st, 2020 by Diana Coman
diana_coman: http://ossasepia.com/2020/04/21/ossasepia-logs-for-07-Apr-2020#1023870 - hopefully none other are missing then, lolz. [04:39]
ossabot: Logged on 2020-04-07 18:23:35 jfw: diana_coman: I got the missing tarballs from asciilifeform btw. (thread) [04:39]
trinque: http://ossasepia.com/2020/04/21/ossasepia-logs-for-07-Apr-2020#1023859 << incidentally I'm engaged in a decrufting of the gcc script ave1 found and modified to support musl. [12:06]
ossabot: Logged on 2020-04-07 16:55:08 billymg: diana_coman: ah, gotcha. i can't say for sure if i'm most familiar with the codebase, my changes/contributions so far have been targeted and the codebase is rather large, however... [12:06]
trinque: for the very same reason. [12:06]
trinque: the thing was in such a terrible state I'll write on just how terrible and why when I'm done, but the summary is yes, maximal optionality at the expense of... being civilized, really. [12:07]
trinque: that was not strictly speaking ave1's doing; it was in that state before him. [12:07]
trinque: but I spent about an hour trying to get the thing to spit out a static *gcc* (which it wasn't) rather than a dynamically linked gcc (which can only emit static build products) [12:09]
trinque: and then I said fuck it all, thanks for the list of dependencies, and started over. [12:09]
trinque: there's a psychological element of it, in that the original author *was not* making the thing "easier to use". he was autistically listing all the things he knew, because he's a very smart boy. [12:12]
trinque: billymg: if I'm reading correctly the idea is to put your item in an existing plugin, and later get rid of the plugin mechanism? [12:20]
trinque: seems a sane progression, sure [12:20]
feedbot: http://trilema.com/2020/qntra-sqntr-closing-statement/ << Trilema -- Qntra (S.QNTR) Closing Statement [13:13]
diana_coman: http://younghands.club/2020/04/06/jfw-plan-week-of-6-apr-2020/#comment-625 - jfw, that's at least an interesting question; as usual though, I'd say the evaluation starts not from imagined benefits but from very real and concrete costs (of keeping it private); for one thing, if I understand correctly that you do make that available to your clients, then how exactly are you going to keep it private ? [16:12]
diana_coman: what, you make them sign papers and then chase the leak or what? [16:12]
diana_coman: dorion: the above is supposedly of equal interest to you too. [16:13]
diana_coman: whaack: did you figure out if all that data collection thing is of much interest to you or not? [16:23]
diana_coman: BingoBoingo: do I understand correctly that you'd rather drop that plan & review stuff as it's not of much use to you? [16:24]
jfw: diana_coman: I don't fancy pushing NDAs and chasing leaks, no. [16:43]
trinque: ^ great way to buy your lawyer a boat [16:43]
jfw: seems like the kind of game that you have to be microsoft to play [16:44]
trinque: even them, the windows source code has leaked how many times [16:44]
diana_coman: jfw: I'd have thought it obvious too, but the question didn't seem to care about it, lolz. [16:45]
jfw: trinque: sure, I imagine the boat floats so long as you can bail faster than the leaks. [16:45]
BingoBoingo: diana_coman: I'm going to try to get back on the review and plan horse, but with the turn to the IRL Sunday has become the day I've been getting restless, putting miles on the shoes, and not coming up with much out of it. [16:47]
diana_coman: still read that "so long as you can boil faster than the leeks" [16:47]
diana_coman: BingoBoingo: I'd have thought Sundays were supposedly the most nothing-happens-days even at best at times, no? [16:48]
diana_coman: I mean I get the restlessness but ... [16:48]
BingoBoingo: diana_coman: Well, now it's the day when there's some people on the street buying groceries at the feria. It was the boring day when everything was closed. [16:49]
BingoBoingo: Now most places are closed every day. [16:49]
jfw: diana_coman: what's less obvious to me is whether there's a middle ground between truly secret and actively published by us, that could work at some scale. Because there's value in ready access, which I figure is part of why a central dealer of a product can make a spread even if his counterparts could otherwise trade privately [16:55]
jfw: perhaps a closer analogy, sometimes people still buy music even if they can torrent it free, because there's more certainty / consistency [16:57]
