Re: Not impressed
I don't care how you pronounce it, it's how I wrote it. ;)
(I didn't actually know that there was an official rule on that, I just went along with the biologists. Thanks for the link, though, I always like to see stuff like that)
187 posts • joined 7 Jul 2017
Isn't Firefox 52 ESR out of support with Mozilla? I thought (and correct me if I'm wrong) they dropped that back in August last year.
I think you may have run into Microsoft's other hobby: trying to get people to keep their browsers on the latest versions for security reasons. Somewhere along the line, MS worked out that they get a lot of flack for all the virii in their ecosystem, so they've sunk resources into fighting that from as many angles as they can.
"Sure the address space could be higher, but no one will ever need two freaking gigs for emails: it's basically plain text, FFS."
I think my personal favourite example of "no-one will ever need that much" coding was in Deluxe Paint back on Ye Olde Amiga. If the RAM used by the picture you were working on exceeded 512, it threw an out of memory warning, because, well... when they wrote it, no-one had more than that, nor did anyone sell more than that, so they set an upper bound on the test.
However, it was just a warning (not an error or a crash), so if you'd gotten one of the 1MB expansions Commodore later introduced, the program would carry on happily after, letting you do all the fancy gradient fills you could want.
A while back (2010 or so), and this is probably the single nerdiest thing I've ever done, I used to play EVE Online using a 3G dongle. Only for a short while waiting for proper internet, but, well, it's an MMO. They pay all the bandwidth costs too, so they make them remarkably low-traffic in normal use and I used hardly any data at all.
Anyway, every now and then, I'd get booted because it lost internet connection for a few milliseconds, and the game was incredibly twitchy about that. I eventually worked out it was because I was living near East Croydon - every time the Gatwick Express shot by, hundreds of active mobile calls were being transferred to and from each base station, and they took priority over data connections (in theory, data connections should handle a momentary disconnection).
Anyway, that's how I ended up playing a fully featured MMO on a _2G_ data connection, since that remained stable and largely unused.
I think a large chunk of it is that early games (especially during the Amiga's shareware boom, when one-man-shops were almost the norm) were playtested primarily by the developers themselves, and their sheer familiarity with their own project lead to the difficulty creeping up.
This created a market full of difficult games, convincing everyone else that the market wanted games to be hard - and as a small child at the time, I was happy to invest hours and hours into becoming good at them.
Then in the late 90s devs starting saying things like "I want players to see the end of this game", and larger studios began to seek wider audiences (including older gamers with less free time), so they slowly dropped the difficulty down or more commonly introduced variable difficulties so that the players could choose their poison.
I think my favourite example of that was the original X-Com game, where a bug silently reset the game difficulty to easy on first save. A bug that was discovered by a fan porting the game over a _decade_ later, by which point the famously super-hard sequel "Terror From the Deep" had already been made based on feedback that the original was "too easy".
Looks like classic Menu pricing to me. If you can sell only at £30, you can't get any of the cash from people with a tenner spare.
This way, they get access to more customers, with an upgrade option for people who go in cheap for whatever reason who later decide to go with the shinier option.
It's like that famous quote about advertising. "I know 50% of my advertising money is wasted, the trouble is I don't know which half".
Except with middle management, the ones you most want rid of are basically only doing one thing: whatever convinces their superiors that they're the Good Ones. The ones who fail the metrics, who lack 'visibility'... those are either utterly useless (which does happen!) or too busy doing important stuff to make sure they look like they're doing important stuff.
Collateral damage, as it were. Airports use a lot of radio across a lot of spectra, so it's hard to jam all possible drone frequencies and still be able to fly planes.
Plus there's a chance it's pre-programmed and only needs GPS, and you really can't fly dozens of airliners if GPS is being jammed.
Our current understanding of low density planets is that they're a *lot* of gas around a smallish rocky core, though at Jupiter-ish sizes there's likely some crazy pressure-related shit going on with metallic, liquid, hydrogen. This one isn't anywhere near to Jupiter scales so it's probably just rocks.
So it's losing all the gas and as the article states it'll eventually end up with just the rocky core.
I *think* what they mean is that this is the first old galaxy we've been able to measure in this way (because of the angle), and previous results suggesting low levels of dark matter in old galaxies were using less precise methods (at a guess that'd be something like gravitational lensing, the rotation at the outer edges only, or trying to infer the effects on nearby galaxies).
I wonder if there's a way to hook into org-charts, with notifications failing up the chain of command until they find a role holder who's actually in the system.
Then you just produce reports that go somewhere higher up when a job role vanishes and isn't reassigned, and make sure HR know how to maintain it. They don't need to know what the report even means- just where it used to go. HR generally has sight of re-orgs, so they'll have very good odds of being able to work out who the new owner should be.
I have seen the "long drawn out proof" come up with a different answer to the original issue, but more often it's "As we suspected, X happened. This was because A, B and C did not stop X happening. We should retrain B, re-code A and correct the config in C so it's monitoring Production Y and not Dev".
Or "the warnings have been going to an unrelated team in Birmingham for the last two years who, after three months of trying to get someone, anyone, to explain the warnings, set a rule to delete them on receipt."
Office's "intelligent" paste can be a real curiosity at times.
If you copy some lines from Access, pasting into an Excel sheet will also give you the column headers above the data.
If you tell Excel to paste as values... you get the table name, the column headers and then the data, but at least it doesn't muck with the font.
I thought bulldog clips were the big wide things you use for paper: crocodile clips being the narrow pointy ones?
Unless this is one of those quirky etymology things where nothing makes and sense, which wouldn't be that surprising *gestures vaguely at the English Language*
At least in theory, they want to snoop on the planning stages. You may say "they can just plan offline", but, well, there's a reason the drones keep hitting weddings 'by mistake': HumInt is tracking when suspects meet, and calling in the hellfires if they think there's enough Target in the collateral.
