Hey! Don't look at the Model 3 delays, look at THIS!!!
This is definitely not a distraction, oh no....
254 posts • joined 9 Jul 2009
This is definitely not a distraction, oh no....
As a glorified kitchen radio. the Dot does an excellent job. Plays radio, sets timers, answers questions all without having to dry your hands. Perfect. Does it constantly send recordings back to Amazon? Demonstrably not, and even if it did they would be the most ineffective spying yet encountered.
Does it warrant the paranoia and naysaying here? Not really. It's a lightweight household gizmo that is marginally more useful than the donut maker we got last year. Sometimes I wonder if the Register has accidentally got forum members shipped over from Luddites-R-Us.
Because then it's easy to automate a program that takes your revenge porn and makes tiny changes so that it's not recognised by the magic algorithm, but still shows all of your unmentionables.
Did that four years ago for an 'infinite scrolling' TV Guide.. didn't make headlines in El Reg.
As a box I can sling in a backpack and take to a client who almost certainly has a monitor around, it's a very handy little device. It can sit on a desk running services for months whilst not costing the premium of a MBP. Doesn't need a power brick, pretty much silent and enough power to run reasonable jobs.
Apple: Put a 1 line LCD on the front, sort out some bare I/O - Mic and speaker so it can run Siri without any additional hardware. Build in wireless dongles for keyboard and mouse and keep the rest pretty much as is.
A true developer just releases the source code and users infer the documentation from that? Remember also that comments in source code are a sure sign your code is not expressive enough.
Our local primary school hands children over to their secondary education with no experience of coding at all. Not even Scratch or any of the other visual programming for kids type things. The secondary school has to start its CS lessons with stuff that should have been taught years earlier.
Why? Because not one teacher in the whole school is confident enough to teach it. They dismiss it as 'not important' and avoid it.
It's hard not to conclude that this article is attempting to demonstrate that Damore is a terrible person because all of these other terrible people have used his memo to spout bile.
If Damore has done something stupid and offensive, it doesn't make it better by compiling a list of other stupid and offensive views. If he hasn't, attempting to pillory him by associating his views with other more stupid and offensive ones is character assassination. If none of those views are stupid and offensive, but just different to your own, then you're going to look bigoted.
Any news item of any appreciable size sets the fringes of the internet off, so an article dedicated to the inevitable doesn't enlighten us, improve the situation or reduce the existence of those fringes.
Since most of the competitors have similar features, and many are releasing open SDKs this year, there can't be that much commercial value in OpenSolo, and not much technical value. As a good will gesture, it keeps them in the news, but the proprietary stuff in SiteScan is where the real value lies.
Yeah, you need to fix the third fence panel on the left, and that bird feeder is hideous.
So it would take two weeks to charge my phone? Hmm...
Like many of the people I work with, my CV shows a healthy history of innovation and near bleeding edge work across the embedded, connected and online spaces. If you want a dev who can do <insert tech here> I can probably point you to one. Happily that means every day is a problem solving day, presents new challenges and new information to pull apart.
So the chances of fiddling around with someone's half baked 'challenge' to see if I'm bright.. um. No.
I'm sure that marks me out as not suitable for the role - cheap and enthusiastic beats cynical and experienced any day.
I've ushered a couple of big data projects into production at a global company, and can say that for the right client, there is real value. The weasel words are 'for the right client'. Big data most often provides means to make marginal improvements in customer retention and spending, but until you're looking at millions of customers those improvements are elusive and the cost outweighs the return. It is possible to run such projects on cost effective 'mini clusters' and with a pretty small dev team, but more often than not, cautious management will load up such attempts with reassuringly expensive hardware and a 'big team' to match.
The other angle is that reporting and compliance can benefit from big data - moving from reports that are typically days (if not weeks) out of date to near realtime, and archive free access to a company's entire history. That level of improved efficiency can be worthwhile, but more often than not, big data runs alongside existing traditional services and so is an additional cost rather than a cost effective replacement.
It remains a very cluttered toolbox, requiring experienced devs who can integrate with the company as much as with the software. Building a solution is still a nuts and bolts job, which leaves employers at the mercy of over optimistic devs and divas.
