However, he usually did have justification for most those programs. Writing code on one project, examining specs on word and excel formats, surfing in MSDN etc. Having his machine automatically "garbage collect" those apps would've meant that he'd been wasting half of his time waiting for those big project files load(WAY more than 5 seconds, no matter what the author of the article claims), or try to re-find the helpful web page he had had open five minutes ago.
The way I imagine it working, the system would only garbage collect programs that had no open windows. Having windows in the background mysteriously vanish because they haven't been activated in x minutes is far more user-hostile than having a "Quit" menu item!
I remember MS talking about this years ago, even before NT came out. They were calling it Cairo back then.
In the end, IIRC, the UI elements of the Cairo project were recycled into the Windows 95 shell and the Object File System concept disappeared entirely... until now, it seems.
Is no-one else disturbed at the short memories in the industry? I was at the launch of Visual Studio.NET in Ireland a few weeks back and there was a Microsoft goon waving a Tablet PC around his head like it was some completely new thing. I mentioned the Go Corp/ Windows for Pen Computing FUD from the early '90s to the guys I was there with and was met with blank stares.
[Open Source exists] because we came in and said there should be a platform that's identical with millions and millions of machines, and the BIOS of that should be open to everybody to use, and all the extensibility should be there.
Sorry, Bill, but that doesn't hold up. Anyone else here remember the DEC Rainbow? The Rainbow was an MS-DOS machine but it wasn't PC-compatible. There were a few machines like this in the early '80s, but they were displaced when the true clones appeared. It seems to me that the early vision for MS-DOS was for it to become the Unix of the microcomputer world: a common API that ran on a number of different architectures where porting applications was (theoretically, anyway) a recompile away. The fact that MS fought IBM for the right to sell DOS to OEMs bears this out.
But the IBM PC succeeded, not because of Microsoft, but because it had IBM written on it and that made the suits all tingly inside.
One could also point to those early MS products like Multiplan (the forerunner of Excel) that were written in some sort of pseudocode so they could be easily ported to different micros. I'm sure Bill would claim that these were compromises in the Great Vision (computer on every desk yadda yadda) and that what he wanted all along was for the PC to succeed. But MS didn't care what machine succeeded, so long as there was MS software running on it and their early strategy of backing every horse in the running demonstrates this.
Re:On correct use of apostrophes
on
God's Debris
·
· Score: 1
I think this is a simple matter of style. Some style guides say you should always put 's after a word, even if it ends in s, some just go with the single apostrophe.
Both conventions have their advocates and their detractors. Advocates of the first convention claim their convention is more correct, while advocates of the second reckon theirs looks more elegant. The pragmatic view is: pick whatever one takes your fancy and use it consistently.
There's good site called The Slot that's aimed at copywriters and discusses this sort of stuff regularly. Also, I've found this book useful; the cutesy pictures of the author's cats scattered through the text make it a somewhat aggrevating read for this geek though...
FYI, Carl Sagan also presents a refutation of Velikovsky's theories in Broca's Brain.
It's been a while since I last read it, but here are a few of Sagan's argument that I remember off-hand:
The energy required to eject a planetary mass from Jupiter's gravity well would be more than enough to vapourise said mass. Also, an ejection event of this size is likely to produce a quantity of bodies, some of which ought to still be raining down on us.
The escape velocity from the Jovian system is very close to the escape velocity of the solar system as a whole. Presumably, if a planet-sized body somehow managed to be ejected from Jupiter without being melted, it is more likely to go flying off into deep space than settle down into orbit (an orbit, furthermore, with one of the lowest eccentricities of any body in the Solar System) around the Sun. The whole "Venus born of Jupiter's brow" shtick is an over-literal (and somewhat forced, IMHO; wasn't it Athene/Minerva who was born from Zeus/Jupiter's brow, not Aphrodite/Venus?) interpretation of Greek myth. An alternative, and somewhat more plausible explanation for this myth can be found here.
The near approaches of Venus to Earth with the consequent slowing of the Earth's rotation violates the law of conservation of angular momentum. Also, the circularisation of Venus' orbit after these transits doesn't jibe with what we know about gravity, tidal effects, etc.
There's other objections too-- I think Sagan has about ten-- but those are the ones I remember.
We're at war with Microsoft?! Holy cow, no-one ever tells me anything!
Free clue to all would-be Web journos: when Linux made that comment about "world domination" all those years ago: he was joking. Really. He was pulling your wire. Jerking your chain. Taking the Michael. Extracting the urine.
'Kay? 'Kay.
No doubt it's a waste of brains and time to even bother refuting this windy gibberish, I'd like to make a few points.
No-one has made money out of Linux and everyone who tries goes to the wall. To paraphrase Bill Hicks: non-Linux businesses go to the wall every day. Bob Cringely has reckoned that 90% of all businesses fail. The Linux has no innate monopoly on business smarts.
A sizeable population of Linux advocates are foul-mouthed social inadequates. Again, so what? I had invective-laden ZX Spectrum/ C64 flamewars with my mates when I was eight years old. While there is a human race, there will always be bigots. Is it impolite? Yes. Is it unprofessional? Surely. Does it amount to two tugs of a dead dog's mickey in the long run? Nope.
