Race to the top: Better data improves everything

IMG_20180429_082212 for blog

By Matthew Fraser, Socitm technical consultant

At the end of April, I completed Etape Loch Ness, my first long distance cycle event. It was 66 miles (106km) on closed roads, around one of the most beautiful parts of Scotland.

Since then I’ve had the same conversation with most of my friends. First, they ask: ‘How did it go?’

I tell them it went well; that the weather was good – albeit a little cold to start – and the whole event was organised well.

When they ask: ‘How long did it take?’ I can give a very precise answer. Thanks to a timing chip provided by the event, I know It took me 3 hours 59 minutes and 31 seconds.

For those who were only being polite, these two questions are sufficient. But others – mostly those I’m cultivating for future cycling events – follow up with: ‘Is that good?’

Now, that is a really tricky question to answer. After all, what constitutes a ‘good’ time? I know that the event organisers weren’t packing up when I crossed the finish line, so that is a helpful indicator – although perhaps that only means that they expected some to be hopeless.

However, I do have some experience objectively determining what ‘good’ should be. In my work with the Socitm Improve service, I’ve spent the last year developing online dashboards that allow organisations to easily compare how ‘good’ they are. So, it made sense to apply similar methods to my finish time.

(I should disclose that, due to my exertions, I lacked the physical capability to get off the couch at this point, so doing anything more productive was impossible.)

Fortunately, I have two friends (we’ll call them Dave and Dan – because those are their names) who also completed the event. If I knew their times, it would tell me if I was ‘good’.

Dave finished in 4 hours 12 minutes. I tell him that is a good time, while inwardly congratulating myself that I’ve beaten him by a massive 13 minutes. Dan’s response provides another emotion, he completed the course in 2 hours 56 minutes!

I was instantly deflated, perhaps Dave and I were amongst the slowest on the day. But then I remembered my work with Improve: we regularly stress that you can’t base your assumptions on just asking a few neighbouring organisations.

Despite being friends living in the same town, Dave, Dan and I are quite different. Dave is now over 50 and did little training; but he’s a former Royal Marine with an active job. Dan has a desk job but is in his early 30s and took his training very seriously. I’m somewhere in the middle.

Whether you are comparing ICT services or athletic prowess, more data provides a better conclusion. Turning to the Etape website, I found I could view all the final times. Here was the data I needed.

The website also had results for the last four years. However, unlike Improve – where you can now compare over a decade of data –  it wasn’t possible to combine the different years together to build a massive dataset.

This year’s results revealed that out of a field of 4,341, I finished in 1401stplace. After my earlier despair, this was pleasing news. I’d just squeezed into the top third.

Of course, my overall position only tells part of the story. I don’t know how many of my fellow cyclists were slowed by punctures, or the desire to take lots of the pictures (I only took the one above).

We overcome this difficulty in Improve by having regular workshops where participants can openly discuss the reasons behind their relative successes and failures. But I think organising something similar for 4,000 cyclists might seem a little excessive (and involve getting off the couch).

Another comparison that we encourage and facilitate with Improve is to compare against yourself. If you’ve taken part in Socitm’s previous benchmarking services, you can quickly see where you are enhancing your service.

Applying a similar approach, I looked over my past data (fortunately, I record all my cycling data using an app on my phone). It reveals that the Etape was my longest ride in both distance and duration. It was also one of my fastest.

After considering all the above, I now have an answer to: ‘Is that good?’

My time was ‘reasonably good’. It’s possible to go faster but I didn’t embarrass myself. I shouldn’t rest on my laurels, but neither should I throw out the bike. Thanks to all this research, if friends question me further I now have the figures to justify my answer.

***

If you were asked if your ICT service was ‘good’, what would you say?  Do you have evidence to back up any feelings that you are already excelling, or could do with better funding?

Socitm’s Improve service can help you compare the size, performance and cost-efficiency of your ICT – helping you to concentrate your efforts for improvement or make a case for improved funding.

 

 

Race to the top: Better data improves everything

Friday roundup: A week in tech

CPUs

CPU-problem-searchers have been busily searching for problems with CPUs – evidently very busily, as they’re found eight brand new security flaws.

Following the Spectre and Meltdown bugs discovered in Intel chips back in January, eager error buffs have been on the hunt for similar gaping issues – and they’ve been duly rewarded in their errand.

According to interestingly-named German tech mag c’t, the fresh flaws are collectively known as the ‘Spectre Next Generation’, and new patches are currently being developed to shore them up.

Though it has all the juicy details, the mag is, admirably, keeping them secret until chip manufacturers have had a chance to sort the security issues out.

And this is no small gesture, for c’t reckons that one of the ‘SPECTRE NG’ flaws eases attacks across system boundaries to such a degree that ‘we estimate the threat potential to be significantly higher than with Spectre. Specifically, an attacker could launch exploit code in a virtual machine (VM) and attack the host system from there – the server of a cloud hoster, for example.

