#editorial | Logs for 2020-04-06

« return
[04:26:18] -!- AzumaHazuki has quit [Remote host closed the connection]
[05:47:35] <Bytram> https://go.theregister.co.uk
[05:47:36] <systemd> ^ 03NASA reveals the new wavy Martian wheels it thinks can crush the red planet ( https://www.theregister.co.uk )
[05:48:48] <chromas> from the wavy-lays dept.
[05:55:12] * Bytram waves
[05:56:30] * chromas transmits chips into the channel
[05:56:48] * Bytram transmits dips into the channel
[05:57:05] * Bytram notes the channel dips in response
[06:00:12] * chromas puts the dips and chips together to get slashdot
[06:00:17] <chromas> Better eat it before it goes beta
[06:01:03] * Bytram thinks there may be an issues of timing, here.
[06:01:11] <Bytram> break
[06:01:21] <chromas> There is. You should be asleep
[06:08:06] <Bytram> =g NASA rss
[06:08:07] <systemd> https://www.nasa.gov - NASA RSS Feeds | NASA
[06:13:54] <Bytram> annnd there's another story for the submissions queue
[06:14:45] <chromas> Ooh, image of the day
[06:14:55] <chromas> I should set that feed to my background in windows
[06:15:12] <chromas> unless they took that out for 10
[06:31:01] <janrinok> Bytram, we were having a discussion regarding git yesterday. Firstly, if you are simply bug squashing then there is no need to use git. There are things that git could do to help you in bug squashing, but the activity does no benefit much by using git.
[06:41:23] <janrinok> If however you want to identify fixes and, having done so, submit them for inclusion into the next release of prod then git is essential. Every piece of code and change can be identified and the reason for it, who wrote it, when it was included in the build, and much more is needed and git does all of this for you. If you are bug squashing - as you often do - you might find that the problem you are investigating was introduced as a
[06:41:23] <janrinok> unintended side-effect of another change elsewhere.
[06:43:00] <janrinok> however, you can use git to keep track of all of your scripts, your documentation, any notes you have made (and you have made lots!) etc etc. It doesn't have to be used for software.
[06:45:46] <janrinok> And if we ever get back having several people actually working in dev rather than just TMB it will become essential to enable the different branches to be tracked and possibly integrated into the production release.
[07:21:00] <chromas> Hmm. One ed leaves, another one takes his place. I'm not fooled by this.
[07:52:40] <janrinok> it's true, you have never seen Bytram, TheMightyBuzzard and me in a room at the same time - go on, admit it
[08:01:07] <chromas> I knew it. Soylent News is person!
[08:02:35] <janrinok> can you explain - if simple terms - why XSS is a stopper for your sub software?
[08:02:52] <janrinok> in simple terms*
[08:04:15] <chromas> It's used for web malwares so some browsers and/or browser extensions block it.
[08:05:06] <chromas> When I try to execute acquire_reskey() by clicking the button, Firefox says the SN server (the one we're making the request to) needs to set "Access-Control-Allow-Origin" in the HTTP header
[08:06:21] <chromas> That way the browser knows it's okay to let you hit up the server in that way. Otherwise, I could have a fake page that looks like SN, then trick you into visiting it and when you submit some sort of data, I can steal it while still passing the data onto SN
[08:07:05] <chromas> Although it doesn't seem to be a problem to submit a form that way without javascript
[08:07:17] <chromas> and I could still use js to steal the form data when you click the button, so I dunno
[08:07:56] <janrinok> so what we would really like to do is compile the sub on your software and when it is transferred to the SN review page get that page to request the reskey itself.
[08:08:42] <janrinok> that way we have the original author, all the publishable material, and a reskey that is valid. Is that summary accurate?
[08:09:44] <janrinok> I only ask as I am trying to do a similar thing and I would rather avoid stumbling in the same places as you. I prefer to stumble in places of my own choosing thank you!
[08:10:02] <chromas> Yeah. The reason I have it grab a reskey is so I can submit directly from my form without previewing on SN. Seems to me that previewing on SN without a reskey would cause SN to generate one during preview, but I could be remembering wrong.