jfw: one obvious cost of keeping private though is lack of feedback from people here (even if we don't go to the 'open source' notion of 'many eyeballs' materializing from the ether) [17:00]
diana_coman: jfw: more to the point the question is what are you selling there exactly? because if you imagine that you are selling the recipe aka the text, the "intellectual property" or similar, then a. you don't have anything worth selling b. you'll play that leeks-for-the-lawyer game. [17:01]
jfw: we're selling training, so focus on the active part of that rather than some imagined income from textbook royalties essentially? [17:04]
diana_coman: jfw: that's pretty much how it looks to me, yes. [17:04]
diana_coman: jfw: sure, you can also sell whatever printed copies/nicely set/certified-by-jwrd copies I suppose, not like it's forbidden or not possible (ie re your buying music even if available otherwise for free) but again, there as with the training, you are selling that additional part & guarantee, not the "recipe" [17:07]
jfw: diana_coman: alright thanks, would be nice to avoid "semi-private document" dances if indeed there's no benefit. I'll let it simmer and see if further objections come to mind. [17:08]
diana_coman: jfw: lol, you are still after the benefits, ok. [17:09]
jfw: d'oh. [17:10]
diana_coman: jfw: I don't know how well you get/are aware of eulora&smg otherwise but in any case, if you want to look at it for this sort of question, what you'd need to look at is ~what's the private part, what are the reasons given for keeping it private and how narrowly/widely is it cut out of the rest otherwise [17:13]
diana_coman: but other than that, ~any "must-be-kept-private" is a significant cost the way I see it and one that needs to be *justified* well to consider taking it on; sure, maybe that's just me, can be. [17:14]
jfw: my picture of eulora is that the server internals are kept private because part of the game is exploring to learn what's in the environment and how it responds [17:16]
diana_coman: quite, yes. [17:16]
jfw: where was it MP wrote that the thing with spam isn't that it pays, but that people imagine it pays because they see others doing it, and so goes the feedback loop. possibly why I imagine that quasi-secrecy can pay. [17:26]
diana_coman: jfw: tbh I'd say the benefit is first of all that you can point to it and clearly say "here it is, first time, made by JWRD"; which goes away the moment someone else publishes the paper they got from you, you know? [17:36]
jfw: that's a point. [17:37]
diana_coman: other than that sure, I can see the usual game with idiots going "oh, but it's public info"; in which case, sure let them see what they can do with the "public info" on their own, that sort of "clients" are going to keep being more trouble than it's worth it as far as I'm concerned; then again, if you want those... [17:38]
diana_coman: dorion: what's your view on this secret documentation anyway? (see the thread above) [17:38]
billymg: http://ossasepia.com/2020/04/21/ossasepia-logs-for-08-Apr-2020#1023882 << that's the general outline, yeah [17:57]
ossabot: Logged on 2020-04-08 12:20:02 trinque: billymg: if I'm reading correctly the idea is to put your item in an existing plugin, and later get rid of the plugin mechanism? [17:57]
billymg: althoug a recent discussion with jfw has me wondering what level of customization should be available in a project like mp-wp, and how that customization should be handled [18:03]
billymg: although* [18:04]
billymg: for example: the footnotes plugin allows for the adjustment of some constants to change e.g. the markers from decimal numbres to roman numerals, or the code itself can be modified to alter the generated markup [18:05]
billymg: and then there's the example of spyked's use of a markdown plugin for his build [18:06]
billymg: i won't get into themes, those to me are 100\% operator owned and meant to be customized wholesale [18:06]
billymg: but jfw brought up a good point, that if one needs/wants to customize parts of mp-wp (which seems entirely valid in th case of a blogging platform -- it's *your* site afterall) then it becomes more difficult to pull in new work from another branch [18:08]
jfw: the footnote identifier case seems simple: there's already code there for decimal or roman numerals, so unless you get rid of one, the optionality is there. [18:11]
billymg: jfw: as in what the constant is set to? [18:11]