The whole troubling report’s here (luckily in English).

***

Languid communications outfit BT has announced plans to cut 13,000 jobs, as it aims to slash costs by £1.5 billion.

The gigantic operation, which likes to do things at its own pace, will make around 12% of its staff, mostly managers and back-office workers, redundant over the next three years.

However, those BT customers thinking ‘great, this means it’ll take them even longer to answer the phone’ should be soothed by the firm’s plans to hire 6,000 new people to ‘support network deployment and customer service’ – and that, readers, very much remains to be seen.

A third of the job losses will be made abroad in the company’s Global Services outfit (‘ou est le ingénieur téléphonique?!’), in the wake of a forecast predicting a 2% revenue drop over 2018-19.

Understandably, the national secretary of the Prospect union, Philippa Childs, has said the plans are ‘a devastating blow to managers and professionals represented by Prospect’.

BT CEO Gavin Paterson, who, incidentally, took home £1,057,000 in bonuses in 2015/16 on top of his £969,000 salary, said: ‘We have the UK’s leading fixed and mobile access networks, a portfolio of strong and well segmented brands, and close strategic partnerships.

‘This position of strength will enable us to build on the disciplined delivery and risk reduction of the last financial year, a period in which we delivered overall in line with our financial and operational commitments whilst addressing many uncertainties.’

***

Bad news, expensive-illegal-powder-of-unknown-provenance fans: US researchers claim they have built a cheap chip that can detect cocaine.

The team behind the breakthrough reckons a portable cocaine breathalyser might not be too far away, so users of the drug may soon have to toe the line.

According to Joshua Harris from UK road safety charity Brake, drug-driving played a part in 81 fatal road accidents in 2016. He said: ‘These findings have the potential to improve the speed and accuracy of roadside drug testing.

‘We are calling upon the government to prioritise the type-approval of roadside screening devices that can detect all banned drugs and step up roads policing levels to deter offending.’

So, if you currently enjoy giving £80 to a cagey stranger on a street corner for the pleasure of nasally consuming an obscure and completely unregulated substance before going for a little drive, it may be time to seek a new form of entertainment.

***

The UK’s communications watchdog has finally revealed some good news about the country’s embarrassingly listless home broadband services.

According to OFCOM, download speeds have climbed by 28% since the beginning of 2018 to a national average of 46.2Mbps (which certainly isn’t anywhere near true round my house but bully for everyone else).

Unsurprisingly, rural internet consumers continue to suffer lower speeds than urbanites – but, then, country folk aren’t being slowly poisoned to death by clouds of toxic exhaust fumes, so maybe that evens things out a bit?

In areas with lots of buildings in them (towns etc.), 59% of connections are considered by OFCOM to be ‘superfast’ – that is, they deliver average speeds of over 30Mbps during the hours of 8-10pm – while in areas with lots of fields in them, only 23% of connections can truly be adorned with that sobriquet

On a national level, England leads the pack with average speeds of 47.8Mbps, while Wales lags at the back of the pack with 33.4Mbps.

And that concludes today’s average broadband speeds update. I very much hope you enjoyed it.

Friday roundup: A week in tech

Swinney highlights Scottish digital accelerator

John Swinney

By SA Mathieson, editor of Socitm In Our View magazine

A digital accelerator is helping Scotland to match public sector problems with entrepreneurial answers, according to deputy first minister John Swinney.

Speaking at Socitm’s President’s Conference in Glasgow on 9 May, Swinney highlighted the CivTech digital accelerator project as a key way Scotland is using technology to improve government.

‘What started as a pilot programme in the Scottish Government has now grown into a flagship innovation programme,’ he told the event. ‘The CivTech model exposes policy-makers to cutting-edge approaches towards innovation and puts young tech entrepreneurs on the public sector’s radar.’

CivTech pays entrepreneurs to pitch solutions to public sector problems, as well as providing access to senior staff and training. Speaking later, CivTech’s head Alexander Holt told the conference that of the nine businesses that took part in the scheme in 2016, eight are still in operation, having created the equivalent of 30 full-time jobs. They have worked with a range of public sector organisations, including Scottish councils.

Holt said that a standard procurement exercise can take 30 months to deliver results, and relies on organisations knowing what to ask for. ‘How can we procure what we don’t know exists?’ he asked, adding that procurement legislation is more flexible than is commonly thought.

Theo Blackwell, London’s chief digital officer, told the event that the UK capital hopes to improve innovation and collaboration through a new Smart London plan, due this summer.

