ericbot: Logged on 2019-11-29 15:20:21 mircea_popescu: lobbes, except if you do that, i'd rather do it on trilema.com first
lobbes: in other news
bracket linking now functional in mpwp logger. Tip of the hat again to asciilifeform, as I ended up lifting his bracket linking code verbatim from the reader.py
lobbes continues first thing tomorrow, but for now sleeps
ossabot: Logged on 2019-11-29 17:10:06 mircea_popescu: do we have a problem there, is it ?
mircea_popescu: well, ideologically the thing stood at "let development be painful-er than entirely painless, it'll produce better code in the end".
mircea_popescu: then again, that was pushed through in a different context, maybe needs reexamination
diana_coman: mircea_popescu: trouble is that it's *not* development that is painful-er, not at all; it's...publishing that is painfuller and I doubt that has any merits.
diana_coman: should add that I think the problem is relatively superficial ie one of tools as they currently stand, not one of principles.
diana_coman: and then ofc something stares me in the face but I can't yet even fully scope it, gah.
mircea_popescu: taylorism ftw, half the time the problem's one of "you know, if you set out the tools on the bench BEFORE you got the hot piece out of the oven, rather than run around the smithy holding a chunk of red iron while looking for things..."
diana_coman: mircea_popescu: mno, it's not just a matter of "tools were not set out properly"; tools were set out and there was no time lost gathering them or whatever; yes, there is some inevitable delay when using key, for obv reason but by now that part is as streamlined as it can currently be.
diana_coman: sure, you can say "you're an idiot, set out your workbench better"; maybe it is that; I think there's more to it though but probably I'll have to find some time to scope it and write it down.
diana_coman: it can be said also that "well, 2 lines of change taking 10 minutes to do should NOT HAVE BEEN a vpatch"
diana_coman: because a vpatch comes with such a large overhead
diana_coman: maybe that is true though I think it shouldn't be.
diana_coman: not to mention that I specifically packed more than just the initial change that you were pushing, precisely because of this issue - the overhead on a vpatch means that small but useful changes will have to wait to accumulate.
diana_coman: at any rate, the hour was spent: ~10 minutes figuring out the change and making it directly on my blog so that the results were clear; ~30 minutes for full vpatch process: retrieve+press current v-tree to head; make code + manifest file changes; test press of result & check + final sign; ~20 minutes for write-up + upload + overall final check.
diana_coman: the grating thing being also that the only alternative I see would be "drop it in the logs/blog" as a clue/how-to and then keep searching for it or something.
mircea_popescu: honestly it doesn't seem to me the breakdown you have is unreasonable. it will take some time to write things up. this time SHOULD be taken, there's no benefit from "oh, let me mention things vaguely and anti-usefully in the log", it's worse than nothing, it's a cost upon the future.
mircea_popescu: now, that time shouldn't be excessive ; but 20 minutes hardly strikes me as excessive, dunno, is the argument here ?
mircea_popescu: arguably a press + vdiff + test press + sign shouldn't be half an hour, but it seems to me by now we're trying to measure the measuring tool, it's not even clear the measurement tolerance is narrower than the width of that "shouldn't". by how much shouldn't it be half hour.
mircea_popescu: maybe the problem here is one of expectations management.
diana_coman: mircea_popescu: no, it's neither about the write up (after all, I did it *because* I considered it needed) nor about the 20 minutes on it; it's about the fact that a vpatch containing ~30 hours worth of changes would have taken ~similar in writeup and the rest.
diana_coman: hence the "will have to wait until they accumulate"
diana_coman: code changes that took 30 minutes, lol; not sure how can one define it more precisely in the general like that, it depends on a lot of things after all.
diana_coman: no, it's not about the loc and I specifically did not mention loc.
mircea_popescu: i guarantee you it'd have taken anyone else > 30 minutes to figure the shit out. just because you have the massiva advantage of perfect familiarity doesn't mean so much,
mircea_popescu: as the man said, "you wanted me to design a typeface, it takes 10 minutes + a lifetime dedicated to preparing for those 10 minutes"
diana_coman: well, the way this goes then is simply this - I should have passed the vpatch making on to a more junior person then; because ok, familiarity -> figure the shit out but the fixed cost of vpatch packaging is what it is
mircea_popescu: incidentally -- there's a divergence in your patch. on the first heading you say and comment_type = '' but on the second heading you say
mircea_popescu: it's really the same conditional, shouldn't it rather be and comment_type = comment' ?
diana_coman: mircea_popescu: that's in mpwp, what can I do if it does same thing in 2 different ways?
diana_coman: in one place it does a direct select in db ; in the other it uses its own function and that has its own parameters
mircea_popescu: you're doing the same thing in two different ways -- one place you ask for an empty type, the other for a "comment" type
mircea_popescu: the function is just a wrapper on the exact same thing, wp_list_comments('type=comment'); is exactly SELECT * FROM $wpdb->comments WHERE comment_approved = '1' and comment_type = 'comment'
diana_coman: it's mpwp that does same thing in two different ways; all I do is to conform to its different ways; and no; in db the comment_type is (as I checked) with 3 distinct values: empty string, "trackback" or "pingback"
diana_coman: no "pings" for instance; but that wp_list_comments method has specific parameters at least according to docs
diana_coman: this sort of shit is what I spent those 10 minutes on.
mircea_popescu: anyway, this is then a mpwp guffaw, should evidently stick to one convention.
ossabot: Logged on 2019-11-30 14:16:26 diana_coman: no, it's not about the loc and I specifically did not mention loc.
ossabot: Logged on 2019-11-30 14:10:35 diana_coman: code changes that took 30 minutes, lol; not sure how can one define it more precisely in the general like that, it depends on a lot of things after all.
mircea_popescu: im thinking... maybe the best kind of patch available is the one that takes the maker ten minutes to make.
mircea_popescu: or in broader words : maybe we deeply misunderstand HOW to score what we do with computers. when to feel good, when not to, what to be happy or sad about
diana_coman: maybe it's just that I remembered why I had taken a break from computers, can be.
diana_coman: because they are sufferable only in small doses really and even then I'm not sure it's not just ..for contrast.
ossabot: Logged on 2019-11-01 22:46:01 mircea_popescu: anyways -- i dun think a blog should list pingbacks either in the comment rss or in the "new comments" navbar.
diana_coman: and more to the point: trackbacks *are* conversation namely via articles so I'd rather see also trackbacks to other blogs that I follow, hm.
mircea_popescu: in other news, this was trilema's biggest month this year, readership-wise. so was october, back then, and september... it's been crawling monotonically all year.