Re: CUDA Setup Failing

[Andrew Zyman]

Eric my last experiments were with it on version 6.so.. check if you have
the original driver installed correctly not the RPM version of it.

On Apr 18, 2017 19:30, "Lofgren, Eric"  wrote:

Hey all,

I’m relatively new to Linux generally and SL in particular. I’m running
SL7, and have run aground trying to follow the setup instructions at
http://developer.download.nvidia.com/compute/cuda/8.0/
secure/Prod2/docs/sidebar/CUDA_Installation_Guide_Linux.
pdf?autho=1492553753_fc0d9c1b985f0ba78200964ccf8f1d
da=CUDA_Installation_Guide_Linux.pdf

I’ve been following the RHEL/CentOS instructions, as my understanding is
that SL7 should just be a drop in for those. Things seem to go well until
the “Verify the Installation” step at 6.2.2 in the document.

Verifying the Driver Version with $cat /proc/driver/nvidia/version returns
‘No such file or directory’

The examples compile fine in 6.2.2.3, but when you run the deviceQuery
binary, rather than finding the Quadro K620 sitting in the machine at the
moment, it returns:

./deviceQuery Starting…

CUDA Device Query (Runtime API) version (CUDART static linking)

cudaGetDeviceCount returned 30
-> unknown error
Result = FAIL

That’s where I stopped, as it appears the GPU isn’t talking nicely with
CUDA. lspci can see the GPU - is this just a driver issue? If so, is sudo
yum install cuda-drivers likely the easiest fix?

Thanks in advance for the help,

Eric

Re: Brainfart: Academic paper rewrite tool

[Andrew Zyman]

Keith,
 It is an interesting subject - one of my "nutz" ideas was creation of a
semantic Web from scientific papers.
 let's assume that there is a market( interest ) for these "liberated"
papers.
Have your thought about business model? Who will pay for liberated papers
and for development of such a process/software?

On Apr 9, 2017 22:17, "Keith Lofstrom"  wrote:

> I watched "The Internet's Own Boy" about Aaron Swartz a few
> days ago.  Wonderful motivation: make academic literature
> publically accessable.  Stupid juvenile implementation:
> download a million papers from JSTOR in an MIT wiring
> closet.  Horrible outcome: Federal prosecution, suicide.
>
> I'll let others march in the streets and demand free ice
> cream and ponies.  I hope that works, but I wouldn't bet
> on it.
>
> I want to make the information contained in academic
> literature publically accessable ... and versioned, and
> updated, and weblinked, and supported.  The academics
> here might have some ideas.
>
> What if:  We (meaning those more capable than me) construct
> a software environment for disassembing the pdf elements of
> an academic paper, which a moderately literate person can
> use to REWRITE and redraw and reformat the paper as a
> substantially different work with an updated version of
> the same information in it?  I am inspired by some of the
> capabilities in Inkscape for reworking graphics into SVG.
>
> What if we improve that process, for example tying a graph
> in an old paper to data from new research that verifies,
> refines, or refutes it?  Move the slider on the graph from
> an original 1960 paper to a new graph that incorporates
> 2017 data?
>
> I have about a dozen published journal papers out there
> (one with over 200 citations) that I would love to
> "de-copyright" out of the clutches of the IEEE, Elsevier,
> etc.  My newest stuff is publically posted as pre-
> publication drafts before I submit it, but I would love
> simple tools that would ease the process of liberating
> my older work.  I'd be glad to help as an alpha-test
> guinea pig, also use the tool for new writing projects.
>
> There are other papers by other authors, some long since
> dead, that I would love to apply the same treatment to,
> so I can cite the liberated version in my open version.
>
> At the end of the rewrite process, the tool can compare
> the original and the liberated versions and estimate
> the legally actionable overlap, which a creative-commons
> community can continue to rework until the overlap is
> zero, and also re-rework if legal threats or court
> decisions add new restrictions to work around.
>
> I expect the Big Content owners will attempt to enact
> legislation to forbid the process, but I believe we can
> rewrite code and evade restrictions faster than they
> can write and pass legislation.  If they are panicky
> in their legislative responses, we can probably trick
> them into passing laws against their own practices.
>
> 95% of the work out there is obscure, and the world
> does not need a rewrite.  If 10 million (WAG) academic
> papers have ever been written, that might mean 500,000
> to be processed.  That would be a lot more work than
> went into wikipedia, but a lot fewer hours than all
> US citizens waste on TV in a year.  It might take a
> few decades, but a small subset of the world's thinkers
> could eventually get this done, and incorporate the
> process into the training of new scholars.
>
> Hopefully, the tools that we write will become the go-to
> tools for the creation of new works, a dual path process
> for authors that produces both a terse, stylized version
> that the JSTORs and Elseviers of the world can greedily
> guard and sell ...
>
> ... and an open, updatable, friendly version of the
> document that 99% of the world will actually use.
> Personally, I would love it if my best papers outlived
> me by centuries, steadily improving and accumulating
> hundreds of coauthors into the far future.
>
> While this would be a very complicated suite of tools to
> write, it's gotta be a lot easier than designing rockets
> and self-driving cars.
>
> Am I nuts?
>
> Keith
>
> --
> Keith Lofstrom  kei...@keithl.com
>