<dwrensha>
oh wait, maybe my non-discount markdown is okay. I had just got into a bad state after it failed the first time. :/
<dwrensha>
jparyani: doesn't jenkins use Dockerfile.dev? how was the build green without the discount dependency?
<jparyani>
nope, we switched to using local builds a while ago
<jparyani>
sandstorm is instlalled in a tmp directory inside the /tests directory, and the test script cleans up after itself
<dwrensha>
so you had to go in and type "sudo apt-get install discount"?
<jparyani>
yep
mcpherrin has joined #sandstorm
gopar has joined #sandstorm
hunterm has quit [Read error: Connection reset by peer]
jancborchardt has quit [Read error: Connection reset by peer]
bpierre has quit [Write error: Connection reset by peer]
jksonc has quit [Read error: Connection reset by peer]
hunterm has joined #sandstorm
jancborchardt has joined #sandstorm
<ocdtrekkie>
I for one like the current tracing method for spk dev.
<ocdtrekkie>
I would probably fail at determining most of what it determines for me.
<ocdtrekkie>
Could spk dev just have a flag to not trace for a certain run?
<ocdtrekkie>
So you trace if you're adding new packages and files and features, but can easily not trace when you want to make sure you add nothing new?
jadewang has quit [Remote host closed the connection]
joshbuddy has joined #sandstorm
jadewang has joined #sandstorm
joshbuddy has quit [Quit: joshbuddy]
jadewang has quit [Ping timeout: 265 seconds]
mort___ has joined #sandstorm
jadewang has joined #sandstorm
jadewang has quit [Ping timeout: 264 seconds]
jadewang has joined #sandstorm
jadewang has quit [Ping timeout: 265 seconds]
jadewang has joined #sandstorm
jadewang has quit [Ping timeout: 264 seconds]
amyers has joined #sandstorm
decipherstatic has quit [Quit: No Ping reply in 180 seconds.]
jadewang has joined #sandstorm
decipherstatic has joined #sandstorm
jadewang has quit [Ping timeout: 256 seconds]
<ocdtrekkie>
Question: EtherDraw doesn't support the user box yet to show participants, nor as far as I know have a read only view yet, so implementing Sandstorm features for the short term is unlikely.
<ocdtrekkie>
Should I put it on the app list for now as is?
<dwrensha>
in my opinion, yes
jadewang has joined #sandstorm
jadewang has quit [Ping timeout: 276 seconds]
paroneayea has quit [Read error: Connection reset by peer]
paroneayea has joined #sandstorm
jadewang has joined #sandstorm
jadewang has quit [Ping timeout: 244 seconds]
<ocdtrekkie>
Well, I got the appMarketingVersion added correctly, and I hit the share/embed button which doesn't work in Sandstorm, so the new package is slightly more polished.
<ocdtrekkie>
kentonv: If I submit EtherDraw to the app list, where do you want it?
<ocdtrekkie>
Random Linux thing I just learned: If your terminal is in a folder, and you delete that folder and replace that folder with another folder at the same location, and try to do something in terminal, it is using the folder in your trash rather than the folder at that location. (Windows does not do this.)
mort___ has left #sandstorm [#sandstorm]
jadewang has joined #sandstorm
jadewang has quit [Ping timeout: 258 seconds]
<XgF>
ocdtrekkie: curdir is maintained by inode reference. Name resolution is done relative to the inode
<XgF>
cd symlinkdir; unset PWD; sh -c pwd
<ocdtrekkie>
XgF: I kinda figured it was some such like that. It just wasn't a behavior I :expected:.
jadewang has joined #sandstorm
jadewang has quit [Ping timeout: 256 seconds]
jadewang has joined #sandstorm
bb010g has quit [Quit: Connection closed for inactivity]
itscassa|away is now known as itscassa
joshbuddy has joined #sandstorm
gopar has joined #sandstorm
<kentonv>
ocdtrekkie: It may seem surprising at a user interface level, but as a systems engineer I am really, really glad it works this way. A lot of things would be harder to code correctly if the current directory were maintained as a string rather than a pointer.
<kentonv>
also there are places in Sandstorm where we intentionally have the current directory set to a dir that isn't visible at any path. :)
<kentonv>
fun fact: the root directory is _also_ just an inode pointer than can be changed per-process.
gopar has quit [Remote host closed the connection]
mort___ has joined #sandstorm
<ocdtrekkie>
kentonv: Honestly, it seems reasonable, I just am used to Windows and such. ;)
<ocdtrekkie>
So I was learning things.
<ocdtrekkie>
kentonv: Nudge on EtherDraw placement question for app list.
natea has joined #sandstorm
<kentonv>
ocdtrekkie: Hmm... Right after HackerSlides?
bb010g has joined #sandstorm
mort___ has quit [Quit: Leaving.]
amyers has quit [Ping timeout: 272 seconds]
<ocdtrekkie>
kentonv: Okay, I'm good with whatever. WTB app store where they're sorted by how many different users click on them or something similar. :P
<kentonv>
yeah trying to judge where to put things only makes me feel bad.
<XgF>
kentonv: Are you still using your Pixel? Finding a PC laptop I actually want to own is a world of hurt...
<XgF>
Because it seems my current laptop's battery has given up, and the other option is handing Apple another obscene wad of cash...
<ocdtrekkie>
XgF: The question is what type of laptop do you want to own?
<ocdtrekkie>
What do you want/not want?
<XgF>
ocdtrekkie: Want: Portable. Good battery life. Good build quality. Good screen. Reasonable performance.
