Posts by happy but not clappy 59 publicly visible posts • joined 9 Jun 2015
Bit players are getting in on this (trooli.com down our way, fibre boxes on telepgraph poles everywhere), but BT is really trying to prevent this.
I got fooled by a 2 year lock-in for a discount, so it will cost me £600 to buy out the contract. I'd rather BT offered the upgrade but nope. No upgrade. No plans. Stuck on ADSL, no doubt until they are forced to offer it.
Most contracts have an IP "duration" clause that basically means there is no such thing in law as "spare time". If you built anything of any interest during the period you were employed, then your employer can claim it is theirs and will almost always win in a British court, even if what you created was of only very limited relationship to your day-job e.g. "runs on a computer and makes money".
Best thing is to ask for an explicit exclusion from those clauses for anything you do in your spare time or risk being sued. For software developers this is a very real likelihood as I understand it.
Amplifying this point.
Lets say we had a bunch of, I dunno, windmills producing electricity no-one can use. Then efficiency is no longer the issue, just make something that can be burned. Could be hydrogen, which can then be combined with CO2 from, say, a massive green-tinted wood-burning station. Or it is possible to do it directly from solar, apparently. Seems more likely than giant battery packs.
The resulting fuel can drive turbines, be they in wings, or in a power station. Rolls Royce might have some spare these days.
Efficiency will suck, but that just means we need more windmills. You can buy a shitload of windmills and infrastructure for £50 billion, or whatever a nuke costs these days.
So, m'lud, I suggest this is all just green posturing by airlines and their political cronies desperate to avoid having to do anything before their gold-plated pensions are due.
Maybe Elon could help before he goes off to Mars?
Much more positive and tactile than the loose chiclet keyboard that we have returned to. If only it had actually worked. Mine lasted about a week before being almost useless. Now I want the best of both worlds, something actually reliable but with that positive clicky-clicky thing.
Well both DB and HSBC are in a world of poo already, so software issues might come as a blessed relief to the endless rounds of shrinkage and headcount reductions. Their entire business model looks like it is broken, held up only by the inertia of their customers.
Also both fired their CEO's in the last year or two, so there is at least a small degree of accountability, though with the usual golden parachutes to soften the blow.
On topic, banks don't do much that is creative, to be honest, with most work being to paper over the cracks. I guess they can probably just stop all the projects and keep trundling along on the same mainframe packages they've relied on for over 50 years.
Well I have had to go over this a lot in my job, and as I understand it (from my legal friends) GDPR is about control, not restriction. So absolutely, you are right, this is PII. The key test is whether Google is clearly identified as the data processor, what that processing is and for what purpose, and finally whether it is equally frictionless to withdraw permission as to give it.
I would say Google skates very close to the wire in that withdrawing permission is tricky, and not terribly sticky (different browsers will change the default as they won't have the "no track" cookie set and you may not always login). Furthermore, it is ubiquitous (much as facebook's is) so giving permission on one web-site is extended to all the rest.
However, they try just a fraction harder than the traditional ad guys (like Criteo and Adroll), and so stay third on the list of priorities.
Like all committee led bollocks, they are looking for the equivalent of a supersonic tank which can swim. Federation gives them the willies, but is essential here. And would cost way less than the 2 shit-ton's of cash quoted. And would be resilient etc etc. What annoys me is there will be a small group of slimy twats egging them on, while collecting huge fees. I'm looking at you Qinetiq, and your spawn.
It's US foreign policy driven from the White House. Look at all the other Chinese manufacturers that have been banned/harassed and generally interfered with over the past few years. Not that the Chinese are being especially targeted, just that the dirty tricks are much more public.
Anyone entering the US market does so with trepidation IME.
As an experiment I once complained to them about the use of the term "pure" to describe adulterated products.
They had a remarkably similar response to the above. That the ingredient in question, despite only being 5% of the product, was itself pure, so calling it "pure" was quite okay. Complaint rejected. Good night.
