And get this: Evolution had left five logic cells unconnected to the rest of the circuit, in a position where they should not have been able to influence its workings. Yet if Thompson disconnected them, the circuit failed. Evidently the chip had evolved a way to use the electromagnetic properties of a signal in a nearby cell. But the fact is that Thompson doesn't know how it works.
Yeah, I've seen source code that worked like this too (remove an unused variable and it stops working), but strangely enough we didn't consider it a great leap forwards... I seem to remember we burnt it and danced on the ashes.
Also:
"Hey, my hardware has stopped working..."
"Sorry, it must have evolved, but we can't fix it cos we don't know how it worked. Have you tried resetting it back to primordial state and waiting 5 years for it to evolve back to useful again ?"
>> Except that some (most?) region 2/3/4/5/6 >> transfers are done quick-and-dirty
I've got news for you, region 1 discs are done quick and dirty (and released cheap), region 2 discs are done to much higher standards, better MPEG encoding and with more extras - partly to justify the higher cost.
Compare US and UK releases of the same material, check the image quality and MPEG artifacts...
>> This guy is just stupid. >> Stupid is as stupid does.
Yeah, right... if only some of the people posting were as "stupid" as Robert Elz.
A lot of very smart people left M.U. CompSci and made lots of money, often with the help of the university, and many of them would have loved to have R.E. join them - but if he's not interested, he's not interested.
Allow other people to have their value systems, and be happy they don't tell you that you can't have yours.
Tim (Melb.Uni. '87)
Not the first time MS hit the Games Console market
on
More on the MS "X-Box"
·
· Score: 1
Microsoft are very good at hiding the bodies from their previous failures.
Back in 90-91 they announced (and developed all the way to release) the same thing, back then it was a 286-tweaked Windows-3.1-in-ROM based console with a CD-ROM.
Anyone remember what it was called ?? I don't think it was the "3DO" failed console, but it was about the same time.
The idea was the same, leverage all those MS developers and tools to dominate the consumer games market.
The only thing left of that is the "autoplay.inf" mechanism that tells Windows machines what to do when you insert a CD-ROM.
Anyone remember Pen Windows, Microsoft Bob, the first attempts at PDA's (jaguar was it called?), the Microsoft Programmers Editor, OS/2, Lan Manager (LanMan for HP-UX), MSN ?
They might recycle the ideas, but don't think there'll be anything new in it. And when it dies, it'll disappear without a trace.
You can say what you like about the technologies, but what I like about Perl, C++ and Linux (amongst others) is exactly the benign dictatorship model that ESR explains. C++ died a death (of sorts) when Bjarne submitted the process to ANSI/ISO standardisation, before that the entire team worked well together - afterwards it became a mess.
These projects work by some gatekeeper keeping control by being first, being reasonable, and being respected, not by threat of courtcase and changing the rules under peoples feet. One of the first things you learn when managing programmers is that being the boss means you've got to let other people be right, be smarter, be quicker than yourself.
The B.D. model works with this maturity - the old way doesn't ("I'm right 'cos I'm mother/teacher/bigger/P.H.B.").
The license agreement, the legalese, the product, well they can be discussed elsewhere, but I trust the future of a product built with a process built on maturity and respect, rather than FUD, bullying and intimidation.
Nobody seen any Ken Russell movies ? "Ladybird, Ladybird" (for example) was filmed without the main actors knowing what comes next, and similarly has a very real, very gritty feel. Uncle Ken's been at it for years, but I don't see Hollywood crumbling. Do his movies even get released in the USA, or is it the art-house circuit only ?
As for one posters comments about "Saving Private Ryan"... spare me. The difference between Lucas and Spielberg is that at least Lucas knows when he's making it up / hamming it up (cf Used Car salesmen and Computer salesmen).
I agree that MS-FUD would be bad, but the unfortunate real world is that the article in question is badly written, but not FUD.
I agree that some posters have remarked on silly issues (like trying to run NT4 programs on Win3.1) but MS backward compatability has always been terrible, and relies on the fact that the old platform dies when the new one is released, so someone will rewrite the old program for the new platform.