It's a war between Microsoft and Linux out there. Oh get a grip, you solipsistic little nonentity. Try to see beyond the VDU on your desk for a minute; in the light of recent events, your inflation of a trend within the IT sector to the status of a war are laughable and tasteless. Sure, there are the windbags on both sides of this MS-vs-Linux thing who read earth-shattering importance into everything, who think installing Linux on their PCs is some sort of subversive act. Nonsense.
I use Linux because it suits my needs. I also use Windows and MacOS. I don't feel any desire to conquer the world. I don't feel like I'm part of some "war for the desktop". No sane person does.
One wonders why WiReD bothered printing this giddy nonsense in the first place. Could it be that no self-respecting techie reads WiReD even though it likes to think of itself as the official organ of tech culture? Is that acid green they favour in their layouts really the colour of sour grapes?
That's all good, but when I'm selling a solution, and they ask about support, what am I going to tell them? "It's support by some geek living in his parents basement" won't fly.
And this is worse than the overall quality of paid Tech Supported exactly how? If you're building a system based on proprietary software and your client asks you about support, honesty dictates that you tell them it's supported by a gang of low-wage graduates living in their parents' basements who get paid to sit in a cube farm all day scratching their arses.
Sure, you'll get the occasional Tech Support person who is courteous and knowledgeable, but most of them are clueless. And who can blame them? They're paid peanuts to be screamed at by irate customers all day. Not my idea of a challenging job. Even if they wanted to know how exactly the product they're supporting really works, they wouldn't be able to find out, because that information is proprietary even within the company itself. How many people in the world do you think really know what happens in Windows 9x between seeing the `Starting Windows 9x' prompt appear on the console and getting your network login prompt? A dozen? Two dozen? Whoever they are, these people aren't sitting in Microsoft Tech Support.
Better to have some geek interested in his work in your corner than some poor wage slave who's just had the ears burned off of her by some snot-nosed MBA because his machine has eaten his PowerPoint slides. But try telling that to the penny-wise-pound-foolish specimens that sign the POs...
Not always the case. I'm a VB programmer and I once had to work with a C++ programmer (or at least, a C coder in a new crunchy OO shell) who simply couldn't get his head around the whole `write a program to write a program' concept, e.g. write a script to spit out a lookup table or whatever. To me, that screamed "cookbook coder".
Ironic as he was using Visual Studio, which is little more than a fancy code generator with a compiler on the back.
I must say that the direction keyboards has gone to, from the clickyclick '80s keyboard to the unusable cheap devices available now, could be one of the sources of RSI.
Right on. I wouldn't spit on most of the keyboards you get with PCs these days. At home, I use an old IBM PS/2 keyboard-- you know, one of those big heavy things that make the nice satisfying clicking noise when you type on them? Found it in a skip (dumpster) out the back of my company's offices. My company still buys its machines from Big Blue, but the keyboards you get with them nowadays aren't worth a curse.
<OLD_FART>
It's not just keyboard quality that's gone down the toilet though. Ten, twelve years back, a name-brand PC would have at least a reasonably good keyboard, a fairly rugged case and a power supply with enough leftover watts to power a small oven, even with all slots full (hey, am I the only one here who's wanted to fit a grill into one of those spare drive bays? I reckon they're wide enough for a good-sized steak!) Nowadays, the keyboard is a joke, the case will crack if you look at it funny, and the power supply is an anaemic piece of shit that barely satisfies the processor's appetite, never mind the cards'.
</OLD_FART>
Personally, I blame Microsoft and Intel (as a/. reader, could I do otherwise?:-). By the time the manufacturer has paid for a grossly overspecced processor and paid the Microsoft Tax, there's nothing left to spend on the peripherals.
The QWERTY layout originally separated keys that are often typed together to minimise the risk of jamming the mechanism. This has actually proved to be a good design; it means that while you type a letter with one hand, your other hand is already acquiring the next one.
My favourite (snort) bug in Word is a problem with page numbering. Sometimes Word won't update its "number of pages" field automatically, so you often get printouts with "Page 1 of 1", "Page 2 of 1", "Page 3 of 1", etc. in the footers. This bug has been in every single version of Word for Windows since at least version 2.
In my experience, the reason people here want use word and excel is for the library of graphics. People love to add fluffy little carton characters to their domunents. Everytime I suggest someone use AbiWord or Gnumeric, they are fine for a bit, but then they get stuck when they click -Insert graphic- and there is no preview catalog of silly images.
That's one of the most insightful things I've ever heard said about MS Office, i.e. that the programs that make up Office are nothing more than glorified turd-polishers! Man, if I only had some mod points...
Sure, Word has great features for indexing, proofing, etc., and Excel has some pretty nice data-massaging tools like the Solver, but how often are these features used? Instead, these products are used to give some superficial "professionalism" to the gibberish of the dumb and half-literate. And don't get me started on PowerPoint; the mind boggles at how much productivity has been lost while junior managers everywhere give every slide in their presentations its own special effect...
I used to find myself in the position of being a guru to friends, family, and neighbours. I'm very reluctant to do so now.
The big problem is complexity, as a lot of other posters have pointed out. Hardware is sourced from all over the place; the system manufacturers go for the bits that give them the biggest margins, and damn the quality of the accompanying documentation and drivers. Then there's the software. A lot of modern software is unforgivably arrogant, sending its tentacles into parts of the system that it has no business going near and demanding the lion's share of the computer's resources. The result is machines that are constantly teetering on the brink of meltdown.
