renaming the git repository directory for the purpose of hiding it from various git-ish tools

In a flash of genius and fuck-wittery I add a git repo at the base of my home directory on my main workstation.

The purpose of this was to find out where various GUI apps were writing their changes, and to sync my dotfiles to github. (I had a similar strategy in place using svn, but I've since mostly abandoned svn from my workflow, and was really only deciding which of hg or git to switch to)

Little did I know that I had just caused myself a bunch of pain.

$ cd
$ git init
$ vi .gitignire
$ git add -A


So I added a wildcard to .gitignore and thought nothing more of it, until....

in eclipse, stuck like this

indexing repo tomhodder....

basically, that I was informed by the folks in #git on freenode that gitignore causes the resources listed to been hidden from git, but not ignored by git. hence they are still in the repository, and hence commands like

$ git status

still return the repo status, even from ignored directories, and this was causing eclipse to have fits.

My work-around is to rename the git repo dir, in order that it is actually ignored by the default tools, and I can access my home dir git repo using an alias like so;

$ cd
$ mv .git .git_alt
$ alias altgit='git --git-dir=.git_alt'
source ~/bashrc


$ git status
fatal: Not a git repository (or any parent up to mount point /home)
Stopping at filesystem boundary (GIT_DISCOVERY_ACROSS_FILESYSTEM not set).

$ altgit status
# On branch master
#
# Initial commit
...


which is fine for my use-case.

The caveat was that, I found if you set the alias first, say in this order;

$ alias altgit='git --git-dir=.git_alt'

$ altgit  init
Initialized empty Git repository in /home/tomhodder/.git_alt/


$ altgit status
fatal: This operation must be run in a work tree

Something fucks up.

I would raise a question, or a bug on that, but it seems that whenever I ask on #git whatever is the issue, that was expected git behaviour, and that my problem is MY FUCKING FAULT, and that somewhere I should have rtfm









one of the most important statistical indicators of quality of life for men

Is whether your girlfriend gets on with your mum.

Think about it.

The great Sun Tzu said "If a challenge offers you no gain, you must refuse it", but you can't can you, it's not your choice. ;-)

There is absolutely nothing you can gain from this situation, and its a gift that will keep giving at every family Christmas and Birthday until forever.


So here is an example of someone increasing the total sum of cobblers talked about alzheimers in the guardian


(ok, I changed the title, but I was very very angry about what someone wrote on the internet. ;-)

So this guy, (MA in zoology, career as a political journalist, http://en.wikipedia.org/wiki/George_Monbiot) read an article in the new scientist reviewing a paper indicating a metabolic connection between AD, and he comes to this conclusion...

"Alzheimer's disease could be another catastrophic impact of the junk food industry" *

"We cannot yet state unequivocally that poor diet is a leading cause of Alzheimer's disease, though we can say that the evidence is strong and growing. But if ever there was a case for the precautionary principle, here it is." *

* (quotes taken out of context, in order to make it looks as damning as possible)


The ban on full body swim suits came at a nice time for all those athletes who set records in 2009... ;-)




After seeing a post about male swim suit technologies... I bet you are hesitant to click on the link now... don't worry there are no NSFW imagery of budgie smuggling... I thought it would be interesting to checkout the Swimming world records mentioned.


http://www.washingtonpost.com/wp-srv/special/sports/olympic-swimming-dressed-for-a-world-record/

So for a tldr; It's an animated graphic illustrating the relationship between the introduction (and withdrawal by banning) of advances in the technology.


But its interesting that banning the all body suits in 2010 came at a great time for all the athletes who used them to set world records using them in the 2009. These records are pretty much set in stone for the time being...