So I guess "pure fibre" means, "fibre somewhere or other, maybe, nothing to see here".
In the early nineties I ordered the David Ferbrache classic book about computer viruses a little before it was properly published.
Within an hour I got a phone call from someone in "intelligence" asking me questions about hacking and trojans. That was instructive.
Can we sack the PAC? MP's is it? Oh well.
Can someone please show them a side-by-side, including the cost of those 700-odd savings professionals?
Central procurement, honestly. Worked so well for the Soviets eh?
Mr Orwell couldn't have been more appropriate, plus a bit of Kafka and maybe a pinch of Philip K Dick. Same everywhere I guess.
For how this will turn out.
Yet our politicos serenely avoid the issues.
I imagine all government systems will now work like our hospitals. Both extremely expensive and crap. Due to a massive mismatch of supply and demand, which turns into a structural poverty of supply. FTW.
A bit like RBS and indeed Lloyds, for a more directly IT angle. Continuously and repeatedly ripped off by consultants from top to bottom, demotivated permies getting fired or off-shored, and criminally poor performance.
I certainly won't be contracting again till all this sorts itself out.
...is your friend.
Microservices et al make this much much easier. They generally do not (should not?) use port 80 (low ports, boo hiss, evil root etc), but that's ok.
Any scaled up microservices setup will also include load-balancing and service liveness/performance monitors, so all that mapping is built in.
InterNetworks can go back to being dumb. Sorry CISCO et al.
check it out: https://github.com/Netflix/vizceral
And we agree. Yawn.
The problem is indeed with specification. It's what I spend most of my time doing these days, pushing agile uphill and telling the "business" to stop using IT as a scapegoat for their laziness and inadequacy. Yet we still have these "my manguage is bigger than yours" threads (saw the typo, liked it :)).
Still, I think we have hit a tipping point of sorts. Previously we had to build our infra, nowadays, not so much. Then we had to build out systems, we still do that, but not for good reasons, and so on. This is quite new. So, at risk of going meta, I imagine we will go to a functional approach, a meaningful CSP with state, on a relatively universal surface. Languages will be rather script'y. So our junior dreams may come true. Do you think it won't happen?
"Junior" he he. Nice.
Object storage is trivial, yes. I like Spring Data. Maybe a little too magical for my taste, but I'm an old timer. Note your argument though, how little human interaction is needed. Why are you behind a keyboard again?
I give it less than three years before Microsoft and Google start punting robo-programming systems and demonstrating that they are provably more correct, more efficient, more whatever.
We shall see eh?
Will program most of it soon.
It boggles my mind that we are still writing CRUD and data mangling back-ends by hand.
I seriously thought about, and part designed, an automatic programming system years ago, but then thought, "nah, might need a job" and quietly shelved it.
Someone may catch on at some point. I would bet good money Googlers know this too.
Meanwhile the Y2K gravy train rolls on, choo choo!
It would be nice if they understood that effective social uplift was the goal, not robots and AI and all that other trickle-down globalised efficiency bollocks.
Unless they replace the politicians of course. I'm all for that. Then they can all do something useful, like building houses or looking after the old and knackered.
This is 50% right, as I have come to understand it. Yes, SQL is crap at expressing these things as it has no explicit looping, and works in sets, not linear chains. There are kludges to fix this of course via stored procedures, and stored tables, and nasty operators (CONNECT anyone?) but it isn't pretty. Impedance mismatch, yes.
It is, however, pretty easy to encode them in relational databases (e.g. two tables, one of nodes, and one of edges). Directions are also easy to express.
So the problem here is one of semantic power, and that is something that is easily added, so I think graph extensions to SQL will appear, and performance will be equivalent if not better, as soon as someone can be bothered to add it to mariaDB. Like all good ideas, it has probably already happened.
So Stevie's problem is that if all the banks expose a standard API, there is no reason for his app to exist. Sorry Stevie.