Now, I'm in this business because I like messing about with computers, but I got sick to the back teeth of sitting in friends' and neighbours' bedrooms, interminably rebooting their balky machines, hunting for drivers, and re-installing Windows, while the person I'm supposedly doing the favour for hovers over my shoulder, sending out vibes that this is all somehow my fault. Eventually, I had to stop. It wasn't worth the heartache.
What pisses me off about the whole thing is that people like me are in a way responsible for this whole cock-up. We're the early adopters who played with the first personal computers in the '80s and told anyone who would listen that computers were The Next Big Thing. It was our evangelism that made the fortunes of companies like Microsoft and IBM. And it is our unpaid tech support, in our roles as gurus, that sustains their fortunes. Think about it: each time you unwedge your next-door neighbour's Windows box, that's one less irate customer onto Microsoft or Corel or IBM, flaming them for the fragile, barely-usable crud they have been inflicting on customers for years. Of course, the execs of these companies have built themselves a nice thick insulating layer of minimum-wage phone jockeys between themselves and their customers so they never have to listen to the anguish.
So, if any friends, neighbours, or relatives ask me to look at their computers anymore, I decline politely. If they persist, I ask to be paid. People do get snotty at that, but I'm fucked if I'm going to let the assholes in Microsoft or Compaq off the hook by providing free technical support for their customers.
Gurus of the world unite! You have nothing to lose but your chains. And, um, most of your friends.:-)
Uh, am I thinking of the same Comal here? The Comal I remember was this crocky BASIC/ Pascal hybrid that had a cult following among academics in Europe for a while.
Either you are smoking some serious rock, Anonymous, or this has to be the Mother of all Trolls...
Let me get this straigt[sic]. You are saying Microsoft should have strongarmed even more and made people make computers to certain standards. Damn, no wonder Microsoft can't win. One minute they are being critized for having questionable practices and then the next being told they should have gone further.
Call it strong-arming if you want. I call it taking responsibility for their platform. We could've had plug and play years ago if MS had had the balls to get on the OEMs' case. We're not talking about putting a gun to anyone's head here. A few rules like, `don't make cards that run on a fixed IRQ or DMA channel' or `always fully decode the address lines' would've gone a long way towards making PC hardware a bit more reliable.
Instead, MS were happy to take the money and run; each cheap-ass PC clone ran MS-DOS, so MS were happy as pigs in shite-- even if their customers weren't.
... The best case you may make is that the system crashes less often than similar MS-based systems, but they are not YOUR system, they will not be identical,...
Well, that's another-- perhaps slightly off-topic-- issue; Windows' habit of running fine on one machine and crashing like a motorway pile-up on another, seemingly identical, machine. Perhaps this is a problem that could be added to the list of misfeatures that cause `demonstrable harm' to customers; who can tell how many man-centuries have been wasted trying to debug mysterious hardware conflicts on Windows machines?
Now you could argue that this is the customer's own fault, buying some shadowy no-name PC clone and expecting it to work fine. But occassionally you'll get a batch of PCs from one of the big-name vendors that simply won't work reliably. 'Course, it's usually eventually tracked back to a last-minute substitution of some key component like the video card in an attempt to shave a few cents off the list price, but Microsoft have allowed this to happen. They could've laid down standards on how PC hardware ought to operate. Instead, they've entered into a silent contract to support anything any random soldering-iron-wielding ding-dong chooses to plug into the PC bus. "What's that? You've bought a no-name sound card made by the Dr. Mbogo Multimedia Corporation? It's made from string and old coathangers, you say? Sure, slap it in and our `Add New Hardware' wizard will figure it out!" In your dreams, pal.
They have further exacerbated this deplorable situation in two ways:
By taking a bigger and bigger chunk of the price of a PC in their OS Tax, manufacturers are forced to cut quality on their components to keep their margins; hence, more Brand Mbogo cards even in the high-end boxes (and don't get me started on shoddy case build quality and weak-ass power supplies!)
The mechanisms they use for hardware detection during system start-up are not documented. What exactly happens during the Windows 9x boot sequence? How many people working for Microsoft really understand it? A dozen? Less? Certainly no-one working in Tech Support knows for sure, so they trot out scripted solutions like, `re-install the OS'.
They saw sense with NT and came up with hardware compatibility lists, but this is a case of `too little, too late'.
I have an old 486 at home that's running twm as its window manager. Twm is pretty `themeable' and it's bloody ancient. You can specify things like the number of buttons that appear on the title bar and the bitmaps (monochrome only!) used to draw them.
I see no justification in Apple being awarded this patent. Ironically, the MacOS has had a thriving niche market in third-party UI-patching utilities for years, e.g. Aurora, Kaleidoscope, Greg's Buttons, Aaron, and countless others before Apple themselves got in on the act.
Almost nothing you cite is open source. Almost everything you cite is the work of private initiatives or corporate consortiums.
<takes deep breath; exhales loudly> What's this got to do with anything? What part of "open source" don't you get? The source code of these programs is openly available and open to modify! Hence: Open. Source. Hello?! McFly?!
