Tuesday 27 August 2013

The Next Google: It's Like Google, But For Search

It is no longer appropriate for search to be under the thumb of private industry. It's a critical part of the national infrastructure. So if I were a real pinko, I'd be advocating for the nationalization of Google, à la Chavez—but I'm not a real pinko. Besides, the American people have already bought and paid for an ideal alternative to Google. That's right: we have the means in hand to create a public, ad-free, totally fair and reasonably transparent search engine with a legal mandate to operate in the public interest, and most of the work is already done. We have also a huge staff of engineers to conclude what little remains on the development and deployment side.
Who are these American heroes, soon to be accepting the thanks of a grateful nation? Why, our fellow citizens, the software engineers and tech gurus and endless numbers of contractors of the NSA! Why don't they make themselves useful and stop spying on everyone and instead, use all that computing power and archived information to make us a fair, fast, ad-free search engine?
They have a copy of the whole Internet, soon to be housed in their giant bunker in Utah!
It is already, apparently, equipped with the latest in search technology! It's probably already better than Google.
Others make their case against Google on antitrust laws. It's not illegal to have a monopoly. According to U.S. courts, it's just not your fault that everybody loves your product! What's illegal is using that power to do bad things, like suppress your own competition. This is why there are ongoing government investigations into Google's anti-competitive business practices in the U.S., in Canada and in Europe.
Probes like these have so far tended to focus on Google's preferential treatment of its own services over those of its competitors in Google search results. Which amounts to ignoring the elephant in the room: Google, with its 67% share of U.S. search traffic (sounds low, tbh), has a potential influence far beyond the industries in which it operates formally. At the moment, Google can legally use its power to make or break any business, or any politician, publication, or public figure it chooses, for any old reason it wants, provided that reason doesn't fall foul of antitrust laws.
For instance, let's suppose one of Google co-founder Sergey Brin's friends were to open a new cafe in Mountain View: there is no legal proscription whatsoever against Google's vaulting the Friends o' Brin Cafe to the top of results on searches for "best cafe Mountain View." Or even "best cafe."
A close reading of Google's ten "Core Principles" appears to suggest, but not quite guarantee, that Google won't simply grant preferential treatment at its own discretion. The fact is, however, that it's entirely up to them. Given the understandably secret nature of Google's algorithms and other techniques for determining search results, it would be impossible to say whether or not this is in fact already happening.
Already companies live or die at the hands of Google. Any update to the Google Panda search ranking algorithm has rippling effects through the Internet. One thing that seems to be the case: older sites, with thousands of internal links and a deep history on the Internet, seem to be constantly downgraded. That's bad news for some non-spam media companies that in part live off search traffic. Google results, in general, overweight newness. It is becoming more and more impossible to find relevant results older than three months.
As well, Google will tell you that active engagement with their product Google+ will be "beneficial" to any publisher as a whole, including in search. Publishers now ignore Google+ at their peril, whether it is relevant to their business or not.
But let's take the real case of 23andMe.com, the "privately-held personal genetics company" whose CEO, Anne Wojcicki, is married to Sergey Brin. According to recent SEC filings, Google invested approximately $1.5 million in the company's Series D round, and Google leases office space to the company. Here's the current results for a search on the phrase "genetic testing":

Friday 9 August 2013

முல்லா குட்டி கதைகள் !

ஒரு தடவை அறவொழுக்கத்தை நேசிக்கும் பிரபலமான தத்துவவாதி ஒருவர் முல்லா வசிக்கும் ஊரை கடந்து செல்ல வேண்டியிருந்தது. அப்போது சாப்பாட்டு நேரமாகையால் அவர் முல்லாவிடம் நல்ல உணவு விடுதி எங்குள்ளது என்று கேட்டார். முல்லா அதற்கு பதில் சொன்னவுடன், தத்துவவாதி போகும் போது பேச ஆள் கிடைத்தால் நல்லது என்ற எண்ணத்தில் முல்லாவையும் தன்னுடன் சாப்பிட வருமாறு அழைத்தார்.
முல்லாவும் நெகிழ்ந்து போய் அந்த படிப்பாளியை அருகிலிருந்த உணவு விடுதிக்கு அழைத்துச் சென்றார். அங்கே போன பிறகு ‘அன்றைய ஸ்பெசல் அயிட்டம் என்ன?’ என்று கடைச் சிப்பந்தியிடம் கேட்டார் முல்லா. ‘மீன்! புதிய மீன்!’ என்று பதில் சொன்னார் சிப்பந்தி. ‘இரண்டு துண்டுகள் நல்லதாக கொண்டு வாருங்கள்’ என இருவரும் ஆர்டர் செய்தனர்.
சிறிது நேரம் கழித்து ஹோட்டல் சிப்பந்தி ஒரு பெரிய தட்டில் இரு மீன் துண்டுகளை வைத்துக் கொண்டு வந்தார். அதில் ஒரு துண்டு பெரியதாகவும், இன்னொரு துண்டு சிறியதாகவும் இருந்தது. அதைக் கண்டவுடன் முல்லா எந்தவொரு தயக்கமில்லாமல் பெரிய மீன் துண்டை எடுத்து தனது தட்டில் போட்டுக் கொண்டார். முல்லாவின் செய்கையால் கடுப்படைந்து போன தத்துவவாதி முல்லாவைப் பார்த்து கடுமையாக முறைத்து விட்டு, ‘முல்லா நீங்கள் நடந்து கொண்ட முறையானது எந்த தர்ம, நீதி, நியாய, மத சாஸ்திரத்துக்கும் ஒத்துவராத ஒன்றாகும்’ என்றார்.
முல்லா, தத்துவவாதி சொல்லுவதையெல்லாம் மிக அமைதியுடன் பொறுமையாக கேட்டுக் கொண்டு வந்தார். கடைசியாக அந்த மெத்தப் படித்தவர் பேசி முடித்தவுடன், “நீங்களாக இருந்தால் என்ன செய்திருப்பீர்கள்?” என்றார் முல்லா. “நான் மனச்சாட்சியுள்ள மனிதனாகையால் சிறு மீன் துண்டை எடுத்திருப்பேன்”. ‘அப்படியா, ரொம்ப நல்லது. இந்தாருங்கள் உங்கள் பங்கு’ என்று சொல்லி சின்ன மீன் துண்டை அந்த தத்துவவாதி தட்டில் வைத்தார் முல்லா.