[08:10:46] <chromas> Well I never had any problems running the script (except since yesterday, when I added js to grab a reskey). Only seems to be a problem for you guys, who have different IPs
[08:10:59] <chromas> So if you're running it on your box then it should be fine
[08:11:19] <chromas> !grab janrinok
[08:11:19] <Bender> Added quote 93
[08:12:41] <janrinok> so we have to find a way of running your software with a reskey from a different origin than the submission itself. If you have the author's ID, couldn't you submit it from your server - or am I missing something?
[08:13:24] <janrinok> as long as it arrives at SN with the correct author he.she will still get the karma.
[08:14:39] <chromas> I don't have your session cookie. I can stick your nick in the field but I don't think it counts the same way. For example, notice exec subs don't show karma in the subs list, while upstart's do. exec isn't logged in
[08:15:13] <janrinok> so, rather than move a reskey around, we can safely transfer the authors creds securely can't we? Including the session cookie
[08:15:27] <chromas> Sometimes you'll see upstart subs with no karma and in the preview it will show "upstart" writes instead of linking the nick. That's because upstart lost its login cookie
[08:16:31] <chromas> You could. Hm, that's another thing I hadn't thought of. When I do a sub with my form, since I'm hitting up the SN server, Firefox is automatically using my login session on SN
[08:16:39] <janrinok> so, for a submission, is the only thing that is needed - apart from the submission itself - the session cookie or does there have to be a user physically logged in to your machine?
[08:17:23] <chromas> Session cookie. When you sub with the form, your browser is directly connecting to SN.
[08:18:00] <chromas> But if I had your session cookie, I could have it run a script (like upstart does) to submit for you
[08:18:06] <janrinok> so if the actual submission is done by your server could it use a session cookie that I had sent to you securely?
[08:18:07] <chromas> on my end
[08:18:09] <janrinok> gotcha'd
[08:18:50] <chromas> Yep. For the bot, I do a login using the rehash API and it just stores a cookie that gets used whenever it does a =submit
[08:19:42] <chromas> So if it had another session cookie it would use that account. Not sure what happens if the username doesn't match up, other than the subs queue won't show karma next to the username
[08:20:25] <janrinok> well we can try it on dev if you knock something together. Prolly best we don't cause prod to barf
[08:20:39] <chromas> If I gave you my cookie then you could have Arthur submit as me :D
[08:23:03] <chromas> I wonder what would happen if the js that grabs the reskey was hosted on sn
[08:23:54] <janrinok> arthur currently submits using the SN sub page and so would have to go via the API directly, - doable but I would need to make some changes.
[08:24:51] <janrinok> gtg - back in a while
[14:15:54] <carny> =submit gentlemen-start-your-factories dept. http://feedproxy.google.com
[14:15:56] <systemd> Submitting "More of us are now supposed to wear a mask"...( 1 modified urls; https://www.cnet.com )
[14:16:17] <systemd> ✓ Sub-ccess! "03More of Us are Now Supposed to Wear a Mask" (8 paragraphs) -> https://soylentnews.org
[14:16:44] <carny> that one needs a new title
[14:17:29] <carny> this saves me from having to build my own hepa filter papr hack
[14:17:55] <carny> they can probably get that ramped up in a month
[14:46:15] -!- AzumaHazuki [AzumaHazuki!~hazuki@the.end.of.time] has joined #editorial
[15:37:12] -!- AzumaHazuki has quit [Remote host closed the connection]
[15:53:16] <Bytram> janrinok: chromas: Interesting discussion! In my automated comment-composing/replying scripts, IIRC, all I needed was a cookie and I was all set. Somehow never got to creating and submitting a story. On the one hand, there is the matter that even if I could automatically create a story (easy enough), and submit a story (fun with reskey and POST), then there was the matter of automating (1) figuring out what date/time to release at
[15:53:16] <Bytram> (mostly done), (2) putting release date/time into the "form", and (3) Previewing, and (4) effectively "pressing the 'Save' button"
[15:56:12] <Bytram> Today we have the first nice day of weather in about a week. /me needs to get out for at least an hour or two (haven't left my apartment/flat in at least a week.