I figure you've got this mental image of "open source" as being the product of the übergeeky, hunched over their machines into the small hours, surrounded by a debris of fast food wrappers and mouldy styrofoam coffee cups. Open Source has nothing to do with this stereotype.
That these programs were developed by "private initiatives" and "corporate consortiums" (sic) is irrelevant. Whether these programs were written by the weird on extreme coffee jags or by 9-to-5 MIS minions in suits is irrelevant. Whether these developments were funded by large corporations or done on one's own dollar is irrelevant. The point is, they've made the source code publically available. These "corporate consortiums" have given something back to the community. They have put their money where their mouths are for open standards. They have not locked up their source code behind NDAs and attempted to pervert open protocols and justified this antisocial behaviour by claiming the right to "innovate" like Microsoft have done. These organisations should be applauded for their public spirit, not sneered at.
Re:I'd rather read about Microsoft
on
Rebel Code
·
· Score: 1
VMS is a monolithic, architecture-dependent operating system written in VAX assembly language
IIRC, the core of VMS is written in BLISS, which was a sort of portable assembly language. BLISS abstracted away some of the machine specifics so programs could be ported between the different Digital machine architectures (PDP, VAX, etc.) Most the libraries that make up the VMS runtime environment are written in higher-level languages.
In what language is OpenVMS written?
OpenVMS is written in a wide variety of languages.
In no particular order, OpenVMS components are implemented using Bliss, Macro,Ada, PLI, VAX and DEC C, Fortran, UIL, VAX and Alpha SDL, Pascal, MDL, DEC C++, DCL, Message, and Document. And this is certainly not a complete list. However, the rumor is not true that an attempt was made to write pieces of OpenVMS in every supported language so that the Run-Time Libraries could not be unbundled. (APL, BASIC, COBOL and RPG are just some of the languages not represented!)
So your assertion that VMS was programmed "on the metal" in assembly language is half-correct. The fact that VMS requires exotic hardware with multiple privilege levels... well, Digital were designing both the hardware and the OS, so it's not surprising there are strong ties between the VAX architecture and the VMS OS. Portability to non-Digital hardware was not an issue when VMS was conceived.
I think what the original poster had in mind is that NT and VMS are similar conceptually (not surprising as one man, Dave Cutler, was responsible for both).
NT is a portable microkernel written in C,
Yadda yadda yadda, we all read this breathless marketroid shite back in 1991 when NT first appeared. If NT is so damn portable, why has it bombed on every platform except x86?
And, IMNSHO, NT ceased to be a "microkernel" when they moved the GDI into kernel space just to squeeze a few more cycles out of the GUI.
This is what makes it so easy to port the entire NT-based Windows OS to new platforms, using a single source tree.
AFAIK, the porting strategy for Linux goes like this:
Fork the kernel source
Tweak what needs to be tweaked to get Linux running on the new architecture
Submit the changes back to the main kernel tree
This is undesirable how?
(in contrast to Linux, where 'ports' tend to use independent source trees, and rarely constitute complete systems)
Depends on what you define as a "complete" system. A lot of these ports are special-purpose projects (e.g. embedded systems), in which case only a fraction of the full functionality of a desktop or server Linux system is required. Not much point in porting a GUI to a system that's intended to be used in a car's fuel injection system, you know.
But good coding style is a very impersonal thing. Code should communicate. Code done in a "personal style" is incomprehensible and IMHO often broken.
When I talked about coding style, I wasn't just referring to badly-indented horrors knocked out by a student who thinks it's kewl to name all his variables after Star Wars characters. Even with coding standards in place, most code does have a "voice" or a "personality" because each developer favours a particular set of language idioms. For example, developer A might always couch his loop constructs in a `repeat while' idiom, while developer B prefers `repeat until' constructions. Things like that.
But the learning curve is the biggest obstacle. Really grokking a program often involves figuring out what was left out, rather than understanding what the programmer put in. So much information is lost in the translation from the programmer's mental model to finished code. Using a production-quality, fully-debugged library is like being handed a chisel and a block of marble and being asked to produce Michelangelo's "David". Most people will want to start with Play-Doh and work their way up to marble eventually.
Why else would someone write a message board when ten already exist? To scratch an itch? Partially, but that only explains why they write it, not why they gift it. They gift it in order to win the acclaim of their peers.
That's not necessarily the case. I reckon the biggest problem is the learning curve. Sure, libraries to do things like message boards are ten a penny from Freshmeat, but what you're left with after you've unpacked the tarball is a big chunk of code designed to a mental model that may or may not be in phase with yours. Digesting this code will be a problem. Even if it's well written, it'll take time. More likely than not-- coding style being a very personal thing-- you'll find some coding idiom that will make you cringe and slow down your absorption. When faced with this, is it any wonder that most people will throw their hands in the air, fire up their editor, and start coding from scratch?
Something I've always wanted was a `time machine' for source code and other documents. Imagine some sort of super source code control system (perhaps integrated into the filesystem so you don't have to set anything up) fitted with a gizmo like a VCR shuttle control, so you can `rewind' the source code through earlier iterations. That way you can see the developer's mental model of the system evolving over time. I would kill for such a system. Well, maim anyway. (-:
The problem Microsoft has is, sure, 90% of the world runs their stuff, but that 90% doesn't give a toss. Your average Microsoft user is an office worker who sits down in the morning, turns on their computer, fires up Word and Excel or whatever and gets on with their job. To them, the computer is just a tool; a tool that is handed down from above to them, and, while frequently balky and unreliable, works sufficiently well to get their job done. If there were no computers, their job would be exactly the same, except they would be using typewriters and paper spreadsheets or whatever.
