#kbsA useful thing jwz points out (once you get past the condescending aggressiveness!) is the distinction between high-availability and backups - it took me quite some time to appreciate it...
#tantekbarely squeezed those both in! Four URLs in a tweet = lots of characters for t.cos
#tantek!tell benwerd, erinjo, j12t great job this week with the projects you've launched! your efforts have really leveled up the #indieweb and #ownyourdata momentum.
#j12ttantek : put something about indie box on your blog, that way I can publicly say thanks, too!
#Loqij12t: tantek left you a message 1 minute ago: great job this week with the projects you've launched! your efforts have really leveled up the #indieweb and #ownyourdata momentum.
#kylewmbret: any more details anywhere about this? "folks from the indieweb and Evan from pump.io have reached a censuses to collaborate!" http://bret.io/2013/08/08/t5/
#bretkylewm: I talked to Evan about that sometime last year in person and he said he wanted to see cross site posting in pump
#breti submitted a simple uf2 patch to pump and let it sit
#tantekbret - no idea. mozilla uses lots of different things. and mozillians themselves use lots of different things. you'd have to check sites and people case-by-case, and you could likely investigate as well as I could.
#rasculaccording to buddycloud.com, mozilla does use it
#kylewmhmm, does anyone know if releasing my project itself as GPL prevents me from factoring out libraries and making them LGPL?
#pauloppenheimkylewm: if you're the only copyright holder, you can do whatever you want whenever you want. As soon as you accept a patch, things change
#pauloppenheimkylewm: why use a share-alike style license in the first place if that's not your intent?
#kylewmgreat question, I think that is what I have to decide
#kylewmI believe in copyleft in theory, but in practice, there are a lot of people just experimenting here, not planning or ready to release their code, and i don't want to disallow that at all
#kylewmwhere is indiewebcamp.com/licensing when I need it :)
pauloppenheim joined the channel
#pauloppenheimkylewm: who would you like to use your code? What is their license? Could they use GPL code?
#rasculit looks similar except that it takes more time to read
#rascultwo big issues i've seen that people have with wtfpl is the language and that it doesn't state there's no warranty
#rasculpersonally i don't feel that stating no warranty should be necessary
#kylewmwtfpl just seems unnecessarily snarky to me
#rasculheh i like it because it's like a jab at all the big licenses
#rasculi keep looking at fair license, but it doesn't really say that you can do whatever you want, it just says to keep the licenses and there's no warranty
#rasculi guess i should correct my earlier statement: i don't like gpl for my stuff
#kylewmah yeah, i think that is my attitude too. i'm glad it exists, i'm glad that important software used it, but i don't think it's going to negatively impact the world if someone uses my blog engine for commercial purposes. :)
#rasculi don't care one bit whatsoever if someone uses my blog engine type thing for commercial purposes
#rasculi write my software for me and me alone, but if someone else finds it useful, i'm ok with that
#rasculmaybe if something i write gets big and widespread my attitude might change on that particular project, but that hasn't happened yet and i don't see it happening in the near future
#kbsrascul: (only as you ask :) love sites that render reasonably without javascript
#kbsas I use noscript. There's some strange shadowing issue at narrower screen widths at http://crash.rascul.io/ (firefox on macos) - be happy to upload a screenshot if that's useful
#rasculkbs i haven't made it work right on smaller screens yet, but does it work right with no javascript? i haven't checked that yet either, and it only uses just a little bit
#kbsrascul: fonts - right. As you can tell, it's not an actual obstacle for most things (unless it uses icon fonts as buttons and one can't figure out what a button is supposed to do)
#rascul(don't get attached to those urls they won't last too long)
#kbsaaronpk: best luck vis-a-vis data recovery. Your notes here have also helped increase Amazon's revenue by a couple hundred bucks vis-a-vis drives and misc peripherals
#kbsrascul: I think it might be one of the plugins, let me poke a bit
#kbsaaronpk: new peripherals are mostly for my friend - she's interested in the auto-archiving thing I've got going for email+facebook+photos from my android phone - I thought I'd risk attaching a pi rather than an openwrt'ed router just so I can play with it
#kbsjust upgraded to firefox 29 and can't find the place where noscript reports its status!