[17:25:49] <janrinok> Bytram, make the most of the good weather and any (limited?) chances that you have to get outside. With regards to submitting stories, you appear to be trying to both submit a story and post it remotely. Both chromas and I are only looking at the submission process. We want to make it easier for more people to submit stories.
[17:29:17] <janrinok> Personally, I can see no benefit to myself in remotely processing subs into stories - if I have to sit at the computer I might as well go directly to the site itself.
[17:29:47] <janrinok> might see/read you later on - if not, have a good one!
[18:01:24] <Bytram> janrinok: thanks for the git info, and weather suggestions
[18:02:41] <Bytram> the benefits I see are resolving URLs so they are no longer site-relative, cleaning off extraneous (to us, at least) tags and attributes, story title capitalization, etc.
[18:03:37] <Bytram> I'm all for making it easier for people to submit stories (and reduce the amount of cleanup we need to do)
[18:09:25] <janrinok> OK, but in that case I didn't understand the reference to 'data/time to release at' - surely that is a processing function, not a submission function?
[18:09:58] <Bytram> Oh, that. Yeah, just carrying on to what I would ultimately like to be able to do.
[18:10:49] <janrinok> ah, OK
[18:11:08] <janrinok> anyway - have you been out and enjoyed the sun a little?
[18:11:13] <Bytram> Automate the creation of a story submission, getting it into the submission queue, promoting to main page (or Nexus... more fun!), and then hanging comments off it and moderating same.
[18:11:14] <Bytram> not yet
[18:11:41] <Bytram> I can already submit comments, replies, and moderate them. Just thinking ahead to cmopleting the process!
[18:14:22] <Bytram> sub queue 2nding completed
[18:38:02] <cmn32480> ~gday bytram
[18:38:10] <cmn32480> ~gday janrinok
[18:38:18] <cmn32480> #smake exec
[18:38:18] * MrPlow smakes exec upside the head with aouto-wash
[18:39:55] -!- exec [exec!~exec@23.24.kp.ip] has joined #editorial
[18:39:55] -!- exec has quit [Changing host]
[18:39:56] -!- exec [exec!~exec@crutchys.brothel] has joined #editorial
[19:35:00] <chromas> ~g'day itself
[19:35:01] * exec extra redundantly hallucinates a ring buffer of bacon-flavored candy canes with itself
[22:41:28] <Bytram> =g SME
[22:41:29] <systemd> https://www.sme.org - SME
[22:41:50] <Bytram> =g SME small
[22:41:51] <systemd> https://www.investopedia.com - Small and Mid-size Enterprise (SME) Definition
[22:42:02] <Bytram> =g SME wikipedia
[22:42:03] <systemd> https://en.wikipedia.org - SME - Wikipedia
[22:43:56] <chromas> Since the Register didn't italicize "sic" should I put a sic next to it?
[22:44:10] <chromas> "cobalt [sic][sic]"
[22:45:23] <Bytram> nah, tht would be sick
[22:45:50] <Bytram> =g sikh
[22:45:50] <systemd> https://en.wikipedia.org - Sikhs - Wikipedia
[22:46:19] <Bytram> ever see six sick sikhs?
[22:46:29] <Bytram> =cite https://www.eurekalert.org
[22:46:30] <systemd> Journal Reference INSEAD research finds how much CEOs matter to firm performance, <cite></cite> (DOI: <a href="https://doi.org/"></a>)
[22:46:31] <exec> └─ 13INSEAD research finds how much CEOs matter to firm performance | EurekAlert! Science News
[22:46:43] <Bytram> yeah, didn't think so...
[22:47:02] <Bytram> i.e. not surprised, paywalled.