I've spent 15 years (since Uni) developing shrink wrap and in-house software, and I'm sorry to say that, contrary to what a number of people have posted;
- well behaved Win16 binaries do not run OK on Win32 platforms (WoW is a virtual subsystem that emulates Win16, it's not the real thing, it's as bad, if not worse, than WABI, SoftPC, etc.) except by luck
- well written 16-bit Windows code did not simply recompile onto Win32 (and the numerous variants). Even with simple 16-32 bit word changes, numerous API calls and messages which MS claimed were "non-order dependant" would only work called (or expected) in one order on win16 and another order under WoW (16on32) or under full Win32. The famous Unix Rabbit book about differences in key API calls would have been invaluable 5 years ago, but nobody knew enough to write one.
- NT 3.5 (especially with the "new shell") and NT 4 were very different and required significant rewrites at source level. I expect W2000 to be as different, if not more so.
- is it 7 different versions, or 7 ways of getting one thing ?? Well, at my current big company, the ONLY way we have found to reliably roll out, for example, Office97 over Office95 is to rebuild every single machine - from fdisk up, because registry keys and DLLs are just so variable across all the permutations of which SP you have now and which you're installing. Is this FUD ? Well, I was at the Office 2000 Deployment conference, and I have the business card of Robert Crissman Jr. (Product Manager for Office) on my desk to say that he agreed - never upgrade Windows or Office, always wipe the disk and start all over again.
- trying to get an inhouse build of NT4 with an inhouse build of IE4 (on a locked down desktop of identical machines) has proved possible, but only with a tweak every month for the last 18 months. Anyone who doesn't mutter "the horror, the horror" in response to the phrase "shlwapi.dll" doesn't really know about deploying NT and Office.
Is all that techie enough ?? I've done cross platform Unix work, and to honest it Unix sounds worse, but ends up easier.
Ho ho ho.... Last I saw HP OpenMail is CC:Mail hacked so the server can run on a Unix machine rather than an NT machine.
It's takes the Mail-Client-Formerly-Known-As-The-Worst-In-The-Wor ld (CC:Mail recently lost this title to Lotus Notes), and breaks it (major areas of functionality such as rules just plain don't work).
Similar to Outlook ?? Well, about as similar as CC:Mail is to Outlook...
Outlook connectivity ?? Yeah, Outlook will read mail via VAPI or whatever it's called now, but then it also talks SMTP and POP3...
>> No type system to speak Ho, ho, ho, seen Perl 5 ? More of a type system than SmallTalk.
>> Non-orthoganal syntax err.. ?? Designed by a linguist to have a natural syntax rather then theoretically perfect but pain to use languages (Pascal, Oberon, ML, etc.) maybe, but that's a plus point for Perl.
>> behaviour isn't well defined As in what behaviour ? Seems as well defined as any other cross platform language to me, if not better (Perl is often more portable than Java IMHO).
>> one cannot write correct programs that use stdio thanks to the lack of an _exit(). Say what ? A million CGI programmers have no problem with stdio. And if you need _exit(), how about use POSIX "_exit";
>> Where should I stop? Perl has its flaws, as does every engineering compromise, but these aren't it...
>> Keanu Reeves is in general, a horrible actor >> but in this role he was cast perfectly, they >> need someone who could act dumb and confused >> real well and he has that covered.
Yeah, cos computer people are really bad at adapting quickly to new concepts, aren't we...
KR is a terrible actor, and wasted the movie. The only time he looked animated was when he did some Kung-Fu, and I'd guess that's cos he claims he's an expert in real-life (from no-acting to over-acting in a couple of frames).
My wife reckons the lead role would've been perfect for Guy Pearce (Aussie actor, last seen in L.A. Confidential), but hey, if you don't have a US lead then you don't get the finance.
I want to see more of Hugo Weaving as the chief baddie, now he can act, maybe in a prequel he could be the original person who broke out of the Matrix, and using him as an agent is the Matrix's idea of a joke....
SCRUM is good, but it's a high-risk methodology, sometimes you get a brilliant solution in a very short time, sometimes you get nothing, sometimes you get nothing and all your best people leave.
Most people look to processes for control and predictability. Project management is risk management - you pick the compromise you're willing to accept and run with it.
"DeGrace, P. and Hulet Stahl, L. 1990. Wicked Problems, Righteous Solutions. Yourdon Press" is a great overview of "project methods" including the pros and cons of each.
>> I remember quite a few dealing with bank >> machines, and although that leads itself to a >> certain stability, it can be hard to get >> excited over such applications.
