<awans>
virtualbox: switch to rsync as default method for synced folders, because vboxfs requires contrib or non-free packages
<awans>
seems it
<awans>
(11 days ago)
<zarvox>
repeat: *sigh*
<aldeka>
Aww.
<zarvox>
okay, I'll look into building an image that doesn't suffer this limitation. In the meantime, there might be a vagrant plugin that solves this.
<zarvox>
Lemme see if I can get you a snippet to add to your Vagrantfile to solve this.
<awans>
thanks!
<awans>
or if i can figure out how to manually install vboxguestadditions on the vm, i'm probably good for now?
<zarvox>
asheesh_: please note the above discussion about Debian/vagrant/virtualbox/DFSG-ness
<zarvox>
awans: oh, you should be able to "vagrant plugin install vagrant-vbguest" and then "vagrant-spk destroy && vagrant-spk up"
<awans>
cool, i'll give it a shot
<zarvox>
the Vagrantfile already supports doing the vagrant-vbguest dance if you have the plugin installed
bb010g has quit [Quit: Connection closed for inactivity]
jadewang has joined #sandstorm
<awans>
(fwiw, seems like I'm all set -- thanks for the help!)
jadewang has quit [Remote host closed the connection]
<ocdtrekkie>
I don't know if HN cares, but I submitted this: https://news.ycombinator.com/item?id=10687459 I just discovered how far really bad reporting on behalf of CNBC spreads.
<ocdtrekkie>
Dozens of major tech blogs picked it up, and the article's entire premise is false, because they didn't read their source.
bb010g has joined #sandstorm
<XgF>
ocdtrekkie: I guess the article needs a title correction to append "bought this year"...
<ocdtrekkie>
It's only bought this quarter even.
<ocdtrekkie>
And only laptops, it excludes desktops entirely.
<ocdtrekkie>
The article is horrifically wrong, and it's on Yahoo Finance, every Android blog, it's been copied to dozens of sites.
<ocdtrekkie>
XgF, this makes me sad to be in the tech industry. Nobody reads stuff, so many people just regurgitate if it sounds like a good thing to them.
<XgF>
ocdtrekkie: Where do you see evidence that the rule only applies to laptops?
<ocdtrekkie>
Fine print at the bottom of the infographic.
<XgF>
Why do you think the figures were taken from the infographic?
<XgF>
(I expect the figures were taken from the "Futuresource K-12 Global quaterly research" - from which the infographic was derived
ArcTanSusan has quit [Quit: ArcTanSusan]
aldeka has quit [Remote host closed the connection]
ocdtrekkie has quit [Remote host closed the connection]
aldeka has joined #sandstorm
ocdtrekkie has joined #sandstorm
<ocdtrekkie>
XgF: The infographic is from FutureSource.
<XgF>
Yes
<XgF>
I don't see any evidence teh article's contents were taken from the infographic
<XgF>
I think that they were taken from Futuresource's data - along with the infographic
<ocdtrekkie>
And CNBC's article contradicts the data from FutureSource, which is who they claim their data is from.
<ocdtrekkie>
The notion that "50% of all classroom devices are Chromebooks" is extraordinary. On the level that requires extraordinary proof, which does not exist.
<ocdtrekkie>
However, the fact that the infographic claims the same basic number, meaning something different, seems a clear indicator that their 'source' is known as 'not reading it'.
<XgF>
Given that the article's source appears to be FutureSource, have you gone and read their data?
<XgF>
Because what you're implying is that CNBC bases articles entirely on infographics, which is an extraordinary claim which requires extraordinary proof
<ocdtrekkie>
XgF: It's an analyst, their data is paid only.
<ocdtrekkie>
I'm not implying that, I'm implying the infographic summarizes FutureSource's data.
<XgF>
It summarizes some subset of FutureSource's data
<ocdtrekkie>
And that data does not jive with the contents of the article.
<ocdtrekkie>
I think you're stretching really far for some reason.
<kentonv>
I think either way the message of "there's a lot of Chromebooks in classrooms" is valid.
<ocdtrekkie>
At best, the article's data represents a very tiny fraction of PCs in classrooms. Of those not covered by that data, many may be Chromebooks, but that isn't shown by this data.
<ocdtrekkie>
If you ballpark assume a four-year replacement cycle, common to organizations, and one quarter of sales, this article is talking about maybe 1/16th of classroom devices, of which they claim half are Google's. So maybe in the 1/32nds of classrooms, having Chromebooks. And that's only if you forget about desktops, upon which this data represents an even smaller subset of classrooms.
