A couple of days ago I had the briefest of looks at Companies House data to see what the extent of Trump’s declared (current) corporate roles are in the UK. Not many, it seems. Of the companies with which Trump has a declared officer interest, the list of co-directed companies in his UK empire seems small:
(u’NITTO WORLD CO., LIMITED’, u’SLC TURNBERRY LIMITED’, 2 common directors)
(u’TRUMP INTERNATIONAL GOLF CLUB SCOTLAND LIMITED’, u’DT CONNECT EUROPE LIMITED’, 3 common directors)
In my code as it currently stands, two directors are the same if they have the same director number according to Companies House records (I think! Need to check… can’t remember if I also added a fuzzy match…).
Unfortunately, Companies House has issues with name authority files (they need to talk to the librarians who’ve been grappling with the question of whether two people with the same, or almost the same, name are actually the same person for ages… “VIAF” is a good keyword to start on…). For example, I strongly suspect these are the same person, given that I found them by mining co-directed companies seeded on two separate Trump companies:
(u’qcmgW-bhHd3TT1MSNuqHIjWBxLI’, 1946, u’TRUMP, Donald J’)
(u’8WlV7G8p1ojhFks_i4ljYwW5WvI’, 1946, u’TRUMP, Donald John’)
(u’65Cc7HAVpXHqcLR_-CczJ80C724′, 1946, u’TRUMP, Donald’)
Or how about:
(u’sj7c-OeX84Ww_JJudaY_D-DZDm4′, 1981, u’TRUMP, Ivanka’)
(u’omdexC3tGVn8JnozQ9ZazJL_MT8′, 1981, u’TRUMP, Ivanka’)
(u’PCrNv-j3ABqrisHsT_PKL3yAlc0′, 1981, u’TRUMP, Ivanka’)
FWIW, Companies House seem to be increasingly of the opinion that month and year discriminators on birthday are plenty, and day doesn’t need to be publicly shared any more (if, indeed, it will still be collected). Occasional name/month/year collisions aside, this may be true (if you’re happy to accept the collisions). But until they sort their authority files out, and use a common director ID (reconcilable to a Person of Significant Control identifier from the PSC register) for the same person, they should be providing as much info as possible to help the rest of us reconcile director identifiers from their inconsistent data.
PS I started to doubt myself that Companies House at least attempts to use the same identifier for the same person, but here’s another example that I’m pretty sure refer to the same person… – note the first result associates 35 appointments with the name:
If you click the top link, you’ll see the appointment dates to the various companies are different, so it’s presumably not as if the commonality arises from the appointments all being declared on the same form. I’m not sure how Companies House reconciles directors, actually? Anyone know (let me know via the comments if you do…). For now, I assume it to be something like a (case insensitive?) exact string match on name, birthdate, and maybe correspondence address (or at least, a recognisable part of it)?
The following records, this time from Formula One co-directed companies, presumably relate to the same person (an accountant…):
(u’keWSNSl6V3Zg2FNV7vPy6BBVPVw’, 1968, u’LLOWARCH, Duncan Francis’)
(u’dzIMC8ot_A9rJThNdKQ5yQC-M3Y’, 1968, u’LLOWARCH, Duncan Francis’)
(u’m5FeeEsclwF0s57UkL2NcB6MIBk’, None, u’LLOWARCH, Duncan’)
(u’S9zuBVuv1LXtbR62_r-x9RzJzRE’, None, u’LLOWARCH, Duncan’)
(u’BTfAza-kduWKPnuUYPDd3w2i9fc’, None, u’LLOWARCH, Duncan’)
(u’3e8laCMUijwG6FdTnqGcDqMsXr4′, None, u’LLOWARCH, Duncan’)
(u’1Qgz-VCSMqjZZgyaibcvBAyGKUU’, None, u’LLOWARCH, Duncan’)
Full of cold and stuck in the biggest rut going – http://xkcd.com/1768/ hits the spot exactly – I stumble across a post from last year, Geographical Rights Management, Mesh based Surveillance, Trickle-Down and Over-Reach, one of the increasingly many dystopian, were it not real, posts on this blog describing stuff that no-one cares about.
I mentally link it to Amazon Go, Amazon’s soon to be opened concept shop where you swipe in, take what you want, and just leave, presumably passing security cameras and a security guards, as your phone automatically picks up the automatically generated bill on your way out, and which just makes me feel cold in a different sense.
(I’m one of the neo-Luddites who refuses to use self-scan tills in supermarkets and self-pay pumps at petrol stations (cos Are Robots Threatening Jobs or Are We Taking Them Ourselves Through Self-Service Automation?).)