[22:47:19] <Bytram> =cite https://onlinelibrary.wiley.com
[22:47:20] <systemd> Journal Reference MORTEN BENNEDSEN, FRANCISCO PÉREZ‐GONZÁLEZ, DANIEL WOLFENZON. Do CEOs Matter? Evidence from Hospitalization Events, <cite>The Journal of Finance</cite> (DOI: <a href="https://doi.org/10.1111/jofi.12897">10.1111/jofi.12897</a>)
[22:47:21] <exec> └─ 13Do CEOs Matter? Evidence from Hospitalization Events - BENNEDSEN - - The Journal of Finance - Wiley Online Library
[22:47:27] <Bytram> woohoo!
[22:49:56] <Bytram> chromas++
[22:49:56] <Bender> karma - chromas: 136
[22:50:15] * Bytram does not recall seeing authors names IN ALL CAPS before.
[22:50:27] * chromas steals the midnight slot
[22:50:32] <chromas> I can filter it if you want
[22:50:47] <Bytram> s/[A-Z]//g
[22:50:49] <exec> <Bytram> does not recall seeing authors names before.
[22:50:54] <Bytram> LOL!
[22:51:09] <Bytram> the you get in to wierdess with person
[22:51:50] <chromas> persons
[22:51:52] <chromas> peopleses
[22:51:55] <Bytram> the you get in to wierdess with person's names, and which parts are capitalized, and ... names are just non-deterministic. So, just let it stick out for us to notice and fix...
[22:52:01] <Bytram> or...
[22:52:37] <Bytram> offer a decapitalized name list on the next line for us to eyeball, and optionally choose to use, instead.
[22:53:50] <chromas> Looks like Eurekalert and the site it copied from don't cite a paper
[22:53:56] <chromas> It just says forthcoming paper
[22:55:07] * chromas reaches the paywall page
[22:55:21] <chromas> =cite 10.1111/jofi.12897
[22:55:22] <systemd> Journal Reference MORTEN BENNEDSEN, FRANCISCO PÉREZ‐GONZÁLEZ, DANIEL WOLFENZON. Do CEOs Matter? Evidence from Hospitalization Events, <cite>The Journal of Finance</cite> (DOI: <a href="https://doi.org/10.1111/jofi.12897">10.1111/jofi.12897</a>)
[22:55:52] <chromas> Ah I thought you were pointing it to one of those off-brand sites like arxiv or whatever
[22:56:32] <chromas> Those dirty, wiley coyotes
[22:56:35] <Bytram> off-brand?
[22:56:37] <Bytram> LOL!
[22:56:48] <Bytram> =g INSEAD
[22:56:48] <systemd> https://www.insead.edu - The Business School for the World | INSEAD
[22:57:03] <Bytram> yeah, and what does INSEAD stand for?
[22:57:39] <chromas> Speaking of caching articles with archive.org
[22:57:52] <chromas> Was that just for editors or for other people to look at?
[22:58:06] <Bytram> =g INSEAD wikipedia
[22:58:07] <systemd> https://en.wikipedia.org - INSEAD - Wikipedia
[22:58:22] <Bytram> yes!
[22:58:39] <chromas> What I know about INSEAD is their site sucks
[22:59:15] <Bytram> javascript require banner covering whole page.
[22:59:31] <Bytram> =yt you can't there from here
[22:59:32] <systemd> https://youtube.com - You Can't Get There From Here with Charls Carroll 15 (41:56)
[22:59:48] <chromas> Haven't seen that but it does make the page invisible for a few seconds after loading
[22:59:48] <Bytram> =yt you can't there from here bert & I
[22:59:49] <systemd> https://youtube.com - Bert and I (5:25)
[23:00:49] <Bytram> Morten Bennedsen, Francisco PÉrez-Gonzalez, Daniel Wolfenzon
[23:00:57] <Bytram> *almost* got it!
[23:06:41] <Bytram> LOL! We both slotted a story into the exact same time! I'll fix mine.
[23:08:27] <chromas> Sweet, I get first post! (of the day)
[23:09:33] <Bytram> Hmm, I wonder if any of those folk seeking unemployment benefits know how to program in COBOL?