As well as innovation, the conference also examined citizen and community engagement, including a presentation on how Sweden benefits from its high level of trust in government; ethical uses of automation such as artificial intelligence, with Maryvonne Hassall of Aylesbury Vale District Council discussing that council’s work; and cybersecurity, with Professor Bill Buchanan of Edinburgh Napier University discussing how technologies including smart contracts could revolutionise public services.

Swinney highlights Scottish digital accelerator

Friday roundup: A week in tech

Closed

It’s a sad week for Cambridge Analytica enthusiasts: the morally whiffy company has announced it’s shutting down – dissolving phantom-like into the mists of time and hazy legality, if you will.

The firm – which has become a byword for unscrupulous internet mischief following revelations that it reaped the personal data of millions of people with the aim of influencing various elections – has reasoned that it cannot go on, as a consequence of the hated mainstream media and its depraved obsession with printing factual accounts of wrongdoing.

According to this Guardian story from which I’ve harvested many of the details, CA is starting insolvency proceedings, and in a statement said: ‘Despite Cambridge Analytica’s unwavering confidence that its employees have acted ethically and lawfully, the siege of media coverage has driven away virtually all of the company’s customers and suppliers.’ Ha, right!

But dry those tears, gigantic dodgy data operation fans, for, according to this other story I’ve been harvesting, CA’s sodden phoenix is set to rise from its ignominious heap of binary ashes as new outfit Emerdata.

And we surely have little reason to doubt that the fresh enterprise will take an entirely different approach, employing irreproachably ethical standards and a Windolene-polished transparency in all of its dealings.

According to former CA employee turned grass Christopher Wylie, the data taken from Facebook was used to influence Donald Trump’s election as US president and the outcome of the UK’s European referendum. Hurray for democracy! What a time to be alive!

***

Twitter have advised their 336 million users to shut their accounts and go out into the world and try to enjoy their short lives instead of getting into fiery, pointless arguments with hot-headed strangers on the network.

No, they haven’t. What they’ve actually done is urged everyone to change their passwords, in the wake of a security cockup.

Due to a bug, users’ passwords were stored ‘unmasked in an internal log’ – though, the firm has said, there is no evidence of any breaches.

In a surely unintentionally laughable statement, Parag Agrawal, Twitter’s CTO, said: ‘We are very sorry this happened. We recognise and appreciate the trust you place in us, and are committed to earning that trust every day [my italics].’

By now, you should be able to insert your own jokes here – unless you’re a Russian bot, of course; in which case, just retweet ‘Hillary practices #witchcraft, vote Trump. EU evil, back #Brexit’ and so on.

***

Broadband providers will soon have to take more care in how they advertise their products, after new rules come in later this month.

Using a practice called ‘lying’, ISPs regularly make bold, fabricated claims about the broadband speeds they are able to provide.

However, from 23 May, the Advertising Standards Authority will insist that providers publish the median speed of the connection they are able to provide between 8-10pm, the time of day when most people are using the internet to watch TV, check bank accounts, find love, write complaining emails to their hopeless ISP, etc.

Furthermore, the frankly ridiculous and mendacity-saturated ‘speeds of up to’ line will be forbidden – a solid staple in the deceptive world of broadband advertising.

Should anyone produce an advert containing misleading codswallop, the ASA will now ban it.

Speaking to the BBC, an ASA spokesman said: ‘If a broadband provider is making speed claims in its ads post-23 May, then it will have to hold the evidence to prove the basis on which that claim is made. Numerical speed claims in broadband ads should be based on the download speed available to at least 50% of customers at peak time and described in ads as “average”.’

Friday roundup: A week in tech

Antivirus: NHS staff set for AI and robot training

Robots.jpg

Thousands of NHS staff could soon be trained to work with artificial intelligence (AI) and robots, as a review looking into how health professionals can best use new technologies begins.

Alarmists concerned that this news heralds the health service’s first step towards providing MOTs and oil changes for our future cyborg masters can relax – the assessment will actually look at what sorts of skillsets the likes of AI will demand of staff in the future.

A man from America called Dr Eric Topol is leading the review, which is set to start at London’s Moorfields Eye Hospital, where machine-learning is currently being scrutinised for its usefulness in analysing eye scans.

Dr Topol, a cardiologist, said: ‘While it’s hard to predict the future, we know artificial intelligence, digital medicine, and genomics will have an enormous impact for improving the efficiency and precision in healthcare. Our review will focus on the extraordinary opportunities to leverage these technologies for the healthcare workforce and power a sustainable and vibrant NHS.’

Our currently human yet figuratively robotic health minister Jeremy Hunt, meanwhile, outputted the following typically insincere mechanical gibberish: ‘These [technologies] give us a glimpse of what the future of the whole NHS could be, which is why in the year of the NHS’s 70th birthday I want to empower staff to offer patients modern healthcare more widely and more quickly.’

Antivirus: NHS staff set for AI and robot training