Nah, it's just exaggeration. There's a sort of arms race that seems to happen as people want to stress something as out of the ordinary in some manner, then the term becomes popular, so even stronger terms need to be used for the out of the ordinary. Rinse and repeat.
This seems to be particularly common in the US. Here in the UK, we often use the expression 'for a long time', which is vague but transparent. In the US, the corresponding expression seems to be 'for the longest time', which is pretty meaningless (after all, the longest time is presumably that between the universe's creation and its eventual destruction, which is either infinite or at least a heck of a lot longer than people actually mean when they're talking about waiting in a queue for something...)
Misuse of 'literally' is another pet hate of mine, as is 'I could care less' to mean exactly the opposite. And don't get me started on bad punctuation...!
Nah. Why show a menu bar all the time? Do what the Psion series do and have a silkscreen button next to the screen to pop up the main menubar. All the benefit, for no screen space at all!
Similarly, there's no real need for a taskbar/dock when you're mainly using standard applications; silkscreen buttons are great for that too.
If you've not got much screen space, then you have to make every pixel count. Some things need a certain amount to be useful; e.g. scrollbars. But prune what's not needed, and it doesn't feel quite so cramped after all.
In future, stuff that never passes out of local scope will be put on the stack anyway. I believe JRE 1.6 will be including escape analysis, which allows it to do stuff like that automatically. And although mark-and-sweep may need the application stopped, I gather there are other GC algorithms which don't have that restriction.
My Beeb was the only machine I really understood inside-out. Its OS was incredibly well-designed, and they squeezed far more stuff into 16KB of ROM than you'd think possible. It was highly extensible; I blew my own EPROM with extra utilities, including a new screen mode. (Yes, that Advanced User Guide was very, very familiar! I had to get the hardback cover for it, coz the paper one was falling apart. Another great book had annotated disassembly of the ROMs.)
It had a phenomenal amount of software; I learned FORTH on it, and had Pascal and BCPL too. Tons of games, of course (the original and best version of Elite was born on the Beeb, as was Repton, for example), but also DTP and lots more. You could get a (three-button) mouse and a ROM that gave you a GEM- and Mac-like GUI interface and various apps.
But it was adding hardware where it excelled. My old Beeb contained a real-time clock, an expansion with 128KB of RAM and space for the same amount of ROM, a hardware speech synthesiser (at least until someone wrote one that worked entirely in software!), and was usually connected to a TV, disk drive, external sound system, analogue and digital joysticks, a music synthesiser, a music keyboard, a printer, and probably more that I've forgotten. Today that doesn't sound like much, but 20 years ago having stuff like that available and affordable wasn't common!
I wrote loads of music on it, a basic loop-based sequencer, games, educational software... Ah, the memories! I wish machines today were as accessible.
I tried an original Mighty Mouse for the first time last month. It's good, and I liked the scroll ball a lot. But one problem made it unusable for me and drove me back to my Logitech Cordless Optical Wheel Mouse: right-clicking. I do that a lot, and although the Mighty Mouse can be set to right-click when pressed on that side, it only seems to work if nothing's touching the other side too. And since I always rest my fingers on both sides, it kept left-clicking when I meant a right-click!
What do you do to a customer with a bot infected computer that he is probably not aware of? What preventative steps can you take that will not interfere with legitimate customer traffic?
How about making users responsible for the spam their PC relays, or the attacks their PC makes?
As a compromise, how about making users responsible, but only if their ISP isn't blocksing infected PCs? That way, users who "want direct access with nothing blocked, dammit" will have to make sure they keep their machines clean, or face the consequences; and most other users will happily accept the possibility of being blocked by their ISP to prevent them being sued. Either way, the rest of the net is safe, and people start taking infection a bit more seriously.
It's a bit like car insurance: if you have a large bank balance and don't want to involve anyone else, you can self-insure; most people play it safe and pay a premium to an insurer. Either way, any damage you do is covered.
Yeah, well, I too was once of the simple, dogmatic view. (That if you pronounce it as a word, it's an acronym; if you spell out the letters, it's an abbreviation.) But then I checked dictionaries.
(For example, Chambers says that an acronym is "usually pronounced as a word". COD similarly says "a word, usu. pronounced as such". Note the suggestive but far from prescriptive 'usually'.)
And what have you against 'DVD'? Does it not stand officially for 'Digital Versatile Disc'? (And originally for 'Digital Video Disc'?)
Yes, MS produces some crap software, but that's not the real problem. If that were all, then users would be free to choose other software, and the free market would decide; MS would either learn to produce genuinely better software, or lose customers.
The real problem has always been MS's unfair business practices: their tying of one market to another, their embracing and extending of formats and protocols, their strongarming suppliers into not providing any alternatives, and so on and so on. Fix THOSE, and the rest will sort itself out.