<XgF>
(Incidentally that article's header video has Google's head of Chrome & Android repeating the same figure - 51% of all devices sold to US schools last year were Chromebooks)
<ocdtrekkie>
Bad writing turned a very tiny footnote into an extraordinary claim.
<kentonv>
It's not like the number suddenly became 51% just last quarter.
<ocdtrekkie>
XgF: The person in the video is reading the FutureSource report.
<ocdtrekkie>
Aka, their placement as being a Googler is irrelevant to the accuracy or lack thereof.
<ocdtrekkie>
Product manager reduced to overglorified title in a marketing video. He's like a paid spokesman here.
<kentonv>
it says 40% in Q3 2014
jadewang has joined #sandstorm
<kentonv>
if I'm reading right
<ocdtrekkie>
And the source claims the 51% is this quarter, but the video says 51% this whole year.
<ocdtrekkie>
So even the Googler reading it seems to be somewhat paraphrasing the data, likely incorrectly.
<XgF>
ocdtrekkie: Erm, even the damn infographic says 51% this year
<ocdtrekkie>
XgF: Q3 2014 vs. Q3 2015.
<ocdtrekkie>
And if in Q3 2014 it was 2014 it was 40%, I doubt it magically became 51% in Q1 2015 and stayed the same for Q2 and Q3 as well. Again, defense of bad reporting is suggesting extraordinary coincidences.
<kentonv>
you're nitpicking, though. Yeah, the words may be inaccurate but if it's gone from 40% to 51% in the last year, and if people get new computers every few years (some more often than others), we can probably conclude that 40%-ish of laptops in schools are in fact chromeos, which is not that far off
<ocdtrekkie>
kentonv: You're missing the other half of this.
<ocdtrekkie>
Desktops.
<ocdtrekkie>
Schools that aren't doing the whole "Chromebooks for everyone" thing, by and large, tend to buy a LOT of desktops.
<ocdtrekkie>
They even got a Google marketing person to regurgitate what is effectively misrepresented data.
<ocdtrekkie>
I feel like I should contact that Googler and let them know, but I doubt they want to admit they vastly overinflated their own product's popularity.
<kentonv>
people put desktops in classrooms?
<XgF>
ocdtrekkie: Because Rajen Sheth is a "marketing person"?
<ocdtrekkie>
XgF: If he's just regurgitating marketing figures that some random analyst hands them? Yeah.
<ocdtrekkie>
I get that random bloggers regurgitate bad figures without reading them, but for a supposedly high-level Googler to do so? That's embarrassing.
<XgF>
ocdtrekkie: Someone who is a PM *wouldn't*. That is a recipe for nasty lawsuites
<ocdtrekkie>
kentonv: Depends what you mean by "in classrooms". Technically, these Chromebooks aren't mostly "in classrooms", a lot are handed out to students to take home. In computer labs, you bet most of it's still all desktop based.
<ocdtrekkie>
XgF: That seems to be what happened here though.
<kentonv>
but a shared computer lab isn't going to buy machines for every student, whereas that's presumably the model with chromebooks
<ocdtrekkie>
kentonv: How's that relevant to the incorrect data presented?
<kentonv>
I'm not sure it really makes sense to put computer lab desktops and take-home laptops in the same category.
<ocdtrekkie>
XgF: Includes the same inaccuracies in the NBC article.
<ocdtrekkie>
kentonv: The claim is "all devices in US classrooms".
<ocdtrekkie>
I didn't declare the broad category, and arguably, the lab PCs are more accurate to that headline than the take-home laptops.
<XgF>
Are they? I bet the laptops get far more use
<kentonv>
I think most people don't think of computer labs when they think of "classrooms". Of course a computer lab is a kind of classroom, but it's a special-purpose one, and most students spend not very much time in there.
<kentonv>
also I think the usage patterns are fundamentally very different
<kentonv>
and therefore putting them in the same category would be weird
<kentonv>
but sure, maybe they should have said "not including computer labs". I don't think it would have changed the impact of the story.
natea has joined #sandstorm
<kentonv>
BTW I'd guess that the vast majority of these kinds of sales happen in Q3, which may be why they'd slip and say "this year".
<XgF>
Yeah - if these are laptop-per-student devices, then you'll have an uberhump in Q3
<XgF>
Where I work we get Q4 & Q1 humps from Christmas device sal
<XgF>
es (since we are paid 1Q in arrears)
<kentonv>
FWIW it also does not surprise me in the slightest that chromebooks would dominate this market. If I were handing out laptops to a school's worth of students I'd give them chromebooks.
<kentonv>
(and my personal dislike of ChromeOS is well-documented.)