<XgF>
Essential: Good trackpad and keyboard
<ocdtrekkie>
Cuz IIRC, Sandstorm.io folks had been looking at the new Dell Mobile Precisions that come with Linux out of the box.
gopar has joined #sandstorm
<ocdtrekkie>
Which, while I'd take the Windows variety, is the sort of laptop I'd salivate over.
<ocdtrekkie>
(We have two Mobile Precisions in the company, and they're beasts.)
<XgF>
Those all seem to be 15"
<ocdtrekkie>
They have a 17" one too. Though that one is not light, I can tell you. :P
<ocdtrekkie>
It's basically a desktop. In a laptop.
<ocdtrekkie>
:D
<XgF>
I have a desktop. In a desktop.
<XgF>
I want a bed/travel machine :-)
<ocdtrekkie>
See, I can't fathom buying a screen smaller than 15" personally. S'why I asked what you were looking for. :D
<XgF>
IME 15" is basically unusable on a train, it wouldn't fit my bag, ...
<ocdtrekkie>
By "good screen", what are you looking for? Resolution, brightness, etc.?
<kentonv>
XgF: I am still using my Pixel, but I do not recommend it.
<kentonv>
XgF: Well, unless you want to run ChromeOS.
<kentonv>
(I'm running Ubuntu on mine.)
<kentonv>
The dell XPS developer edition is 13" and ships with Ubuntu
<kentonv>
apparently the 2015 edition is quite nice? Though I haven't actually tried one.
<ocdtrekkie>
Though in fairness, I've seen plenty of issues with the Lenovo Yogas I've dealt with.
<ocdtrekkie>
And Lenovo is going to be borderline useless supporting Linux, I assume.
<zarvox>
I went with the Precision m3800 (15") with the 3840x2160 panel.
<XgF>
I have a Lenovo through work, the trackpad is crap
<ocdtrekkie>
I personally have yet to desire a screen greater than 1080p.
<zarvox>
Dell managed to accidentally ship it with broadcom chipsets for wifi and bluetooth, which wasn't great.
<ocdtrekkie>
Of the... eight screens I use, none of them are greater than 1080p, and that doesn't bother me.
<ocdtrekkie>
Actually, my phone screen, for some mystifying reason, is above 1080p, and I mostly just assume that's a waste of battery life, because 840x480 was fine.
<XgF>
I wish the screen on my Lenovo T440s wasn't 1080p, but only because Windows 7 can't handle different DPIs on different displays
<ocdtrekkie>
XgF: How does the XPS 13 look to you?
<ocdtrekkie>
(See, I'm the sort that'd want an M4800 or M6800 because they have disc drives.)
<XgF>
No 16GB of RAM config (:() and the trackpad has an ominous looking mark on it as if to indicate the bottom bit is virtual buttons
<kentonv>
XgF: In Linux, you can almost certainly configure the presence and positioning of virtual buttons.
<XgF>
Probably
gopar has quit [Quit: Leaving]
<ocdtrekkie>
EtherDraw -> app list PR submitted. I tagged John McLear in case he wants it credited differently. It seems to be "his repo", but the Etherpad blog post about it made it sound like a team project.
<paulproteus>
It'd be nice if we could test systems for the "Your kernel will lock up, this is really bad" issue, rather than locking up their systems.
mattl has quit [Ping timeout: 252 seconds]
<kentonv>
paulproteus: Yes, though someone will have to spend time developing that, and maybe the time is better spent chasing down whatever distros/services are still shipping this kernel.
<kentonv>
paulproteus: also, how would we test it? Simply checking the kernel version doesn't work, since we can't tell if it's been patched. Triggering the problem would not be a good way to test. :)
mquandalle has quit [Ping timeout: 276 seconds]
<zarvox>
FWIW, kernel.org still hasn't released 4.0.5 :/
natea has quit [Quit: natea]
itscassa is now known as itscassa|away
<paulproteus>
kentonv: Yeah it occurred to me re: "triggering the problem would not be a good way to test"
<paulproteus>
However, it _almost_ would be.
<paulproteus>
Maybe, like:
<paulproteus>
* If the current kernel version has not successfully started a grain yet, and this is the first time we're starting a grain with that kernel version, make a file like
<paulproteus>
* At start, if the former filename exists, but the latter doesn't, declare 'Your kernel seems broken, be careful, read <a href="somewhere">here</a>' in the UI somewhere.
<paulproteus>
This would also help people who somehow don't have user namespaces enabled, although it's a bit coarse-grained.
<paulproteus>
I'm not actually convinced this is worth doing, but that's one way it could be done.
<paulproteus>
</ramble>
<kentonv>
yeah it's possible but seems like a lot of work...
mattl has joined #sandstorm
gopar has joined #sandstorm
mquandalle has joined #sandstorm
paroneayea has quit [Read error: Connection reset by peer]
<paulproteus>
Maybe we can just ask nicely and they'll change their very academic ways and, like, offer a dual-license with AGPLv3 or something.
<paulproteus>
I figure if they want to reserve rights to use it proprietarily, AGPLv3 is a pretty safe way to avoid that.
ragesoss_ is now known as ragesoss
<zarvox>
give it a shot!
<zarvox>
also I'd potentially be willing to do the clean-room reverse-engineering/reimplementation game with someone if we think it'll be worthwhile/isn't way way way too much crufty academic code
natea has joined #sandstorm
<paulproteus>
Oh I was hoping you would contact them zarvox since you know some of the Berkeley people at least.