August 2014 - luv-talk

Where is Keech House? (Sun 7th sustainable house day)
by Douglas Ray 01 Sep '14

01 Sep '14

sustainablehouseday.com shows Richard Keech's house googlemapped as a church in Essendon ("Presbyterian Church of Vic."), and no address. Can anyone help? ta, Douglas

2 2

Re: [luv-talk] Secure and Bug Free ?
by David E Payne 31 Aug '14

31 Aug '14

On Sat, 30/8/14, Rohan McLeod <rhn(a)jeack.com.au> wrote: Subject: [luv-talk] Secure and Bug Free ? To: luv-talk(a)luv.asn.au Received: Saturday, 30 August, 2014, 9:26 AM SNIP "... control and from there to fly-by-wire and then fly-by-fiber networks; (removing hundreds of kilo-meters of fire prone complex, heavy wiring in the process); ..." DP: I once heard somebody referring to the fact that while mechanical, electrical OR optical control wires/fibres are often severed by bullets or shrapnel, leading to a plane crash, because a <insert name of the military jet> had a boron fuselage the whole thing could transmit optical packet signals which would still get through as long as there was any continuous skin between the communicants. SNIP "... anyway the assumption is that the compiler may introduce bugs, security holes. " DP: Many LUVees will have read or heard of Ken Thompson's infamous Turing Award Lecture This link enables downloading a PDF http://dl.acm.org/ft_gateway.cfm?id=1283940&type=pdf&ip=202.45.117.50 . or if you prefer select it yourself eg. from http://amturing.acm.org/award_winners/thompson_4588371.cfm Given Ken's explanation & the layers of software inside hardware, (was there EVER a X86 desktop microprocessor if that means where hardware executed X-86 instructions?), I suspect confidence that no one CAN spy on one's computer(s) is rarely justified! " ... Apparently the resulting RT OS has very little functionality, which is remedied by supplying that functionality via Linux VM's ..." DP: The RT OS runs on a VM which runs on Linux or have I misunderstood? (I REALLY don't think embedded system gurus will regard that as hard real time any-more, at least on a normal Linux kernel.) "... Apparently the OS is available as open-source and binary for ARM and x86 CPU's, but not in the above secure form at: http://sel4.systems " DP: To paraphrase (yes this time I will in a teasing fashion), Ken in the above lecture: simply examining the source code & compiling it yourself won't guarantee that the resulting software won't have intentionally programmed effects you don't want but someone else does. Not that I'm suggesting NICTA would would destroy your purity of essence with nasty spyware additives; - no sir! On a related note, Genode released a new framework last Thursday. "Genode is an offspring of the L4 community" whatever that means. Somebody decided to fork off & do their own project? http://genode.org/about/index

2 2

Secure and Bug Free ?
by Rohan McLeod 30 Aug '14

30 Aug '14

Assembled Cognoscenti; I thought the following tale in which Linux turned out to be the insecure element; might be of amusement. In a conversation with a gentleman at the Nicta "TechShowCsae2014" yesterday; http://www.eventbrite.com.au/e/nicta-tech-showcase-2014-tickets-12119157727… he was explaining his group's interest in secure and bug-free software; in the context of 'mission-critical' applications; particularly OS's. I thought 'mission-critical' had something to do with the military and large corporation's; but it turned out they are primarily looking at the OS's in vehicles and air-craft. Apparently there have been recent cases of the "OS's' of cars being cracked remotely; via the wireless immobilizing system and the brakes etc. being operated remotely ! Regardless; the story seemed to be that as aircraft move from mechanical to electrical control and from there to fly-by-wire and then fly-by-fiber networks; (removing hundreds of kilo-meters of fire prone complex, heavy wiring in the process); a huge requirement for secure and bug-free OS's arose. Since sequential bug removal is not a practical option :-) What he seemed to be claiming was his group had written a RT OS in C; and then compiled it for ARM processors. Where things seemed to become interesting was 1/ the C code was apparently subject to some kind of mathematical analysis which 'somehow' checked the code's logical integrity against the specification ? (Is this possible ?) 2/ the machine code from the compiler was then checked against the specification ? - I have no idea what this means !; anyway the assumption is that the compiler may introduce bugs, security holes. Apparently the resulting RT OS has very little functionality, which is remedied by supplying that functionality via Linux VM's Anyway to get to the end of the tale; where Linux turned out to be the insecure element !; they are employing 'Red-Teams' (Tiger-teams ?) of hackers to 'crack' the Linux VM's. This is not just pie-in-the sky stuff either; they had a quad-copter with such an add-on OS; as proof of concept. Apparently the OS is available as open-source and binary for ARM and x86 CPU's, but not in the above secure form at: http://sel4.systems regards Rohan McLeod

3 3

Should have told their lawyers to take an icy cold shower?
by David E Payne 30 Aug '14