Rumbaugh's "classic" book was overly simplistic, and Booch's "management overview" book suffered from the same. They're great at describing "tangible" object systems (hot water systems, vehicles etc.) and systems where the major classes are obvious (customer, account, transaction, etc.), but fall apart on real-world examples of abstract problem spaces. I know these "physical examples" are easier to use as illustrations (familiarity of the reader etc.) but they assume that all problems are like this.
Is the new book any better, or are they still relatively naive ?
If not, I'd recommend DeGrace and Stahl ("Wicked Problems, Righteous Solutions" and "The Olduvai Imperative: CASE and the State of Software Engineering Practice") as balancing the hype of all those processes which feature steps like "and then a design miraculously appears". Maybe I should pull my finger out and submit a review;^)
Tele-marketers effectively save money when you hang up quick, it frees them up to call the next person (ie a cheap negative). What hurts telemarketers is when they spend time on you and then don't make a sale (ie an expensive negative).
So waste their time, either ask them inane questions for ages (depending how bored you are this can be quite fun) or I usually just tell them "I'll go get the person you want to speak to", then go back to what I was doing. Pop back every 5 minutes or so and say "he's just coming now..." - this really cheeses them off. I had one guy hanging on for about an hour one time while I watched TV. In Australia this used to actually tie up their phone line, they could hang up but they couldn't actually get another dial tone until I hung up too. Of course, if you're expecting a call, then your options are limited.
I don't get any near as many calls since I started doing this, may be coincidence of course.
Why is it every time I start a book, slashdot posts a review and I lose the urge to continue ?
But anyway, I was wondering about errors, eg page 27, line 22 of the code QLCDNumber * mylcdnum = new QWidget(1,mywidget); surely this should be "new QLCDNumber(...)" (see the rest of the code). Is this an isolated instance, or does it continue ?
As for the "diehard MFC programmers, do they exist" comment of one of the replies, yes, unfortunately they do exist, legions of them. I wrote my first C++ Windows wrapper layer in 1990, before MS released a C++ compiler, and I've refused to use that brain-dead excuse for a class library since. But unfortunately all those superb VB programmers out in the world want to progress, and MFC lets them write VB-quality code in C++ (The horror, the horror...).
>> Its hard to beat that with a stick. All religion aside, the mobile market is where the PPC truly blows the doors off the x86.
Point taken, I didn't want a CE machine, I wanted something I could run/develop Perl on and enough of an OS to make it useful. Right now, that means an Intel machine, if PPC or the LinuxCE project gives me a usable environment on something better, well, I'll be even happier. If I could get a realistic dev env (Perl, C++ and email) on a Psion 5... well that'd be enough for me.
I bought an Acer Travelmate 312T sub-notebook, it's got a 233 MMX Pentium, 32 Mb RAM, 3 Gb disk, 2 PC card slots and includes floppy disk and cdrom for about UKP 1,000 (US Price - $ 1000 - 1500 ??). The screen is 8.4 inch TFT and runs 800x600. The keyboard is fine for me (JAPH typing speed), and it has a drag-pad thing rather than one of those terrible "nipple" things. The built in 56k modem is a WinModem, so only works under MS environments - time for another PC card I guess.
It's almost exactly the same size and weight as a decent O'Reilly book (slightly thinner than The Perl Cookbook). The power supply is small enough to stick in a pocket - I just leave spare leads at work and home, the battery normally lasts about 90 minutes for me, which is fine.
But I haven't got Linux on it... quite yet (;^(). The PCMCIA needs a quite new version of the PCMCIA package (I forget which, but RedHat 5.2 won't work, whereas RedHat 6.0 will), but the included PCMCIA CD-ROM is not quite standard and I need to rebuild the kernel with some fixes (see Linux Laptops pages for more, basically an extra delay in ide_cs.c) before I can install from CD. But as soon as I get a spare weekend I'll start work on that again, or get a SCSI card and hook up my Jaz drive maybe.
I'd recommend it heavily, the only real competition right now I can see is the Sony Vaio Picturebook, but that costs at least 50% more.
Anyone know why there's not a DVD drive in there too ? I would have thought seeing they've got all the rest of the circuitry in place, an optional DVD player would be cheap, reliable, and would add a lot of value... (how many boxes can you fit on the top of a TV anyway ?)