If these 12 steps actually address those, then they're a Good Thing(TM). (How far MS will actually act on them is another matter entirely, of course, but they might at least be using the right words.
My point is that criminals don't obey gun control laws
Some criminals don't obey gun control laws, true. But that's a very small 'some'; a tiny tiny proportion of them here in the UK. You certainly can't imply that every criminal you meet will have a gun. Whether that's because they can't get them, can't easily get them, don't want to risk carrying them, don't want to risk using them, or whatever, in general you won't be up against a firearm.
The gun control argument already has too many oversimplifications; please don't make it worse:)
In many cases, the HUMONGOUS BIG FONT price on the item is the price after the rebate -- not what you actually have to pay to the cashier.
Yeah, this is off-topic, but isn't that the way things are usually priced in the USA? Here in the UK, the price you show most prominently must be the one inclusive of VAT; it was a nasty shock when I visited the USA and found the cashier asking more than I expected, due to the sales tax.
But 'becoming the dominant OS' isn't Linux's only goal; in fact, it's probably not even the goal that would benefit most people.
As far as I'm concerned, the main goal should be to stop any OS becoming dominant to that level. When web designers, software houses, etc. can't just say "We write for <OS> only coz that's what the huge majority of people use," then they'll have to start considering cross-platform standards instead. And we all win: OS X users like me, Linux users like lots of you here, and yes, arguably Windows users too.
Maybe it was revised; it's an electronic version of, er, uncertain provenance, so I can't really tell. It's also interesting that through the entire canon of 14 books (all from the same source), I could only find that one mention of 'vodka martini'. Would make sense if it was simply one of many drinks he took in the books, and then in the films it became the single 'trademark', after, as you say, a sponsorship deal.
So you don't think it has anything to do with the fact that he drinks a vodka martini in Ian Fleming's original book of From Russia, With Love, six years earlier?
Even if you're right (and I suspect that in a large degree you are), that doesn't mean that virtualisation isn't useful; maybe even necessary.
Look at it this way: a decade or two ago, people would have been saying almost exactly the same things about pre-emptive multitasking, protected memory, and virtual memory. And they'd be right: pre-emption is less efficient than co-operative multitasking, and a memory manager does introduce extra complexity and delays. We should be capable of writing applications that co-operate properly. We should be able to write applications that never try to read or write outside their allocated memory.
But the fact is that we can't, as countless buffer-overflow exploits and dodgy drivers show. Inefficient as it is, providing processes with their own environment where they can pretend they have the memory and processor time all to themselves makes for a) simpler applications that are easier to write, and b) much safer machines, where application errors don't bring down the machine. In effect, it gives them their own virtual environment.
So pretty much the same arguments apply to virtualising OSs too: it shouldn't be necessary, but it makes the OSs simpler to write and safer to run and administer.
Agreed. However, I can understand why many people want to reject the big methodologies: because they've seen them applied wrongly.
A methodology is a framework for thinking, not a replacement for it. Too many people seem to think that a methodology lets you fill in the boxes, join the dots, perform the right steps, and Hey Presto! your project will be perfect. Some methodologies even seem to advertise themselves that way. And that's a recipe for over-long projects, pointless busywork, overengineering, and difficulty in changing anything.
Instead, I find it most useful to see a methodology as set of tools. Get to know them, pick the ones that work for you, and use them as you see fit.
You mentioned UML. Lots of people understand UML class diagrams, for example, which makes them very useful for communicating an OO design. But you still need to have a design worth communicating! Use cases are a great way of thinking about requirements and UI design, but you still need to do that thinking. And so on.
If you're taking time at the start of a project to do some good design work, and you happen to be using UML to help that, then fine; but I 'd be less happy at the idea of taking time at the start of a project just for the sake of 'doing the UML' merely because that's what you've been told to do.
This is one of the (few) points that really worries me. Microsoft have a huge pile of cash. And they've shown many times before that their market-entry strategy is simply to keep trying, keep pouring money in, and wear their opponents down. How long did it take before Windows was any good? IE? Word? In fact, any number of products, file format, protocols, etc.
Of course, that's no guarantee of success. But it would be dangerous to write off a product from someone with their cash reserves, determination, ruthlessness, and failure to understand the meaning of anti-trust legislation...
My gut feeling is that if it has to stand or fall on its own merits, it's doomed. But they'll find some way to tie it in with Windows,
make it easier to use that and harder to use an iPod or other device, and they'll dig in and keep pouring in cash, and in a year or two's time people might be wondering why anyone ever doubted it:(
Erm, you are aware that iPods play MP3s, and unprotected AAC files as well as the protected ones that the iTMS sells, aren't you?