Freestyle world records
wikipedia

50 m freestyle  20.91 
César Cielo  Brazil
18 Dec 2009
Brazilian Championships São Paulo, Brazil [5][6]
100 m freestyle  46.91 
César Cielo  Brazil
30 Jul 2009
World Championships Rome, Italy [7][8]
200 m freestyle  1:42.00 
Paul Biedermann  Germany
28 Jul 2009
World Championships Rome, Italy [9][10]
400 m freestyle  3:40.07 
Paul Biedermann  Germany
26 Jul 2009
World Championships Rome, Italy [11][12]

50 m freestyle  23.73 
Britta Steffen  Germany
2 Aug 2009
World Championships Rome, Italy [43][44]
100 m freestyle  52.07 
Britta Steffen  Germany
31 Jul 2009
World Championships Rome, Italy [45][46]
200 m freestyle  1:52.98 
Federica Pellegrini  Italy
29 Jul 2009
World Championships Rome, Italy [47][23]
400 m freestyle  3:59.15 
Federica Pellegrini  Italy
26 Jul 2009
World Championships Rome, Italy [48][49]
-









Other sources on this topic;
http://ivorysofa.blogspot.co.uk/2011/07/end-of-swimsuit-wars.html
(more detail)


Original post;
http://flowingdata.com/2012/07/26/swimsuit-technology-and-breaking-world-records/



Corollary to "If you are not paying for the service, then you are the product being sold"

This is totally unresearched first draft post, so it might contain some gross misunderstandings. I am going to look into the history of that quote and my layman economics and update my notes below at some point.
My highly unoriginal observation on the pithy one-liner:

"If you are not paying for the service, then you are the product being sold"
Is that it follows that it is in the best interests [1] of you  to be a good quality product.

My reasoning follows thusly:

What next for the HIV derived cure for cancer?

In August of 2011, the New England Journal of Medicine published an editorial article "Redirecting T Cells", which introduced an interim report from a Phase 1 clinical trial of an Anti-Cancer treatment derived from the HIV virus on 3 patients with refractory chronic lymphocytic leukemia (CLL).

The results were extremely positive, with 2 of the 3 patients having undetectable levels of cancer after a follow up at 6 months. That is, in a very limited sense, they had been cured of Cancer.

As usual with important scientific matters the best place to go for an informed view on the subject is comedy, and xkcd hit the nail on the head with a perfect mixture of optimism and scepticism in this comic.

http://xkcd.com/938/


From a medical genetics point of view, retroviruses like HIV include a viral enzyme called reverse-transcriptase (RT) of which the function is to incorporate the viral nucleic acid into the genome of the host cell, hence not only does the virus make copies of itself using your own cells, any daughter cells of the original also have copies of the viral genome installed and running.
Very cheeky indeed!

The treatment (nice video from UPenn team here) involves infusing back into patients modified versions of their own T-cells, that have been tinkered with by using the HIV RT enzyme to make the T-cells recognise and attack cancerous cells. However crucially, the modified T-cell genome remains persistent in the host and goes on killing cancer cells beyond the death of original infusion. (see xkcd for a succinct review of the risks and side-effects).


Given the theoretical importance of this type of treatment, I think its worth mentioning how the 3 patient trial was funded:
"The researchers did manage to get a grant from the Alliance for Cancer Gene Therapy, a charity founded by Barbara and Edward Netter after their daughter-in-law died of cancer. The money was enough to finance the trials on the first three patients."

Anyway... so during a conversation about cancer and clinical trials a few days ago I was trying to make the case that despite the cost of phase IV trials being bonkers money, and getting more expensive each year, that there is a productive area for relatively small scale charitable funding which if successful can leverage up to big-pharma funding.

However when I tried to reference a couple of the "facts" that I had used in my claim, it turns out that my memory had served me poorly and I needed to revisit my sources.  It becomes clear why my poor memory is an asset if you read on...

(These mis-remembering fuckups suggest I might do well to enlist on some early stage treatment for memory loss ASAP, or pay more attention the first time... whatever ;-)



Quantitative finance stole all the math grads.

"While they are booming, these industries draw in resources at a phenomenal rate. It is only when they crash, after the bust, that we realise the extent of the overinvestment that occurred. "
BIS Annual Report  
If your chosen industry sector relies on research, and your country has a growing financial sector, then you lose your best brains.
"That is to say, a sector with high R+D intensity located in a country whose financial system is growing rapidly grows between 1.9 and 2.9% a year slower than a sector with low R+D intensity located in a country whose financial system is growing slowly."
BIS Annual Report


I've been saying to anyone who will listen, that one of the things that came out of my dalliances with mathematically orientated acedemia, is that I now have an appreciation of quite how much effort goes into knowing stuff. (if nothing else)

Its also pretty clear that society, and possibly genetics, only produces a limited number of people who are willing to devote years of their lives to learning in depth about mathematics.