<ocdtrekkie>
kentonv: Labs/classrooms take many forms. Computer classrooms, graphic arts classrooms, etc. A lot of classes are taught exclusively in rooms with computers.
<ocdtrekkie>
Again, I think you may be using weasel terms to try to justify a simple failure on the part of the CNBC reporter to actually read.
<ocdtrekkie>
XgF: Again, just laptop per student devices disproportionately favors Google statistics again.
<kentonv>
I'm not justifying anything, I'm just saying I think you're overstating the impact of the error.
<ocdtrekkie>
Because very few to almost nobody else promotes that style of distribution.
<ocdtrekkie>
Maybe Apple.
<XgF>
ocdtrekkie: And I think you may be jumping to conclusions based upon your interpretation of the article. Unless one of us is going to buy FutureSource's data, I doubt we will come to an agreement on this
<ocdtrekkie>
By focusing this on Q3, it's likely the statistic is even more distorted.
<ocdtrekkie>
During Q1, Q2, and Q4, Windows may win out on all of them.
<XgF>
Not really, the bulk of all purchasing is done in Q3
<XgF>
You upgrade all of your stuff during the holiday
<ocdtrekkie>
Wouldn't you upgrade before/after the year? Traditional PC deployments at that scale, maybe not a two week project.
<XgF>
Two week project?
<XgF>
Dunno about the US, but in the UK schools basically close for 2 months
<XgF>
(OK, its 6 weeks, but still)
<kentonv>
in the US they close for three months. :P
<kentonv>
most of which is in Q3
<kentonv>
but I suppose it starts in Q2
<XgF>
Closing^Wkicking out all the students and doing research for 3 months is for the Universities here :P
<kentonv>
but when you're passing out devices to kids, it makes sense to purchase them as the school year starts. You don't need the devices already in a storage room to set up the infrastructure.
<XgF>
And you probably don't want 1000 laptops cluttering up your storage closet all Summer either
<kentonv>
yeah, these aren't the old days where the local IT guy had to configure each and every one.
<ocdtrekkie>
XgF: We close for two weeks, if that, in the winter. Yeah, we close for 2-3 months in the summer, but if you ordered in Q3, you already missed your window.
<ocdtrekkie>
Desktops would have to be ordered in Q2 at the latest to be installed in summer break.
<XgF>
ocdtrekkie: Uh, what?
<XgF>
The summer break will be June-ish to September-ish, which is almost entirely Q3
<ocdtrekkie>
June is Q2, is it not?
<XgF>
Sure, but tail end of June
<XgF>
And these things are normally done based upon when invoices are actually paid, so that probably pushes things to Q3
<ocdtrekkie>
If based on invoices maybe.
<XgF>
Also it wouldn't surprise me at all if the staff take a well deserved holiday right after closure
jadewang has joined #sandstorm
jadewang has quit [Ping timeout: 260 seconds]
ArcTanSusan has joined #sandstorm
|jemc| has quit [Ping timeout: 245 seconds]
bb010g has quit [Quit: Connection closed for inactivity]
simonv3 has quit [Quit: Connection closed for inactivity]
bb010g has joined #sandstorm
todayman has quit [Remote host closed the connection]
TC01 has quit [Remote host closed the connection]
natea has quit [Quit: natea]
jadewang has joined #sandstorm
TC01 has joined #sandstorm
jadewang has quit [Ping timeout: 246 seconds]
enthd has left #sandstorm [#sandstorm]
|jemc| has joined #sandstorm
M-eternaleye has quit [Changing host]
M-eternaleye has joined #sandstorm
M-eternaleye is now known as eternaleye
jadewang has joined #sandstorm
mnutt_ has quit [Quit: mnutt_]
|jemc| has quit [Ping timeout: 245 seconds]
|jemc| has joined #sandstorm
KCinJP2 has joined #sandstorm
jadewang has quit [Remote host closed the connection]
|jemc| has quit [Ping timeout: 246 seconds]
KCinJP2 has quit [Ping timeout: 260 seconds]
|jemc| has joined #sandstorm
bb010g has quit [Quit: Connection closed for inactivity]
|jemc| has quit [Ping timeout: 245 seconds]
jadewang has joined #sandstorm
KCinJP2 has joined #sandstorm
jadewang has quit [Remote host closed the connection]
aldeka has quit [Remote host closed the connection]
ocdtrekkie has quit [Remote host closed the connection]
aldeka has joined #sandstorm
ocdtrekkie has joined #sandstorm
ArcTanSusan has quit [Quit: ArcTanSusan]
KCinJP2 has quit [Ping timeout: 260 seconds]
KCinJP2 has joined #sandstorm
cbaines has quit [Ping timeout: 264 seconds]
cbaines has joined #sandstorm
KCinJP2 has quit [Ping timeout: 260 seconds]
xet7 has joined #sandstorm
aldeka has quit [Remote host closed the connection]
ocdtrekkie has quit [Remote host closed the connection]
aldeka has joined #sandstorm
ocdtrekkie has joined #sandstorm
paroneayea has quit [Read error: Connection reset by peer]
mnutt_ has joined #sandstorm
paroneayea has joined #sandstorm
* asheesh_
stretches.