Like Smart watches, Smart home stuff seems to be falling into the 'solution looking for a problem' category. Home automation has been around for decades, since X10 came out in the 70s - yes, that long ago. The slow uptake is not due to lack of shininess, just the simple fact that we don't live our lives expecting inanimate objects to gain independent thought.
Our ape brains still work around the idea that if we want an object to do something, we do it ourselves; if we're not in the room, we don't expect or need things to 'control' themselves and if we are in the room, we like clearly defined cause and effect. I press the wall switch and the light comes on. The level of complexity involved in getting a smart light to come on massively outweighs the actual end result.
That's not true. Not least because you must fly within line of sight unless you have the appropriate license. Small drones get impossible to see at quite short distances. The drone code is quite clear about restrictions to operators: http://dronesafe.uk/drone-code/
Unfortunately, since drones have become high profile, they're the first thing that get blamed when pilots see something. Notably UFO sightings have dropped in inverse proportion...
In this case something 2m across, 'balloon like' and at 5,000+ feet is NOT a drone of any sort that you can buy or even build without serious commitment. Any one of those descriptions rules out just about anything available commercially unless you're spending tens of thousands of pounds or work for the military. The largest commercial drones are under half that size, and most would struggle to reach 5,000 due to battery limitations.
So either, it's someone with Bond-villain levels of commitment and finance, or it's not a drone. The most likely suspect is either a hobbyist playing with high altitude baloons, or a stray party prop.
Of course, we could just ban/license/arrest everyone flying a drone - but it has to be pointed out that the commitment to stupidity that would be needed to cause an incident like this would not be stopped by harassing kids playing with toys, or licensed drone operators doing their jobs.
Perhaps a first step would be to put a gimbal mounted high resolution camera on each plane so that unexpected objects can be accurately identified and the threat assessed? Compared with the cost of any of the alternatives, a few hundred quid for a camera would not be so unreasonable.
.. for the money.
In the consumer space, China owns the vast majority of drones. But hey, if we mix in regulation maybe we can force a subscription out of all of those lucrative owners!
At present, I'm not sure there is genuine justification for end to end flight management, and the start ups in this space will only get traction if they are handed it by the government. Meanwhile, the flight controllers are developed and coded by companies in China, Zurich and the US. By the time regulation is needed, the solution will be handed to us by the companies that dominate the market. They already manage telemetry, GPS tracking and No fly zones, so the step of traffic management is theirs to make.
"And the Problem is with aircraft is that they habit of flying abroad, and the most likely destination for UK aircraft is Europe. So you can have one set for UK and one for the EU, but in the end you are just duplicating effort, cost and regulation, so why do it?"
I believe that international rules about aircraft are well established. This article however is about UAVs, for which rules and regulations are still in a prototypical state at best. The thing to note here is that in the main UAVs do not have a habit of flying abroad (except in the case where they have armaments on the tip).
A better comparison would be with the drivers license - most drivers drive locally, some need to go further afield, when you would hope their license is accepted. Should we get rid of the DVLA on the grounds that we might want to drive in Europe and those nice people in Brussels can probably do it better than us?
This appears to be a Brexit issue because post-Brexit the CAA have more latitude as to which regulations they adopt and whether they hand over some authority to an external agency. Prior to that, the EASA and CAA pretty much ran in agreement. As I understand it, the CAA have been quite proactive in this space and have driven a lot of the decisions later made by the EU. The subject of the article appears to believe that Europe have the authority that we lack. I guess we should hand over drivers licenses to them as well.
On the other hand YOU have made it a Tory issue because you hate the Tories. I have no idea whatsoever what the funding for the NHS has to do with the CAA.
Of course, we can't propose anything because the 'Evil Tories' won't let us?.. and people wonder why the Brexit debate went the way it did.
You know we're talking about regulations for pilots, not drone manufacturers? Unless you can get a Kite Mark tatoo...
Not that we have that many manufacturers outside of the usual military contractors.
So... because we might want to fly in the EU, we should adopt their regulations? That makes no sense whatsoever. What rules we follow domestically does not dictate the rules we follow in other countries. If you wish to fly abroad (which for drone pilots is not a given), there are so many other rules that you need to be aware of (public/private spaces, privacy rules, insurance and identification) that adopting flight regulations is possibly the least of our worries.