This leads to the observation that one of the uncounted costs (at least at the societal level) of the recent financial boom and crash, was that the rewards offered in the financial industry attracted many more of the top math students into quantitative finance, to the detriment of the other mathematically orientated fields.

Hence that the Bank of International settlements published their annual report, in which they quantified (you know, in numbers) the actual cost to society of the brain drain to hedge-funds to real science (that stuff that cures cancer) and surprisingly enough, they agree with me.


Ultimatum & Dictator


At a family dinner on Sunday the subject came up of corrupt officials offering tickets to the olympics and everyone being sooo shocked that these naughty foreigners would do such a thing.
 
But it reminded me of the games that economists (?) devised that can be used as some "objective" test of how participants value fairness, and tolerance for unfairness (i.e. tolerance of corruption) and how this varies between nations, societies and cultures.  

Basically the point of these games is to arrange that random people to play the games in various countries and record the results, the idea being that its simple enough that its meaningful to compare across.

Why developers will rule the world.


I did this post on why sysadmins will just get more powerful, by server and system, until they control the very air that we breathe.

Clearly in response to that actual post Forbes has decided to fight back, on the side of the developers with this article... "The Rise of Developeronomics" and make the case that is in fact the developers who will rule the earth.

Obviously I disagree mightily, but it is a good read none the less for anyone in a particularly technical career path.


http://www.forbes.com/sites/venkateshrao/2011/12/05/the-rise-of-developeronomics/

the value of being hacked.


I am great believer in the idea that experience is the best teacher, and a shocking experience is the most teaching of them all.

Hence I strongly approve of the behaviour and message recounted in these here posts about the hacking of a ATM machine in a public demonstration, and how the company responded.

The security guy from the ATM company blogs about his companies machine getting "jackpotted" in front of 5000 screaming black-hat conference attendees:
http://henryschwarz.blogspot.co.uk/2012/06/black-hatted.html

BGR on the matter:
http://www.bgr.com/2012/06/07/hacked-insider-account-atm-black-hat/

The youtube video of the hacking:
http://youtu.be/Ss_RWctTARU



I am coming round to the view that the Leveson enquiry was a stroke of genius

in a time of endemic stupidly.

Really it has been more successful that I could possibly have hoped, and not just in a criminal justice sense.


I think one of the un-anticipated legislative changes for the future will be vehicle and driving licensing

So below are a bunch of links that have some details about recent-ish autonomous and integrated driving technologies.
 
Each individually is interesting, Volvo working on mesh networks to provide "Linked" vehicles, google is developing autonomous vehicle decision making and route finding, the US Black Box bill makes data collection mandatory, and ubiquitous.

Basically what I am saying, is that TAKEN TOGETHER, all these technologies converge to create a world where in 10 years un-crashable, self driven vehicles are the norm in a crazy futuristic madness. It would be mostly electric, mostly quite boring etc.

But some of the implications are interesting. 

1) How do you steal a car, when all the cars and all their owners know exactly where it is at any time?

2) If its almost impossible to crash, why bother with drink driving laws?

3) Fuck it, might as well let kids drive themselves to school....

123-reg and google reviews, I think someone needs to "Quit making stuff up"

So the 123-reg.co.uk nameservers stopped working for a period today, not totally surprising as they have been pretty poor as a domain host, all things considered.

zero-day exploits, my finest serverfault.com moment (according to the points table)

A question was asked on serverfault about responding to zero-day exploits, and I managed to pitch in with a slightly pontificating answer

However it caused me to hit to 200 a day reputation limit, so it seems it is definitely the sort of thing that chaps on serverfault.com like to read.

