From kmsujit at gmail.com  Tue Aug  1 04:54:03 2017
From: kmsujit at gmail.com (Sujit K M)
Date: Tue, 1 Aug 2017 14:24:03 +0530
Subject: [talk] Personalization II
In-Reply-To: <20170731185545.GG84110@ayvali.org>
References: <CA+6mNeru6b4itt-oLT0usrLExCM2UdM4_aPMgW3NMEE4mQ4Pww@mail.gmail.com>
 <20170731185545.GG84110@ayvali.org>
Message-ID: <CA+6mNerUCahF55ggDE8VoQ0k53yv8xw8D7cHWBPVb=3W86BCgA@mail.gmail.com>

> He suspects the credit card transaction data was sold to Google.

Who sold the data? The merchant/bank or the payment gateway
software was tweaked to give this data.



From njt at ayvali.org  Tue Aug  1 12:41:10 2017
From: njt at ayvali.org (N.J. Thomas)
Date: Tue, 1 Aug 2017 09:41:10 -0700
Subject: [talk] Personalization II
In-Reply-To: <CA+6mNerUCahF55ggDE8VoQ0k53yv8xw8D7cHWBPVb=3W86BCgA@mail.gmail.com>
References: <CA+6mNeru6b4itt-oLT0usrLExCM2UdM4_aPMgW3NMEE4mQ4Pww@mail.gmail.com>
 <20170731185545.GG84110@ayvali.org>
 <CA+6mNerUCahF55ggDE8VoQ0k53yv8xw8D7cHWBPVb=3W86BCgA@mail.gmail.com>
Message-ID: <20170801164110.GA12555@ayvali.org>

* Sujit K M <kmsujit at gmail.com> [2017-08-01 14:24:03+0530]:
> > He suspects the credit card transaction data was sold to Google.
> 
> Who sold the data? The merchant/bank or the payment gateway
> software was tweaked to give this data.

We suspect the bank sold it to Google. Nothing to back that up though.

Thomas



From kmsujit at gmail.com  Fri Aug  4 10:58:11 2017
From: kmsujit at gmail.com (Sujit K M)
Date: Fri, 4 Aug 2017 20:28:11 +0530
Subject: [talk] Personalization II
In-Reply-To: <20170801164110.GA12555@ayvali.org>
References: <CA+6mNeru6b4itt-oLT0usrLExCM2UdM4_aPMgW3NMEE4mQ4Pww@mail.gmail.com>
 <20170731185545.GG84110@ayvali.org>
 <CA+6mNerUCahF55ggDE8VoQ0k53yv8xw8D7cHWBPVb=3W86BCgA@mail.gmail.com>
 <20170801164110.GA12555@ayvali.org>
Message-ID: <CA+6mNeqEwfEQ-nvd1gO8ZcfcdUPU_nNkXVrfg5sfKz0N92K+Kg@mail.gmail.com>

> We suspect the bank sold it to Google. Nothing to back that up though.

I don't think so. If you buy something in I.E and something else on Chrome.
You would see a preference which is different on both browsers. How are
they enabling the ads in browser? One of the simplest ways is using browser
cache instead of cookies.

https://betterexplained.com/articles/how-to-optimize-your-site-with-http-caching/



From kmsujit at gmail.com  Fri Aug  4 11:21:07 2017
From: kmsujit at gmail.com (Sujit K M)
Date: Fri, 4 Aug 2017 20:51:07 +0530
Subject: [talk] Personalization II
In-Reply-To: <CA+6mNeqEwfEQ-nvd1gO8ZcfcdUPU_nNkXVrfg5sfKz0N92K+Kg@mail.gmail.com>
References: <CA+6mNeru6b4itt-oLT0usrLExCM2UdM4_aPMgW3NMEE4mQ4Pww@mail.gmail.com>
 <20170731185545.GG84110@ayvali.org>
 <CA+6mNerUCahF55ggDE8VoQ0k53yv8xw8D7cHWBPVb=3W86BCgA@mail.gmail.com>
 <20170801164110.GA12555@ayvali.org>
 <CA+6mNeqEwfEQ-nvd1gO8ZcfcdUPU_nNkXVrfg5sfKz0N92K+Kg@mail.gmail.com>
Message-ID: <CA+6mNepUe-fn+s=n6PfcMcwFQYGOyMOp+c8pQxwcMLBhfvbCEg@mail.gmail.com>

> I don't think so. If you buy something in I.E and something else on Chrome.
> You would see a preference which is different on both browsers. How are
> they enabling the ads in browser? One of the simplest ways is using browser
> cache instead of cookies.
>
> https://betterexplained.com/articles/how-to-optimize-your-site-with-http-caching/

more interesting notes on Ad Technology on google below.

https://support.google.com/adsense/topic/1319753?hl=en&ref_topic=3373519



From mirimir at riseup.net  Fri Aug  4 21:17:30 2017
From: mirimir at riseup.net (Mirimir)
Date: Fri, 4 Aug 2017 14:17:30 -1100
Subject: [talk] Personalization II
In-Reply-To: <20170801164110.GA12555@ayvali.org>
References: <CA+6mNeru6b4itt-oLT0usrLExCM2UdM4_aPMgW3NMEE4mQ4Pww@mail.gmail.com>
 <20170731185545.GG84110@ayvali.org>
 <CA+6mNerUCahF55ggDE8VoQ0k53yv8xw8D7cHWBPVb=3W86BCgA@mail.gmail.com>
 <20170801164110.GA12555@ayvali.org>
Message-ID: <bd2a7a1c-4ab6-2c2a-191b-8b2a4f4fc029@riseup.net>

On 08/01/2017 05:41 AM, N.J. Thomas wrote:
> * Sujit K M <kmsujit at gmail.com> [2017-08-01 14:24:03+0530]:
>>> He suspects the credit card transaction data was sold to Google.
>>
>> Who sold the data? The merchant/bank or the payment gateway
>> software was tweaked to give this data.
> 
> We suspect the bank sold it to Google. Nothing to back that up though.
> 
> Thomas

| Google says this new system is driven by machine learning and
| combines data from Google's regular advertising campaigns with
| "third-party partnerships, which capture approximately 70% of
| credit and debit card transactions in the United States."

https://www.bleepingcomputer.com/news/google/google-will-track-your-offline-credit-card-payments-to-make-advertisers-happy/



From mcevoy.pat at gmail.com  Sat Aug  5 15:44:47 2017
From: mcevoy.pat at gmail.com (Pat McEvoy)
Date: Sat, 5 Aug 2017 15:44:47 -0400
Subject: [talk] Tweet by Dir. Of Special Ops on Twitter
Message-ID: <3F3120DB-23A7-401E-84DC-1D1D513145A2@gmail.com>

Just an FYI as I think there a few on this list who would be interested or possibly want in on this:


	Dir. Of Special Ops (@ScorpionsAfrica)
7/29/17, 3:29 AM
Hello @q5sys! I read your blog entry q5sys.sh/?p=397. Can you and @bsdnow assist me to start the first BUG on the African continent? twitter.com/ScorpionsAfric?

Download the Twitter app


Patrick

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.nycbug.org:8443/pipermail/talk/attachments/20170805/9cd6feac/attachment.htm>

From george at ceetonetechnology.com  Tue Aug  8 11:28:00 2017
From: george at ceetonetechnology.com (George Rosamond)
Date: Tue, 08 Aug 2017 15:28:00 +0000
Subject: [talk] upcoming hackathon proposal: NYC BSD Tor bridges
Message-ID: <508ea7ec-8a41-6c03-6138-1c7fcbbe82e5@ceetonetechnology.com>

Most of you are probably familiar with the Tor BSD Diversity Project
(https://torbsd.github.io/). We ported Tor Browser to OpenBSD, conducted
a number of BoFs, presentations and workshops over the years, and have a
positive net impact on Tor land.

The operating system diversity numbers in the Tor network are ugly, and
the more we poke, the uglier it gets. All statistics point to an
overwhelming Linux monoculture, with just under 6% of total public
network bandwidth provided by one BSD variant or another.

These are some of the stats:

https://torbsd.github.io/oostats.html

While public Tor relays have some non-Linux/BSD presence, it's really
disturbing when it comes to bridges. Bridges are non-public entry nodes
for users blocked from the Tor network.

https://torbsd.github.io/oostats/bridges-bw-by-os.txt

Yes... about 99.1% of bridge bandwidth is Linux.

And the absolute numbers reveal more:

https://torbsd.github.io/oostats/bridges-os-count.txt

There's only 31 *BSD bridges.  That's a number we could quickly change
just based on a small portion of people around NYC*BUG and on the talk@
list.

Bridges are easy to run from any residential home network. Bridge IPs
are not publicly listed (as relays are), and the IP will not be
blacklisted. Bridges are purely for entry, and don't push any exit traffic.

Many people in and around NYC on this list have decent bandwidth at
home, whether with one of the cable providers, FIOS, etc. Allowing up to
5 or even 10 megabytes of traffic will have no impact on your home
network, and it's not under consistent usage.

I want to put this workshop together to conduct a hands-on session that
will put another five or ten *BSD bridges into the Tor ecosystem. You
don't need to go an buy some fat 2U Supermicro box that will make ConEd
happy and give you the constant droning of a data center. Any small
system should be fine, whether you have it sitting around or you buy an
APU2, Soekris, RPi, BeagleBone, an old thin client desktop.