I've long explained to non-computer people that computing in practice is too much like magic - it is a collection of arcance sequences and irrational invocations, memorised by the practitioners as "spells that work", without them neccesarily knowing how or why.
I'm looking forward to the day when I don't get told "you must follow these exact steps, if you get any two in the wrong order it won't work but we don't know why".
Any but the most specialist computing niches are now too large and too genereal for deep understanding to be feasible. So we rely on too many arbitrary black boxes, hence the popular success of OpenSource (yeah, preaching to the converted I know) so that instead of having to remember "that's the way it is" we can read the source and replace some of the magic with a bit of science.
I think this idea was inspired if not blatantly stolen from Weinberg's "The Psychology of Computer Programming" (1973 and still true).
As for why the magic/mysticism imagery is so popular right now ? Fin-de-siecle always produces a swing towards mysticism, this coming at the peak of quite a sustained period of vicious technological acceleration, and I'm surprised the "backlash" (I think "whiplash" may be more accurate, but probably that word is banned for our friends in Oz;^)) is still as weak as the X-Files, new age hippies, down-shifting and the odd eastern-inspired religious cult phenomena.
>> have warred in a very public way against hackers... [snip] >> hackers don't steal, vandalize or damage. They are most often freedom-loving and generous problem solvers and information sharers
Hackers may be, but they're not who're being chased.
The problem is crackers, and "real crackers" are not [gush, gush], they're anti-social malicious self-righteous spolit brats.
I know the article goes on to draw a difference, but it implies that all crackers are hackers "crossed over to the dark side" (my quote, not JK), I'd disagree... hackers tend to have a deep understanding of the systems they work on, crackers tend to just know surface details and incantations. Crackers (in general, obvious exceptions apply) feed off the teachings of hackers, they're rarely the source of new information themselves.
Once you've started down the path of computing, the choice is hacking vs. cracking, not one as a specialisation of the other.
>> Game developer that's overclocking? >> Pretty dumb I would say
Why so, if it runs on my overclocked CPU, then a "normal" CPU should be no problem...
As a (non-games) developer I run loads of Beta OS code etc. I test under a "normal" platform too, but I find it less frustrating to develop under a less tolerant environment (debugging kernels, heap libraries etc. anyone) and test under a more tolerant env. than vice versa.
>> why do these things seem to be centred around the US?
Because the US news gets reported everywhere else (CNN anyone), whereas australian etc. news is rarely reported overseas.
Australia has ahuge number of "berzerker" incidents when considered per head of population (16 million = less than 10% of US population) - Hoddle Street anyone ? Backpackers murders ? The postal workers in Melbourne ? The Tasmanian incident that is so rarely mentioned that I forget it's name (not to mention the determined extermination of aboriginals in Tasmania).
Australia suffers the same "beat the crap out of anyone different" mentality as the US, but because so many kids are immigrants they learn pretty quick to abandon their own personality and join in with the mob-think.
The author of the parent comment raises exactly the points taht the scientists in question are rasing and examining.
Most of the other comments here are as uninformed as the source article. I heard an interview with the head of the department on Today this morniing (BBC Radio 4 for US readers) and yes, the correct figures are 6,3 and 9 years, NOT 6, 3 and 6.
The bloke said (as memory serves) "this is one measure of cell age (out of many), and we don't know how important it is over many generatiosn, but in a single case it is not important BUT WORTH INVESTIGATING, and the test size is only one, but this is the point of a test case".
The more I read news on the web the more I despair for anything even approaching accurate reporting.
Australians are under the illusion that they're living in the free-est of democracies, whereas they are one of the most dominated of the western nations. The Aus government continues to treat its citizens as "naughty children", or dare I say convicts.
Guns wasn't too bad, the dangerous signs were compulsory tax-file numbers, compulsory voting, compulsory carrying of drivers licenses (with photos)... treat your citizens as idiots and you end up with idiots for citizens.
Slashdot Mirror
Yeah, I've seen source code that worked like this too (remove an unused variable and it stops working), but strangely enough we didn't consider it a great leap forwards... I seem to remember we burnt it and danced on the ashes.
Also:
"Hey, my hardware has stopped working..."