jfw: however it's implemented, there's a switch that's meant to be operator-exposed. [18:13]
jfw: whereas in the case of the footnote delimiters, http://logs.ossasepia.com/log/trilema/2020-01-27#1957382 [18:15]
ossabot: (trilema) 2020-01-27 mircea_popescu: http://logs.ossasepia.com/log/trilema/2020-01-26#1957358 << honestly, no. i think the "option" to "pick the special chars" was useful at the onset, when i was figuring out how the world should work, but double-parens is so well established by now, even etymologically! ("every time you go on a tanget, you have to decide -- is your paranthetical is simple or double?") that it needn't be misrepresented as an option anymore, a [18:15]
billymg: jfw: yes, and in that same thread i described removing the code that writes the options to the db (and exposes them to operator), with the idea that "operator" in this case can edit a constant and doesn't need checkboxes in UI to update db [18:18]
jfw: my 'simple' is a bit too simple above: it makes sense for the identifier style to be configurable because experience showed that both styles worked well in different themes. [18:18]
jfw: billymg: possibly he didn't realize this was among the db-configured options? [18:19]
billymg: yes, possibly [18:20]
jfw: personally I'd prefer more config through... config files, and leave the database for... data! hence my suggesting a wp-config.php variable. it'd be a question of whether one cares about admin GUIs. [18:20]
billymg: jfw: i agree, i would favor the flags in wp-config.php instead of the db (which then requires the extra db and gui code for adjusting them( [18:21]
billymg: jfw: it sounds like i had it wrong with my initial reasoning of "user can edit footnotes.php". yes, user can edit php file, but options/settings/config should be pulled as much as possible into wp-config.php so that when one presses a new patch only one file needs to be edited to restore their customizations [18:34]
billymg: and by that logic, this value should perhaps also be pulled into wp-config.php [18:35]
billymg: or not, and have users overwrite in theme css [18:36]
billymg: but if not that then maybe a flag for whether pre/post identifiers are inside/outside the <a> tag (referencing your tweak for the footnote markers) [18:37]
billymg: i'm curious what others think about this list of options from the footnotes plugin: http://paste.deedbot.org/?id=XSi4 [18:46]
billymg: it could be that this whole options map can be set in wp-config.php, and then the plugin would read from what's there and set defaults for missing keys (similar to what it did previously with the database) [18:48]
billymg: ^ diana_coman whaack trinque spyked BingoBoingo dorion lobbes [18:54]
jfw: billymg: could that 670px be done as a percentage of the parent element's width? [20:18]
trinque: billymg: I'd say first yes, pull all the optionality dorkitude into one place, but then aggressively prune said optionality. [20:21]
jfw: afaik I'm the only one using the pre/post identifier knob for my (i) footnotes, and I'd live without 'em [20:22]
jfw: what I like about them is that when pasting you don't end up with wordsi like thisii [20:25]
trinque: the roman numeral / digit distinction can just as well be custom css [20:26]
trinque: and probably ought [20:27]
jfw: bbl [20:27]
trinque: I'm pretty skeptical that there's any need for theming other than custom css. [20:27]
BingoBoingo: finds the current footnote behavior acceptable [20:46]
billymg: http://ossasepia.com/2020/04/21/ossasepia-logs-for-08-Apr-2020#1023947 << good point, will set to 100\% and users can always override if they want it something other than the width of their paragraphs [23:26]
ossabot: Logged on 2020-04-08 20:18:02 jfw: billymg: could that 670px be done as a percentage of the parent element's width? [23:26]
billymg: trinque: makes sense re: consolidate then prune strategy [23:28]
billymg: as for css roman numerals, the footnotes in the <ol> at the bottom of an article are set with css, but the footnotes interspersed in the article text are generated individually, not sure if those can be set via css [23:32]
billymg: http://ossasepia.com/2020/04/21/ossasepia-logs-for-08-Apr-2020#1023954 << i think this is what i'd like to get to eventually as well [23:33]
ossabot: Logged on 2020-04-08 20:27:37 trinque: I'm pretty skeptical that there's any need for theming other than custom css. [23:33]

Comments feed: RSS 2.0

Leave a Reply