This only makes sense if we intend the EU to manage pilot registration and certification on our behalf. Do we want to do that? I understand the CAA is over-worked and under-staffed, but if this is our approach, we might as well close them down completely.
Why not call for some sensible investment in the CAA post-Brexit, and a close working relationship not only with Europe, but also with Canada, America and Asia? Europe is a long way from being the biggest market for drones, America treats Europe as a 'second option' when it comes to delivering new technology and a huge proportion of global innovation and investment occurs in China.
Frankly, if this is the spokesman for drone development in the UK, it explains why we are lightyears behind the real innovators in this space.
It says a lot about this industry that an article describing what appears to be a severely dysfunctional company is followed by long discussions about whether or not that constitutes socially unacceptable behaviour.
Hint: totally unacceptable.
It doesn't matter if you once heard a woman say a rude word, or the girl you fancied told you to get lost - nothing makes being an asshat acceptable. That applies just as much to women as to men.
> That doesn't matter...
From what I've heard, it's far from a 'vanity address'. Most of the original development of the Echo was done in Cambridge, though not in quite as shiny offices. From what I've heard, the drone stuff is not really related - this is all about their home gizmos and smart speech work. Amazon are building on the success of Echo by consolidating their team.
But of course, if you've got a chip on your shoulder and think that Manchester should have been chosen as a location for a team that started in Cambridge, then you're going to be righteously annoyed.
DJI alone has sold more than 20 times as many drones as all of the planes (commercial and civil) in the world. Can you suggest how the currently stretched air traffic systems cope with that many 'flight plans' (most of which will be along the lines of "I dunno, just thought I'd fly over there for a bit, then maybe look at that tree")?
The firm got a whiff of bad news when the first reports of terrorists drones were inevitably illustrated with pictures of those familiar white jelly moulds. Not too surprising then that they take steps to avoid any more direct links.
A bit like the concept of 'smart guns' of course, the problem is that you can make drones smart enough to not get involved in a land war in Asia, but the terrorists will simply build their own dumb drones to use instead. There's no magic solution here. Though it wouldn't hurt to limit the supply of conveniently reliable mass produced ordinances (grenades, land mines) into war zones.
No mention of Scala or Spark ML? Not convinced. Not that they're the 'best choice', but if you're listing options, I'd expect them to be there.
My experience is we get a bunch of guys who've used a range of tools in an academic environment, and thus have used whatever language or package happened to come to hand to most easily solve their particular problem.
Then they have to face a production system designed around all the usual requirements for uptime, release management, security, versioning, testing and so on. The range of options when you have petabytes of data being generated in real-time is somewhat smaller, especially if you have any sane constraints on hardware (after all, you probably have a few hundred machines in your cluster, you can't justify doubling that just because a particular project runs a bit slow). Businesses at this sort of scale only want limited exposure to tools that cannot be proven to be robust. DevOps have no desire to install yet another package just to satisfy your needs, when it probably needs yet another license audit, release cycle and further support.
Suddenly, you have a bunch of guys who are jumping through hoops not to give up their language of choice because they've got into the mindset that you have to have the right language to solve the problem. And in the process they've completely failed to understand the architectural techniques needed to deliver robust, performant solutions - believing that because they once saw Perl running quickly, it's the only way to deliver at scale.
Then you look at the business analysts who have demanded some horrendous SQL shim over your 'big data' just because they're incapable of learning a new language, and you can see where this is all going.
You missed the important minus sign there... A(minus)99 is big... very big
Why (should they be paying the 'same' tax)?
Seriously, it's a question worth asking. We've already established that a company is ok to pay two different individuals different amounts of money to do the same job. We've established that the contractor is likely to be paying more tax (as an absolute amount) than the permanent employee. We've established that the workers have different employment contracts, and different obligations towards risk and career development.
So, other than the fact that they're sitting in similar coloured chairs, why should they pay the same proportion of tax? The contractor is almost certainly paying more tax whilst doing the same job, so why should they pay more still? Contractors aren't holding anyone to ransom here, they clearly fulfil a function (and a useful one at that judging by the rates companies are willing to pay for them). So why should they actually have to pay something that you, as a permie, are not paying?