"Sorry, it must have evolved, but we can't fix it cos we don't know how it worked. Have you tried resetting it back to primordial state and waiting 5 years for it to evolve back to useful again ?"
T
>> Except that some (most?) region 2/3/4/5/6
>> transfers are done quick-and-dirty
I've got news for you, region 1 discs are done quick and dirty (and released cheap), region 2 discs are done to much higher standards, better MPEG encoding and with more extras - partly to justify the higher cost.
Compare US and UK releases of the same material, check the image quality and MPEG artifacts...
Tim
>> This guy is just stupid.
... if only some of the people posting were as "stupid" as Robert Elz.
>> Stupid is as stupid does.
Yeah, right
A lot of very smart people left M.U. CompSci and made lots of money, often with the help of the university, and many of them would have loved to have R.E. join them - but if he's not interested, he's not interested.
Allow other people to have their value systems, and be happy they don't tell you that you can't have yours.
Tim (Melb.Uni. '87)
Microsoft are very good at hiding the bodies from their previous failures.
Back in 90-91 they announced (and developed all the way to release) the same thing, back then it was a 286-tweaked Windows-3.1-in-ROM based console with a CD-ROM.
Anyone remember what it was called ?? I don't think it was the "3DO" failed console, but it was about the same time.
The idea was the same, leverage all those MS developers and tools to dominate the consumer games market.
The only thing left of that is the "autoplay.inf" mechanism that tells Windows machines what to do when you insert a CD-ROM.
Anyone remember Pen Windows, Microsoft Bob, the first attempts at PDA's (jaguar was it called?), the Microsoft Programmers Editor, OS/2, Lan Manager (LanMan for HP-UX), MSN ?
They might recycle the ideas, but don't think there'll be anything new in it. And when it dies, it'll disappear without a trace.
Tim
You can say what you like about the technologies, but what I like about Perl, C++ and Linux (amongst others) is exactly the benign dictatorship model that ESR explains. C++ died a death (of sorts) when Bjarne submitted the process to ANSI/ISO standardisation, before that the entire team worked well together - afterwards it became a mess.
These projects work by some gatekeeper keeping control by being first, being reasonable, and being respected, not by threat of courtcase and changing the rules under peoples feet. One of the first things you learn when managing programmers is that being the boss means you've got to let other people be right, be smarter, be quicker than yourself.
The B.D. model works with this maturity - the old way doesn't ("I'm right 'cos I'm mother/teacher/bigger/P.H.B.").
The license agreement, the legalese, the product, well they can be discussed elsewhere, but I trust the future of a product built with a process built on maturity and respect, rather than FUD, bullying and intimidation.
Should be 2 competitions, the other to find the Anti-Patterns in the project too.
Projects are improved by a bit of push-me pull-you, too much self-congratulation makes you go blind.
Nobody seen any Ken Russell movies ? "Ladybird, Ladybird" (for example) was filmed without the main actors knowing what comes next, and similarly has a very real, very gritty feel. Uncle Ken's been at it for years, but I don't see Hollywood crumbling. Do his movies even get released in the USA, or is it the art-house circuit only ?
... spare me.
As for one posters comments about "Saving Private Ryan"
The difference between Lucas and Spielberg is that at least Lucas knows when he's making it up / hamming it up (cf Used Car salesmen and Computer salesmen).
I agree that MS-FUD would be bad, but the unfortunate real world is that the article in question is badly written, but not FUD.
I agree that some posters have remarked on silly issues (like trying to run NT4 programs on Win3.1) but MS backward compatability has always been terrible, and relies on the fact that the old platform dies when the new one is released, so someone will rewrite the old program for the new platform.
I've spent 15 years (since Uni) developing shrink wrap and in-house software, and I'm sorry to say that, contrary to what a number of people have posted;
- well behaved Win16 binaries do not run OK on Win32 platforms (WoW is a virtual subsystem that emulates Win16, it's not the real thing, it's as bad, if not worse, than WABI, SoftPC, etc.) except by luck
- well written 16-bit Windows code did not simply recompile onto Win32 (and the numerous variants). Even with simple 16-32 bit word changes, numerous API calls and messages which MS claimed were "non-order dependant" would only
work called (or expected) in one order on win16 and another order under WoW (16on32) or under full Win32. The famous Unix Rabbit book about differences in key API calls would have been invaluable 5 years ago, but nobody knew enough to write one.