In a further round of consolidation, I have a quick peek around what other news I may have missed over the last year or so, seeing who else might know I’ve popped into the physical Amazon store: Google, perhaps (see SearchEngineLand on Google Launches “Store Visits” Metric In AdWords, To Help Prove Online-To-Offline Impact or Under The Hood: How Google AdWords Measures Store Visits, for example), or Facebook (Facebook’s new ads will track which stores you visit).
I also happen across a seriously f*****d up piece of shop furniture, the Skinny Mirror, a fitting room fitting that makes you look thinner than you are, so you feel better and buy whatever it is you’re trying on… (well, not you trying it on, obviously, some weirdly distorted f*****d up re-presentation of yourself). I imagine folk will then grab a selfie using something like the updated version of Facetune, an app that lets you photoshop, (verb), a live preview of yourself before you actually take the photo.
And if they walk out of the not Amazon store without paying, they’ll maybe try to explain it away with “I’ve got the app, so I thought I could just go…”.
PS I really need to put a distorted reality tag on a chunk of stuff on the Digital Worlds blog…
PPS ish via @kpfssport, I note a recent report from the University of Leicester that suggests that Mobile Scan and Pay Technology could promote supermarket theft. See also a review of a pay-to-read Australian study (Emmeline Taylor, Supermarket self-checkouts and retail theft: The curious case of the SWIPERS): Are supermarket self-checkouts turning shoppers into swipers?.
I’ve always found Amazon’s AWS tools really fiddly to use – settings all over the place, the all too easy possibility of putting things into the wrong zone and then forgetting about them/having to try to track them down as you get billed for them, etc etc – but that’s partly the way of self-service, I guess.
Anyway, last week, amongst a slew of other announcements (AI services, new hardware platforms that include FPGAs), Amazon announced a range of developer/devops productivity tools that shows they’re now looking at supporting workflows as well as just providing raw services.
Here’s a quick summary of the ones I spotted:
- AWS Batch: run batch jobs on AWS;
- AWS CodeBuild: “a managed build service” that will “build[s] in a fresh, isolated, container-based environment”, incorporating:
- Source Repository – Source code location (AWS CodeCommit repository, GitHub repository, or S3 bucket).
- Build Environment – Language / runtime environment (Android, Java, Python, Ruby, Go, Node.js, or Docker).
- IAM Role – Grants CodeBuild permission to access to specific AWS services and resources.
- Build Spec – Series of build commands, in YAML form.
- Compute Type – Amount of memory and compute power required (up to 15 GB of memory and 8 vCPUs).
- Amazon X-Ray: a debug tool that allows you track things across multiple connected Amazon services. Apparently, Amazon X-Ray provides:
… follow-the-thread tracing by adding an HTTP header (including a unique ID) to requests that do not already have one, and passing the header along to additional tiers of request handlers. The data collected at each point is called a segment, and is stored as a chunk of JSON data. A segment represents a unit of work, and includes request and response timing, along with optional sub-segments that represent smaller work units (down to lines of code, if you supply the proper instrumentation). A statistically meaningful sample of the segments are routed to X-Ray (a daemon process handles this on EC2 instances and inside of containers) where it is assembled into traces (groups of segments that share a common ID). The traces are segments are further processed to create service graphs that visually depict the relationship of services to each other.
- AWS Shield: a tool that protects your service against DDoS attacks. In waggish mood, @daveyp suggested that many DDoS attacks he’s aware of come from AWS IP addresses. This feels a bit like a twist on an operating system vendor also selling security software to make up for security deficiencies in their base O/S? That said, “AWS Shield Standard is available to all AWS customers at no extra cost” and seems to be applied in basic mode automatically. Security essentials, then?!
Amazon are also starting to offer segmented alert targeting services for your mobile apps with Amazon Pinpoint. The service lets you “define target segments from a variety of different data sources” and more:
You can identify target segments from app user data collected in Pinpoint. You can build custom target segments from user data collected in other AWS services such as Amazon S3 and Amazon Redshift, and import target user segments from third party sources such as Salesforce via S3.
Once you define your segments, Pinpoint lets you send targeted notifications with personalized messages to each user in the campaign based on custom attributes such as game level, favorite team, and news preferences for example. Amazon Pinpoint can send push notifications immediately, at a time you define, or as a recurring campaign. By scheduling campaigns, you can optimize the push notifications to be delivered at a specific time across multiple time zones. For your marketing campaigns Pinpoint supports Rich Notifications to enable you to send images as part of your campaigns. We also support silent or data notifications which allow you to control app behavior and app config on the background.
