Re: I have a local library ...
I have a local library
Can you get it in your suitcase for an extended vacation?
7402 publicly visible posts • joined 28 Oct 2011
less than $10 – a fifth of the price of gasoline
Only because gasoline for road use is taxed much more than electricty. At present.
a matter of how quickly you could get the energy into the car
Which is one place where flow-technology batteries might actually be useful. If recharging became simply a case of draining the used electrolyte and refilling with fresh, refuelling could be as quick and easy as it is today for gasoline vehicles. Small filling stations could be refilled by tanker using a similar system, larger ones with good grid access might have on-site electrolyte recharging facilities. It would be interesting to see numbers on fluid quantity versus range.
clicking on a button where you don't read the details
That would be the button that says "click here to upgrade", where there is no button that says "No, thanks" ? Sure, if you know enough to realise that closing the window acts as "not now, nag me again later" you could argue that it's not forced, but without a "Just STFU and leave me alone" button it most certainly does seem like a forced upgrade to must of its victims.
Supersonic does nothing to address the key constraint in air travel. To put it bluntly, the sky is full.
Other than immediately over airports like LHR the sky is pretty empty. The key constraint in air travel is the time it takes to get to the airport, park, faff around for an hour's checkin/bag drop delay and two hours security theatre, all for a 1 hour flight.
Supersonic transport is never going to be a replacement for 4-hour holiday flights, but for 10-18 hour business longhaul flights there's a definite market for something that can shave a day or two off a business trip.
I'm not convinced the bird will be any more a money-maker than its illustrious predecessor.
Operationally Concorde made money, it just couldn't recover its development costs because there were so few viable routes because of the boom. Fix that so that US West coast and Middle East are in range of Europe, and they could potentially sell many more. With 150 seats or so it could be comparable with current subsonic business class prices.
Do you really believe that the people working on self-driving cars have not realised this?
I'm sure they have, and I''m sure that somewhere in their design document there's a small footnote saying something like:
¹This presupposes the availability of 100% network coverage.
The actual implementation is, of course, somebody else's problem.
To avoid confusion this work is normally called OpenZFS and Oracle has no claim to it - it cannot have any, since its has been made open source before they bought Sun.
IANAL either, but I don't think that statement is correct. Releasing a product as open source doesn't mean surrendering the rights to it, it means licensing those rights. Sun chose to license them under CDDL, and Oracle (having bought Sun) will likely have acquired the ownership of those licensed rights as Sun's legal successor, unless the acquisition contracts said otherwise. For the puposes of the legal arguments here, Oracle ≡ Sun.
There's definitely a "mindset" problem. 20 years ago you might have spent $250K on a minicomputer, and $50K on the database package, and considered it a good investment.
Today you'll get more powerful hardware for only $10K, but inflation and extra functionality could have pushed the software licence to $100K. You're still paying less than half the cost in total (much less if you consider inflation-adjusted dollars) yet people get upset because the software costs more than the hardware, and it "used not to be like that".
There's no logical reason that the prices should track. Hardware development is largely automated and manufacture is cheap. Software development is very labour-intensive, and the price of labour has skyrocketed compared to the cost of electronics. We just have to get used to the idea the the hardware is now the irrelevant part of the system's purchase price, and accept that good software costs a lot of money to write and support.
No-one's mentioned the horrors of having to port between different endian architectures yet.
Like the network code that works fine on big-endian like SPARC or M68K, but fails on intel, because the programmer had never head of htonl() and friends. Or worse, had heard of them, but forgot one in a critical place.
Or the weird things that happen when someone has carelessly passed a 32-bit address to a library that expects 16, which works as expected on little-endian systems, but not on big-endian ones.
Brings back memories of hours of not-so happy debugging
Dunno about bidets, but lived in countries that use bum guns and you still need some paper unless you want a funny wet patch on your trouser seat.
Yeah, I'd like to see a Dyson Airblade deal with that.
No, on second thoughts I would not like to see that.
At about the same time we'd just got some DEC LN03 laser printers
I remember them. Worked great on the DEC paper, but our office went on a 'green' kick and bought boxes of made-from-recycled paper. The LN03s all started to jam within a few weeks, leading to new rollers & a warning from DEC not to use anything but their paper. Always seemed like clever design to me, and a trick that HP obviously missed.
All of my engagements to reverse-engineer any fortran code ended up in a complete re-write in C, as it was faster.
For you, maybe, but that just shows your lack of skills as a programmer.
That's why only computer scientists write good code
In my experience they write absolutely crap code.
Sure, it's elegant, well-structed, well-commented, and may even do what it's meant to, but since computer scientists know little or nothing about the hardware, and are generally terrified they might break it, the code tends to be mind-boggling slow and inefficient. It has little error handling, since they don't know what might go wrong, and is often prone to race conditions because they have no idea what's happening in the real world of the system. Resource management is non-existent, since resources are assumed to be either infinite, or somebody else's problem (Java garbage collection, anyone?).
A ghood programer can write fast, efficient, well-commented code without making it cryptic or unmaintainable. That can rarely be said of "computer scientists". In the real world, computing isn't a science, it's engineering.
I used to number my punched cards in pencil on the back just in case I dropped the stack. That was considered to be a useful innovation at the time...
That's why Fortran only uses columns 1-72, you punch sequence numbers in the last 8 columns (going up in 10s so you can insert missing cards later, just like you used to do for BASIC line numbers). Then when you drop the stack you can load them in any order and have the system reorder them for you.
I find it impressive that they managed to identify the cold caller behind a withheld number.