Any thoughts on this?  Concerns?

g

Note that by the time of the hackathon, we hope to have obfs4proxy in
both the FreeBSD and OpenBSD ports.

https://github.com/torbsd/freebsd-ports

https://github.com/torbsd/openbsd-ports



From shawn.webb at hardenedbsd.org  Tue Aug  8 11:50:43 2017
From: shawn.webb at hardenedbsd.org (Shawn Webb)
Date: Tue, 8 Aug 2017 11:50:43 -0400
Subject: [talk] upcoming hackathon proposal: NYC BSD Tor bridges
In-Reply-To: <508ea7ec-8a41-6c03-6138-1c7fcbbe82e5@ceetonetechnology.com>
References: <508ea7ec-8a41-6c03-6138-1c7fcbbe82e5@ceetonetechnology.com>
Message-ID: <20170808155043.cbxdqrehkaqghfqe@mutt-hbsd>

On Tue, Aug 08, 2017 at 03:28:00PM +0000, George Rosamond wrote:
> Most of you are probably familiar with the Tor BSD Diversity Project
> (https://torbsd.github.io/). We ported Tor Browser to OpenBSD, conducted
> a number of BoFs, presentations and workshops over the years, and have a
> positive net impact on Tor land.
> 
> The operating system diversity numbers in the Tor network are ugly, and
> the more we poke, the uglier it gets. All statistics point to an
> overwhelming Linux monoculture, with just under 6% of total public
> network bandwidth provided by one BSD variant or another.
> 
> These are some of the stats:
> 
> https://torbsd.github.io/oostats.html
> 
> While public Tor relays have some non-Linux/BSD presence, it's really
> disturbing when it comes to bridges. Bridges are non-public entry nodes
> for users blocked from the Tor network.
> 
> https://torbsd.github.io/oostats/bridges-bw-by-os.txt
> 
> Yes... about 99.1% of bridge bandwidth is Linux.
> 
> And the absolute numbers reveal more:
> 
> https://torbsd.github.io/oostats/bridges-os-count.txt
> 
> There's only 31 *BSD bridges.  That's a number we could quickly change
> just based on a small portion of people around NYC*BUG and on the talk@
> list.
> 
> Bridges are easy to run from any residential home network. Bridge IPs
> are not publicly listed (as relays are), and the IP will not be
> blacklisted. Bridges are purely for entry, and don't push any exit traffic.
> 
> Many people in and around NYC on this list have decent bandwidth at
> home, whether with one of the cable providers, FIOS, etc. Allowing up to
> 5 or even 10 megabytes of traffic will have no impact on your home
> network, and it's not under consistent usage.
> 
> I want to put this workshop together to conduct a hands-on session that
> will put another five or ten *BSD bridges into the Tor ecosystem. You
> don't need to go an buy some fat 2U Supermicro box that will make ConEd
> happy and give you the constant droning of a data center. Any small
> system should be fine, whether you have it sitting around or you buy an
> APU2, Soekris, RPi, BeagleBone, an old thin client desktop.
> 
> Any thoughts on this?  Concerns?

I'd love to drive up for this. Would someone be able to offer their couch
for me to surf on? If not, I'll see if I can get a hotel approved by my
employer.

Thanks,

-- 
Shawn Webb
Cofounder and Security Engineer
HardenedBSD

GPG Key ID:          0x6A84658F52456EEE
GPG Key Fingerprint: 2ABA B6BD EF6A F486 BE89  3D9E 6A84 658F 5245 6EEE
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 833 bytes
Desc: not available
URL: <https://lists.nycbug.org:8443/pipermail/talk/attachments/20170808/6c6ef100/attachment.bin>

From bcully at gmail.com  Tue Aug  8 12:31:52 2017
From: bcully at gmail.com (Brian Cully)
Date: Tue, 8 Aug 2017 12:31:52 -0400
Subject: [talk] upcoming hackathon proposal: NYC BSD Tor bridges
In-Reply-To: <508ea7ec-8a41-6c03-6138-1c7fcbbe82e5@ceetonetechnology.com>
References: <508ea7ec-8a41-6c03-6138-1c7fcbbe82e5@ceetonetechnology.com>
Message-ID: <2104DF95-5043-4D4D-B14B-4C84F84DAB78@gmail.com>


> On 8-Aug-2017, at 11:28, George Rosamond <george at ceetonetechnology.com> wrote:
> I want to put this workshop together to conduct a hands-on session that
> will put another five or ten *BSD bridges into the Tor ecosystem. You
> don't need to go an buy some fat 2U Supermicro box that will make ConEd
> happy and give you the constant droning of a data center. Any small
> system should be fine, whether you have it sitting around or you buy an
> APU2, Soekris, RPi, BeagleBone, an old thin client desktop.

	I?d be up for this. I?ve got a FreeBSD box at home for my NAS that?s been pretty over-provisioned and basically idle.

-bjc


From espen at tagestad.no  Tue Aug  8 13:03:27 2017
From: espen at tagestad.no (Espen Tagestad)
Date: Tue, 8 Aug 2017 13:03:27 -0400
Subject: [talk] upcoming hackathon proposal: NYC BSD Tor bridges
In-Reply-To: <2104DF95-5043-4D4D-B14B-4C84F84DAB78@gmail.com>
References: <508ea7ec-8a41-6c03-6138-1c7fcbbe82e5@ceetonetechnology.com>
 <2104DF95-5043-4D4D-B14B-4C84F84DAB78@gmail.com>
Message-ID: <9E51D99F-99FB-4602-BF1A-86E1529C0F53@tagestad.no>

On Aug 8, 2017, at 12:31 PM, Brian Cully <bcully at gmail.com> wrote:
> 
>> On 8-Aug-2017, at 11:28, George Rosamond <george at ceetonetechnology.com> wrote:
>> I want to put this workshop together to conduct a hands-on session that
>> will put another five or ten *BSD bridges into the Tor ecosystem. You
>> don't need to go an buy some fat 2U Supermicro box that will make ConEd
>> happy and give you the constant droning of a data center. Any small
>> system should be fine, whether you have it sitting around or you buy an
>> APU2, Soekris, RPi, BeagleBone, an old thin client desktop.
> 
> 	I?d be up for this. I?ve got a FreeBSD box at home for my NAS that?s been pretty over-provisioned and basically idle.


Second that! I also have a couple of boxes around which could potentially be used to host a jail or two for this. I have some concerns running an exit node, though.

br.
Espen


From _ at thomaslevine.com  Tue Aug  8 13:37:28 2017
From: _ at thomaslevine.com (Thomas Levine)
Date: Tue, 08 Aug 2017 17:37:28 +0000
Subject: [talk] upcoming hackathon proposal: NYC BSD Tor bridges
In-Reply-To: <9E51D99F-99FB-4602-BF1A-86E1529C0F53@tagestad.no>
References: <508ea7ec-8a41-6c03-6138-1c7fcbbe82e5@ceetonetechnology.com>
 <2104DF95-5043-4D4D-B14B-4C84F84DAB78@gmail.com>
 <9E51D99F-99FB-4602-BF1A-86E1529C0F53@tagestad.no>
Message-ID: <20170808173730.627EC7E4EA@mailuser.nyi.internal>

I am unlikely to attend, but I would like to contribute to the effort.

It should be pretty straightforward to set up, no? I install the tor
package and edit the tor configuration file. That said, if you expect
that the hackathon will produce advice on how to configure the bridges,
I will wait until that is ready.

Now that I have settled on using disposible computers to avoid troubles
at airports, I have several small computers that I don't know what to do
with.  I was going to set them up with *BSD systems in hackerspaces so
people could play with them, and it is much more interesting if they are
doing something like this. They don't have good internet access, but the
bridges will at least be useful in the morning when nobody is in the
hackerspaces.



From george at ceetonetechnology.com  Tue Aug  8 13:42:00 2017
From: george at ceetonetechnology.com (George Rosamond)
Date: Tue, 08 Aug 2017 17:42:00 +0000
Subject: [talk] upcoming hackathon proposal: NYC BSD Tor bridges
In-Reply-To: <9E51D99F-99FB-4602-BF1A-86E1529C0F53@tagestad.no>
References: <508ea7ec-8a41-6c03-6138-1c7fcbbe82e5@ceetonetechnology.com>
 <2104DF95-5043-4D4D-B14B-4C84F84DAB78@gmail.com>
 <9E51D99F-99FB-4602-BF1A-86E1529C0F53@tagestad.no>
Message-ID: <936b75fd-561c-5369-f6e0-18d75f257af6@ceetonetechnology.com>

Espen Tagestad:
> On Aug 8, 2017, at 12:31 PM, Brian Cully <bcully at gmail.com> wrote:
>> 
>>> On 8-Aug-2017, at 11:28, George Rosamond
>>> <george at ceetonetechnology.com> wrote: I want to put this workshop
>>> together to conduct a hands-on session that will put another five
>>> or ten *BSD bridges into the Tor ecosystem. You don't need to go
>>> an buy some fat 2U Supermicro box that will make ConEd happy and
>>> give you the constant droning of a data center. Any small system
>>> should be fine, whether you have it sitting around or you buy an 
>>> APU2, Soekris, RPi, BeagleBone, an old thin client desktop.
>> 
>> I?d be up for this. I?ve got a FreeBSD box at home for my NAS
>> that?s been pretty over-provisioned and basically idle.
> 
> 
> Second that! I also have a couple of boxes around which could
> potentially be used to host a jail or two for this. I have some
> concerns running an exit node, though.

No need to run an exit node.. which would be a bad idea at home anyways.

Bridges are non-public entries into the Tor network, with the IPs
carefully distributed to users who request them.

They are basically hidden nodes for users blocked from the public Tor
network.

g

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 833 bytes
Desc: OpenPGP digital signature
URL: <https://lists.nycbug.org:8443/pipermail/talk/attachments/20170808/9760913a/attachment.bin>

From george at ceetonetechnology.com  Tue Aug  8 13:44:00 2017
From: george at ceetonetechnology.com (George Rosamond)
Date: Tue, 08 Aug 2017 17:44:00 +0000
Subject: [talk] upcoming hackathon proposal: NYC BSD Tor bridges
In-Reply-To: <20170808173730.627EC7E4EA@mailuser.nyi.internal>
References: <508ea7ec-8a41-6c03-6138-1c7fcbbe82e5@ceetonetechnology.com>
 <2104DF95-5043-4D4D-B14B-4C84F84DAB78@gmail.com>
 <9E51D99F-99FB-4602-BF1A-86E1529C0F53@tagestad.no>
 <20170808173730.627EC7E4EA@mailuser.nyi.internal>
Message-ID: <4b188749-1786-f96d-c576-105d4a533146@ceetonetechnology.com>

Thomas Levine:
> I am unlikely to attend, but I would like to contribute to the effort.
> 
> It should be pretty straightforward to set up, no? I install the tor
> package and edit the tor configuration file. That said, if you expect
> that the hackathon will produce advice on how to configure the bridges,
> I will wait until that is ready.

Yes... all true.  Configuring a bridge is easy enough, plus dealing with
your own port forwarding.

> 
> Now that I have settled on using disposible computers to avoid troubles
> at airports, I have several small computers that I don't know what to do
> with.  I was going to set them up with *BSD systems in hackerspaces so
> people could play with them, and it is much more interesting if they are
> doing something like this. They don't have good internet access, but the
> bridges will at least be useful in the morning when nobody is in the
> hackerspaces.
> 

Sure... any stable internet connection, even with a non-static IP,
should be fine.

g



From ike at blackskyresearch.net  Tue Aug  8 14:05:25 2017
From: ike at blackskyresearch.net (Isaac (.ike) Levy)
Date: Tue, 08 Aug 2017 14:05:25 -0400
Subject: [talk] upcoming hackathon proposal: NYC BSD Tor bridges
In-Reply-To: <508ea7ec-8a41-6c03-6138-1c7fcbbe82e5@ceetonetechnology.com>
References: <508ea7ec-8a41-6c03-6138-1c7fcbbe82e5@ceetonetechnology.com>
Message-ID: <1502215525.4004675.1067125920.20686A0C@webmail.messagingengine.com>


On Tue, Aug 8, 2017, at 11:28 AM, George Rosamond wrote:
> The operating system diversity numbers in the Tor network are ugly, and

... yep, right now diversity is bad all over- anecdotally I point to the
monoculture of Linux that is emerging on the net at large, (servers and
networking gear), not to mention the monoculture of mac.