That sort of thing (which, IMO, is probably some manner of war-crime) strongly discourages personal meetings. So SigInt has to try and fill in the gaps. Your average bomb-maker isn't going to roll their own crypto, but they're happy to use one of the off-the-shelf ones that the Five-Eyes types complain about not being able to crack.
To be clear: I have some sympathy for the aims! Terrorists are bad. Child abuse rings, also bad.
But the NSA, GCHQ and pals took a calculated risk by violating the privacy of millions, and the dice didn't work out for them. Nobody held a gun to their heads and said "you must spy on your own citizens, en mass, on dubious legal grounds". This is their screw-up, and it'd be nice if they took the consequences like adults.
I've got to be honest, I'm curious as to what goes wrong with this sort of thing, because it's the same pretty much everywhere.
MMOs, ISPs, cloud providers, train companies, banks: the time lag between "customers notice en mass" and "the service desk are told/permitted to treat it as a major incident" is almost always achingly long.
Almost all of them should have easily visible metrics that reveal the issue: your logins drop off a cliff. All the trains are behind on their routes. Transactions/minute have doubled or halved or whatever.
So why don't they have a way for the poor grunts on the helldesk to be told "it's buggered, tell people we're working on it and will provide an update at 14:00 UTC"? Why are they left fielding angry customers armed only with "try rebooting" scripts? It's not just bad customer service, it's down-right unfair to your staff.
"And why do tellers have the ability to bring down a whole system? No permission control?"
Either the architect in 198X didn't think it's would come up (seriously, just don't tell them the command exists, they'll never type that by accident, right), or they did and their manager said "we don't have a week for you to code in permissions".
A lot of what we now see as obvious best practice was learned the hard way.
They're essentially illegal to fly outdoors in towns and cities.
The rules say you can't have a drone within 50m of people of buildings (excluding buildings "you control", I think the wording is, which is why drone racing can be legally done), nor can you fly it within 150m of crowds or built-up areas.
There's obviously a band between 150m up and the 400m cap, but how'd you get them there? You want to play with these toys, you need to aska farmer nicely or see if the National Trust will let you fly 'em (I expect they will in some of the less sensitive reserves, but I've not checked).
"Just because your compiler and debugger knows it is a boolean, does not mean that the storage for it is only a single bit, it is going to be at least a byte and can hold a range of values."
You know, that probably is what was happening. I don't mess with bytes much myself, so it never occurred to me! Thanks for the insight there.
In IBM's PCOMMS API, at least the aged version my illustrious employer still uses, there are functions that return phantom "true" values that do not trigger "if Blah= True then" statements.
But... they aren't strings, or ints, or anything else you might see kicking around pretending to be a boolean. They're very explicitly Boolean values. You can see this clear as day in your Locals, they happily work with declared Boolean variables, but they remain unholy ghost-values.
I mean, when they're False, they can be tested for "= False" like you'd expect. But if they're True, they can't be tested for "= True". I spent hours trying to work out what I'd done wrong - surely that had to be a mistake - maybe I wasn't declaring the type properly, or had misspelled "true" in the test or something - but nothing made it behave.
So now, buried in some of my old automation code next to some rather explicit comments is:
If Cbool(Cstr([foo])) != False then ...
Companies only get money from customers. If they're paying money on things, the customers pay for it.
It feels unfair that companies can 'earn money' and not pay tax, but... when you boil it down, corporate tax is basically just VAT. VAT that you only have to pay if the corp you buy from is honest with it's book-keeping. Amazon are basically doing a cash-in-hand 'discount' via tax avoidance, but because it's "legal" that translates to a massive advantage to the more socially responsible and less tricky competitors... funded by the tax payer.
I was about to say something about the volume making checking everything tricky, but then I remembered PACER exists.
It wouldn't be that expensive for Google to get a subscription to court document services and make sure that the court order exists, is of the right time-frame, and names something vaguely like the right people. The documents have very controlled formatting, at least for naming the parties, so that's probably not even hard to parse, even from a scanned image.
Or hell, given the photoshopping involved here, just check if this is the same file.
I've been with Virgin since about 2002, and almost no problems until the current place. After I had five multi-day outages in a few months, I switched to one of the BT resellers (not plugging them 'till I know they're actually good).
Sometimes, you get wired through a bad cabinet (in my case, mucked up by years of landlords turning houses into 4-7 bed flats for students until the cabinet is 60% splitters by volume) and it'll just keep failing. Not much you can do, since they're not willing to put the effort in to actually fix it properly - they just unplug someone else and hope it's not in active use. Most of the customers on the box will move out next summer anyway.
"Surely there should be a healthy market for scanners and printers incorporating a metal detector that'll complain *before* potentially self-harming if fed a stash containing staples and paperclips?"
They make plastic, wooden and even paper fasteners these days. There's even a clever tool that punches out tabs in the paper itself and pushes it though, making a 'staple' out of the very paper you're 'stapling'.
The nifty trick I saw in the newer machines we had was having contra-rotating rollers under the feed, along-side the ones that rolled inwards.
Not enough force in them to over-power the in-feed normally, but if more than one sheet went in at once the bottom sheet would slip (even If stapled, as the sheets could crumple), and the change in tensions tells the machine to stop and object. This caught staples early, but also solved a much more common problem with double-feeding sheets.
I used to operate a scanner in a post-room, and after a few weeks I got quite good at fishing staples out of the paper-guides.
I'd have complained more, but people will staple things in really strange was if you let them. Half-way though a stack of letters, there'll be one receipt stapled at the bottom.
Biting the hand that feeds IT © 1998–2019