30 Aug '14

Not long after a tragedy surely destined to inspire tasteless limericks (here ... pick your own media outlet) https://duckduckgo.com/?q=ice-bucket-challenge-cofounder-corey-griffin-drow… the Amyotrophic Lateral Sclerosis Association applied for a trademark on the term "Ice Bucket Challenge" despite it's earlier use by other charities. Maybe they doubted they could patent the "process" due to such prior art as the sauna. They've now characterised their reasons in a way which reminds me of the Free Software Foundation arguments for the greater restrictions of GNU Public Licences compared to other licences such as MIT. I have mixed feelings about the GPL but this trademark application just offends and annoys me. Maybe more businesses (including non-profits), when their lawyers suggest they can (& therefore should?) do something like this, ought to "Just Say No". (I was going to follow THAT phrase with "TM Nancy Raygun" but I think making false claims of trademark is illegal. Unlike applying for one you aren't entitled to;- Microsoft's trademarks on "Windows" & other generic terms being other examples IMNSHO).

1 1

Secure and Bug Free ?
by Rohan McLeod 29 Aug '14

29 Aug '14

Adam ; on the assumption you are not on luv-talk, I have forwarded the following: I thought the tale below; in which Linux turned out to be the insecure element; might be of amusement. In a conversation with a gentleman at the Nicta "TechShowCsae2014" yesterday; http://www.eventbrite.com.au/e/nicta-tech-showcase-2014-tickets-12119157727… he was explaining his group's interest in secure and bug-free software; in the context of 'mission-critical' applications; particularly OS's. I thought 'mission-critical' had something to do with the military and large corporation's; but it turned out they are primarily looking at the OS's in vehicles and air-craft. Apparently there have been recent cases of the "OS's' of cars being cracked remotely; via the wireless immobilizing system and the brakes etc. being operated remotely ! Regardless; the story seemed to be that as aircraft move from mechanical to electrical control and from there to fly-by-wire and then fly-by-fiber networks; (removing hundreds of kilo-meters of fire prone complex, heavy wiring in the process); a huge requirement for secure and bug-free OS's arose. Since sequential bug removal is not a practical option :-) What he seemed to be claiming was his group had written a RT OS in C; and then compiled it for ARM processors. Where things seemed to become interesting was 1/ the C code was apparently subject to some kind of mathematical analysis which 'somehow' checked the code's logical integrity against the specification ? (Is this possible ?) 2/ the machine code from the compiler was then checked against the specification ? - I have no idea what this means !; anyway the assumption is that the compiler may introduce bugs, security holes. Apparently the resulting RT OS has very little functionality, which is remedied by supplying that functionality via Linux VM's Anyway to get to the end of the tale; where Linux turned out to be the insecure element !;they are employing 'Red-Teams' (Tiger-teams ?) of hackers to 'crack' the Linux VM's. This is not just pie-in-the sky stuff either; they had a quad-copter with such an add-on OS;as proof of concept. Apparently the OS is available as open-source and binary for ARM and x86 CPU's, but not in the above secure form at: http://sel4.systems regards Rohan McLeod

1 0

Re: [luv-talk] Playing LP's with a laser pickup ?
by David E Payne 29 Aug '14

29 Aug '14

Exact citation thanks toDavid I believe the device described and sold at the URL below is real; I was really just exploring whether 1/ anyone had every had any experience with such and 2/ whether any other varieties existed regards Rohan I don't have any reason to doubt that $15k+ device is/was real. I expect some hackers could make their own LASER pickup for a turntable for less than $500-$1000. (Maybe less than $50, I dunno.) My only doubt is whether an UNMODIFIED LaserDisc player could really play UNMODIFIED LPs. The audio and video on LaserDiscs was analog but span much faster than 33 1/3 RPM. A C.onstant A.ngular V.elocity Disc for PAL would spin at 25 revs per SECOND = 1500 RPM as each track held one video frame. (30RPS = 1800 RPM for an NTSC disc.) So for a start, the sound would have to BE buffered, say by writing to & replaying from a computer file, otherwise the treble section would be ultrasonic & only microscopic creatures could dance to the beat! The Wikipedia LaserDisc article is pretty comprehensive. In referring to LASER Rot however, it doesn't mention that apparently the metal recording surface was exposed at the outside edge, which I read back then made the discs more vulnerable than CDs etc. It also doesn't mention a reason for adopting a smaller form-factor for the Digital A.udio D.is(c or k?) standard which became the Compact Disc. Manufacturers wanted to make a player that would fit in the standard car radio/cartridge/cassette dashboard space. (There was a least one phonograph produced for cars I think.) I would have liked to see the "Soundstream Audio File" system adopted as the DAD standard but there wasn't much chance of a little company beating the winning alliance of Philips & Sony. The article reference 14 in http://en.wikipedia.org/wiki/Soundstream is not available at http://www.eetimes.com/ but basically AudioFile, VideoFile & DataFile used a photographically reproducible optical track read by an orbiting lens which swept past & tracked along the length of the rectangular card. (Or the card could have tracked past the lens but that's not how I remember it. Wouldn't affect compatibility either way.) Because the medium wasn't spinning, all the arc-shaped tracks were the same length, also the medium could be various lengths & potentially transparent so both sides could be played without flipping. (I read about AudioFile in an article by J. Hansen called "The Record that Doesn't Go Round" in January 1983 "Hi-Fi News & Record Review" magazine presumably still at the State library of Victoria. Maybe in the stacks now. I still had a photocopy not long ago, somewhere. Exact citation thanks to http://arpjournal.com/2140/soundstream-the-introduction-of-commercial-digit… )