<asheesh_>
Glad awans got his problem solved, seemingly.
asheesh_ is now known as asheesh
mrshu has quit [Ping timeout: 265 seconds]
mrshu has joined #sandstorm
pod has joined #sandstorm
cbaines has quit [Ping timeout: 264 seconds]
cbaines has joined #sandstorm
geofft has joined #sandstorm
mnutt_ has quit [Quit: mnutt_]
<asheesh>
https://getsentry.com/welcome/ simonft and I were talking about Sentry last night. It'd be cool if there were a Sentry package.
<asheesh>
There are a few sticking points he mentioned that I thought were interesting.
<asheesh>
- For a company, it's not clear he should rely on something whose future is less certain than (say) Debian's future, which seems pretty solid, even if slow-moving.
<asheesh>
- Doing the work of packaging Sentry isn't clearly easier than "just" running it himself.
mrdomino has quit [Remote host closed the connection]
mnutt_ has joined #sandstorm
mrdomino has joined #sandstorm
<asheesh>
I would use a git-annex Sandstorm app that looked as nice as FileDrop.
<asheesh>
I mean, as nice as Davros.
<asheesh>
I don't have a clear mental model of under what circumstances owncloudcmd is willing to delete files yet.
<mrdomino>
agreed re git-annex
<mrdomino>
are there any plans, theories, or pie-in-the-sky stories about durable storage and computation in sandstorm? things like raft, paxos, gfs?
<asheesh>
"durable" storage, well, make backups of your filesystem!!
<asheesh>
Semi-joking, semi-not.
<asheesh>
One could imagine a Tahoe-LAFS package within Sandstorm, if you want that. Sandstorm is optimized for single-machine use-cases, but you could federate a couple of Sandstorms this way.
<dwrensha>
apparently I can get openjdk-7's java to crash with "*** buffer overflow detected ***: /usr/lib/jvm/java-7-openjdk-amd64/bin/java terminated"
<|jemc|>
plus they let you SSH into the build to poke around if you want to debug what went wrong
<dwrensha>
I'm thinking now that the build failures have to do with some problem in SBT.
aldeka has quit [Remote host closed the connection]
ocdtrekkie has quit [Remote host closed the connection]
aldeka has joined #sandstorm
ocdtrekkie has joined #sandstorm
notevil has joined #sandstorm
funwhilelost has joined #sandstorm
hunterm__ has quit []
hunterm__ has joined #sandstorm
<simonft>
asheesh: I'll add another. I don't know what scaling it would look like on sandstorm.
jadewang has joined #sandstorm
funwhilelost has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
funwhilelost has joined #sandstorm
erikmaarten has joined #sandstorm
amyers has joined #sandstorm
amyers has quit [Remote host closed the connection]
amyers has joined #sandstorm
funwhilelost has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
funwhilelost has joined #sandstorm
<dwrensha>
my storage bar is all colored in on testrock
<dwrensha>
17.2 MB used out of 100 GB total
<dwrensha>
but visually, it looks like I'm at capacity
<asheesh>
dwrensha: I wonder if I introduced some NaNs.
<asheesh>
Let me check.
<asheesh>
BTW <3 the purple "test" text.
<asheesh>
width: NaN%
<asheesh>
Let me fix that now.
<asheesh>
I thought I had fixed that but clearly I was wrong.
<asheesh>
Fix ETA ~20 min or less
<asheesh>
I have no point, I just want to say:
<asheesh>
Crouching Tiger Hidden Sidebar
<asheesh>
Ooh nice new sidebar.
amyers has quit [Read error: Connection reset by peer]
amyers has joined #sandstorm
<asheesh>
Yeah, honestly very nice new sidebar.
rustyrazorblade has joined #sandstorm
rustyrazorblade has quit [Quit: rustyrazorblade]
rustyrazorblade has joined #sandstorm
rustyrazorblade has quit [Quit: rustyrazorblade]
mnutt_ has joined #sandstorm
<mnutt_>
asheesh: if you only care about read-only file access, you could probably plop davros’ frontend on top of git-annex and call it a day, as davros currently just stores files and directories as-is. you’d just want a git post-receive hook to check out a non-bare copy and point davros at that.