Once your campaign is running, Amazon Pinpoint provides metrics to track the impact of your campaign, including the number of notifications received, number of times the app was opened as a result of the campaign, time of app open, push notification opt-out rate, and revenue generated from campaigns.
One thing I didn’t spot were any announcements about significant moves into “digital manufacturing” and 3D print-on-demand (something I wondered about some time ago: Amazon “Edge Services” – Digital Manufacturing).
They do seem to be moving into surveilled, auto-checkout, real-world shopping though… Amazon Go.
I’ve been thinking about all those terms and conditions that the big web corps use to justify doing what they want with the data they collect about our actions. And also the way that Facebook, particularly, does abusive stuff and then just apologises, says sorry, it won’t happen again…
From the UK Serious Crime Act 2015, c. 9, Part 5, s. 76:
76 Controlling or coercive behaviour in an intimate or family relationship
(1) A person (A) commits an offence if—
(a) A repeatedly or continuously engages in behaviour towards another person (B) that is controlling or coercive,
(b) at the time of the behaviour, A and B are personally connected,
(c) the behaviour has a serious effect on B, and
(d) A knows or ought to know that the behaviour will have a serious effect on B.
(2) A and B are “personally connected” if—
(a) A is in an intimate personal relationship with B, or
(b) A and B live together and—
(i) they are members of the same family, or
(ii) they have previously been in an intimate personal relationship with each other.
(3) But A does not commit an offence under this section if at the time of the behaviour in question—
(a) A has responsibility for B, for the purposes of Part 1 of the Children and Young Persons Act 1933 (see section 17 of that Act), and
(b) B is under 16.
(4) A’s behaviour has a “serious effect” on B if—
(a) it causes B to fear, on at least two occasions, that violence will be used against B, or
(b) it causes B serious alarm or distress which has a substantial adverse effect on B’s usual day-to-day activities.
(5) For the purposes of subsection (1)(d) A “ought to know” that which a reasonable person in possession of the same information would know.
(6) For the purposes of subsection (2)(b)(i) A and B are members of the same family if—
(a) they are, or have been, married to each other;
(b) they are, or have been, civil partners of each other;
(c) they are relatives;
(d) they have agreed to marry one another (whether or not the agreement has been terminated);
(e) they have entered into a civil partnership agreement (whether or not the agreement has been terminated);
(f) they are both parents of the same child;
(g) they have, or have had, parental responsibility for the same child.
(7) In subsection (6)—
“civil partnership agreement” has the meaning given by section 73 of the Civil Partnership Act 2004;
“child” means a person under the age of 18 years;
“parental responsibility” has the same meaning as in the Children Act 1989;
“relative” has the meaning given by section 63(1) of the Family Law Act 1996.
(8) In proceedings for an offence under this section it is a defence for A to show that—
(a) in engaging in the behaviour in question, A believed that he or she was acting in B’s best interests, and
(b) the behaviour was in all the circumstances reasonable.
(9) A is to be taken to have shown the facts mentioned in subsection (8) if—
(a) sufficient evidence of the facts is adduced to raise an issue with respect to them, and
(b) the contrary is not proved beyond reasonable doubt.
(10) The defence in subsection (8) is not available to A in relation to behaviour that causes B to fear that violence will be used against B.
(11) A person guilty of an offence under this section is liable—
(a) on conviction on indictment, to imprisonment for a term not exceeding five years, or a fine, or both;
(b) on summary conviction, to imprisonment for a term not exceeding 12 months, or a fine, or both.
To what extent could the sorts of thing that recommendation services do, (recommendation services that model a great deal about us), start to appear coercive? Can the asymmetric (power) relationship we are in with this services be defined as “intimate”?
PS by the by, I’ve started looking at laws again that might be used as the basis of “robot laws” (laws relating to slavery, animal rights, accessibility, limits on behaviour as a result of mental (in)capacity etc) and also started trying to note the laws that companies use to weasel their way out of various corporate responsibilities. Things like the Innocent publication defence in The Business Protection from Misleading Marketing Regulations 2008, for example; how easy is it to look up whether Google or Facebook have availed themselves of this sort of defence, I wonder?