In this hand, I hold my iPod, chock full of music despite never having been within yelling distance of the iTMS. In the other hand, I hold a cluestick. Take your pick:)
a far better example would be the instances of journalists repeatedly and successfully smuggling weapons through TSA security, onto commercial flights. Absolutely no real harm is done by it
Is that so? Hmmm. "No, officer, I'm a journalist, honest. I know I'm wearing a turban and have a foreign-sounding name, but I wasn't going to use these explosives strapped to my chest. It's just for a newspaper story. Sorry, what? 'Press card'? Er, no, I left that at home, sorry. But I really wasn't going to set these off. Honest."
Exactly the same applies to computer hacking. If an intruder was good enough, you might not be able to tell whether any serious damage was done; which means you must always assume the worst. Grey-hat hacking just doesn't seem justifiable.
Anyway, doesn't this whole story boil down to "We need hackers, because if there were no hackers we wouldn't know how to protect against, er, hackers..."?
Slashdot Mirror
This seems to be particularly common in the US. Here in the UK, we often use the expression 'for a long time', which is vague but transparent. In the US, the corresponding expression seems to be 'for the longest time', which is pretty meaningless (after all, the longest time is presumably that between the universe's creation and its eventual destruction, which is either infinite or at least a heck of a lot longer than people actually mean when they're talking about waiting in a queue for something...)
Misuse of 'literally' is another pet hate of mine, as is 'I could care less' to mean exactly the opposite. And don't get me started on bad punctuation...!
Thank you. I feel better now.
Similarly, there's no real need for a taskbar/dock when you're mainly using standard applications; silkscreen buttons are great for that too.
If you've not got much screen space, then you have to make every pixel count. Some things need a certain amount to be useful; e.g. scrollbars. But prune what's not needed, and it doesn't feel quite so cramped after all.
In future, stuff that never passes out of local scope will be put on the stack anyway. I believe JRE 1.6 will be including escape analysis, which allows it to do stuff like that automatically. And although mark-and-sweep may need the application stopped, I gather there are other GC algorithms which don't have that restriction.
My Beeb was the only machine I really understood inside-out. Its OS was incredibly well-designed, and they squeezed far more stuff into 16KB of ROM than you'd think possible. It was highly extensible; I blew my own EPROM with extra utilities, including a new screen mode. (Yes, that Advanced User Guide was very, very familiar! I had to get the hardback cover for it, coz the paper one was falling apart. Another great book had annotated disassembly of the ROMs.)
It had a phenomenal amount of software; I learned FORTH on it, and had Pascal and BCPL too. Tons of games, of course (the original and best version of Elite was born on the Beeb, as was Repton, for example), but also DTP and lots more. You could get a (three-button) mouse and a ROM that gave you a GEM- and Mac-like GUI interface and various apps.
But it was adding hardware where it excelled. My old Beeb contained a real-time clock, an expansion with 128KB of RAM and space for the same amount of ROM, a hardware speech synthesiser (at least until someone wrote one that worked entirely in software!), and was usually connected to a TV, disk drive, external sound system, analogue and digital joysticks, a music synthesiser, a music keyboard, a printer, and probably more that I've forgotten. Today that doesn't sound like much, but 20 years ago having stuff like that available and affordable wasn't common!
I wrote loads of music on it, a basic loop-based sequencer, games, educational software... Ah, the memories! I wish machines today were as accessible.
Does this new one fix that?
How about making users responsible for the spam their PC relays, or the attacks their PC makes?
As a compromise, how about making users responsible, but only if their ISP isn't blocksing infected PCs? That way, users who "want direct access with nothing blocked, dammit" will have to make sure they keep their machines clean, or face the consequences; and most other users will happily accept the possibility of being blocked by their ISP to prevent them being sued. Either way, the rest of the net is safe, and people start taking infection a bit more seriously.
It's a bit like car insurance: if you have a large bank balance and don't want to involve anyone else, you can self-insure; most people play it safe and pay a premium to an insurer. Either way, any damage you do is covered.
[fx: resists]
[fx: resists]
[fx: yields]
Yeah, well, I too was once of the simple, dogmatic view. (That if you pronounce it as a word, it's an acronym; if you spell out the letters, it's an abbreviation.) But then I checked dictionaries.
(For example, Chambers says that an acronym is "usually pronounced as a word". COD similarly says "a word, usu. pronounced as such". Note the suggestive but far from prescriptive 'usually'.)
And what have you against 'DVD'? Does it not stand officially for 'Digital Versatile Disc'? (And originally for 'Digital Video Disc'?)
(* I'll leave the argument about whether SEO is an acronym or merely an abbreviation for another day...)
Yes, MS produces some crap software, but that's not the real problem. If that were all, then users would be free to choose other software, and the free market would decide; MS would either learn to produce genuinely better software, or lose customers.
The real problem has always been MS's unfair business practices: their tying of one market to another, their embracing and extending of formats and protocols, their strongarming suppliers into not providing any alternatives, and so on and so on. Fix THOSE, and the rest will sort itself out.