Oh, and I am in this business, and it is looking like the banks will not be dis-intermediated by this as described. It's worse. They'll lose the piles of cash they've made by controlling the channel. And if that happens, then someone else will need to pay. Probably us.
Then people will see that push payments are intrinsically safer, and therefore less risky, and therefore cheaper, than pull payments. Cue Visa and Mastercard's business exiting stage left, followed swiftly by all those payment providers and probably a bunch of banks. Part of me is pleased.
So docker will, I hope, try to avoid being a bridging technology. We have porky VM's for that.
A modular approach, with lightweight automated ops on a commoditised fine-grained utility platform is obviously the way forward. Give it twenty years and a business disruption cycle and all this will be seen as academic. Some of us are already building this componentry.
Bye bye Oracle and friends. FTW.
most of the use-cases for devops to live are only for systems where a bug in live is a low risk. Furthermore live devops to stateful systems is ridiculously immature.
One poster suggested devops to pre-live, and that's good, though unpicking one change from a bunch is difficult.
The key here is what is called "QA", though really they are talking about QC i.e. testing. This is the hurty point now, not the technology, and is why the business are nervous. They want good arse-coverage, and this doesn't give it to them, yet.
"Transformation"
Any IT program with "Transformation" in the title leads to pain and failure, mainly due to top-down consultant-led fuck-wittery. Extra points for adding "digital" at the front.
Ooh, what is this I see before me: https://www.gov.uk/transformation. I rest my case m'lud.
They've rewritten Apache Derby to use HBASE as a storage layer, and added some MVCC. Nothing wrong with that, although that implies optimistic locking which sucks under pressure, but otherwise it is a big bag of open-source stuff, which they are then trying to sell. I find that offensive.
I'm currently reviewing "big data" tech, and it isn't pretty. Most of it (with the honourable exception of Cassandra) just doesn't scale well. And none of them cover the standard use-cases well without significant complexity, though they are trying. Look at the mess that is the Apache ecosystem. How many products?
Any medium size business that throws money at this is probably going to lose large chunks of it, same as they did in the nineties with "data warehouses". Still, this is SNAFU, so not so surprising.
FYI, for extra CV points, add "data lakes". Oh yes. I've got three out back...
Innovate and all those others desperately seek to glue academia to industry (universities try really hard to justifdy their own existence, and they have copious free time to stuff the boards of these organisations).
Thus partnerships with Universities give you a huge uplift. Add some spurious IP, some eco "woo woo" digital fairy-dust, pick a number that hits the average (about £1M by the looks of things) and you are good to go. So you can see why the mentioned project got loads of cash, it is perfectly adapted.
The London catapult is empty of serious applications for this reason, academia are largely irrelevant in the software space, and the amounts of money offered are paltry. £5K gets you nothing really. £800K gets you a little more, but not much. Consider instead what it costs to develop a new car and you are getting closer.
Don't forget, we are collateral damage to the powers that be.
What they expect, evidently, is not to prevent attacks (though they would love to have that too and spend a fortune trying), but the ability to find perpetrators and bring them to justice after the fact, and their friends, and anyone else who was nearby at the time, in a form of judicial drone strike of a wedding party.
The fact that this also kills the false-positives is really not something they care about. Thus anyone who has ever downloaded the ancient and largely shite anarchists' cookbook becomes a terrorist, and goes to jail, and so on. It is the very definition of thought crime. Which is kinda scary.
If you read their stuff, it's just a bit too marketroid. Buzz-word bingo abounds. They support every language going, every devops going, open standards, have a laser focus on the user, use agile, talk about delivering at pace and so on, but there is nothing to illustrate how they do this, or display status of how they are getting on. Lack of transparency/visibility is the kiss of death IMO, and smells of dishonesty.
Good luck to them, but agile is more than dev. Dev is the easy 80%. They need to get this.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