- NT 3.5 (especially with the "new shell") and NT 4 were very different and required significant rewrites at source level. I expect W2000 to be as different, if not more so.
- is it 7 different versions, or 7 ways of getting one thing ?? Well, at my current big company, the ONLY way we have found to reliably roll out, for example, Office97 over Office95 is to rebuild every single machine - from fdisk up, because registry keys and DLLs are just so variable across all the permutations of which SP you have now and which you're installing. Is this FUD ? Well, I was at the Office 2000 Deployment conference, and I have the business card of Robert Crissman Jr. (Product Manager for Office) on my desk to say that he agreed - never upgrade Windows or Office, always wipe the disk and start all over again.
- trying to get an inhouse build of NT4 with an inhouse build of IE4 (on a locked down desktop of identical machines) has proved possible, but only with a tweak every month for the last 18 months. Anyone who doesn't mutter "the horror, the horror" in response to the phrase "shlwapi.dll" doesn't really know about deploying NT and Office.
Is all that techie enough ??
I've done cross platform Unix work, and to honest it Unix sounds worse, but ends up easier.
Tim
Ho ho ho....
r ld (CC:Mail recently lost this title to Lotus Notes), and breaks it (major areas of functionality such as rules just plain don't work).
Last I saw HP OpenMail is CC:Mail hacked so the server can run on a Unix machine rather than an NT machine.
It's takes the Mail-Client-Formerly-Known-As-The-Worst-In-The-Wo
Similar to Outlook ?? Well, about as similar as CC:Mail is to Outlook...
Outlook connectivity ?? Yeah, Outlook will read mail via VAPI or whatever it's called now, but then it also talks SMTP and POP3...
This is a bit of a non-story....
>> No type system to speak
Ho, ho, ho, seen Perl 5 ?
More of a type system than SmallTalk.
>> Non-orthoganal syntax
err.. ?? Designed by a linguist to have a natural syntax rather then theoretically perfect but pain to use languages (Pascal, Oberon, ML, etc.) maybe, but that's a plus point for Perl.
>> behaviour isn't well defined
As in what behaviour ?
Seems as well defined as any other cross platform language to me, if not better (Perl is often more portable than Java IMHO).
>> one cannot write correct programs that use stdio thanks to the lack of an _exit().
Say what ?
A million CGI programmers have no problem with stdio.
And if you need _exit(), how about
use POSIX "_exit";
>> Where should I stop?
Perl has its flaws, as does every engineering compromise, but these aren't it...
And remember, this was in comparison to TCL.
Tim
>> Keanu Reeves is in general, a horrible actor
>> but in this role he was cast perfectly, they
>> need someone who could act dumb and confused
>> real well and he has that covered.
Yeah, cos computer people are really bad at adapting quickly to new concepts, aren't we...
KR is a terrible actor, and wasted the movie. The only time he looked animated was when he did some Kung-Fu, and I'd guess that's cos he claims he's an expert in real-life (from no-acting to over-acting in a couple of frames).
My wife reckons the lead role would've been perfect for Guy Pearce (Aussie actor, last seen in L.A. Confidential), but hey, if you don't have a US lead then you don't get the finance.
I want to see more of Hugo Weaving as the chief baddie, now he can act, maybe in a prequel he could be the original person who broke out of the Matrix, and using him as an agent is the Matrix's idea of a joke....
Hmmm, yeah, right.
So Rob and the boys are Cobol nerds eh ?
VB is the Cobol of the 90's.
Perl is the C or maybe the Lisp of the 90's. Python looks like it may end up being the Eiffel of the 90's.
SCRUM is good, but it's a high-risk methodology, sometimes you get a brilliant solution in a very short time, sometimes you get nothing, sometimes you get nothing and all your best people leave.
Most people look to processes for control and predictability. Project management is risk management - you pick the compromise you're willing to accept and run with it.
"DeGrace, P. and Hulet Stahl, L. 1990. Wicked Problems, Righteous Solutions. Yourdon Press" is a great overview of "project methods" including the pros and cons of each.
>> I remember quite a few dealing with bank
;^)
>> machines, and although that leads itself to a
>> certain stability, it can be hard to get
>> excited over such applications.