> There's only 31 *BSD bridges.  That's a number we could quickly change
... UGH

Wow.

> I want to put this workshop together to conduct a hands-on session that
> will put another five or ten *BSD bridges into the Tor ecosystem. You

Yes- and count me in to arrive with a bag of ethernet switches and a
router doing DHCP, to make group setup of various gear fast and easy.

/salute

Best,
.ike



From jkeenan at pobox.com  Tue Aug  8 15:00:58 2017
From: jkeenan at pobox.com (James E Keenan)
Date: Tue, 8 Aug 2017 15:00:58 -0400
Subject: [talk] Disposible computers
In-Reply-To: <20170808173730.627EC7E4EA@mailuser.nyi.internal>
References: <508ea7ec-8a41-6c03-6138-1c7fcbbe82e5@ceetonetechnology.com>
 <2104DF95-5043-4D4D-B14B-4C84F84DAB78@gmail.com>
 <9E51D99F-99FB-4602-BF1A-86E1529C0F53@tagestad.no>
 <20170808173730.627EC7E4EA@mailuser.nyi.internal>
Message-ID: <f4baceba-bb71-8279-97dc-bf00f6bacac8@pobox.com>

On 08/08/2017 01:37 PM, Thomas Levine wrote:
> I am unlikely to attend, but I would like to contribute to the effort.
> 
[snip]
> 
> Now that I have settled on using disposible computers to avoid troubles
> at airports, ...

Now, *there* is a topic I'd like to hear a presentation on at NYCBUG!

jimk



From _ at thomaslevine.com  Tue Aug  8 16:39:13 2017
From: _ at thomaslevine.com (Thomas Levine)
Date: Tue, 08 Aug 2017 20:39:13 +0000
Subject: [talk] Disposible computers
In-Reply-To: <f4baceba-bb71-8279-97dc-bf00f6bacac8@pobox.com>
References: <508ea7ec-8a41-6c03-6138-1c7fcbbe82e5@ceetonetechnology.com>
 <2104DF95-5043-4D4D-B14B-4C84F84DAB78@gmail.com>
 <9E51D99F-99FB-4602-BF1A-86E1529C0F53@tagestad.no>
 <20170808173730.627EC7E4EA@mailuser.nyi.internal>
 <f4baceba-bb71-8279-97dc-bf00f6bacac8@pobox.com>
Message-ID: <20170808203915.A06047F992@mailuser.nyi.internal>

I will be happy to present the system once I have tested it on a few trips.

James E Keenan writes:
> On 08/08/2017 01:37 PM, Thomas Levine wrote:
> > I am unlikely to attend, but I would like to contribute to the effort.
> > 
> [snip]
> > 
> > Now that I have settled on using disposible computers to avoid troubles
> > at airports, ...
> 
> Now, *there* is a topic I'd like to hear a presentation on at NYCBUG!
> 
> jimk
> 



From george at ceetonetechnology.com  Tue Aug  8 16:49:00 2017
From: george at ceetonetechnology.com (George Rosamond)
Date: Tue, 08 Aug 2017 20:49:00 +0000
Subject: [talk] Disposible computers
In-Reply-To: <20170808203915.A06047F992@mailuser.nyi.internal>
References: <508ea7ec-8a41-6c03-6138-1c7fcbbe82e5@ceetonetechnology.com>
 <2104DF95-5043-4D4D-B14B-4C84F84DAB78@gmail.com>
 <9E51D99F-99FB-4602-BF1A-86E1529C0F53@tagestad.no>
 <20170808173730.627EC7E4EA@mailuser.nyi.internal>
 <f4baceba-bb71-8279-97dc-bf00f6bacac8@pobox.com>
 <20170808203915.A06047F992@mailuser.nyi.internal>
Message-ID: <2ae692ae-a82c-83e3-f117-b2fc06cc5c6c@ceetonetechnology.com>

Thomas Levine:
> I will be happy to present the system once I have tested it on a few trips.
> 

There is a lot of thoughts on this over the years... particularly today.

Some travel with little data, and pull it on the 'other side'.  Sync and
then delete before traveling back.

Some use VPNs.

Some encrypt and use passwd as a key, and maybe a digital private key.

Of course how are you carrying that private key is also a question.

There are a host of approaches in terms of 'hidden' partitions, which
aren't hidden from dd(1).

It all really depends on the threat model.  That should be the starting
point, and not the technology and tools. What are you trying to protect
and from whom, and how high are the stakes? That determines everything,
including how much hassle you are willing to endure for a solution.

Then there's always the most effective approach: don't bring anything,
which I've read is even impervious to a full dd.

g


-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 833 bytes
Desc: OpenPGP digital signature
URL: <https://lists.nycbug.org:8443/pipermail/talk/attachments/20170808/b06c1dd8/attachment.bin>

From ahpook at mitnal.com  Tue Aug  8 17:07:51 2017
From: ahpook at mitnal.com (Ah Pook)
Date: Tue, 8 Aug 2017 17:07:51 -0400
Subject: [talk] upcoming hackathon proposal: NYC BSD Tor bridges
In-Reply-To: <20170808155043.cbxdqrehkaqghfqe@mutt-hbsd>
References: <508ea7ec-8a41-6c03-6138-1c7fcbbe82e5@ceetonetechnology.com>
 <20170808155043.cbxdqrehkaqghfqe@mutt-hbsd>
Message-ID: <be708959-3757-4eda-4020-7c3202525848@mitnal.com>

On 08/08/2017 11:50 AM, Shawn Webb wrote:
> I'd love to drive up for this. Would someone be able to offer their couch
> for me to surf on? If not, I'll see if I can get a hotel approved by my
> employer.

I have lots of room in Brooklyn, and just got Fios ?bergig or whatever
they're calling it.  Symmetric 940M.  :-D  Happy to host.



From raulcuza at gmail.com  Tue Aug  8 18:30:57 2017
From: raulcuza at gmail.com (Raul Cuza)
Date: Tue, 8 Aug 2017 18:30:57 -0400
Subject: [talk] upcoming hackathon proposal: NYC BSD Tor bridges
In-Reply-To: <be708959-3757-4eda-4020-7c3202525848@mitnal.com>
References: <508ea7ec-8a41-6c03-6138-1c7fcbbe82e5@ceetonetechnology.com>
 <20170808155043.cbxdqrehkaqghfqe@mutt-hbsd>
 <be708959-3757-4eda-4020-7c3202525848@mitnal.com>
Message-ID: <CAOnfn2zPJ6k+rCNdQtvrBkQt9N33E63cFdBCZfZGXD-w3228Uw@mail.gmail.com>

On Aug 8, 2017 17:09, "Ah Pook" <ahpook at mitnal.com> wrote:

On 08/08/2017 11:50 AM, Shawn Webb wrote:
> I'd love to drive up for this. Would someone be able to offer their couch
> for me to surf on? If not, I'll see if I can get a hotel approved by my
> employer.

I have lots of room in Brooklyn, and just got Fios ?bergig or whatever
they're calling it.  Symmetric 940M.  :-D  Happy to host.

_______________________________________________
talk mailing list
talk at lists.nycbug.org
http://lists.nycbug.org/mailman/listinfo/talk


Love this topic. Just the boost I need to make my home bandwidth do
something useful.

R.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.nycbug.org:8443/pipermail/talk/attachments/20170808/1ad0aa04/attachment.htm>

From _ at thomaslevine.com  Tue Aug  8 19:06:45 2017
From: _ at thomaslevine.com (Thomas Levine)
Date: Tue, 08 Aug 2017 23:06:45 +0000
Subject: [talk] Disposible computers
In-Reply-To: <2ae692ae-a82c-83e3-f117-b2fc06cc5c6c@ceetonetechnology.com>
References: <508ea7ec-8a41-6c03-6138-1c7fcbbe82e5@ceetonetechnology.com>
 <2104DF95-5043-4D4D-B14B-4C84F84DAB78@gmail.com>
 <9E51D99F-99FB-4602-BF1A-86E1529C0F53@tagestad.no>
 <20170808173730.627EC7E4EA@mailuser.nyi.internal>
 <f4baceba-bb71-8279-97dc-bf00f6bacac8@pobox.com>
 <20170808203915.A06047F992@mailuser.nyi.internal>
 <2ae692ae-a82c-83e3-f117-b2fc06cc5c6c@ceetonetechnology.com>
Message-ID: <20170808230646.C61D37F9A8@mailuser.nyi.internal>

George Rosamond writes:
> It all really depends on the threat model.  That should be the starting
> point, and not the technology and tools. What are you trying to protect
> and from whom, and how high are the stakes? That determines everything,
> including how much hassle you are willing to endure for a solution.

Breaking and losing my stuff is mostly problematic because it is
expensive and annoying to purchase and assemble it, and this is why I am
switching to disposible computers. One threat is thus the untrustworthy
airport staff working for airlines and airports who have strong
incentives to give you wrong information. Such staff have multiple times:

* confiscated my things that followed regulations
* broken things inside my checked bags
* given me wrong directions that cost me a lot of money

The other threat, which I am more scared of even though it hasn't
happened much to me, is that a government officer will create trouble
for me because I am suspicious or because I have commited some crime
that I in fact didn't commit.



I recently tried checking a bag in hopes that it would make things easier.
Of course, the airline representative gave me wrong directions about
where to pick up the bag, so it took several days to get the bag, which
was of course not easier. On the other hand, it turned out that
attending a programming conference with only my purse and the clothes
I was wearing was pretty okay. It really would have been nice to have
a computer, but I realized that I can add a second change of clothes and
a small disposible computer and still keep it to two purses.



I have not really concerned myself with surveillance by governments
because I think they have surely owned me by now if they wanted to.
For example, I sometimes run software without fully reading its source
code or installing it in a separate user, I usually don't cover my
keyboard when I type passwords in public, and I have never checked for
tampering with the electronics that I use.

Also, I think that government surveillance organizations must quite
incompetent at computing, just like technology companies, so I think I
am inconveniencing them enough by simply using slightly unusual
software. They'll own me if they try, but they'll have to exploit
something just for weird people like me, and that would be a lot of
custom work, not mass surveillance.

It happens that my setup might do pretty well against mass surveillance
for reasons other than obscurity. I can easily modify my setup such that
I can travel with nothing, not even the knowledge of the password that
I will need in order to rebuild everything.