<mnutt_>
there may be a little bit of mucking around to figure out how to combine the bare and non-bare repos to save on storage
<asheesh>
FWIW I'm not so much interested in "read-only" as much as "If I run owncloudcmd once, then ^C it, then re-run it, will it now go delete my local files?"
<asheesh>
This possible failure mode is on my mind due to memories of offlineimap bugs.
<asheesh>
For some reason I already trust the design of git-annex, so I suspect it won't have that bug.
<asheesh>
then again git-annex does support webdav remotes!
rustyrazorblade has joined #sandstorm
rustyrazorblade has quit [Client Quit]
funwhilelost has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
funwhilelost has joined #sandstorm
amyers has quit [Ping timeout: 256 seconds]
aldeka has quit [Remote host closed the connection]
ocdtrekkie has quit [Remote host closed the connection]
mnutt_ has quit [Ping timeout: 256 seconds]
aldeka has joined #sandstorm
ocdtrekkie has joined #sandstorm
notevil has quit [Quit: +++ ATZ]
erikmaarten has quit [Ping timeout: 250 seconds]
mnutt_ has joined #sandstorm
<mnutt_>
asheesh: I am also terrified of that, I’m working on a separate client that will just walk the filesystem and do a push-only update. it may be possible to modify owncloudcmd, but I think at some point they stated this wasn’t a priority for the command line syncer.
<mnutt_>
and regarding read-only I meant that the web UI would be read-only, from the command line you’d use git-annex as usual
<asheesh>
Ah, gotcha re: read-only
<asheesh>
You wanna see an awesome log message?
<asheesh>
"Will switch certs, probably to /var/sandcats/https/rose.sandcats.io/1449306826631 , at time Invalid Date via real-time SNI re-keying system.
<asheesh>
"
<asheesh>
A free almond to the first person who can state something suspicious about the above.
<dwrensha>
did you swap two variables?
<asheesh>
I think I'm adding a new String(new Date()) and an integer? not sure, but the answer should be clear soonish.
simonv3 has quit [Quit: Connection closed for inactivity]
amyers has joined #sandstorm
<asheesh>
My kingdom for a typechecker.
<asheesh>
Although this thing could really be refactored into an npm module and that is generated from typescript.
<asheesh>
I think I have a winner. Testing now.
<zarvox>
asheesh: man, I know that feeling ("My kingdom for a typechecker!")
<asheesh>
Will switch certs, probably to /var/sandcats/https/rose.sandcats.io/1449306826631 , at time Sat Dec 05 2015 04:33:57 GMT+0000 (UTC) via real-time SNI re-keying system.
<asheesh>
We have a winner.
<asheesh>
Re-keying resulting in a new certificate. Good.
<asheesh>
Yay.
<zarvox>
Yay!
amyers has quit [Remote host closed the connection]
amyers has joined #sandstorm
amyers has quit [Read error: Connection reset by peer]
amyers has joined #sandstorm
<ocdtrekkie>
mnutt_ did you see my question about forcing text only preview in Davros? It was on a closed issue.
<mnutt_>
ocdtrekkie: just saw it now, and I think that makes sense. I comes down to some file types probably needing both a “raw” mode and a rendered mode. I’ll have to look into how that might work.
<ocdtrekkie>
Yeah, with HTML and MD and stuff, a raw text view would be nice.
<mnutt_>
it would be trivial to show raw text and not render html tags, I just need to figure out how it fits into the UI.
<ocdtrekkie>
I'd love to force text so we can do like php and other code files that wouldn't render/run properly but might have HTML tags in it.
<mnutt_>
I’m not actually sure there’s much of a use case for rendering html inside of davros’s UI anyway. if you want to display as html, turn on web publishing.
<mnutt_>
markdown is right down the middle, though. I’d probably prefer to see it rendered in the UI, especially since I can’t edit it there at the moment.
simonv3 has joined #sandstorm
<asheesh>
BTW zarvox you may not know it but
<asheesh>
"Re-keying resulting in a new certificate. Good."
<asheesh>
is a literal log message from sandstorm.log.
<asheesh>
(-:
asmyers has joined #sandstorm
amyers has quit [Remote host closed the connection]
mnutt_ has quit [Read error: Connection reset by peer]
mnutt_ has joined #sandstorm
jadewang has quit [Remote host closed the connection]
funwhilelost has quit [Quit: My Mac has gone to sleep. ZZZzzz…]