[23:09:50] <Bytram> =g gold of kobol
[23:09:51] <systemd> https://www.discogs.com - Kobol Electronics - Computer Pervaded Society (2007, Vinyl) | Discogs
[23:09:56] <Bytram> =g gods of kobol
[23:09:57] <systemd> https://neoencyclopedia.fandom.com - Lords of Kobol | Neo Encyclopedia Wiki | Fandom
[23:10:11] <Bytram> =g battlestar galactica gods of kobol
[23:10:12] <systemd> https://en.wikipedia.org - Battlestar Galactica - Wikipedia
[23:10:46] <chromas> Aw, you put the [] outside the superscript
[23:11:09] <Bytram> hrrrm? Oh! Good catch!
[23:11:29] <chromas> I dunno if it matters
[23:11:39] * chromas saves
[23:11:49] <Bytram> could go either way, but I like your idea.
[23:11:56] <Bytram> well, no
[23:12:41] <Bytram> on second thought, [] are editorial additions and should remain in normal size to *clearly* identify editroial additions
[23:12:43] <chromas> I've thought about learning COBOL, but then I look at some example code and then I no longer think about it.
[23:12:54] <chromas> True. but so is the $
[23:13:14] <chromas> Actually I thought I took the tags out
[23:13:24] <Bytram> nope, that is the "message" that the "[" and "]" are calling out to the reader
[23:13:24] <chromas> from the bot template
[23:13:38] <Bytram> chromas: s/at//
[23:13:39] <exec> <Bytram> <chromas> from the bot temple
[23:15:11] <chromas> I probably should've capitalized Cobol in the title. I'll leave that for someone else. Job preservation!
[23:15:34] <Bytram> fixing
[23:17:53] <chromas> I'm raising your bonus by 11%
[23:18:00] <Bytram> updated
[23:18:06] <Bytram> is that in binary?
[23:18:17] <Bytram> or hex?
[23:18:31] <Bytram> =g base 36
[23:18:31] <systemd> https://en.wikipedia.org - Base36 - Wikipedia
[23:18:35] <Bytram> lol
[23:18:36] <chromas> Hm. I remember one language used % prefix for binary I think
[23:21:41] <Bytram> It
[23:22:17] <Bytram> it's been a *long* times, but I want to say that % denoted an integer in ... BASIC?
[23:22:26] <Bytram> =g HEXATRIGESIMAL
[23:22:27] <systemd> https://trustconverter.com - Hexatrigesimal (base 36) (hexa) Conversion - Base Number ...
[23:22:38] <chromas> Yeah
[23:22:48] <chromas> %10011011 for binary in Pascal
[23:23:05] <chromas> And in BASIC there was variable% for ints
[23:26:12] <Bytram> yep, that's it for BASIC
[23:26:24] <Bytram> foo$ was a string?
[23:26:29] <Bytram> or was it $foo?
[23:26:37] <Bytram> I think it was a suffix
[23:26:52] <Bytram> made it easier for the interpreter.
[23:28:00] <Bytram> just gather chars until a 'signifier' (e.g. "$" or "%") so the you have the symbol name, and when you get to the signifier,, then you know how much space you will need to allocated for it.
[23:28:40] <chromas> A$ = INKEY$
[23:29:07] <Bytram> B$ = OUTLOCK$
[23:29:22] <chromas> Some BASICs would just take the first three or so letters of the variable name and ignore the rest, much like VNC with passwords
[23:30:02] <Bytram> I've seen similar at different lengths. The trouble comes when they fail, silently.
[23:30:12] <chromas> I don't remember OUTLOCK$ and the web just wants me to look at Mcrisoft Outlook
[23:31:35] <Bytram> varname(MoneyAmountDollar) == varname(MoneyAmountCent) == varname(MoneyAmountSymbol) if you only look at, say, the first 8 chars.
[23:31:42] * Bytram made it up
[23:31:46] <Bytram> =)
[23:31:53] <Bytram> short break
[23:31:56] <Bytram> afk
[23:32:27] <chromas> Well now I'm disappointed
[23:42:18] <Bytram> so sorry to dis-appoint