According to the Encyclopedia Britannica (online edition) there are several sorts of electronic countermeasure used against opponents’ radar:
Electronic countermeasures (electronic warfare)
The purpose of hostile electronic countermeasures (ECM) is to degrade the effectiveness of military radar deliberately. ECM can consist of (1) noise jamming that enters the receiver via the antenna and increases the noise level at the input of the receiver, (2) false target generation, or repeater jamming, by which hostile jammers introduce additional signals into the radar receiver in an attempt to confuse the receiver into thinking that they are real target echoes, (3) chaff, which is an artificial cloud consisting of a large number of tiny metallic reflecting strips that create strong echoes over a large area to mask the presence of real target echoes or to create confusion, and (4) decoys, which are small, inexpensive air vehicles or other objects designed to appear to the radar as if they are real targets. Military radars are also subject to direct attack by conventional weapons or by antiradiation missiles (ARMs) that use radar transmissions to find the target and home in on it. A measure of the effectiveness of military radar is the large sums of money spent on electronic warfare measures, ARMs, and low-cross-section (stealth) aircraft.
These are worth bearing in mind when using Twitter and other social media, as well as keyword driven news search alerts, as your own, personal news radar. In this analogy, the things I want to detect are “true” news stories (whatever that means…); here are some countermeasures you could take to try to prevent high quality news signals, or news signals that inform me about the things you are doing that you don’t want me to know about, or that you need to spin because they paint you in an unfavourable light, getting through to me:
- noise jamming: pollute my feed with noise that makes me filter out certain forms of traffic (your noise) and, as a side effect, legitimate news; reference me in e.g. tweets and swamp my mentions feed with noise; if I’ve subscribed to one of the accounts you control, feed that stream with random retweets, auto-generated rubbish, etc;
- false target generation: try to get me to subscribe to an account you control, thinking it’s a legitimate news source;
- chaff: chaff masks your current “location”, or a story about you; if I make a search or want to follow a particular topic, try to make sure all I can ever find are empty pages that attract those search terms, or your spin on the story;
- decoys: push out your own news story or, even better, a ridiculous claim that gets widely reshared and that pulls interest away form a legitimate story breaking about you; if I’m only going to read one thing about you today, better it’s the one you put out rather than the one that shows you for what you are…
(If you can think of better examples, please share them in the comments; this was just a quick coffee break post… didn’t really try to think the examples through…)
Remember, folks, this is information war… We should all be reading up on psyops too…
Way back when, companies such as Amazon and Google realised that they could leverage the large amounts of computing infrastructure developed to support their own operations by selling their spare compute and memory capacity as self-service resources.
The engineering effort used to guarantee the high service quality levels for their core businesses could be sold on to startups, and established companies alike, who did not have the engineering expertise to develop and run their own scalable, and resilient, cloud services. (You’d know if Amazon Web Services (AWS) went down completely: so would large parts of the web that are hosted there.)
In the last couple of years, the likes of Google, Amazon and IBM have moved up a level, and now offer “commodity AI” services – recognising faces and and objects in photographs, performing entity extraction on the contents of large texts, generating speech from text and text from speech, and so on. (Facebook seems to prefer to remain inward looking.)
In a spate of announcements today, Amazon joined the part with the release of their own AI services, reviewed in a post by Amazon CTO, Werner Vogels, Bringing the Magic of Amazon AI and Alexa to Apps on AWS. (I’ll post my own summary review when I’ve had a chance to play with them…)
But it seems that AWS have been shopping too. As well as providing a range of different server sizes and base operating systems, the machine instances that Amazon provides now includes FPGAs (Field Programmable Gate Arrays; which is to say, programmable chips…) and (soon) GPUs.
The FPGA machine instance, the suitably named F1 includes one to eight [Xilinx UltraScale+ VU9P?] FPGAs dedicated to the instance, isolated for use in multi-tenant environments. to support the development the machine instance also incudes
a 2.3GHz Intel Broadwell E5 2686 v4 processors, up to 976 GiB of memory and up to 4 TB of NVMe SSD storage. So that looks alright, then… Gulp. (For more, see the product announcement, Developer Preview – EC2 Instances (F1) with Programmable Hardware.)
The pre-announcement for the GPU instances (In the Works – Amazon EC2 Elastic GPUs), which have been a long time coming, look set to offer Windows support for Open GL, followed by support for other versions of OpenGL, DirectX and Vulkan. This means you’ll be able to render and stream your own 3D models, at scale. (Anyone think this may be gearing up to support AR and VR apps, as well as online streaming games? Or support for GPU crunched Deep Learning/AI models?)
(All the new machine instance offerings are described in the summary announcement post, EC2 Instance Type Update – T2, R4, F1, Elastic GPUs, I3, C5</a.)
As well as offering more physical machine types, Amazon have also upgraded their Aurora relational database product so that it is now compliant with PostgreSQL as well as MySQL (Amazon Aurora Update – PostgreSQL Compatibility).
