**** BEGIN LOGGING AT Wed Jan 27 02:59:59 2016
Jan 27 03:09:32 <jonwil> hi
Jan 27 03:36:13 <Vajb> hey jonwil
Jan 27 05:14:45 <Maxdamantus> jonwil: looks like they might not fix the & thing after all.
Jan 27 05:14:57 <jonwil> damn
Jan 27 05:15:00 <jonwil> What did they say?
Jan 27 05:16:03 <Maxdamantus> “Based on our initial investigation this doesn't look like it would allow an injection of HTML metacharacters which could lead to XSS (i.e. this would require < to be unescaped as well). However, if you are able to cause the server output to contained such unescaped characters, it would definitely be a problem -- if that's the case, please let us know!”
Jan 27 05:16:31 <Maxdamantus> I suspect it probably is still a problem, but I don't really have evidence for it.
Jan 27 05:16:41 <Maxdamantus> It would be very strange if it just happened to be an issue with &.
Jan 27 05:17:30 <Maxdamantus> I saw there was a supposed company on Google maps in China somewhere called "<script>alert(1)</script>", but it was marked as invalid.
Jan 27 05:18:33 <Maxdamantus> Maybe that's what they mean by "unable"—there's an assumption that the people who validate places that are able to appear in search results are able to identify potentially malicious company names.
Jan 27 05:19:10 <Maxdamantus> but really, all you need to do is create an actual company called something like "Web <script>s ltd."
Jan 27 05:20:04 <jonwil> Yeah its quite possible they already have stuff at the Google Maps level to block (or rename) marker points with <> symbols in the name
Jan 27 05:20:38 <jonwil> Anyhow it seems like they are saying "its not a security issue therefore we aren't going to put any priority into it" not "we will never fix the problem"
Jan 27 05:20:56 <Maxdamantus> Yeah, that's probably the main thing.
Jan 27 05:21:20 <Maxdamantus> Maybe it'll still happen some time, since at least two people at Google know about it.
Jan 27 05:25:42 <jonwil> I am going to continue using my microb patch for now
Jan 27 05:26:02 <Maxdamantus> I use Opera, so it doesn't really affect me anyway.
Jan 27 05:26:14 <jonwil> lots of people use microb :)
Jan 27 05:26:56 <jonwil> and I have suggested (in the tmo thread) that we consider adding such a patch to microb-engine (IMO its better to have a small hack in microb-engine than to get strange "random" errors from Google)
Jan 27 05:27:19 <Maxdamantus> An alternative MicroB-centric fix would be to just change the user agent.
Jan 27 05:27:42 <Maxdamantus> that's why it doesn't affect Opera, not because of invalid XML workarounds.
Jan 27 05:29:51 <jonwil> but what user agent to use? And how to be sure it isn't going to result in Google (or anyone else) sending the N900 pages it cant handle (or pages that use more mobile data than the current ones)
Jan 27 05:32:25 <jonwil> Its not like there is some web page out there serving XHTML with an unescaped & symbol where parsing the page and ignoring the & instead of throwing an error is going to break things
Jan 27 08:20:46 <Sicelo> so swap fragmentation can cause slowness?
Jan 27 08:22:18 <Sicelo> finally hit 10 days uptime on my N900 (first in a loooong time), but it was becoming too slow. swapping on/off between sd card & mmcblk0p3 seems to have restored sanity. or just placebo effect?
Jan 27 08:37:01 <Sicelo> on another note .. for some weird reason, auto-complete text suddenly does not work with hwkbd, but works with vkbd. what could be the cause?
Jan 27 08:43:24 <bencoh> Sicelo: it's not a placebo effect :)
Jan 27 09:09:23 <Maxdamantus> It's probably not a placebo effect.
Jan 27 09:09:56 <Maxdamantus> When you swapoff, it'll put more pages back into RAM.
Jan 27 09:10:15 <Maxdamantus> You can't really tell it to move all swap from one device to another.
Jan 27 09:11:21 <Maxdamantus> All you can do is tell it to move everything off a particular swap device, using swapoff .. some of those pages might end up staying in RAM, some might end up going to another swap device.