But I don't really like talking about government surveillance because
I think the topic has become trendy enough that the discussion is
usually mostly a marketing tool at this point.



George Rosamond writes:
> Then there's always the most effective approach: don't bring anything,
> which I've read is even impervious to a full dd.

This is a very bad idea because your lack of stuff would make you look
very suspicious; even though they can't get your data, they can make
trouble for you in different ways. If I take only the things I need with
me, then I have few enough things I look like a drug smuggler. To avoid
fitting this profile, I have thus been carrying extra, otherwise
unnecessary, things in order to look less suspicous, and I have been
doing this even before I decided to switch to disposible computers.
It seems to help.

So, I think you should bring at least some clothes and one suitcase or
two backpacks. If you want to be safe, also bring a smartphone and a
very normal laptop with Windows or Mac installed (and without any data
that you care about). This would be very expensive and annoying for me
as I presently own neither of these computers. More importantly, it
would feel really stupid and wasteful to acquire two computers, clothes,
and extra bags just in order to avoid troubles at airports; it's better
just to avoid going to airports.



So it's not just about the stuff; you need to fit your profile into
something normal.

One problem for me is that I don't have a good answer for when a police
officer asks, "what do you do [for work]?" The true answers to this
loaded question are that are that I don't work or that I am retired
(in my 20s), and they never seem to like these answers. I am considering
writing a book so I can say I am an author or getting a part-time
normal-sounding job in order that I may give a normal answer.
So I suggest that you also get a normal-sounding job.

Another thing that makes me suspicious is that I am male and usually
alone.  When going through checkpoints, especially customs, it is good
for me to strike up a conversation with a similarly aged female so it
looks like we are a couple traveling together.



Part of the convenience of having only two purses is that it will become
more practical to walk instead of flying, so I will probably do that
more often instead. And even though regulations are pretty scary for
other modes of transit, I think they're the worst in air travel, so
I will probably take a boat for the next inter-continental trip.



Hmm I have gotten far off the original tangent of disposible computers.
I will some day write up my recommendations for safe air travel, but
note for now that bringing nothing with you is a bad idea.



From edlinuxguru at gmail.com  Tue Aug  8 19:22:44 2017
From: edlinuxguru at gmail.com (Edward Capriolo)
Date: Tue, 8 Aug 2017 19:22:44 -0400
Subject: [talk] Disposible computers
In-Reply-To: <20170808230646.C61D37F9A8@mailuser.nyi.internal>
References: <508ea7ec-8a41-6c03-6138-1c7fcbbe82e5@ceetonetechnology.com>
 <2104DF95-5043-4D4D-B14B-4C84F84DAB78@gmail.com>
 <9E51D99F-99FB-4602-BF1A-86E1529C0F53@tagestad.no>
 <20170808173730.627EC7E4EA@mailuser.nyi.internal>
 <f4baceba-bb71-8279-97dc-bf00f6bacac8@pobox.com>
 <20170808203915.A06047F992@mailuser.nyi.internal>
 <2ae692ae-a82c-83e3-f117-b2fc06cc5c6c@ceetonetechnology.com>
 <20170808230646.C61D37F9A8@mailuser.nyi.internal>
Message-ID: <CAENxBwy=7wQUwGhQWF9em3iSgVMn6YYorA8LN9UY4whe+V9RzA@mail.gmail.com>

On Tuesday, August 8, 2017, Thomas Levine <_ at thomaslevine.com> wrote:

> George Rosamond writes:
> > It all really depends on the threat model.  That should be the starting
> > point, and not the technology and tools. What are you trying to protect
> > and from whom, and how high are the stakes? That determines everything,
> > including how much hassle you are willing to endure for a solution.
>
> Breaking and losing my stuff is mostly problematic because it is
> expensive and annoying to purchase and assemble it, and this is why I am
> switching to disposible computers. One threat is thus the untrustworthy
> airport staff working for airlines and airports who have strong
> incentives to give you wrong information. Such staff have multiple times:
>
> * confiscated my things that followed regulations
> * broken things inside my checked bags
> * given me wrong directions that cost me a lot of money
>
> The other threat, which I am more scared of even though it hasn't
> happened much to me, is that a government officer will create trouble
> for me because I am suspicious or because I have commited some crime
> that I in fact didn't commit.
>
>
>
> I recently tried checking a bag in hopes that it would make things easier.
> Of course, the airline representative gave me wrong directions about
> where to pick up the bag, so it took several days to get the bag, which
> was of course not easier. On the other hand, it turned out that
> attending a programming conference with only my purse and the clothes
> I was wearing was pretty okay. It really would have been nice to have
> a computer, but I realized that I can add a second change of clothes and
> a small disposible computer and still keep it to two purses.
>
>
>
> I have not really concerned myself with surveillance by governments
> because I think they have surely owned me by now if they wanted to.
> For example, I sometimes run software without fully reading its source
> code or installing it in a separate user, I usually don't cover my
> keyboard when I type passwords in public, and I have never checked for
> tampering with the electronics that I use.
>
> Also, I think that government surveillance organizations must quite
> incompetent at computing, just like technology companies, so I think I
> am inconveniencing them enough by simply using slightly unusual
> software. They'll own me if they try, but they'll have to exploit
> something just for weird people like me, and that would be a lot of
> custom work, not mass surveillance.
>
> It happens that my setup might do pretty well against mass surveillance
> for reasons other than obscurity. I can easily modify my setup such that
> I can travel with nothing, not even the knowledge of the password that
> I will need in order to rebuild everything.
>
> But I don't really like talking about government surveillance because
> I think the topic has become trendy enough that the discussion is
> usually mostly a marketing tool at this point.
>
>
>
> George Rosamond writes:
> > Then there's always the most effective approach: don't bring anything,
> > which I've read is even impervious to a full dd.
>
> This is a very bad idea because your lack of stuff would make you look
> very suspicious; even though they can't get your data, they can make
> trouble for you in different ways. If I take only the things I need with
> me, then I have few enough things I look like a drug smuggler. To avoid
> fitting this profile, I have thus been carrying extra, otherwise
> unnecessary, things in order to look less suspicous, and I have been
> doing this even before I decided to switch to disposible computers.
> It seems to help.
>
> So, I think you should bring at least some clothes and one suitcase or
> two backpacks. If you want to be safe, also bring a smartphone and a
> very normal laptop with Windows or Mac installed (and without any data
> that you care about). This would be very expensive and annoying for me
> as I presently own neither of these computers. More importantly, it
> would feel really stupid and wasteful to acquire two computers, clothes,
> and extra bags just in order to avoid troubles at airports; it's better
> just to avoid going to airports.
>
>
>
> So it's not just about the stuff; you need to fit your profile into
> something normal.
>
> One problem for me is that I don't have a good answer for when a police
> officer asks, "what do you do [for work]?" The true answers to this
> loaded question are that are that I don't work or that I am retired
> (in my 20s), and they never seem to like these answers. I am considering
> writing a book so I can say I am an author or getting a part-time
> normal-sounding job in order that I may give a normal answer.
> So I suggest that you also get a normal-sounding job.
>
> Another thing that makes me suspicious is that I am male and usually
> alone.  When going through checkpoints, especially customs, it is good
> for me to strike up a conversation with a similarly aged female so it
> looks like we are a couple traveling together.
>
>
>
> Part of the convenience of having only two purses is that it will become
> more practical to walk instead of flying, so I will probably do that
> more often instead. And even though regulations are pretty scary for
> other modes of transit, I think they're the worst in air travel, so
> I will probably take a boat for the next inter-continental trip.
>
>
>
> Hmm I have gotten far off the original tangent of disposible computers.
> I will some day write up my recommendations for safe air travel, but
> note for now that bringing nothing with you is a bad idea.
>
> _______________________________________________
> talk mailing list
> talk at lists.nycbug.org <javascript:;>
> http://lists.nycbug.org/mailman/listinfo/talk
>

A few of my buddies travled to China for a wedding. Devices can be
inspected/confiscated etc. At the time, the company we worked at had
'media' in the name. Even though it was an adtech company they believed
they received extra scrutiny because the government believed they were
affiliated with a news organization.

I personally would not bring a device with me outside the us that had
personal information or that of my company.


-- 
Sorry this was sent from mobile. Will do less grammar and spell check than
usual.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.nycbug.org:8443/pipermail/talk/attachments/20170808/be717745/attachment.htm>

From george at ceetonetechnology.com  Sat Aug 12 23:59:00 2017
From: george at ceetonetechnology.com (George Rosamond)
Date: Sun, 13 Aug 2017 03:59:00 +0000
Subject: [talk] mickey@ OpenBSD RIP
Message-ID: <a226c090-0893-62fc-29b8-528454103321@ceetonetechnology.com>

Mickey Shalayeff passed recently.

He was a longtime OpenBSD developer and had his fingerprints throughout
the operating system. He lived in Brooklyn for many years and was
involved in infrastructures around NYC. He was at Calyx (.com) when Nick
Merrill got the NSL.

He did a great NYC*BUG meeting for us many years ago.

I'll give more details later, and there's so many crazy stories about
Mickey.

He'll be sorely missed. He was talented and eternally devoted to good
code. Personally, I'll miss the long rambling emails we had so many
times hitting tech, politics and food.

He was the only person I ever met who put salt and pepper in coffee. He
communicated in mickeyese. Unintelligible to many, but hysterical to
read and listen to.

mickey.lucifer.net is filled with some of his gems, including his take
on the alleged OpenBSD IPSec backdoor.

http://images.kd85.com/mickey/IMAG0484.jpg

RIP mickey@

g



From jim at netgate.com  Sun Aug 13 13:26:35 2017
From: jim at netgate.com (Jim Thompson)
Date: Sun, 13 Aug 2017 12:26:35 -0500
Subject: [talk] mickey@ OpenBSD RIP
In-Reply-To: <a226c090-0893-62fc-29b8-528454103321@ceetonetechnology.com>
References: <a226c090-0893-62fc-29b8-528454103321@ceetonetechnology.com>
Message-ID: <378725E1-BAA0-4251-AA70-55427A6D49E7@netgate.com>



> On Aug 12, 2017, at 10:59 PM, George Rosamond <george at ceetonetechnology.com> wrote:
> 
> [...]

> mickey.lucifer.net is filled with some of his gems, including his take on the alleged OpenBSD IPSec backdoor.

http://mickey.lucifier.net/ 

(two 'i's in "lucifier")

Jim


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.nycbug.org:8443/pipermail/talk/attachments/20170813/d67432a6/attachment.htm>

From raulcuza at gmail.com  Thu Aug 24 14:36:59 2017
From: raulcuza at gmail.com (Raul Cuza)
Date: Thu, 24 Aug 2017 14:36:59 -0400
Subject: [talk] LISA'17 and FreeBSD Foundation
Message-ID: <CAOnfn2xkBoyFNkZSCb=qUh3vdtQQr0vXb1JUge6VNR4yNuqCyA@mail.gmail.com>

Hola,

I'm angling to go to LISA in SF this year. I noticed that the FreeBSD
Foundation is a sponsor. To whom should I reach out to offer my
volunteer services? I did a shift in the booth at the 2600 conf a
little whiles back and want to help again.