In other words, Microsoft is successful, but it's not the sort of success that Bill Gates or his executives want. Microsoft occupies the same space in the minds of its customers as a traditional energy utility, rather than the bringers of hacktastic innovation to the masses that Bill and the boys would like to be seen as. Electricity? Plug your appliance into a wall socket; it's there. Water? Turn on the tap; it's there. Microsoft software? Switch on your machine; it's there. Ho hum.
No wonder poor Bill and Steve and the rest are so mad! They've very nearly realised their original goal of a computer on every desk running Microsoft software, but practically no-one cares they're running Microsoft software. The only time anyone cares is when it bluescreens for the nth time that day and come 5pm they clock off and go home. No, the only ones that do care are long-haired leftie weirdos who read Slashdot-- and they hate Microsoft and bitch about the technical inelegance of their products! The one segment of the computer-using population that still cares about OSes and APIs and gnarly hacks thinks Bill's software sucks! That's gotta hurt. Bill wants to be a Thomas Edison; not the faceless CEO of some electricity or gas utility. If not, then why all the endless carping about `innovation' during the trial?
Microsoft had better admit this truth pretty quickly-- that their perception of themselves is grossly out of kilter with how they are perceived in the Real World by their customers. C'mon, Bill, there are worse things than being merely adequate; not everyone can be a trailblazer. If they don't wise up, the whole Microsoft edifice will disintegrate, with Bill screaming to the end about `innovation' and the `right to compete', even as the orderlies from Happy Acres Home for the Incorrigibly Bewildered bust down the door of his office, restraints and tasers in hand. Because remember, kids, the paranoid do survive, but only because we lock them up so they can't hurt themselves.
Slashdot Mirror
The way I imagine it working, the system would only garbage collect programs that had no open windows. Having windows in the background mysteriously vanish because they haven't been activated in x minutes is far more user-hostile than having a "Quit" menu item!
Let the system handle it. When it's idle, it can clean the processes up itself. Think of garbage collection under Java or flushing a disk cache...
I remember MS talking about this years ago, even before NT came out. They were calling it Cairo back then.
In the end, IIRC, the UI elements of the Cairo project were recycled into the Windows 95 shell and the Object File System concept disappeared entirely... until now, it seems.
Is no-one else disturbed at the short memories in the industry? I was at the launch of Visual Studio.NET in Ireland a few weeks back and there was a Microsoft goon waving a Tablet PC around his head like it was some completely new thing. I mentioned the Go Corp/ Windows for Pen Computing FUD from the early '90s to the guys I was there with and was met with blank stares.
Guys, can we please have a YHBT. YHL. HAND moderation option? I can't mod this post otherwise...
Sorry, Bill, but that doesn't hold up. Anyone else here remember the DEC Rainbow? The Rainbow was an MS-DOS machine but it wasn't PC-compatible. There were a few machines like this in the early '80s, but they were displaced when the true clones appeared. It seems to me that the early vision for MS-DOS was for it to become the Unix of the microcomputer world: a common API that ran on a number of different architectures where porting applications was (theoretically, anyway) a recompile away. The fact that MS fought IBM for the right to sell DOS to OEMs bears this out.
But the IBM PC succeeded, not because of Microsoft, but because it had IBM written on it and that made the suits all tingly inside.
One could also point to those early MS products like Multiplan (the forerunner of Excel) that were written in some sort of pseudocode so they could be easily ported to different micros. I'm sure Bill would claim that these were compromises in the Great Vision (computer on every desk yadda yadda) and that what he wanted all along was for the PC to succeed. But MS didn't care what machine succeeded, so long as there was MS software running on it and their early strategy of backing every horse in the running demonstrates this.
I think this is a simple matter of style. Some style guides say you should always put 's after a word, even if it ends in s, some just go with the single apostrophe.
Both conventions have their advocates and their detractors. Advocates of the first convention claim their convention is more correct, while advocates of the second reckon theirs looks more elegant. The pragmatic view is: pick whatever one takes your fancy and use it consistently.
There's good site called The Slot that's aimed at copywriters and discusses this sort of stuff regularly. Also, I've found this book useful; the cutesy pictures of the author's cats scattered through the text make it a somewhat aggrevating read for this geek though...
FYI, Carl Sagan also presents a refutation of Velikovsky's theories in Broca's Brain.
It's been a while since I last read it, but here are a few of Sagan's argument that I remember off-hand:
There's other objections too-- I think Sagan has about ten-- but those are the ones I remember.
We're at war with Microsoft?! Holy cow, no-one ever tells me anything!
Free clue to all would-be Web journos: when Linux made that comment about "world domination" all those years ago: he was joking. Really. He was pulling your wire. Jerking your chain. Taking the Michael. Extracting the urine.
'Kay? 'Kay.
No doubt it's a waste of brains and time to even bother refuting this windy gibberish, I'd like to make a few points.