Jan 27 09:12:50 <Maxdamantus> It's probably not really a matter of fragmentation though.
Jan 27 09:13:56 * Maxdamantus only uses SD cards for swap.
Jan 27 09:14:08 * Maxdamantus is still wary of flash storage.
Jan 27 09:31:24 <jonwil> I use whatever the default swap settings are :)
Jan 27 10:20:07 <kerio> yeah because SD cards store data on unicorn farts
Jan 27 10:20:14 <kerio> :v
Jan 27 10:30:03 <Maxdamantus> No. Because SD cards can be replaced.
Jan 27 10:34:06 <Wizzup> you can also boot from sd cards if emmc breaks
Jan 27 10:36:22 <Maxdamantus> Indeed, though I don't intend to break it.
Jan 27 10:38:27 <Vajb> how to chain dbus commands in shortcutd?
Jan 27 10:41:58 <Vajb> trying with &&
Jan 27 10:43:10 <Vajb> hah works
Jan 27 10:43:44 <Vajb> now when i use camerabutton to pause panucci it also rewinds it 10 seconds
Jan 27 11:22:15 <Sicelo> hi. my hwkbd 'problem' is getting more serious. Noticed the symbo table doesn't come up with Fn key anymore 
Jan 27 11:22:29 <Sicelo> how do i restart/reload hildon-im-*?
Jan 27 11:38:39 <Vajb> maybe just kill it?
Jan 27 11:45:42 <Sicelo> good idea. ah, let me check the init
Jan 27 11:48:36 <Sicelo> nothing there. i wonder how it gets started.
Jan 27 13:03:16 <Sicelo> hmm, nothing seems to be restoring normal functionality for my hwkbd
Jan 27 13:03:39 <Vajb> did u try to kill hildon input method?
Jan 27 13:03:44 <DocScrutinizer05> moin
Jan 27 13:04:32 <Vajb> hey DocScrutinizer05
Jan 27 13:05:54 <Sicelo> yes Vajb
Jan 27 13:07:50 <Sicelo> in vkbd all works normally. only hwkbd acting up: no auto-complete, and Fn key doesn't bring up symbol table.
Jan 27 13:12:18 <Vajb> hmm maybe u changed some settings accidentally?
Jan 27 13:13:05 <Sicelo> no. to make sure, i've changed between all the options in control panel. no difference
Jan 27 13:13:30 <Sicelo> guess N900 just asking for a reboot, lol
Jan 27 13:14:00 <Vajb> haha
Jan 27 13:14:12 <Vajb> maybe kill whole hildon desktop?
Jan 27 13:14:40 <Sicelo> i did :)
Jan 27 13:14:53 <Vajb> oh and still no joy...
Jan 27 13:15:10 <Sicelo> yeah
Jan 27 13:16:19 <Vajb> can u make special characters if u longpress keyb?
Jan 27 13:17:11 <Vajb> ah but there is different keys there...
Jan 27 13:17:48 <Sicelo> long press works fine. even Fn + character
Jan 27 13:18:02 <Sicelo> only Fn+Ctrol does not bring up symbol table
Jan 27 13:31:57 <Vajb> btw does ur vkbd work otherwise?
Jan 27 13:34:44 <Vajb> i see lots of posts reporting that vkbd not working after installing SCIM
Jan 27 13:37:42 <Sicelo> vkbd is working just fine.
Jan 27 13:40:05 <Vajb> well im not an expert, but i'll say u have a weird problem :)
Jan 27 13:41:11 <Sicelo> and i don't want to reboot just for this, haha
Jan 27 13:41:49 <Vajb> did u do something odd?
Jan 27 13:42:23 <Sicelo> possibly. i don't know. it's been a couple of days
Jan 27 13:43:03 <Sicelo> not a very big problem for me .. i don't really use the symbols
Jan 27 13:44:33 <Vajb> have u read related wiki article?
Jan 27 13:44:48 <Vajb> special_characters_view
**** ENDING LOGGING AT Thu Jan 28 02:59:58 2016