Ra?l



From george at ceetonetechnology.com  Thu Aug 24 14:55:00 2017
From: george at ceetonetechnology.com (George Rosamond)
Date: Thu, 24 Aug 2017 18:55:00 +0000
Subject: [talk] LISA'17 and FreeBSD Foundation
In-Reply-To: <CAOnfn2xkBoyFNkZSCb=qUh3vdtQQr0vXb1JUge6VNR4yNuqCyA@mail.gmail.com>
References: <CAOnfn2xkBoyFNkZSCb=qUh3vdtQQr0vXb1JUge6VNR4yNuqCyA@mail.gmail.com>
Message-ID: <71e103f9-a869-2a7d-43b6-c475cb6e6a0b@ceetonetechnology.com>

Raul Cuza:
> Hola,
> 
> I'm angling to go to LISA in SF this year. I noticed that the FreeBSD
> Foundation is a sponsor. To whom should I reach out to offer my
> volunteer services? I did a shift in the booth at the 2600 conf a
> little whiles back and want to help again.

Nice.  We need more of that... having individuals from NYC*BUG has a big
impact on events like that.

Email deb@ the Foundation and she should be able to direct you.  Others
are encouraged to do the same if they're attending.

Then we have HOPE again next July I think.

g



From george at ceetonetechnology.com  Fri Aug 25 13:10:00 2017
From: george at ceetonetechnology.com (George Rosamond)
Date: Fri, 25 Aug 2017 17:10:00 +0000
Subject: [talk] crazy hardware porting example
Message-ID: <50f60d80-1e18-ff32-46f7-df8d21712358@ceetonetechnology.com>

A long while ago someone on-list (Ike) gave me this dinky wireless AP
back from .jp.

2 1/4" x 1.5" x .5"

It's cool and functional in itself, and Ike mentioned how this hardware
should be in focus for building consumer BSD-based firewall devices.

While drifting through dmesgd this AM, I found this device, an Elecom
WRH-300, has NetBSD running on it, thanks to Jun for his work:

http://dmesgd.nycbug.org/index.cgi?do=view&id=3251

This thing is *small*.

g



From ike at blackskyresearch.net  Fri Aug 25 13:23:09 2017
From: ike at blackskyresearch.net (Isaac (.ike) Levy)
Date: Fri, 25 Aug 2017 13:23:09 -0400
Subject: [talk] crazy hardware porting example
In-Reply-To: <50f60d80-1e18-ff32-46f7-df8d21712358@ceetonetechnology.com>
References: <50f60d80-1e18-ff32-46f7-df8d21712358@ceetonetechnology.com>
Message-ID: <1503681789.487631.1085095864.137F3DFE@webmail.messagingengine.com>

On Fri, Aug 25, 2017, at 01:10 PM, George Rosamond wrote:
> A long while ago someone on-list (Ike) gave me this dinky wireless AP
> back from .jp.
> 
> 2 1/4" x 1.5" x .5"
> 
> It's cool and functional in itself, and Ike mentioned how this hardware
> should be in focus for building consumer BSD-based firewall devices.
> 
> While drifting through dmesgd this AM, I found this device, an Elecom
> WRH-300, has NetBSD running on it, thanks to Jun for his work:
> 
> http://dmesgd.nycbug.org/index.cgi?do=view&id=3251

Woah woah I'm confused- Jun got the *tiny* device I brought back to run
NetBSD?!?!

Or is the dmesg just from something with the same chipset?

Best,
.ike


> 
> This thing is *small*.
> 
> g
> 
> _______________________________________________
> talk mailing list
> talk at lists.nycbug.org
> http://lists.nycbug.org/mailman/listinfo/talk



From george at ceetonetechnology.com  Fri Aug 25 17:34:00 2017
From: george at ceetonetechnology.com (George Rosamond)
Date: Fri, 25 Aug 2017 21:34:00 +0000
Subject: [talk] crazy hardware porting example
In-Reply-To: <1503681789.487631.1085095864.137F3DFE@webmail.messagingengine.com>
References: <50f60d80-1e18-ff32-46f7-df8d21712358@ceetonetechnology.com>
 <1503681789.487631.1085095864.137F3DFE@webmail.messagingengine.com>
Message-ID: <b3bcee81-aac4-8021-43a3-b54f8721631d@ceetonetechnology.com>

Isaac (.ike) Levy:
> On Fri, Aug 25, 2017, at 01:10 PM, George Rosamond wrote:
>> A long while ago someone on-list (Ike) gave me this dinky wireless AP
>> back from .jp.
>>
>> 2 1/4" x 1.5" x .5"
>>
>> It's cool and functional in itself, and Ike mentioned how this hardware
>> should be in focus for building consumer BSD-based firewall devices.
>>
>> While drifting through dmesgd this AM, I found this device, an Elecom
>> WRH-300, has NetBSD running on it, thanks to Jun for his work:
>>
>> http://dmesgd.nycbug.org/index.cgi?do=view&id=3251
> 
> Woah woah I'm confused- Jun got the *tiny* device I brought back to run
> NetBSD?!?!
> 
> Or is the dmesg just from something with the same chipset?

It's the same "model" with a trailing BK at the end... but who knows on
the chipset.

Also, anyone tinker around with:

* Intel NUC

* GPD Pocket

* Pine64 port is in decent shape with the BSDs IIRC, but how about the
Pinebook?

g



From kmsujit at gmail.com  Sat Aug 26 02:17:17 2017
From: kmsujit at gmail.com (Sujit K M)
Date: Sat, 26 Aug 2017 11:47:17 +0530
Subject: [talk] crazy hardware porting example
In-Reply-To: <50f60d80-1e18-ff32-46f7-df8d21712358@ceetonetechnology.com>
References: <50f60d80-1e18-ff32-46f7-df8d21712358@ceetonetechnology.com>
Message-ID: <CA+6mNeouCDAERK1DbXh-EcFREUWF-ZkaCm+8WHzep7mPcputHQ@mail.gmail.com>

On Fri, Aug 25, 2017 at 10:40 PM, George Rosamond
<george at ceetonetechnology.com> wrote:
> A long while ago someone on-list (Ike) gave me this dinky wireless AP
> back from .jp.
>
> 2 1/4" x 1.5" x .5"
>
> It's cool and functional in itself, and Ike mentioned how this hardware
> should be in focus for building consumer BSD-based firewall devices.
>
> While drifting through dmesgd this AM, I found this device, an Elecom
> WRH-300, has NetBSD running on it, thanks to Jun for his work:
>
> http://dmesgd.nycbug.org/index.cgi?do=view&id=3251
>
> This thing is *small*.

I was always interested in getting devices that are plug and play so
that a device running NetBSD
can use it across an interface that doesn't require porting.



From kmsujit at gmail.com  Sat Aug 26 04:49:46 2017
From: kmsujit at gmail.com (Sujit K M)
Date: Sat, 26 Aug 2017 14:19:46 +0530
Subject: [talk] Machine Learning
Message-ID: <CA+6mNepyeMAygmc3WuESmO2aEwu46KxcYUw+t0sUiW4Ns9fihA@mail.gmail.com>

Hi All,

I have had regular debates with friends in top companies in USA, Who say without
an ML tag they won't have got a job in these companies. My response to that is
If you are aware of Data Issues which plague a give system you would be able to
work on ML. I find it funny that these companies really give so much Emphasis on
ML. My simple Use case in this scenario is this.

CODE + DATA   ----> Non Deterministic (No Result/Negative Result)
                                                          |
                                                          |
                                                          |
                                               Domain Knowledge --->
Machine Learning Algo --> Result

So according to me this use case is sufficient for ML Jobs.

Regards,
Sujit K M



From _ at thomaslevine.com  Sat Aug 26 07:18:59 2017
From: _ at thomaslevine.com (Thomas Levine)
Date: Sat, 26 Aug 2017 11:18:59 +0000
Subject: [talk] Machine Learning
In-Reply-To: <CA+6mNepyeMAygmc3WuESmO2aEwu46KxcYUw+t0sUiW4Ns9fihA@mail.gmail.com>
References: <CA+6mNepyeMAygmc3WuESmO2aEwu46KxcYUw+t0sUiW4Ns9fihA@mail.gmail.com>
Message-ID: <20170826111908.6661C7F982@mailuser.nyi.internal>

I don't follow your use case chart; could you give an example?

It sounds, though, like you and your friends are talk past each other.
They discuss job requirements, and you discuss software design. Dijkstra
commented on this common conflict.

  Simplicity is a great virtue but it requires hard work to achieve it
  and education to appreciate it. And to make matters worse: complexity
  sells better.
  https://www.cs.utexas.edu/users/EWD/transcriptions/EWD08xx/EWD896.html

Ignoring the need to sell newfangled complexity, I find generic search
engines, strict text parsers, and summary statistics to be far more
practical, effective, and reliable than the overwhelming majority of
branded machine learning products. This is just like how I prefer the
reliability and ease-of-use of OpenBSD over the novelty, complexity, and
opacity of most of the other popular contemporary operating systems.



From kmsujit at gmail.com  Sat Aug 26 08:53:46 2017
From: kmsujit at gmail.com (Sujit K M)
Date: Sat, 26 Aug 2017 18:23:46 +0530
Subject: [talk] Machine Learning
In-Reply-To: <20170826111908.6661C7F982@mailuser.nyi.internal>
References: <CA+6mNepyeMAygmc3WuESmO2aEwu46KxcYUw+t0sUiW4Ns9fihA@mail.gmail.com>
 <20170826111908.6661C7F982@mailuser.nyi.internal>
Message-ID: <CA+6mNerHOA9UQ-kkPdWw77fZoRZwW2kSZ23eXW=w4SN+hW46-Q@mail.gmail.com>

On Sat, Aug 26, 2017 at 4:48 PM, Thomas Levine <_ at thomaslevine.com> wrote:
> I don't follow your use case chart; could you give an example?

I am much more interested in virtual landscaping than any thing else,
be it a game
you play or garden you want to see, might be thousands of mile away.