"Withheld" just means that the called person's exchange was asked not to present the number. The telco still has the number, and can release it when asked by some suitably-authorized person. Since this lowlife was running his companies in the UK he didn't even have the 'protection' of a faked international number to help him. Bet he'll not make that mistake again.
The fine is a joke, he'll just put it in the 'expenses' column of his cooked books and move on to the next scam.
If the software wasn’t ready – and it clearly wasn’t – it wouldn’t have been expensive to inform customers that the release date was being delayed. I doubt anyone would have cared. But by pushing it out the door unfinished they created such a fiasco...
I've had that argument so many times, sometimes with managers who listren, sometimes with ones who wont.
Late delivery of a project annoys the hell out of a customer, but if it works they soon forget. Deliver something on time but broken and they'll be reminded every day they have to use the stuff.
The article says "We're told 40ms of latency is approximately half way across the USA. Transmitting 1GB through an unaccelerated link with that latency took 5 mins 20 secs. WANrockIT accelerated it to 11.8 secs."
Since latency has no actual impact on one-way data transfer speed, any improvement is surely down to delaying acknowledgements, either by using large packets or large windows? Using large packets has a problem on noisy links, since even a small number of retries can kill performance. You're still subject to the laws of physics, and I suspect that a well-tuned TCP/IP stack with judicious packet and window sizes could match these figures, at least for a bulk transfer copy protocol.
When the round trip latency was increased to 360ms, simulating a longer WAN distance, the raw transmission time for 1GB was 42.6 mins while with WANrockIT it decreased to 12 seconds.
So they're using a bulk-transfer protocol with large window sizes, and possibly separate streams within the physical link? Is this really all that novel?
How does it compare with FTPing a tape image? Or even the classic "747 full of DVDs (or Bluerays)" ?
He's had his 15 minutes of fame, and trying to get notoriety as a fugitive hadn't worked out for him (although admittedly it's hard to play the romantic fugitive when the charges he's sought on include rape) so I suppose playing the "martyr" card was an obvious next step.
Sorry Julien, but you were last decade's leaker, Snowden is the fashion these days.
I do want to address the implication that the company was somehow conducting less than above-board operations. The word ‘scam’ was used
He then goes on to say how hurtful it was for this ex-employee to use that word. Nowhere that I can see does he get even close to denying the accusation though, let alone refuting it.
SLS's "Block 1" configuration, capable of lifting 70 metric tonnes. Ultimately, NASA hopes the evolved system will provide "an unprecedented lift capability of 130 metric tons"
50 years on, and they've only just managed to reinvent the Saturn V. I'll bet it costs a tad more, though.
we used to customize our product for individual customers, so we maintained a DAT tape of the source code specific to each customer.
Rarely a good idea. Over time as the codebases diverge even the simplest bug fix becomes harder and harder to to apply to each custom build, and takes longer each time. Eventually you end up doing nothing but housekeeping and no new development (I've seen it happen, just before the company in question ran out of money).
It's also risky, we once had a supplier that did this with #ifdef-type processing. When looking at a new code drop one day I saw code that was unrelated to our product. Not only that, it was clearly code for a competitor that I should not have seen. Relations with the supplier were very cordial, one phone call later produced a new code drop, and an embarrassed request to delete every copy of the old one and forget we ever saw it. Trust works both ways, so of course we did so, nothing more ever came of it.
Small teams do work better, just as small meetings do, but that has nothing to do with DevOps or its principles.
Have a small team do the design, and especially the inter-component interface design, and distribute that to a number of smal teams each of which owns a component. Have the testing also divided into small teams, each of which tests some sub-assembly. You'll get far better results than you would with a single large team trying to do everything, since in my experience such a team will keep "improving" the interfaces, inconsistently. Whether those teams use waterfall, DevOps, or any other methodology is largely irrelevant.
Of course, it takes good management to get those teams working together constructively. All too often "DevOps" is just an excuse for "nobody wants to manage this properly so we'll just let the smart guys do their own thing and cross our fingers". Certainly, with that approach "small teams" makes it easier to spot the fuckups early, but that's no reason to equate "small teams" to "DevOps"
In fact most of the time a twin is so over-powered to allow for the engine failure case that it has much higher performance after take-off than a 4-engine aircraft -
There are exceptions. I've watched a 747 take off at Heathrow, ambling down almost the whole runway so slowly and for such a long time that it seemed it was never going to get airborne. A minute or two later a Concorde entered the same runway. The usual earth-shaking roar, and what seemed like 1/4 of the runway later it pointed its nose into the sky and climbed effortlessly away. Hardly a fair comparison, I know, even though they did both have 4 engines.
I won't miss 747's at all, they certainly put the "coach" into "coach class". Concorde, though... :(
I remember something similar in a VMS Fortran program. VAXen had a block move instruction, somewhat similar in operation to the C memcpy() function. Someone (I think it might have been me, but I can't remember) got the source & destination addresses backwards. Instead of rotating a buffer by 4 bytes when that line was reached, it shifted the whole program address space *except* the buffer by 4 bytes. Neeedless to say the program crashed almost immediately, but the resultant stack trace was incomprehensible, since everything was 4 bytes off. It took ages to figure out what was happening.
Windows 10 Home-based machines accounted for 78 per cent of retail PC sales and the OS was running on 35 per cent of business machines
Truly "running on", or just "licensed on"?
It's true that W10 is driving desktop sales, I'm about to buy a new PC with W7 installed, while I still can. It'll come with a W10 license + downgrade rights, but W7 preinstalled. How will that show up in Context's figures?