Why is it you get away with paying less (absolute) tax than the contractor sitting next to you? How is that fair? How is it you don't have to re-apply for the same job every few months, with a full interview and references? How is it you can take a day off sick when you feel like it and still get paid? Who's taking the p*** here?
I'm not sure that justifies the (admittedly fairly rare) 'verbal diarrhoea' articles that have to be picked apart in the comments. I do come here for the stories, but that's because I expect the authors to provide some insight into the subjects - or at least some industry standard cynicism :D
Before you crow too much, El. Reg should remember that the quality of its own reporting has been known to go down as well as up. Some of your articles over the last few months have been notable for the lazy journalism, poorly thought out reasoning and apparent lack of editing.
Medium is indeed very Silicon Valley, but I'm not sure 'quality journalism' is out of the woods yet.
Terrorists also use mobile phones. Let's ban advertising for mobile phones!!
If a channel is hateful or offensive, then fair enough, let's make sure it's taken down under Google's guidelines. Banging on about what is advertised on that channel is the sort of hysterical nonsense that the Daily Mail would normally be proud of.
...that's why we teach science in school.
Not a bad intro, but could have done with a little editing and a couple of illustrations. El. Reg should be able to do a little better than this - the information and understanding is clearly there, but the presentation is a bit behind.
For the last hundred years or more, there has been more than a few people who're desperate to believe that magical mind control can be moved from myth and fiction onto some sort of scientific basis. See the book The Men Who Stare At Goats to see how deep the belief goes.
It's true that you can understand the people in ever greater detail with big data, but you only have to see the political upset on both sides when Brexit/Trump won to realise that there's no uber conspiracy here, just the normal fallible humans finding new and interesting ways to screw things up. Not that the changes being ushered in are necessarily bad for our deeply embedded political systems, but no-one could really claim that there is any evidence of a mastermind at work...
""Junior" he he. Nice."
For sure it was a cheap shot, but if you will insist on posting "I solved world peace the other day, but then thought.. nah...", that's the sort of response you'll get :)
I can fully understand the issues with 'magical', and have railed against it enough in my time, but if there's a tipping point, it's when frameworks like this can make pragmatic decisions and pretty much be sure of being right. There's no AI involved, just a bunch of tools that are mature enough that when you want a collection of objects back from a database. the 'solution' is as uncontentious as a Taylor Swift album.
My feeling on robo-sytems is that we've not yet cracked the problem of specifying what we want (as anyone who's delivered a project will know), so there's no start point for robo-programmers to get purchase. What we will see is more of the Spring-like 'component declaration DSLs' that get munged into working code by 'magic' frameworks.
All those big data services must be running on some sort of magic sauce then. And those web servers. Oh and I guess that's why Android battery life is so much shorter than iOS.
"and part designed, an automatic programming system years ago, but then thought, "nah, might need a job" and quietly shelved it."
I don't know any junior programmer who hasn't gone through that phase. It's funny that no-one has actually managed to write such a thing (that worked) so far.
On the other hand, for the (many) posters here who clearly don't keep up with Java frameworks, it's worth pointing out that implementing a CRUD repository for an arbitrary Java object involves no more than declaring an interface (yes, literally one line of code) in Spring Boot. With two lines of code (annotations), the same Java object is your data definition, generating an SQL schema, indexes, constraints and sequences as appropriate. One more annotation turns any method into a REST interface, with form decoding, object transformation, output encoding, validation and error capture all happening automatically. Metrics don't take any more effort.
Of course if your only exposure to Java is through using Stack Overflow to find answers, you'll probably be thinking that PHP makes things far easier.
It's Friday and I'm standing in a school playground.
"My language is better than your language!"
Anyone who thinks the success of Java is down to the language should not be allowed to make tech choices for any company. Similarly for anyone who thinks that feature-X in a language is a reason to choose that language for a project.
"I was coding Z when you were still in nappies!"
Having a beard doesn't make you wise, old man. I pre-ordered the first public Java distribution (sent on CD from America no less), used Turbo pascal before it was Delphi, coded Z-80 and 6502 by hand and once fought a bear bare handed.. or something. That doesn't mean I can't learn new things, and discover new tools and techniques. As the article points out, what was true for Java five years ago isn't necessarily true now, which is why it stays relevant. When that is no longer the case, a competent developer is more than able to pick another toolset.