1 2

Curious CD audio burning behavior !
by Rohan McLeod 25 Aug '14

25 Aug '14

Assembled cognoscenti; just wondering about an explanation (if any); for anomalous results when burning an audio-CD. The problem was to burn a compilation audio-CD; so it could be played on the audio-CD player on my mini-hifi system There are two "boxes" identical OS and application; application on box I aborts burn twice; seemingly complaining about discontinuous audio stream; application on box II completes without problem. Box I: 2.8Ghz dual core CPU 2GB RAM ; IDE CD/DVD burner SATA HD Box II : old Intel server-board 2.4GHz dual core x 2 Xeon CPU's and 2GB of RAM; CD/DVD burner is IDE with adaptor card so that HD and CD burner; are both on the SCSI bus connected to Adaptec 39320 LVD SCSI controller; in PCI-X slot. thanks Rohan McLeod

4 10

2 gb ddr2 ram stick
by Daniel J Jitnah 23 Aug '14

23 Aug '14

Hi Luvers, I have a rather old MSI amd-am2 cpu motherboard which was originally rated to take max 2gb ram. However there are reports on the web that it can actually take 4 gb, but I can't work out if it is exactly the same motherboard or a later revision (same model number) - it looks the same in photos though. There are no bios upgrade that I can find neither. My guess is that when it was first released, 2gb ddr2 ram sticks did not exist, so it was rated to 2gb max (ie 2 x 1gb sticks). Would that make sense? (we are talking around 2007?) One way to find out is to try! Any one has 1 (or preferably 2 matching) 2gb ddr2 ram sticks that I can borrow please? Thanks in advance. Daniel.

2 1

Re: [luv-talk] Android 4.4 external SD card
by Tim Connors 23 Aug '14

23 Aug '14

On Fri, 15 Aug 2014, Andrew McGlashan wrote: > On 14/08/2014 7:19 PM, Russell Coker wrote: > > How reliable is the process of rooting a phone using instructions such as the > > above? Can I expect it to work without wiping the phone data? > > When I get the time, I would like to ditch Samsung /lack of support/ for > my i9300 and i9505 devices that are 100% stock and only getting 4.3 at > this stage -- the devices are fully capable of running the latest > Android version. > > One of my justifications for going stock Samsung was to get updates that > were not carrier branded, nor carrier delayed. And given all the horror > of today's Android App permissions, I might even go to the dark side and > choose an iPhone next. Suffice to say, I'm not at all happy with the > way things are going Android now. Grr, I've just been looking at this, since my own android phone is a pile of shit, and appears to be monitored by the Mafia, RIAA, Russians, and ASIO (or maybe Android is just shit, and 20% of the battery really does go to "Android OS" and another 19% of battery goes to "Android system" and another 20% of battery goes to Flaky-as-fuck WiFi in 8 hours of uptime while it sits idle in my pocket). My requirements: 1) can easily (no downloading rootkits from the Russians) run latest stable version of clockworkmod, and likely to keep running latest stable (because android permission models are absolutely shit by default - surely CWM's is slightly better, even though I've never seen it in practice). Probably also bypasses all of the google and carrier issues. Do you lose any functionaility I care about? I dunno. 2) has an sdcard slot. No google, I can't move all my music into the cloud, because I tend to want to listen to it 1000km from the nearest mobile phone tower, and I'm too cheap to have any more than 200MB of data per month connected to my phone account (internode, $10 per month. Told you I was cheap. No, you can't buy that plan anymore). HTC one m8 can do it. Maybe. But it will probably suck, because of the 3 golden rules 1) all software sucks 2) all hardware sucks 3) all phones suck 4) it's too expensive ok, 4 golden rules. Sorry I'll come in again. -- Tim Connors

4 4

Fw: Trans Pacific Partnership
by David E Payne 20 Aug '14

20 Aug '14

SNIP > In ~ 30 minutes, (2014 Aug 20th Wed 6:30pm), there will be a public meeting at Brighton Town Hall entitled "How could the TPP affect your future?" Here's where I deliberately don't provide the inbody characterisation or even expansion of acronyms that some eg. Rohan want. > Here's the general website http://tppaustralia.org/ > It includes a search function to find events in your area. > > Monash Uni's Kevin Korb IS concerned about TPP but seems to feel the DCTA legislation may be even worse than TPP & that's saying a lot! http://bayesian-intelligence.com/bwb/2014-02/reaction-to-the-dtca-begins/ > Dav* - (I'm so pessimistic sometimes I think I'll NEVER manage to read everything on the World Wide Web!)

2 1