So here is my currently highest scoring answer in all its glory...


reverse engineering the placebo effect


I've read about this one a few times before, but this blog chats about some of the studies that suggest various biological pathways involved in producing the placebo effect:
  • Researchers treated a group of people who were expected to show a placebo effect for pain tolerance, except for one important difference: people were also unknowingly administered naloxone. Unlike the first group, the placebo effect on pain tolerance vanished: people did not have a significantly increased pain tolerance.
  • These results suggest that after being “conditioned” with an opiate drug like morphine, people were capable of producing their own natural opiate-like chemicals that bind to some of the same receptors as morphine. These are called the brain’s endogenous opioids, a class that includes well-known natural painkillers like endorphins, which are released for example during exercise.

  • Researchers had known opioids were involved in pain tolerance, but these studies were amongst the first to show they can be involved in the brain’s placebo response to pain.

log-splitting within a virtual-host in apache

If like me, you see nginx as a flash in the pan newcomer (released in 2004) and mod_balancer as entirely everything one would require in a load balancer.

Then you might well have some Apache config files that have grown over the years into vast monstrosities of Alias and ProxyPass, then no-one understands what they all do.

I think the inevitable next step after "cancer management" is cartoonish monstrosities




"First Gene Therapy Successful Against Aging-Associated Decline: Mouse Lifespan Extended Up to 24% With a Single Treatment"
http://www.sciencedaily.com/releases/2012/05/120514204050.htm




So the future is nearly here. Again. (Its getting boring now. Just tell me when its safe to take up smoking, please.)

Software domains/industries with translatable solutions?

Recently I was looking into some ideas for a project based around the idea of collecting body data from customers, and using it to either generate visualizations of fashion, or to make recommendations based on body-type and fit.

After a bit of research (googling), it was pretty obvious that there is a lot of technology and software around the subject of biomechanical modeling for medical and academic purposes, that might be possible to re-purpose for some retail fashion ends.

If Java and RDBS/SQL are like 2 rival empires, then ORM is the UN.

Or maybe Java and SQL/RDMS are like 2 parents in divorce court, and ORM is their fucked up child?

Responsible adult?

I think someone with some spare time should start a "coalition party" which runs campaigns on the basis that both the conservatives and the labour party are demonstrably incompetent and should be at all times chaperoned by a responsible adult. i.e. Menzies Campbell


What's in a photo?

While doing some research for another blog post it struck me that I could quite unfairly slap-in any image that suited my purposes, but that might display the subject in a quite unflattering light and I have almost unbounded (within the limits of copyright and photoshop) scope for naughtiness.
I've included copies of the images under the principle of "Research and Private study" of Section 29 of CDPA 1988 with links-back and acknowledgement in the Notes & References.)


I started writing some ideas for a post about the first episode of a BBC 4  radio programme called "The prisoner and the Bishop" by the Rt Rev James Jones, the Bishop of Liverpool, and as I've decided on a more structured approach for the article than previous efforts, this involved some research.

In order to create some visual impact I'm going to try to incorporate at least one chart or photo. I also wanted to get a sense of whether the reverend was an old style godist, whether he was Jeans and t-shirts down with the kids type, or some other as yet to be discovered species of senior institutional religionist.

So I proceeded to the web (google image search) to see what was available.

google image search (original)

R dot plots for Maternal mortality for WHO nation data 2008

Queens hospital in Romford and King Georges in Goodmayes were recently criticised by the health watchdog for extremely poor standards of care, and this quote from the guardian article suggests a level of nasty, that is beyond simple incompetence:

"The CQC report highlighted verbally abusive staff in the maternity unit. One of the midwives was heard to say to a woman in labour: "Hurry up, or I'll cut you." - http://is.gd/dxVosz
 
Surely this person should be arrested, or did i miss something?

The BBC and Guardian both covered the report from the care commission, and its makes for pretty grim reading given that they are my 2 nearest general hospitals.

The article in the Daily-Mail mentions a statistic that they may well have pulled out of their collective arse (who knows they seem to just lie about everything), that the maternal death rate at Queens hospital is 3 times the national average.

So I thought it would be worth to see graphically how that would look on the global charts if that was representative of the UK as a whole. The WHO data provides an age-standardized rate of 12/100,000 births, hence the chart is plotted with 2 lines at 12 and 36.