The Idea behind this is that if there an non deterministic problem you
can always solve
the problem by using an algorithm by needing other parameters. Say you
want to see
the garden when flooded with water, you need to know the depth of the
water. You need
parameters like depth of the water/how much water is present in the
garden and real time
plotting the depth of the water in an x y z coordinates. or Say there
is an deluge on an
water filed dam and you want to inspect it, you will see that you need
not only the water level
or amount of water, you will also need to check what is the wind
speed, force in an x y z
coordinate, acceleration in x y z coordinates and complex 3d imaging
concepts. This is not a
scenario you are testing the dams strength in any sense. This is an
example of As A Service
concept, like PaaS or IaaS, At an generic level.



From edlinuxguru at gmail.com  Sat Aug 26 09:21:44 2017
From: edlinuxguru at gmail.com (Edward Capriolo)
Date: Sat, 26 Aug 2017 09:21:44 -0400
Subject: [talk] Machine Learning
In-Reply-To: <20170826111908.6661C7F982@mailuser.nyi.internal>
References: <CA+6mNepyeMAygmc3WuESmO2aEwu46KxcYUw+t0sUiW4Ns9fihA@mail.gmail.com>
 <20170826111908.6661C7F982@mailuser.nyi.internal>
Message-ID: <CAENxBwz-Zt1-8JQnmbNVKhbMKSurxCtzhr-2qQeJWT33vWvQtg@mail.gmail.com>

On Sat, Aug 26, 2017 at 7:18 AM, Thomas Levine <_ at thomaslevine.com> wrote:

> I don't follow your use case chart; could you give an example?
>
> It sounds, though, like you and your friends are talk past each other.
> They discuss job requirements, and you discuss software design. Dijkstra
> commented on this common conflict.
>
>   Simplicity is a great virtue but it requires hard work to achieve it
>   and education to appreciate it. And to make matters worse: complexity
>   sells better.
>   https://www.cs.utexas.edu/users/EWD/transcriptions/EWD08xx/EWD896.html
>
> Ignoring the need to sell newfangled complexity, I find generic search
> engines, strict text parsers, and summary statistics to be far more
> practical, effective, and reliable than the overwhelming majority of
> branded machine learning products. This is just like how I prefer the
> reliability and ease-of-use of OpenBSD over the novelty, complexity, and
> opacity of most of the other popular contemporary operating systems.
>
> _______________________________________________
> talk mailing list
> talk at lists.nycbug.org
> http://lists.nycbug.org/mailman/listinfo/talk
>

Ignoring the need to sell newfangled complexity, I find generic search
engines, strict text parsers, and summary statistics to be far more
practical, effective, and reliable than the overwhelming majority of
branded machine learning products. This is just like how I prefer the
reliability and ease-of-use of OpenBSD over the novelty, complexity, and
opacity of most of the other popular contemporary operating systems.

One of the reasons machine learning is becoming popular is because big data
is commodity now. Users can store vast amounts of data in "the cloud"
BigQuery, Hadoop, Redshift, Oracle RAQ, etc. Ten years ago the dominant
challenge was cheaply storing and querying data, but now that is becoming
commodity. An enterprise can buy tableau and use Amazon Redshift, and have
an analyst (or a non technical product manager) give them summary
statistics untill all parties are blue in the face, 500 scheduled reports
running a day.

You can not replace summary statistics with machine learning. A classic
machine learning tool is linear regression which you can use to make
predictions. You take a dataset and you train a model. That model can be
used to make predictions.

For example given: users with tens/hundreds/thousands of attributes (age,
gender,...) and a bid request with (tens/hundreds/thousands) of attributes
(time of day, url,...), what attributes can be used to predict the final
bid price? Running one process (Linear Regression) that tells what
attribute or combinations of attributes predicts the price, COULD BE
easier/more simple then having humans attempt to figure it out by producing
different sets of summary statistics and collectively deciding what to
optimize on, and constantly re-evaluating the rules as the landscape
changes.

Obviously there is hype cycle, not every problem needs machine learning to
solve. But get readdyyy for a ::shocker:: not everything is solved by BSD
port tree.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.nycbug.org:8443/pipermail/talk/attachments/20170826/fdd5eb57/attachment.htm>

From kmsujit at gmail.com  Sat Aug 26 09:55:31 2017
From: kmsujit at gmail.com (Sujit K M)
Date: Sat, 26 Aug 2017 19:25:31 +0530
Subject: [talk] Machine Learning
In-Reply-To: <CAENxBwz-Zt1-8JQnmbNVKhbMKSurxCtzhr-2qQeJWT33vWvQtg@mail.gmail.com>
References: <CA+6mNepyeMAygmc3WuESmO2aEwu46KxcYUw+t0sUiW4Ns9fihA@mail.gmail.com>
 <20170826111908.6661C7F982@mailuser.nyi.internal>
 <CAENxBwz-Zt1-8JQnmbNVKhbMKSurxCtzhr-2qQeJWT33vWvQtg@mail.gmail.com>
Message-ID: <CA+6mNeqxnOFCWquBmCuvQQyq+Eajpe7oFszWL9pyq_tzEiBwUg@mail.gmail.com>

> One of the reasons machine learning is becoming popular is because big data
> is commodity now. Users can store vast amounts of data in "the cloud"
> BigQuery, Hadoop, Redshift, Oracle RAQ, etc. Ten years ago the dominant
> challenge was cheaply storing and querying data, but now that is becoming
> commodity. An enterprise can buy tableau and use Amazon Redshift, and have
> an analyst (or a non technical product manager) give them summary statistics
> untill all parties are blue in the face, 500 scheduled reports running a
> day.

Machine Learning might be that. But these don't solve an non
deterministic problem.

>
> You can not replace summary statistics with machine learning. A classic
> machine learning tool is linear regression which you can use to make
> predictions. You take a dataset and you train a model. That model can be
> used to make predictions.

Agree with this too. But you are scoping yourself primarily on present or past
you are not taking care of future. Which Machine Learning should be
used to solve.

>
> For example given: users with tens/hundreds/thousands of attributes (age,
> gender,...) and a bid request with (tens/hundreds/thousands) of attributes
> (time of day, url,...), what attributes can be used to predict the final bid
> price? Running one process (Linear Regression) that tells what attribute or
> combinations of attributes predicts the price, COULD BE easier/more simple
> then having humans attempt to figure it out by producing different sets of
> summary statistics and collectively deciding what to optimize on, and
> constantly re-evaluating the rules as the landscape changes.

This forms the basis of computing, We had a Calculator invented for
this purpose.

>
> Obviously there is hype cycle, not every problem needs machine learning to
> solve. But get readdyyy for a ::shocker:: not everything is solved by BSD
> port tree.

I would be only happy to see the day we can solve a problem say increase every
.001% of transactions.



From jun at soum.co.jp  Mon Aug 28 22:04:39 2017
From: jun at soum.co.jp (Jun Ebihara)
Date: Tue, 29 Aug 2017 11:04:39 +0900 (JST)
Subject: [talk] crazy hardware porting example
In-Reply-To: <1503681789.487631.1085095864.137F3DFE@webmail.messagingengine.com>
References: <50f60d80-1e18-ff32-46f7-df8d21712358@ceetonetechnology.com>
 <1503681789.487631.1085095864.137F3DFE@webmail.messagingengine.com>
Message-ID: <20170829.110439.1528659580440536817.jun@soum.co.jp>

From: "Isaac (.ike) Levy" <ike at blackskyresearch.net>
Subject: Re: [talk] crazy hardware porting example
Date: Fri, 25 Aug 2017 13:23:09 -0400

> On Fri, Aug 25, 2017, at 01:10 PM, George Rosamond wrote:
>> While drifting through dmesgd this AM, I found this device, an Elecom
>> WRH-300, has NetBSD running on it, thanks to Jun for his work:
>> http://dmesgd.nycbug.org/index.cgi?do=view&id=3251

http://www2.elecom.co.jp/products/WRH-300CRWH.html
Mediatek MT7620 

> Woah woah I'm confused- Jun got the *tiny* device I brought back to run
> NetBSD?!?!

Ike got http://www2.elecom.co.jp/products/WRH-300BK.html
Last evening,I found WRH-300BK at Jyannpara-Kichijoji.
have a pint of bitter and open the closure.

WRH-300BK contains RTL8196E.

from card edge: cu -38400bps?
-----------
O  <-- GND
O  <-- RX
O  <-- TX
O  <-- 5V

--
Jun Ebihara



From max at stucchi.ch  Sat Aug 26 07:55:48 2017
From: max at stucchi.ch (Massimiliano Stucchi)
Date: Sat, 26 Aug 2017 13:55:48 +0200
Subject: [talk] crazy hardware porting example
In-Reply-To: <b3bcee81-aac4-8021-43a3-b54f8721631d@ceetonetechnology.com>
References: <50f60d80-1e18-ff32-46f7-df8d21712358@ceetonetechnology.com>
 <1503681789.487631.1085095864.137F3DFE@webmail.messagingengine.com>
 <b3bcee81-aac4-8021-43a3-b54f8721631d@ceetonetechnology.com>
Message-ID: <74e1925d-a9a7-86d8-edda-70c16fe3f459@stucchi.ch>



On 25/08/2017 23:34, George Rosamond wrote:

> 
> * Pine64 port is in decent shape with the BSDs IIRC, but how about the
> Pinebook?

Peter Hessler tweeted yesterday about it:
https://twitter.com/phessler/status/901172141881778176

Apparently it works.

Ciao!

-- 

Massimiliano Stucchi
MS16801-RIPE

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 496 bytes
Desc: OpenPGP digital signature
URL: <https://lists.nycbug.org:8443/pipermail/talk/attachments/20170826/4f0c7f94/attachment.bin>

From pete at nomadlogic.org  Tue Aug 29 13:11:10 2017
From: pete at nomadlogic.org (Pete Wright)
Date: Tue, 29 Aug 2017 10:11:10 -0700
Subject: [talk] Updated FreeBSD Graphics For Intel GPU's
Message-ID: <4f4a39f5-f2b5-9635-7219-bff4db4157c7@nomadlogic.org>

hey everyone - so i've been helping test the work that the FreeBSD folks 
have been doing to get support for modern Intel GPU's based on the i915 
chipset (which is a majority of the laptops shipped these days).? The 
initial work was to improve the linux KBI interface in FreeBSD which 
allowed the developers to easily import the linux i915 code into 
FreeBSD.? The decision was also made to create a port of this work - 
much in the same way that the Nvidia Xorg drivers are handled.

I've been dog-fooding this code for quite a while now and while there 
are some rough edges it looks like this work has been accepted by the 
ports team:

https://reviews.freebsd.org/rP448928

I think this is a major milestone, and hopefully will make it easier for 
FreeBSD devs and users to finally ditch macbooks and start using FreeBSD 
as their daily driver!