No-one has made money out of Linux and everyone who tries goes to the wall. To paraphrase Bill Hicks: non-Linux businesses go to the wall every day. Bob Cringely has reckoned that 90% of all businesses fail. The Linux has no innate monopoly on business smarts.
A sizeable population of Linux advocates are foul-mouthed social inadequates. Again, so what? I had invective-laden ZX Spectrum/ C64 flamewars with my mates when I was eight years old. While there is a human race, there will always be bigots. Is it impolite? Yes. Is it unprofessional? Surely. Does it amount to two tugs of a dead dog's mickey in the long run? Nope.
It's a war between Microsoft and Linux out there. Oh get a grip, you solipsistic little nonentity. Try to see beyond the VDU on your desk for a minute; in the light of recent events, your inflation of a trend within the IT sector to the status of a war are laughable and tasteless. Sure, there are the windbags on both sides of this MS-vs-Linux thing who read earth-shattering importance into everything, who think installing Linux on their PCs is some sort of subversive act. Nonsense.
I use Linux because it suits my needs. I also use Windows and MacOS. I don't feel any desire to conquer the world. I don't feel like I'm part of some "war for the desktop". No sane person does.
One wonders why WiReD bothered printing this giddy nonsense in the first place. Could it be that no self-respecting techie reads WiReD even though it likes to think of itself as the official organ of tech culture? Is that acid green they favour in their layouts really the colour of sour grapes?
And this is worse than the overall quality of paid Tech Supported exactly how? If you're building a system based on proprietary software and your client asks you about support, honesty dictates that you tell them it's supported by a gang of low-wage graduates living in their parents' basements who get paid to sit in a cube farm all day scratching their arses.
Sure, you'll get the occasional Tech Support person who is courteous and knowledgeable, but most of them are clueless. And who can blame them? They're paid peanuts to be screamed at by irate customers all day. Not my idea of a challenging job. Even if they wanted to know how exactly the product they're supporting really works, they wouldn't be able to find out, because that information is proprietary even within the company itself. How many people in the world do you think really know what happens in Windows 9x between seeing the `Starting Windows 9x' prompt appear on the console and getting your network login prompt? A dozen? Two dozen? Whoever they are, these people aren't sitting in Microsoft Tech Support.
Better to have some geek interested in his work in your corner than some poor wage slave who's just had the ears burned off of her by some snot-nosed MBA because his machine has eaten his PowerPoint slides. But try telling that to the penny-wise-pound-foolish specimens that sign the POs...
Not always the case. I'm a VB programmer and I once had to work with a C++ programmer (or at least, a C coder in a new crunchy OO shell) who simply couldn't get his head around the whole `write a program to write a program' concept, e.g. write a script to spit out a lookup table or whatever. To me, that screamed "cookbook coder".
Ironic as he was using Visual Studio, which is little more than a fancy code generator with a compiler on the back.
Right on. I wouldn't spit on most of the keyboards you get with PCs these days. At home, I use an old IBM PS/2 keyboard-- you know, one of those big heavy things that make the nice satisfying clicking noise when you type on them? Found it in a skip (dumpster) out the back of my company's offices. My company still buys its machines from Big Blue, but the keyboards you get with them nowadays aren't worth a curse.
<OLD_FART>
It's not just keyboard quality that's gone down the toilet though. Ten, twelve years back, a name-brand PC would have at least a reasonably good keyboard, a fairly rugged case and a power supply with enough leftover watts to power a small oven, even with all slots full (hey, am I the only one here who's wanted to fit a grill into one of those spare drive bays? I reckon they're wide enough for a good-sized steak!) Nowadays, the keyboard is a joke, the case will crack if you look at it funny, and the power supply is an anaemic piece of shit that barely satisfies the processor's appetite, never mind the cards'.
</OLD_FART>
Personally, I blame Microsoft and Intel (as a /. reader, could I do otherwise? :-). By the time the manufacturer has paid for a grossly overspecced processor and paid the Microsoft Tax, there's nothing left to spend on the peripherals.
The QWERTY layout originally separated keys that are often typed together to minimise the risk of jamming the mechanism. This has actually proved to be a good design; it means that while you type a letter with one hand, your other hand is already acquiring the next one.
My favourite (snort) bug in Word is a problem with page numbering. Sometimes Word won't update its "number of pages" field automatically, so you often get printouts with "Page 1 of 1", "Page 2 of 1", "Page 3 of 1", etc. in the footers. This bug has been in every single version of Word for Windows since at least version 2.
That's one of the most insightful things I've ever heard said about MS Office, i.e. that the programs that make up Office are nothing more than glorified turd-polishers! Man, if I only had some mod points...
Sure, Word has great features for indexing, proofing, etc., and Excel has some pretty nice data-massaging tools like the Solver, but how often are these features used? Instead, these products are used to give some superficial "professionalism" to the gibberish of the dumb and half-literate. And don't get me started on PowerPoint; the mind boggles at how much productivity has been lost while junior managers everywhere give every slide in their presentations its own special effect...
I used to find myself in the position of being a guru to friends, family, and neighbours. I'm very reluctant to do so now.