If these 12 steps actually address those, then they're a Good Thing(TM). (How far MS will actually act on them is another matter entirely, of course, but they might at least be using the right words.
(Of course, you'd then need to write a program to write the programs...)
Some criminals don't obey gun control laws, true. But that's a very small 'some'; a tiny tiny proportion of them here in the UK. You certainly can't imply that every criminal you meet will have a gun. Whether that's because they can't get them, can't easily get them, don't want to risk carrying them, don't want to risk using them, or whatever, in general you won't be up against a firearm.
The gun control argument already has too many oversimplifications; please don't make it worse :)
Ah, but in the USA everyone is a victim of something!
Sounds a bit like the experiments that produced the Hawthorne Effect.
Erm... warehouses? Whalehouses? Whorehouses? You know those sailors...
Yeah, this is off-topic, but isn't that the way things are usually priced in the USA? Here in the UK, the price you show most prominently must be the one inclusive of VAT; it was a nasty shock when I visited the USA and found the cashier asking more than I expected, due to the sales tax.
As far as I'm concerned, the main goal should be to stop any OS becoming dominant to that level. When web designers, software houses, etc. can't just say "We write for <OS> only coz that's what the huge majority of people use," then they'll have to start considering cross-platform standards instead. And we all win: OS X users like me, Linux users like lots of you here, and yes, arguably Windows users too.
Maybe it was revised; it's an electronic version of, er, uncertain provenance, so I can't really tell. It's also interesting that through the entire canon of 14 books (all from the same source), I could only find that one mention of 'vodka martini'. Would make sense if it was simply one of many drinks he took in the books, and then in the films it became the single 'trademark', after, as you say, a sponsorship deal.
Look at it this way: a decade or two ago, people would have been saying almost exactly the same things about pre-emptive multitasking, protected memory, and virtual memory. And they'd be right: pre-emption is less efficient than co-operative multitasking, and a memory manager does introduce extra complexity and delays. We should be capable of writing applications that co-operate properly. We should be able to write applications that never try to read or write outside their allocated memory.
But the fact is that we can't, as countless buffer-overflow exploits and dodgy drivers show. Inefficient as it is, providing processes with their own environment where they can pretend they have the memory and processor time all to themselves makes for a) simpler applications that are easier to write, and b) much safer machines, where application errors don't bring down the machine. In effect, it gives them their own virtual environment.
So pretty much the same arguments apply to virtualising OSs too: it shouldn't be necessary, but it makes the OSs simpler to write and safer to run and administer.
[fx: checks the calendar]
It's now another day :)
A methodology is a framework for thinking, not a replacement for it. Too many people seem to think that a methodology lets you fill in the boxes, join the dots, perform the right steps, and Hey Presto! your project will be perfect. Some methodologies even seem to advertise themselves that way. And that's a recipe for over-long projects, pointless busywork, overengineering, and difficulty in changing anything.
Instead, I find it most useful to see a methodology as set of tools. Get to know them, pick the ones that work for you, and use them as you see fit.
You mentioned UML. Lots of people understand UML class diagrams, for example, which makes them very useful for communicating an OO design. But you still need to have a design worth communicating! Use cases are a great way of thinking about requirements and UI design, but you still need to do that thinking. And so on.
If you're taking time at the start of a project to do some good design work, and you happen to be using UML to help that, then fine; but I 'd be less happy at the idea of taking time at the start of a project just for the sake of 'doing the UML' merely because that's what you've been told to do.
Of course, that's no guarantee of success. But it would be dangerous to write off a product from someone with their cash reserves, determination, ruthlessness, and failure to understand the meaning of anti-trust legislation...
My gut feeling is that if it has to stand or fall on its own merits, it's doomed. But they'll find some way to tie it in with Windows, make it easier to use that and harder to use an iPod or other device, and they'll dig in and keep pouring in cash, and in a year or two's time people might be wondering why anyone ever doubted it :(
In this hand, I hold my iPod, chock full of music despite never having been within yelling distance of the iTMS. In the other hand, I hold a cluestick. Take your pick :)
Is that so? Hmmm. "No, officer, I'm a journalist, honest. I know I'm wearing a turban and have a foreign-sounding name, but I wasn't going to use these explosives strapped to my chest. It's just for a newspaper story. Sorry, what? 'Press card'? Er, no, I left that at home, sorry. But I really wasn't going to set these off. Honest."
Exactly the same applies to computer hacking. If an intruder was good enough, you might not be able to tell whether any serious damage was done; which means you must always assume the worst. Grey-hat hacking just doesn't seem justifiable.
Anyway, doesn't this whole story boil down to "We need hackers, because if there were no hackers we wouldn't know how to protect against, er, hackers..."?