But it doesn’t stop there. For the consumer, just wanting to run their oiwn web hosted instance of WordPress, Amazon virtual personal servers are now available: Amazon Lightsail – The Power of AWS, the Simplicity of a VPS (though it looks a bit pricey compared to something like Reclaim Hosting…)
Back to the big commercial users, another of the benefits of using Amazon Web Services, whose resources far exceed the capacity of all but the largest technology operating companies, is that you can avail yourself of the large amounts of computing resource that might be required to analyse and process large datasets. Very large datasets. Huge datasets, in fact. Datasets so huge that you need a freight container to ship the data to Amazon because you’re unlikely to have the bandwidth to get it there via any other means. Freight containers like AWS Snowmobile (H/T Les Carr for the pointer).
According to the FAQ, each Snowmobile is a secure data truck with up to 100PB storage capacity in a 45-foot long High Cube tamper-resistant, water-resistent, temperature controlled and GPS-tracked shipping container. On arrival at your datacentre, it needs a 350KW power supply (Amazon can supply a generator, if required). Physical access to your datacentre is achieved using the supplied removable connector rack (up to two kilometers of networking cable are provided too).
Once you have completed the data transfer using your local data connect, the Snowmobile is returned to a designated AWS region datacentre. It’s not clear how the data is then uploaded – maybe they just wheel the container into a spare bay and hook it up?
This is all starting to get really silly now…
With a media who failed to hold jokers to account when they had their chance, preferring “balanced” reporting that biases news reports and gives equal measure to unequally validated ideas, and social media opting for truthiness rather than fact to generate momentum for spreading (fake) news, it seems we’re told by commentators we’re now in a “post-truth”/”post-factual” world.
As the OED define it, truthiness is The quality of seeming or being felt to be true, even if not necessarily true.
Although the definition could be debated…
A few years ago, at the dawn of the age of Big Data, the idea that segmenting and modelling large datasets in a “theory-free” way (Big data and the end of theory?) perhaps gave an inkling that truthiness was on its way in, big time. (Compare this also with anti-expert rhetoric over the last couple of years. I’m all for slamming certain classes of academic outlook and activity, but I also think there are reasons for trusting certain sorts of claims more than others…)
The fact that data processing algorithms are likely to have ever increasing power of what we read – not only in terms of selecting which stories to show us in our personalised news feeds, but also because other machines may themselves have written the stories we’re reading – means that we need to start getting a feel for what sorts of biases are likely to be baked into these algorithms.
In contrast to earlier generation of rile based expert systems that could be asked to “explain” their reasoning, today’s systems are often black box statistical machines. Whereas rule based systems used logical reasoning to come up with answers, Deep Learning algorithms and their ilk have gut reactions: rule based expert systems reasoned towards a truth associated with the logical statements asserted into them in an explainable way; black boxes have gut reactions and deal in truthiness.
But whereas we might be suspicious about a person making a truthy claim (“that doesn’t sound quite right to me…”) once we start to trust machine – because they appear to be right-ish, most of the time – we start to over-trust them. I think – I haven’t checked. Sounds truthy to me…
So with a tech news report doing the rounds at the moment that a “Neural Network Learns to Identify Criminals by Their Faces”, it seems that the paper authors “have demonstrated that via supervised machine learning, data-driven face classifiers are able to make reliable inference on criminality” as well as identifying “a law of normality for faces of noncriminals. After controlled for race, gender and age, the general law-biding public have facial appearances that vary in a significantly lesser degree than criminals”. (It’s not hard to imagine this being used a ranking factor for something…) The (best) false positive rate looked on one of the charts (figure 4 in the paper) to be around 6%. Are the decisions “true”, then, or just “truthy”? What level of false positivity makes the difference? (Bear in mind behaviourist training – partial reinforcement can be really powerful…) I also wonder if the researchers ran the same training schedule against IQ? Or etc etc
(In passing, another recent preprint report on arXiv – Lip Reading Sentences in the Wild reports on an automated lip reading system trained on several hours of people talking on BBC television (the UK based researchers were license fee payers, I suspect, but the Google Deepmind sponsor..?!) (If you’d rather read a pop sci write up, New Scientist has one here: Google’s DeepMind AI can lip-read TV shows better than a pro.) For reference, the best word error rate the researchers report is 3.3%. So are the outputs true or truthy?)
So… I’m wondering… algorithmic truthiness: the extent to which the outputs of an algorithm feel as if they could be true, even if not necessarily true. … a useful conceit, or not?
Or maybe we need an alt definition, such as “The extent to which you believe the output of an algorithm to be true rather than what you know to be true”?!