Cheers,

-pete


-- 
Pete Wright
pete at nomadlogic.org
@nomadlogicLA



From scottro11 at gmail.com  Tue Aug 29 13:28:09 2017
From: scottro11 at gmail.com (Scott Robbins)
Date: Tue, 29 Aug 2017 13:28:09 -0400
Subject: [talk] Updated FreeBSD Graphics For Intel GPU's
In-Reply-To: <4f4a39f5-f2b5-9635-7219-bff4db4157c7@nomadlogic.org>
References: <4f4a39f5-f2b5-9635-7219-bff4db4157c7@nomadlogic.org>
Message-ID: <20170829172809.GA18844@scott1.scottro.net>

On Tue, Aug 29, 2017 at 10:11:10AM -0700, Pete Wright wrote:
> hey everyone - so i've been helping test the work that the FreeBSD folks
> have been doing to get support for modern Intel GPU's based on the i915
> chipset (which is a majority of the laptops shipped these days).? The
> initial work was to improve the linux KBI interface in FreeBSD which allowed
> the developers to easily import the linux i915 code into FreeBSD.? The
> decision was also made to create a port of this work - much in the same way
> that the Nvidia Xorg drivers are handled.
> 
> I've been dog-fooding this code for quite a while now and while there are
> some rough edges it looks like this work has been accepted by the ports
> team:
> 
> https://reviews.freebsd.org/rP448928

So, in theory, I should be able to install 11.1, install drm-next-kmod port
and intel driver, and it should work?

Are there instructions anywhere?  The previous instructions for testing
drm-next were outdated, last time I looked.  I have my own
http://srobb.net/freebsdintel.html page, which goes through the steps of
installing CURRENT then using the (at that time) drm-next kernel. 


Thanks.





> 
> 
> Cheers,
> 
> -pete
> 
> 
> -- 
> Pete Wright
> pete at nomadlogic.org
> @nomadlogicLA
> 
> _______________________________________________
> talk mailing list
> talk at lists.nycbug.org
> http://lists.nycbug.org/mailman/listinfo/talk

-- 
Scott Robbins
PGP keyID EB3467D6
( 1B48 077D 66F6 9DB0 FDC2 A409 FA54 EB34 67D6 )
gpg --keyserver pgp.mit.edu --recv-keys EB3467D6



From jpb at jimby.name  Tue Aug 29 16:00:32 2017
From: jpb at jimby.name (Jim B.)
Date: Tue, 29 Aug 2017 16:00:32 -0400
Subject: [talk] Updated FreeBSD Graphics For Intel GPU's
In-Reply-To: <4f4a39f5-f2b5-9635-7219-bff4db4157c7@nomadlogic.org>
References: <4f4a39f5-f2b5-9635-7219-bff4db4157c7@nomadlogic.org>
Message-ID: <20170829200032.GA59571@jimby.name>

* Pete Wright <pete at nomadlogic.org> [2017-08-29 13:11]:
> hey everyone - so i've been helping test the work that the FreeBSD folks 
> have been doing to get support for modern Intel GPU's based on the i915 
> chipset (which is a majority of the laptops shipped these days).?? The 
> initial work was to improve the linux KBI interface in FreeBSD which 
> allowed the developers to easily import the linux i915 code into 
> FreeBSD.?? The decision was also made to create a port of this work - 
> much in the same way that the Nvidia Xorg drivers are handled.
> 
> I've been dog-fooding this code for quite a while now and while there 
> are some rough edges it looks like this work has been accepted by the 
> ports team:
> 
> https://reviews.freebsd.org/rP448928
> 
> I think this is a major milestone, and hopefully will make it easier for 
> FreeBSD devs and users to finally ditch macbooks and start using FreeBSD 
> as their daily driver!
> 

Good to hear.  Have you been able to test with hashcat?

Earlier drivers did not work.I've had to go the SuperMicro motherboard
route with Nvidia GPUs a la 
http://www.netmux.com/blog/how-to-build-a-password-cracking-rig


Would be interresting to see if hashcat can work with my lappie's Intel
chip.

Jim B.



From pete at nomadlogic.org  Tue Aug 29 16:02:20 2017
From: pete at nomadlogic.org (Pete Wright)
Date: Tue, 29 Aug 2017 13:02:20 -0700
Subject: [talk] Updated FreeBSD Graphics For Intel GPU's
In-Reply-To: <20170829172809.GA18844@scott1.scottro.net>
References: <4f4a39f5-f2b5-9635-7219-bff4db4157c7@nomadlogic.org>
 <20170829172809.GA18844@scott1.scottro.net>
Message-ID: <eee53870-29a7-644a-d2fd-f133d25f62d7@nomadlogic.org>



On 08/29/2017 10:28, Scott Robbins wrote:
> On Tue, Aug 29, 2017 at 10:11:10AM -0700, Pete Wright wrote:
>> hey everyone - so i've been helping test the work that the FreeBSD folks
>> have been doing to get support for modern Intel GPU's based on the i915
>> chipset (which is a majority of the laptops shipped these days).? The
>> initial work was to improve the linux KBI interface in FreeBSD which allowed
>> the developers to easily import the linux i915 code into FreeBSD.? The
>> decision was also made to create a port of this work - much in the same way
>> that the Nvidia Xorg drivers are handled.
>>
>> I've been dog-fooding this code for quite a while now and while there are
>> some rough edges it looks like this work has been accepted by the ports
>> team:
>>
>> https://reviews.freebsd.org/rP448928
> So, in theory, I should be able to install 11.1, install drm-next-kmod port
> and intel driver, and it should work?
>
> Are there instructions anywhere?  The previous instructions for testing
> drm-next were outdated, last time I looked.  I have my own
> http://srobb.net/freebsdintel.html page, which goes through the steps of
> installing CURRENT then using the (at that time) drm-next kernel.
>

Hi Scott,
I am not %100 clear on the status of 11.1-RELEASE support to be honest. 
It looks like in the Makefile that CURRENT is needed (which I'm guessing 
is due to lkpi changes which aren't present in 11.x yet):

<snip>
.if ${OPSYS} == FreeBSD && ${OSVERSION} < 1200039
IGNORE=??????????????? not supported on 11.x or older, no kernel support
.endif
</snip>

My understanding is that on CURRENT you should be able to build/install 
the pkg and things should start working.? I haven't had a chance to test 
the port itself yet unfortunately due to some time constraints on my 
end, altho i plan on rebuilding my skylake system later this afternoon.? 
I'll report on how that goes and will let you all know.

Cheers,
-pete



-- 
Pete Wright
pete at nomadlogic.org
@nomadlogicLA



From pete at nomadlogic.org  Tue Aug 29 16:06:27 2017
From: pete at nomadlogic.org (Pete Wright)
Date: Tue, 29 Aug 2017 13:06:27 -0700
Subject: [talk] Updated FreeBSD Graphics For Intel GPU's
In-Reply-To: <20170829200032.GA59571@jimby.name>
References: <4f4a39f5-f2b5-9635-7219-bff4db4157c7@nomadlogic.org>
 <20170829200032.GA59571@jimby.name>
Message-ID: <4dd3778a-49b6-a545-ca52-2d0d872eea2a@nomadlogic.org>



On 08/29/2017 13:00, Jim B. wrote:
> * Pete Wright <pete at nomadlogic.org> [2017-08-29 13:11]:
>> hey everyone - so i've been helping test the work that the FreeBSD folks
>> have been doing to get support for modern Intel GPU's based on the i915
>> chipset (which is a majority of the laptops shipped these days).?? The
>> initial work was to improve the linux KBI interface in FreeBSD which
>> allowed the developers to easily import the linux i915 code into
>> FreeBSD.?? The decision was also made to create a port of this work -
>> much in the same way that the Nvidia Xorg drivers are handled.
>>
>> I've been dog-fooding this code for quite a while now and while there
>> are some rough edges it looks like this work has been accepted by the
>> ports team:
>>
>> https://reviews.freebsd.org/rP448928
>>
>> I think this is a major milestone, and hopefully will make it easier for
>> FreeBSD devs and users to finally ditch macbooks and start using FreeBSD
>> as their daily driver!
>>
> Good to hear.  Have you been able to test with hashcat?
>
> Earlier drivers did not work.I've had to go the SuperMicro motherboard
> route with Nvidia GPUs a la
> http://www.netmux.com/blog/how-to-build-a-password-cracking-rig
>
>
> Would be interresting to see if hashcat can work with my lappie's Intel
> chip.

Oh that's pretty awesome.

I know that when mmacy starting hacking on this his original goal was to 
get support for AMDGPU working for some boards he had on hand.

The testing I've done is only on Intel i915 chips you'd find on Skylake 
or Kabylake systems.? Since the rig you link to seems to use NVidia 
GPU's I'd guess that this port wouldn't change anything. Although 
hopefully with the refactored lkpi layer it'll be easier to port over 
Linux GPU code for Nvidia down the road.

Cheers,
-pete

-- 
Pete Wright
pete at nomadlogic.org
@nomadlogicLA



From mmatalka at gmail.com  Wed Aug 30 03:49:07 2017
From: mmatalka at gmail.com (Malcolm Matalka)
Date: Wed, 30 Aug 2017 07:49:07 +0000
Subject: [talk] Updated FreeBSD Graphics For Intel GPU's
In-Reply-To: <4f4a39f5-f2b5-9635-7219-bff4db4157c7@nomadlogic.org> (Pete
 Wright's message of "Tue, 29 Aug 2017 10:11:10 -0700")
References: <4f4a39f5-f2b5-9635-7219-bff4db4157c7@nomadlogic.org>
Message-ID: <868ti15wcs.fsf@gmail.com>

Pete Wright <pete at nomadlogic.org> writes:

> hey everyone - so i've been helping test the work that the FreeBSD folks have been doing to get support for modern Intel GPU's based on the
> i915 chipset (which is a majority of the laptops shipped these days).? The initial work was to improve the linux KBI interface in FreeBSD
> which allowed the developers to easily import the linux i915 code into FreeBSD.? The decision was also made to create a port of this work - 
> much in the same way that the Nvidia Xorg drivers are handled.
>
> I've been dog-fooding this code for quite a while now and while there are some rough edges it looks like this work has been accepted by the
> ports team:

Do you know if this means it'll start showing up TrueOS UNSTABLE woon?