"Java is slow/bloaty/crippled"
Yet it's still here. What could it be that you are missing?
Personally, I'm not trying to come up with a "better argument". I'm saying that it's only worth the hassle and risk of contracting if you're paid appropriately. If HMRC want to increase the hassle and risk whilst decreasing the pay, then many more will do what I've done and get out.
I'm not trying to convince anyone that complex tax arrangements are a good idea. I'm not even trying to convince anyone that "i'm worth it". I am saying that a flexible, mobile workforce is a good thing, and this is running absolutely contrary to this. If you try to level the field between permanent and contract staff, then who in their right mind would want to take on the added burden of being a contractor? On a personal level, it doesn't matter to me - I have and always will follow the rules. However, I'm quite happy to have withdrawn my labour from companies that otherwise would quite like it (judging by the regular recruitment calls).
As an aside, the same applies to travel subsides and housing costs. If you make it difficult to travel to work, and expensive to move house, then you have a less mobile workforce which causes poor population distribution and strangles companies needing workers. It's easier to bring in overseas workers than help someone relocate in this country - and let's not even talk about the pleasures of commuting.
..those that haven't got the drive or skillset to get up and leave keep working on projects that desperately need people with drive and skills.
IR35 isn't fixed by creating some hideous online tool to manage its complexities. It's fixed by ending this obfuscation and witch-hunt mentality.
Someone in Government needs to remember that the job of Government is to support the people, not tax the people. In this case, we desperately need a mobile, highly skilled workforce - both in public and private sector. Enabling people to switch jobs and apply their skills where they generate the most value is a key to building a healthy, competitive economy. Things like IR-35 are the complete antithesis of that, though I'll bet that all those large corporate advisors will be whispering the complete opposite in the ear of any MP that'll listen.
As with so much crystal ball gazing, this article falls into the trap of extrapolating 'disruption' from today's norms. Before cars, if you'd asked anyone what the future held, it would have been 'faster horses'.
The sheer material and energy wasted in having a fleet of cars deliver a chain of goods to your house is immense. The logistics of the 'last three yards' are not magically solved by solving the problem of a car capable of autonomous driving to your doorstep (and that itself is far from a solved problem - whatever the AI pundits may claim). It seems to me the likely solution is further consolidation towards human driven vans doing the rounds.
Besides, the disruption mantra is a nonsense. Techies and investors fall time and time into the trap of believing that because you can digitally disrupt services like hotel booking and book selection, you can digitally disrupt the physical world. For some reason people believe Tesla can disrupt the car industry 'because digital' - when they're surrounded by companies with vastly more experience and infrastructure for delivering physical machines that customers want. Why does an electric motor make the slightest bit of difference? Consider how the car industry evolved 'consumer' diesel in very short order in the 90's and you'll realise that motive power is not an issue for the incumbents.
But hey, investor hype, wild optimism about AI and in inability to distinguish between physical and virtual problems are a great source of articles.
A small, easy to type on device with endless battery power and a decent telnet client - any day.
It booted quickly, could interact reasonably with standard documents and kept a basic calendar, spreadsheet and editor.
Doesn't need to be colour, or run windows or Linux (or Android for that matter). Doesn't need to play games, call my mum or take photos.
Does need WiFi (possibly bluetooth), a top notch keyboard and these days USB/memory card slots. A bonus if it could run a half decent browser, GIT and basic Java stack (but only when you wanted to kill the battery life to save a client's bacon).
EPOC had got far too convoluted to maintain and improve, but was key to battery life and the foundations of the excellent document handling. Not sure what could replace it.
It says a lot that I've not found a combination of Android device/keyboard/app(s) that can match the Psion for 'just working'.
If you were relying on a document in a court of law and it had been signed with SHA1, then my assumption is that if two parties produced two documents with the same hash, there would be some examination of the documents. The guess is that the original would be 'clean' and the doctored one would have the doctored text (the subtle addition of the word 'not' where appropriate) PLUS some additional garbage to bring the signatures back into line. As it's astronomically unlikely to produce the same signature when only making a meaningful semantic change to the text, it should be possible for a forensic examination of the documents to identify which is 'clean'.