#rascul24 esr here but i'll install some more browsers later on to test em out and stuff
#kbsfwiw - KartikPrabhu goes bananas with svg rather than using icon fonts - which works nicely with noscript. He'll probably have some neat ideas, should definitely ask him when he's awake
#kbsidly wondering whether "the web" should learn from the brain and have some way of auto-forgetting some things. eg: an infrequently clicked link reduces the probability of that link remaining over time. Conversely, a frequently clicked link begins to cache the linked-to page, to increase the likelihood of the content remaining
#kbs(blame it on feeling sleepy on this sunny day and procrastinating on getting things done...)
#kylewminterested in electronic music and/or cyborgs :)
#kbsaaronpk: feel more comfortable adding to /IndieArchive if/when I or someone actually implements some sort of local caching proxy - right now the 'idea' such as it is, feels too vague to me...
#aaronpkthat's what a "Brainstorming" header is for :)
#rasculoh i guess i could just look at the wiki...
#kbsrascul: fwiw - tantek has his polemic re databases here http://indiewebcamp.com/databases-antipattern - while I'm less strident :) for archival purposes at the least, it's seems useful to always have a bunch of simple files that can be masaged into new database (versions/formats/types) etc over time.
#rasculi see stuff for html to json but not the other way
#rasculi'm thinking up my storage idea, i should note this stuff down somewhere
#rasculi used to be of the though to do markdown stuff, but it seems to me that since i know html then maybe i should skip the markdown and store the html for my posts in a bunch of files
#kbsOne slightly less obvious thing (for some definition of obvious? :) is filesystems themselves are databases - they too can become obsolete...
#Loqirascul meant to say: i used to be of the thought to do markdown stuff, but it seems to me that since i know html then maybe i should skip the markdown and store the html for my posts in a bunch of files
#kbs(until you find that the last backup you had was on a sunos ufs filesystem, but I digress... :)
#aaronpkbtw if you know of any significant differences between the way mysql and postgres work with regards to things on that page, that would be worth noting
#kbs*nod* funnily enough, that's the URL I visit the most often :) was starting to think there's usually enough interesting stuff there that it's worth grepping [for me anyway] every once in a while
#aaronpkit wouldn't take much for me to add a .txt version of the irc logs to the filesystem
#aaronpkok so since we have a bunch of wiki pages that are the same name but different case, how should I be handling that?
#aaronpki spoze I could check everything in to a git repo all the time, but that seems less automated and more likely to fail at some point
#kbswas thinking more doing that just when tarballing it (should you ever decide to tarball your downloads I guess, dunno :) I know little about mediawiki, and should stop random peanut gallery commentary...
#aaronpki'm going to have to do some fancy logic for only creating the non-redirect version
#aaronpkhere's a question, is it worth storing the redirect pages in the archive at all?
#kbsout of curiosity - does mediawiki in essense just have a (btsync'able I guess) directory of content? So restore == copy back this directory and point the software at this?
#kbspretty sure you've taken this into account (but having burnt myself a few times :) would be awesome to allow for snapshots of some kind rather than only a 'latest backup'. (c.f. my scars - some script runs rm -r on something, not noticed for a while, backup runs, and all copies are now useless.)
#aaronpkdo you think that should be part of the shared folder?
#aaronpklike, would you mind hosting weekly snapshots of the wiki?
#kbsI'm not sure about the actual mechanics to be honest - not least because I don't understand mediawiki (and its internal versioning of content as well)
#kbsbut that's also why I was initially wondering whether git might serve that purpose - but not fully sure :) also, I guess mediawiki has some internal understanding of versioning as well
#aaronpkyeah mediawiki keeps a full history of each page at every revision
#kbsaaronpk: and sure - goes without saying, be happy to suck down copies into whatever random boxes I have scattered around :) I can also potentially keep the various versions regardless naturally
#snarfedhey gRegor`, remember when you wanted the twitter user's URL from bridgy mf2 pages that showed their web site instead?
#snarfeddid you end up extracting their Twitter handle from the u-id? or do you still want their twitter URL?
#snarfedalso, yeah, pin13.net is aaronpk's and it does use php-mf2
#KartikPrabhukbs: rascul: i use icon fonts too. For all the little icons in the menu items for example. Though even there I prioritise SVG over others :)
#j12tjust noticing that wikipedia has no entry on indieweb
#gRegor`I haven't done any more on that, no. I've thought about it some and was thinking I might just use the same name on the same post as a pretty good guess it's the same person. Lower priority, though.
emmak joined the channel
#gRegor`But yes, it would be awesome if the same profile URL (Twitter, FB) showed up in all wm, in addition to the author's site