>
> https://reviews.freebsd.org/rP448928
>
> I think this is a major milestone, and hopefully will make it easier for FreeBSD devs and users to finally ditch macbooks and start using
> FreeBSD as their daily driver!
>
>
> Cheers,
>
> -pete



From kmsujit at gmail.com  Wed Aug 30 04:39:42 2017
From: kmsujit at gmail.com (Sujit K M)
Date: Wed, 30 Aug 2017 14:09:42 +0530
Subject: [talk] Updated FreeBSD Graphics For Intel GPU's
In-Reply-To: <4f4a39f5-f2b5-9635-7219-bff4db4157c7@nomadlogic.org>
References: <4f4a39f5-f2b5-9635-7219-bff4db4157c7@nomadlogic.org>
Message-ID: <CA+6mNepAVdgJbZK9rmzKxa6GVXee_D-JCybw+0=-ocn3Thhbiw@mail.gmail.com>

On Aug 29, 2017 10:41 PM, "Pete Wright" <pete at nomadlogic.org> wrote:

hey everyone - so i've been helping test the work that the FreeBSD folks
have been doing to get support for modern Intel GPU's based on the i915
chipset (which is a majority of the laptops shipped these days).  The
initial work was to improve the linux KBI interface in FreeBSD which
allowed the developers to easily import the linux i915 code into FreeBSD.
The decision was also made to create a port of this work - much in the same
way that the Nvidia Xorg drivers are handled.

I've been dog-fooding this code for quite a while now and while there are
some rough edges it looks like this work has been accepted by the ports
team:

https://reviews.freebsd.org/rP448928

I think this is a major milestone, and hopefully will make it easier for
FreeBSD devs and users to finally ditch macbooks and start using FreeBSD as
their daily driver!

Cheers,

-pete


-- 
Pete Wright
pete at nomadlogic.org
@nomadlogicLA

_______________________________________________
talk mailing list
talk at lists.nycbug.org
http://lists.nycbug.org/mailman/listinfo/talk

I understand the Linux kpi let's you integrate with Linux drm. How about
vesa. Looks like the Linux code is better.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.nycbug.org:8443/pipermail/talk/attachments/20170830/defe0923/attachment.htm>

From scottro11 at gmail.com  Wed Aug 30 07:14:37 2017
From: scottro11 at gmail.com (Scott Robbins)
Date: Wed, 30 Aug 2017 07:14:37 -0400
Subject: [talk] Updated FreeBSD Graphics For Intel GPU's
In-Reply-To: <868ti15wcs.fsf@gmail.com>
References: <4f4a39f5-f2b5-9635-7219-bff4db4157c7@nomadlogic.org>
 <868ti15wcs.fsf@gmail.com>
Message-ID: <20170830111437.GA11313@scott1.scottro.net>

On Wed, Aug 30, 2017 at 07:49:07AM +0000, Malcolm Matalka wrote:
> Pete Wright <pete at nomadlogic.org> writes:
> 
> > hey everyone - so i've been helping test the work that the FreeBSD folks have been doing to get support for modern Intel GPU's based on the
> > i915 chipset (which is a majority of the laptops shipped these days).? The initial work was to improve the linux KBI interface in FreeBSD
> > which allowed the developers to easily import the linux i915 code into FreeBSD.? The decision was also made to create a port of this work - 
> > much in the same way that the Nvidia Xorg drivers are handled.
> >
> > I've been dog-fooding this code for quite a while now and while there are some rough edges it looks like this work has been accepted by the
> > ports team:
> 
> Do you know if this means it'll start showing up TrueOS UNSTABLE woon?
> 
I was under the impression that TrueOS is already using the drm-next stuff.
(I'm going on a vague memory of a message from Warren Block after I'd
mentioned my issues with an Intel card on FreeBSD forums, but I can't find
it.)

If you do try with TrueOS, please update us. 


-- 
Scott Robbins
PGP keyID EB3467D6
( 1B48 077D 66F6 9DB0 FDC2 A409 FA54 EB34 67D6 )
gpg --keyserver pgp.mit.edu --recv-keys EB3467D6



From pete at nomadlogic.org  Wed Aug 30 12:48:17 2017
From: pete at nomadlogic.org (Pete Wright)
Date: Wed, 30 Aug 2017 09:48:17 -0700
Subject: [talk] Updated FreeBSD Graphics For Intel GPU's
In-Reply-To: <868ti15wcs.fsf@gmail.com>
References: <4f4a39f5-f2b5-9635-7219-bff4db4157c7@nomadlogic.org>
 <868ti15wcs.fsf@gmail.com>
Message-ID: <b5447805-0f34-e618-2746-9c3772ae159c@nomadlogic.org>



On 08/30/2017 00:49, Malcolm Matalka wrote:
> Pete Wright <pete at nomadlogic.org> writes:
>
>> hey everyone - so i've been helping test the work that the FreeBSD folks have been doing to get support for modern Intel GPU's based on the
>> i915 chipset (which is a majority of the laptops shipped these days).? The initial work was to improve the linux KBI interface in FreeBSD
>> which allowed the developers to easily import the linux i915 code into FreeBSD.? The decision was also made to create a port of this work -
>> much in the same way that the Nvidia Xorg drivers are handled.
>>
>> I've been dog-fooding this code for quite a while now and while there are some rough edges it looks like this work has been accepted by the
>> ports team:
> Do you know if this means it'll start showing up TrueOS UNSTABLE woon?

TrueOS has been incorporating the patches from drm-next for several 
months now, so if you are running UNSTABLE you should be good to go.

-pete

-- 
Pete Wright
pete at nomadlogic.org
@nomadlogicLA



From mmatalka at gmail.com  Thu Aug 31 02:11:17 2017
From: mmatalka at gmail.com (Malcolm Matalka)
Date: Thu, 31 Aug 2017 06:11:17 +0000
Subject: [talk] Updated FreeBSD Graphics For Intel GPU's
In-Reply-To: <20170830111437.GA11313@scott1.scottro.net> (Scott Robbins's
 message of "Wed, 30 Aug 2017 07:14:37 -0400")
References: <4f4a39f5-f2b5-9635-7219-bff4db4157c7@nomadlogic.org>
 <868ti15wcs.fsf@gmail.com> <20170830111437.GA11313@scott1.scottro.net>
Message-ID: <86ziagjmgq.fsf@gmail.com>

Scott Robbins <scottro11 at gmail.com> writes:

> On Wed, Aug 30, 2017 at 07:49:07AM +0000, Malcolm Matalka wrote:
>> Pete Wright <pete at nomadlogic.org> writes:
>> 
>> > hey everyone - so i've been helping test the work that the FreeBSD folks have been doing to get support for modern Intel GPU's based on the
>> > i915 chipset (which is a majority of the laptops shipped these days).? The initial work was to improve the linux KBI interface in FreeBSD
>> > which allowed the developers to easily import the linux i915 code into FreeBSD.? The decision was also made to create a port of this work - 
>> > much in the same way that the Nvidia Xorg drivers are handled.
>> >
>> > I've been dog-fooding this code for quite a while now and while there are some rough edges it looks like this work has been accepted by the
>> > ports team:
>> 
>> Do you know if this means it'll start showing up TrueOS UNSTABLE woon?
>> 
> I was under the impression that TrueOS is already using the drm-next stuff.
> (I'm going on a vague memory of a message from Warren Block after I'd
> mentioned my issues with an Intel card on FreeBSD forums, but I can't find
> it.)
>
> If you do try with TrueOS, please update us. 

I've been running TrueOS UNSTABLE on an Asus UX350CA (i think) ZenBook,
which is Skylake for awhile.  I don't know if the contents of this
thread are in what I'm running but my experience is:

- Video flickers.  Watching something full screen will have the screen
  half-updated when it changes a lot.

- Screen brightness doesn't work.

- Suspend/resume worked for a month and then broke, which makes me sad.

Crossing my fingers all of these will be solved in the near future,
especially screen brightness.  It's like staring at the sun for too
long.

And those three things left are after a long time of having to run scfb,
so it's come a long way and I'm very grateful for the hard work others
have done.

/Malcolm



From pete at nomadlogic.org  Thu Aug 31 12:53:45 2017
From: pete at nomadlogic.org (Pete Wright)
Date: Thu, 31 Aug 2017 09:53:45 -0700
Subject: [talk] Updated FreeBSD Graphics For Intel GPU's
In-Reply-To: <86ziagjmgq.fsf@gmail.com>
References: <4f4a39f5-f2b5-9635-7219-bff4db4157c7@nomadlogic.org>
 <868ti15wcs.fsf@gmail.com> <20170830111437.GA11313@scott1.scottro.net>
 <86ziagjmgq.fsf@gmail.com>
Message-ID: <c2462c93-8e43-bff9-08ab-227a41c75d14@nomadlogic.org>



On 08/30/2017 23:11, Malcolm Matalka wrote:
> Scott Robbins <scottro11 at gmail.com> writes:
>
>> On Wed, Aug 30, 2017 at 07:49:07AM +0000, Malcolm Matalka wrote:
>>> Pete Wright <pete at nomadlogic.org> writes:
>>>
>>>> hey everyone - so i've been helping test the work that the FreeBSD folks have been doing to get support for modern Intel GPU's based on the
>>>> i915 chipset (which is a majority of the laptops shipped these days).? The initial work was to improve the linux KBI interface in FreeBSD
>>>> which allowed the developers to easily import the linux i915 code into FreeBSD.? The decision was also made to create a port of this work -
>>>> much in the same way that the Nvidia Xorg drivers are handled.
>>>>
>>>> I've been dog-fooding this code for quite a while now and while there are some rough edges it looks like this work has been accepted by the
>>>> ports team:
>>> Do you know if this means it'll start showing up TrueOS UNSTABLE woon?
>>>
>> I was under the impression that TrueOS is already using the drm-next stuff.
>> (I'm going on a vague memory of a message from Warren Block after I'd
>> mentioned my issues with an Intel card on FreeBSD forums, but I can't find
>> it.)
>>
>> If you do try with TrueOS, please update us.
> I've been running TrueOS UNSTABLE on an Asus UX350CA (i think) ZenBook,
> which is Skylake for awhile.  I don't know if the contents of this
> thread are in what I'm running but my experience is:
>
> - Video flickers.  Watching something full screen will have the screen
>    half-updated when it changes a lot.
>
> - Screen brightness doesn't work.
>
> - Suspend/resume worked for a month and then broke, which makes me sad.
>
> Crossing my fingers all of these will be solved in the near future,
> especially screen brightness.  It's like staring at the sun for too
> long.
>
> And those three things left are after a long time of having to run scfb,
> so it's come a long way and I'm very grateful for the hard work others
> have done.

out of curiosity - are you seeing that behavior using the modesetting or 
intel-i915 xorg driver?? my experience using modesetting (and glamor for 
GL acceleration) has been pretty good on both skylake and kabylake 
systems and has been more stable than the intel-i915 xorg driver.? for 
example i frequently watch youtube and other html5 videos with 
acceptable playback.

suspend and resume has been working on my end for quite a while as well 
- although there are certainly some *very* rough edges in this regard.? 
one sysctl knob that may help is:

hw.acpi.reset_video=1


-pete

-- 
Pete Wright
pete at nomadlogic.org
@nomadlogicLA