The big problem is complexity, as a lot of other posters have pointed out. Hardware is sourced from all over the place; the system manufacturers go for the bits that give them the biggest margins, and damn the quality of the accompanying documentation and drivers. Then there's the software. A lot of modern software is unforgivably arrogant, sending its tentacles into parts of the system that it has no business going near and demanding the lion's share of the computer's resources. The result is machines that are constantly teetering on the brink of meltdown.
Now, I'm in this business because I like messing about with computers, but I got sick to the back teeth of sitting in friends' and neighbours' bedrooms, interminably rebooting their balky machines, hunting for drivers, and re-installing Windows, while the person I'm supposedly doing the favour for hovers over my shoulder, sending out vibes that this is all somehow my fault. Eventually, I had to stop. It wasn't worth the heartache.
What pisses me off about the whole thing is that people like me are in a way responsible for this whole cock-up. We're the early adopters who played with the first personal computers in the '80s and told anyone who would listen that computers were The Next Big Thing. It was our evangelism that made the fortunes of companies like Microsoft and IBM. And it is our unpaid tech support, in our roles as gurus, that sustains their fortunes. Think about it: each time you unwedge your next-door neighbour's Windows box, that's one less irate customer onto Microsoft or Corel or IBM, flaming them for the fragile, barely-usable crud they have been inflicting on customers for years. Of course, the execs of these companies have built themselves a nice thick insulating layer of minimum-wage phone jockeys between themselves and their customers so they never have to listen to the anguish.
So, if any friends, neighbours, or relatives ask me to look at their computers anymore, I decline politely. If they persist, I ask to be paid. People do get snotty at that, but I'm fucked if I'm going to let the assholes in Microsoft or Compaq off the hook by providing free technical support for their customers.
Gurus of the world unite! You have nothing to lose but your chains. And, um, most of your friends. :-)
Uh, am I thinking of the same Comal here? The Comal I remember was this crocky BASIC/ Pascal hybrid that had a cult following among academics in Europe for a while.
Either you are smoking some serious rock, Anonymous, or this has to be the Mother of all Trolls...
Call it strong-arming if you want. I call it taking responsibility for their platform. We could've had plug and play years ago if MS had had the balls to get on the OEMs' case. We're not talking about putting a gun to anyone's head here. A few rules like, `don't make cards that run on a fixed IRQ or DMA channel' or `always fully decode the address lines' would've gone a long way towards making PC hardware a bit more reliable.
Instead, MS were happy to take the money and run; each cheap-ass PC clone ran MS-DOS, so MS were happy as pigs in shite-- even if their customers weren't.
Well, that's another-- perhaps slightly off-topic-- issue; Windows' habit of running fine on one machine and crashing like a motorway pile-up on another, seemingly identical, machine. Perhaps this is a problem that could be added to the list of misfeatures that cause `demonstrable harm' to customers; who can tell how many man-centuries have been wasted trying to debug mysterious hardware conflicts on Windows machines?
Now you could argue that this is the customer's own fault, buying some shadowy no-name PC clone and expecting it to work fine. But occassionally you'll get a batch of PCs from one of the big-name vendors that simply won't work reliably. 'Course, it's usually eventually tracked back to a last-minute substitution of some key component like the video card in an attempt to shave a few cents off the list price, but Microsoft have allowed this to happen. They could've laid down standards on how PC hardware ought to operate. Instead, they've entered into a silent contract to support anything any random soldering-iron-wielding ding-dong chooses to plug into the PC bus. "What's that? You've bought a no-name sound card made by the Dr. Mbogo Multimedia Corporation? It's made from string and old coathangers, you say? Sure, slap it in and our `Add New Hardware' wizard will figure it out!" In your dreams, pal.
They have further exacerbated this deplorable situation in two ways:
They saw sense with NT and came up with hardware compatibility lists, but this is a case of `too little, too late'.
I have an old 486 at home that's running twm as its window manager. Twm is pretty `themeable' and it's bloody ancient. You can specify things like the number of buttons that appear on the title bar and the bitmaps (monochrome only!) used to draw them.
I see no justification in Apple being awarded this patent. Ironically, the MacOS has had a thriving niche market in third-party UI-patching utilities for years, e.g. Aurora, Kaleidoscope, Greg's Buttons, Aaron, and countless others before Apple themselves got in on the act.
<takes deep breath; exhales loudly> What's this got to do with anything? What part of "open source" don't you get? The source code of these programs is openly available and open to modify! Hence: Open. Source. Hello?! McFly?!
I figure you've got this mental image of "open source" as being the product of the übergeeky, hunched over their machines into the small hours, surrounded by a debris of fast food wrappers and mouldy styrofoam coffee cups. Open Source has nothing to do with this stereotype.
That these programs were developed by "private initiatives" and "corporate consortiums" (sic) is irrelevant. Whether these programs were written by the weird on extreme coffee jags or by 9-to-5 MIS minions in suits is irrelevant. Whether these developments were funded by large corporations or done on one's own dollar is irrelevant. The point is, they've made the source code publically available. These "corporate consortiums" have given something back to the community. They have put their money where their mouths are for open standards. They have not locked up their source code behind NDAs and attempted to pervert open protocols and justified this antisocial behaviour by claiming the right to "innovate" like Microsoft have done. These organisations should be applauded for their public spirit, not sneered at.