Tuesday 6 August 2013

Blood test to tells U REALLY need antibiotics

A simple three-minute blood test could tell GPs whether a patient needs antibiotics.
Not only could this help avoid patients suffering nasty side-effects from taking unnecessary drugs, but it could also tackle one of the greatest threats to modern health — antibiotic resistance. The test tells a doctor whether the patient is suffering from a viral or a bacterial infection — that way, they know whether or not to prescribe antibiotics.
Antibiotics are only effective against bacteria — they do not kill viruses. Currently, the type of infection can only be confirmed with a blood test which must analysed in a lab, a process that can take two to three days.
However, GPs say they often give antibiotics as a fail-safe measure, and that patients pressure them for the pills.
According to Public Health England (formerly the Health Protection Agency), if patients ask their GP for an antibiotic, the vast majority will get one.
Over-prescribing has consequences for both the patient and the population. As well as causing side-effects, over-use can lead to bacteria becoming resistant, making antibiotics less effective at fighting infections. The Government’s chief medical officer has described this as one of the greatest threats to modern health.
Over the past five years alone, the number of antibiotic prescriptions has risen by 10 per cent to 41  million prescriptions at a cost of £170 million to the NHS, and a third of all Britons have taken them in the past 12 months.
But now a simple fingerstick test could solve this ‘catastrophic threat’. The test — which involves taking a drop of blood from the finger — can tell doctors within three minutes whether an illness is caused by a bacterial infection which requires antibiotics, or a virus, which does not.
It measures a substance called C-Reactive Protein (CRP) in the blood. The amount of this protein increases when the body is fighting a bacterial infection, but not when it is fighting a virus, which triggers a different immune response.

So a doctor would know that if the CRP level was shown to be low, antibiotics would not be required.
Studies show that providing ‘proof’ that they are unnecessary to patients who demand antibiotics can significantly reduce the number of prescriptions.
One EU-funded study, presented at the World Association of Family Doctors conference in 2010, looked at how respiratory infections which are generally caused by viruses were treated by 600 GPs in six different countries. It found that antibiotic prescriptions fell by 25 per cent when doctors used the CRP test.
British experts say the test could be a useful tool for significantly reducing antibiotic prescriptions.
Dr Nick Francis, a senior clinical research fellow at Cardiff University and expert in antibiotic resistance in primary care, says: ‘Unfortunately, it is very difficult to accurately determine whether an infection is viral or bacterial.

‘Markers such as CRP have evolved to help where there is lingering uncertainty after a clinical assessment or where the patient has strong beliefs that antibiotics are needed.’
However, not everyone needs the test. ‘Patients who appear very unwell should be treated with antibiotics or admitted to hospital without the test because they could develop complications,’ explains Dr Francis.
‘But for those patients where there is doubt, or where the GP feels antibiotics are not needed but the patient is putting pressure on to prescribe them, the test can be helpful.’
The test is currently only available in laboratories in the UK (it can be carried out privately for around £50) because the NHS does not yet fund it in GP practices (the machine to analyse the test would initially cost £1,000 and then £3 per test).
Sid Dajani, a community pharmacist in Southampton and spokesman for the Royal Pharmaceutical Society, says it is likely to be some time before the test is routinely available on the NHS.
‘CRP testing would be a natural extension to the clinical services we offer but it will be two or three years before there is enough evidence for it to be made widely available.’
Other concerns about the test are that the results are not always clear — levels of CRP also increase as a result of inflammation caused by other conditions such as rheumatoid arthritis as well as infections.
Dr Francis explains: ‘The test is only a guide. It does not tell you categorically that this patient has a bacterial infection and this patient does not. It gives you a number that has to be interpreted in light of the other symptoms and the patient’s overall risk profile.
‘For example, you are more likely to treat an elderly person with diabetes with antibiotics than a young healthy person, as the elderly person is more likely to develop complications.’
Experts say the best way to reduce antibiotic prescriptions is to educate doctors and patients about common complaints and when antibiotics are necessary.
Mr Dajani warns: ‘To see this test as the panacea is an extremely dangerous idea. We need to work hard to educate patients about when it is appropriate to take antibiotics and make sure they take them correctly and finish the course.’
Public Health England adds: ‘CRP may be useful in a small range of infections provided the test is robustly quality controlled.
‘But nothing can replace taking a detailed patient history and thorough examination.’