This is the maternal mortality dataset currently available on the WHO datasets websites. The 2 lines are first the United Kingdom, and where the United Kingdom would be if Queens hospital in Romford and King Georges in Goodmayes were representative of the UK as a whole.





UK Stats: Deaths during "Pregnancy, childbirth and the puerperium"

I was fairly surprised how badly the UK fares in the Cancer Mortality statistics compared to our continental and Transatlantic cousins. So when I heard that some UK hospitals were reputed to have a very poor comparative record for maternal survival rates, I thought it might be informative to dig out the national numbers, given I had the spread sheet downloaded already.

Basically, the Office of National Statistics collates data from the causes of death as certified by a doctor or coroner and releases the data periodically. Typically there is good quality coverage in the UK, and the data is classified using ICD-10 disease codes for helpful comparative purposes. The last full year was 2010, for the which the data was released in Oct-2011.

So the head-line numbers for this rather unfortunate way to perish are as follows;

Of the 236,732 women whose deaths were registered in 2010 there were 35 for whom the final cause of death was recorded in the category "Pregnancy, childbirth and the puerperium".  (Def. "puerperium" Medline)



The absolute counts of deaths (other age segments are zero):




TotalUnder 11-45-910-1415-1920-2425-2930-3435-4940-44

O00-O99
XV Pregnancy, childbirth and the puerperiumF 35----23121233

Mars should be right there on the list of things to do...

I ended up on the wrong end of an argument where I was forced to defend myself against the accusation that I am "opposed to the principle of expanding human knowledge" or something like that.

I totally think that sending a Person to Mars would be a good thing, also it would be very cool to put on your CV. So it should be right there on the list of things Humans should be doing.

artists impression of "manned" mars mission.


Right there next after "Cure cancer".

That's actually pretty high on the list.

"Cure cancer" is already at number 2.

Behind number 1) "Save all those people who are dying from things - that WE ALREADY KNOW HOW TO CURE or prevent" like malaria, TB, etc etc

WHO IS TO BLAME FOR THIS OUTRAGE? (Network error: XXX)


This is a very basic set of troubleshooting steps for remote TCP services, which include a couple of tests to determine the important thing in these situations...

"WHO IS TO BLAME FOR THIS OUTRAGE?"

So its not working.

What is not working? Just some random TCP service on the network, MySQL-port 3306, MS SQL 1433, tomcat 8080, apache 80, you can check the IANA database of registered port assignments for 57333 million other examples.

(in fact this has run a bit long, so this is just Part 1 - the common error messages, the payoff, ie who to blame has got pushed back to part 2, or 3...)


Money for old drugs

The wikipedia article on drug trial phases which indicates sizes and groups types is an interesting starter on clinical trials, and there are some links to approval and success rates.


This is a table of clinical trial approval rates at each stage:

Stage of Development               Probability of Approval
  • Phase 1                                                       9%
  • Phase 2                                                     15%
  • Phase 3                                                    44%
  • Submission                                            80%
So basically only 9% of candidate drugs make it through the phase 1 trial, so there is massive attrition. my rudimentary calculations make that if you start with 1000 phase 1 trials then 90 make it to phase 2, of those 13.5 make it to phase 3, and then .44 of those.

Hence only ~6 make it to the application for approval to the FDA (this is US data, but its broadly similar over here)


Tools for debugging DNS fuckup-ery

If you have a hang-over, and you are trying to show someone who is not a command-line freak, why their web-site is down for DNS reasons, you can do worse than this on-line graphical tool;
http://www.zonecut.net/dns/

Looks to have been setup by Bjorn Isaksson some time in the ancient past internet terms, i.e. before last Tuesday.



"Whitelisting applications" - I think we already have it.

http://www.channelregister.co.uk/2012/04/24/intelligent_whitelisting/
"The firm is marketing its brand of trust-based application control and whitelisting as a better way of tackling the growing malware menace posed by targeted attacks on security suites from the likes of Symantec and McAfee."



I think we already have it, its;

yum  --disablerepo="*" --enablerepo=fedora --enablerepo=updates list



Respiratory disease in the UK during the Clinton administration?


What is going on in the middle of this chart?