<rjo>
whitequark: i can image overflow on shifts would have made the barrel shifter quite a bit larger.
_rht has joined #m-labs
ylamarre has joined #m-labs
ylamarre has quit [Remote host closed the connection]
sandeepkr has quit [Ping timeout: 244 seconds]
sb0 has joined #m-labs
<sb0>
what, Qt only calls the model once with ToolTipRole when a tooltip is displayed, and not 537 times at application startup, 345 times when scrolling, 12340 when the tooltip is shown and 234 when it is hidden?
<sb0>
surprising
<GitHub17>
[artiq] sbourdeauducq pushed 2 new commits to master: https://git.io/vVujy
<GitHub17>
artiq/master d9b2968 Sebastien Bourdeauducq: gui/models/DictSyncTreeSepModel: support tooltips
<GitHub17>
artiq/master 2dfc1a8 Sebastien Bourdeauducq: gui/explorer: display filenames and classes in tooltips. Closes #371
<rjo>
all electrons were cage free and none were harmed. they were all safely returned.
<rjo>
lets divide the two numbers by another factor of two and just change the labels accordingly. seems fair to me.
_rht has quit [Quit: Connection closed for inactivity]
<whitequark>
ack
key2 has joined #m-labs
rohitksingh has joined #m-labs
<sb0>
whitequark, pff whatsapp
<sb0>
do you know they banned me for using a reverse-engineered client?
<sb0>
if they actually cared about privacy and not buzzwords such as "end to end encryption", "open whisper systems", "snowden", etc. they would not have done that
<sb0>
and reverse engineering would not have been required in the first place
<larsc>
but you might be using your client to spy on other people!!!
<rjo>
open whisper systems is actually good and open.
<whitequark>
calling "end to end encryption" a buzzword just shows that you have no clue what you're talking about
<sb0>
it's not a buzzword in itself.
mumptai has joined #m-labs
<whitequark>
as for the rest... if you want a FOSS solution, you can use Signal
<whitequark>
they brought end-to-end encryption to 1B of people, most of which don't give a fuck about FOSS or even know what that is
<whitequark>
that counts
<whitequark>
if your solution to privacy starts with "well, use linux" then you are going nowhere very fast.
<sb0>
and insist you must use their proprietary client, and actively ban FOSS users?
<whitequark>
yes. that's bad. no question here.
<whitequark>
and ultimately seems rather pointless on top of that.
<sb0>
bad to the point they're untrustable imo
<whitequark>
how many people have lost their whatsapp account vs how many people gained an important measure against an adversary that's active right now?
<whitequark>
my issue with trusting whatsapp is different. the US government cannot demand their keys, but they can compel them to silently push a backdoored update
<sb0>
yes.
<whitequark>
which is a problem that's shared by the majority of privacy tools today, and pretty much all the ones that are actually usable by people who cannot compule shit from source
<whitequark>
so... yeah? but we don't have anything better right now.
<whitequark>
compile*
<sb0>
but those other privacy tools 1) are often not run by silly valley companies whose typical business model is selling user data (yes, whatsapp claim they're not doing that) 2) do not have this nonsensical attitude of simultaneously attacking FOSS clients
<whitequark>
they are also not used.
<whitequark>
a deployment of a high-quality end-to-end protocol of this scale is historically unprecedental, and it presents massive, immediate improvement even in light of its flaws
<whitequark>
the next step is to improve on that.
<sb0>
"high-quality", who says? does it have public specs?
<whitequark>
it uses Axolotl
<whitequark>
it has public specs and it's pretty much the state of the art right now. far better than OTR
<whitequark>
one possibility of making money for a messenger that's not involving selling user data is the LINE model
<whitequark>
have you seen LINE?
<whitequark>
its business model is very funny. they let people purchase "stickers", which are custom emoticons, mostly based on anime and movie characters. which they can then send to each other.
<whitequark>
it's wildly profitable. something like $0.5B USD in 2014, if I remember correctly
key2 has quit [Ping timeout: 276 seconds]
<whitequark>
and the thing is, WhatsApp doesn't even need much revenue. it has a team of ~50 engineers. let's give each a generous silly valley salary of $300k. given how most of the operating costs for a tech company is salaries, just over $15m can keep it afloat indefinitely
<whitequark>
(as an aside: how can 50 people serve the communication needs of 1B? Erlang, mostly)
sandeepkr_ has joined #m-labs
sandeepkr has quit [Read error: No route to host]
sandeepkr__ has joined #m-labs
sandeepkr_ has quit [Ping timeout: 252 seconds]
rohitksingh has quit [Ping timeout: 240 seconds]
<GitHub167>
[artiq] jordens pushed 2 new commits to master: https://git.io/vV2uR
<GitHub167>
artiq/master 4bc5eaf Robert Jordens: test_rtio: scale speed test results to 'event' intervals
<GitHub167>
artiq/master 1a04cc0 Robert Jordens: browser: add thumbnail icon provider