IIRC, the core of VMS is written in BLISS, which was a sort of portable assembly language. BLISS abstracted away some of the machine specifics so programs could be ported between the different Digital machine architectures (PDP, VAX, etc.) Most the libraries that make up the VMS runtime environment are written in higher-level languages.
From the OpenVMS FAQ
So your assertion that VMS was programmed "on the metal" in assembly language is half-correct. The fact that VMS requires exotic hardware with multiple privilege levels... well, Digital were designing both the hardware and the OS, so it's not surprising there are strong ties between the VAX architecture and the VMS OS. Portability to non-Digital hardware was not an issue when VMS was conceived.
I think what the original poster had in mind is that NT and VMS are similar conceptually (not surprising as one man, Dave Cutler, was responsible for both).
Yadda yadda yadda, we all read this breathless marketroid shite back in 1991 when NT first appeared. If NT is so damn portable, why has it bombed on every platform except x86?
And, IMNSHO, NT ceased to be a "microkernel" when they moved the GDI into kernel space just to squeeze a few more cycles out of the GUI.
AFAIK, the porting strategy for Linux goes like this:
This is undesirable how?
Depends on what you define as a "complete" system. A lot of these ports are special-purpose projects (e.g. embedded systems), in which case only a fraction of the full functionality of a desktop or server Linux system is required. Not much point in porting a GUI to a system that's intended to be used in a car's fuel injection system, you know.
If you're going to indulge in spelling flames, at least amend your .sig. The correct spelling is Póg mo thóin!
When I talked about coding style, I wasn't just referring to badly-indented horrors knocked out by a student who thinks it's kewl to name all his variables after Star Wars characters. Even with coding standards in place, most code does have a "voice" or a "personality" because each developer favours a particular set of language idioms. For example, developer A might always couch his loop constructs in a `repeat while' idiom, while developer B prefers `repeat until' constructions. Things like that.
But the learning curve is the biggest obstacle. Really grokking a program often involves figuring out what was left out, rather than understanding what the programmer put in. So much information is lost in the translation from the programmer's mental model to finished code. Using a production-quality, fully-debugged library is like being handed a chisel and a block of marble and being asked to produce Michelangelo's "David". Most people will want to start with Play-Doh and work their way up to marble eventually.
That's not necessarily the case. I reckon the biggest problem is the learning curve. Sure, libraries to do things like message boards are ten a penny from Freshmeat, but what you're left with after you've unpacked the tarball is a big chunk of code designed to a mental model that may or may not be in phase with yours. Digesting this code will be a problem. Even if it's well written, it'll take time. More likely than not-- coding style being a very personal thing-- you'll find some coding idiom that will make you cringe and slow down your absorption. When faced with this, is it any wonder that most people will throw their hands in the air, fire up their editor, and start coding from scratch?
Something I've always wanted was a `time machine' for source code and other documents. Imagine some sort of super source code control system (perhaps integrated into the filesystem so you don't have to set anything up) fitted with a gizmo like a VCR shuttle control, so you can `rewind' the source code through earlier iterations. That way you can see the developer's mental model of the system evolving over time. I would kill for such a system. Well, maim anyway. (-:
The problem Microsoft has is, sure, 90% of the world runs their stuff, but that 90% doesn't give a toss. Your average Microsoft user is an office worker who sits down in the morning, turns on their computer, fires up Word and Excel or whatever and gets on with their job. To them, the computer is just a tool; a tool that is handed down from above to them, and, while frequently balky and unreliable, works sufficiently well to get their job done. If there were no computers, their job would be exactly the same, except they would be using typewriters and paper spreadsheets or whatever.
In other words, Microsoft is successful, but it's not the sort of success that Bill Gates or his executives want. Microsoft occupies the same space in the minds of its customers as a traditional energy utility, rather than the bringers of hacktastic innovation to the masses that Bill and the boys would like to be seen as. Electricity? Plug your appliance into a wall socket; it's there. Water? Turn on the tap; it's there. Microsoft software? Switch on your machine; it's there. Ho hum.
No wonder poor Bill and Steve and the rest are so mad! They've very nearly realised their original goal of a computer on every desk running Microsoft software, but practically no-one cares they're running Microsoft software. The only time anyone cares is when it bluescreens for the nth time that day and come 5pm they clock off and go home. No, the only ones that do care are long-haired leftie weirdos who read Slashdot-- and they hate Microsoft and bitch about the technical inelegance of their products! The one segment of the computer-using population that still cares about OSes and APIs and gnarly hacks thinks Bill's software sucks! That's gotta hurt. Bill wants to be a Thomas Edison; not the faceless CEO of some electricity or gas utility. If not, then why all the endless carping about `innovation' during the trial?
Microsoft had better admit this truth pretty quickly-- that their perception of themselves is grossly out of kilter with how they are perceived in the Real World by their customers. C'mon, Bill, there are worse things than being merely adequate; not everyone can be a trailblazer. If they don't wise up, the whole Microsoft edifice will disintegrate, with Bill screaming to the end about `innovation' and the `right to compete', even as the orderlies from Happy Acres Home for the Incorrigibly Bewildered bust down the door of his office, restraints and tasers in hand. Because remember, kids, the paranoid do survive, but only because we lock them up so they can't hurt themselves.
Here endeth the lesson.