However, if SHA1 is only being used in an automated system (Git etc.), then the issue is much more pressing - the assumption being that the system will not be coded to identify a collision, which would be processed and actioned unnoticed.
...a sack full of spanners. Sure, it can do some amazing work, but it requires a lot of self assembly, and you need to have the right sort of problem that the tools will actually fit.
If your data is big enough, and you have the means to act on the results, you can indeed deliver real value. I was lucky enough to work on a project that returned significant revenue (I offered to work on a percentage basis, but they seemed to think that was unreasonable... sigh).
On the other hand, if you've got the infrastructure there, and guys smart enough to bolt it all together (which seems to more or less be a prerequisite at the moment), you can roll your own stuff out of the wider 'big data' ecosystem that's likely to be a far better fit than trying to put in a screw with the Spark hammer. We replaced clunky Hadoop batch processing with Kafka + microservices + Cassandra and delivered the same results but in near real time and with much less reliance on 'magic coordination'. Despite the hype, Spark was not a good enough fit to justify mangling the problem just to be able to use it.
The thing about reporting is that you end up with reports that tell you the same thing each time so you can compare like for like to spot trends and patterns. Having AI interpret what you need to know doesn't fill me with confidence.
And then there's the spectre of misunderstanding:
"Deploy the production release, Hal"
"Deleting the production database, Dave"
"No, no, no! Stop!"
"Phoning your mother, Dave"
Not sure about that.
1. (Seeing as we're doing numbered lists). The executive orders are temporary. That's brilliant - it gives him 90 (120) days for people to 'get used to' the idea, then he can put something in place that's not as severe and everyone thinks they've won concessions.
2. The 'Government isn't business' line concerns me. This is the Hillary Clinton approach - "this job is too small for your small ape brains, let me do it". Both America and the UK have built up a huge government machine that is dedicated to oiling its own machinery. Laws only get more complex, lawyers only get richer and the civil service has so many offshoots and branches that genuine change is almost impossible. The end result is that successive governments are increasingly only able to tinker around the edges because "this is how it's always been done". That's not to advocate Twitter diplomacy, but to recognise that government has become incredibly self serving lately - one of the reasons we got Brexit and Trump.
"If you have a choice between disabling it in a harmless way by taking control of it yourself, or even by simply preventing it from being controlled at all while still a good distance from any target, why on Earth would you not choose to do that and instead start wildly firing live ammunition into the air?"
You're assuming that sending radio signals to a modified drone will do what you expect. Most drones can fly predefined paths, so don't need continuous control. Any attempt at signalling to them could have completely unexpected results - a weapon could be tied to any given behaviour so 'safely' stopping it involves double guessing that behaviour. Will it drop a grenade if you ask it to slow down, speed up, lower altitude, gain altitude, fly north?
I won't go into the ways a consumer drone can be modified, nor the challenges of gaining access to it "while still a good distance from any target". The operational issues with telling a flying bomb to be harmless are immense.
So a drone comes over that's been modified in an unknown way, and the first thing you want to do is mess with it's radio signal? Isn't that rather like the bomb squad myth that someone goes in and cuts the green wire?
Back in the real world, I would have thought a serious threat would more likely be shot out of the sky.
"It's ok chief, we've got you covered. Not a single threat can get in"
"What sort of thing are we talking about?"
"Oh, you know, drones!"
"Drones? Those military bastards!"
"Uh, no.. like, kids drones. They can carry a mean GoPro"
"Oh.. I see. What about grenade launchers? Guns? Trucks running through crowds?"
"What?! That's a hardware problem! We're here to ensure no-one can take an evil selfie!"
The point is that at the time it was launched, it wasn't a me too product. It was very innovative.
They took so long to get it to market that the rest of the industry caught up with all of the practical bits, leaving only the impossible features for them to implement. So it took even longer to get to market whilst they figured out that the reason no-one else was doing that stuff was because it was very, very hard.
So they ended up with a me too product, late to market. But that's not where they started.
The issue here is that innovation is hard, and the competition is fierce. That's not at all the same as the dot com boom, which majored on things with imaginary profits. We know drones are profitable - they turned DJI into a billion dollar company in very short time. The challenge is replicating that fierce pace of innovation.
Biting the hand that feeds IT © 1998–2017