Rumbaugh's "classic" book was overly simplistic, and Booch's "management overview" book suffered from the same. They're great at describing "tangible" object systems (hot water systems, vehicles etc.) and systems where the major classes are obvious (customer, account, transaction, etc.), but fall apart on real-world examples of abstract problem spaces. I know these "physical examples" are easier to use as illustrations (familiarity of the reader etc.) but they assume that all problems are like this.
Is the new book any better, or are they still relatively naive ?
If not, I'd recommend DeGrace and Stahl ("Wicked Problems, Righteous Solutions" and "The Olduvai Imperative: CASE and the State of Software Engineering Practice") as balancing the hype of all those processes which feature steps like "and then a design miraculously appears". Maybe I should pull my finger out and submit a review
Tele-marketers effectively save money when you hang up quick, it frees them up to call the next person (ie a cheap negative). What hurts telemarketers is when they spend time on you and then don't make a sale (ie an expensive negative).
So waste their time, either ask them inane questions for ages (depending how bored you are this can be quite fun) or I usually just tell them "I'll go get the person you want to speak to", then go back to what I was doing. Pop back every 5 minutes or so and say "he's just coming now..." - this really cheeses them off. I had one guy hanging on for about an hour one time while I watched TV. In Australia this used to actually tie up their phone line, they could hang up but they couldn't actually get another dial tone until I hung up too.
Of course, if you're expecting a call, then your options are limited.
I don't get any near as many calls since I started doing this, may be coincidence of course.
Why is it every time I start a book, slashdot posts a review and I lose the urge to continue ?
But anyway, I was wondering about errors, eg page 27, line 22 of the code
QLCDNumber * mylcdnum = new QWidget(1,mywidget);
surely this should be "new QLCDNumber(...)" (see the rest of the code). Is this an isolated instance, or does it continue ?
As for the "diehard MFC programmers, do they exist" comment of one of the replies, yes, unfortunately they do exist, legions of them. I wrote my first C++ Windows wrapper layer in 1990, before MS released a C++ compiler, and I've refused to use that brain-dead excuse for a class library since. But unfortunately all those superb VB programmers out in the world want to progress, and MFC lets them write VB-quality code in C++ (The horror, the horror...).
>> Its hard to beat that with a stick. All religion aside, the mobile market is where the PPC truly blows the doors off the x86.
... well that'd be enough for me.
Point taken, I didn't want a CE machine, I wanted something I could run/develop Perl on and enough of an OS to make it useful. Right now, that means an Intel machine, if PPC or the LinuxCE project gives me a usable environment on something better, well, I'll be even happier.
If I could get a realistic dev env (Perl, C++ and email) on a Psion 5
I bought an Acer Travelmate 312T sub-notebook, it's got a 233 MMX Pentium, 32 Mb RAM, 3 Gb disk, 2 PC card slots and includes floppy disk and cdrom for about UKP 1,000 (US Price - $ 1000 - 1500 ??).
... quite yet (;^(). The PCMCIA needs a quite new version of the PCMCIA package (I forget which, but RedHat 5.2 won't work, whereas RedHat 6.0 will), but the included PCMCIA CD-ROM is not quite standard and I need to rebuild the kernel with some fixes (see Linux Laptops pages for more, basically an extra delay in ide_cs.c) before I can install from CD. But as soon as I get a spare weekend I'll start work on that again, or get a SCSI card and hook up my Jaz drive maybe.
The screen is 8.4 inch TFT and runs 800x600.
The keyboard is fine for me (JAPH typing speed), and it has a drag-pad thing rather than one of those terrible "nipple" things.
The built in 56k modem is a WinModem, so only works under MS environments - time for another PC card I guess.
It's almost exactly the same size and weight as a decent O'Reilly book (slightly thinner than The Perl Cookbook). The power supply is small enough to stick in a pocket - I just leave spare leads at work and home, the battery normally lasts about 90 minutes for me, which is fine.
But I haven't got Linux on it
I'd recommend it heavily, the only real competition right now I can see is the Sony Vaio Picturebook, but that costs at least 50% more.
Tim
Anyone know why there's not a DVD drive in there too ?
I would have thought seeing they've got all the rest of the circuitry in place, an optional DVD player would be cheap, reliable, and would add a lot of value... (how many boxes can you fit on the top of a TV anyway ?)
I've long explained to non-computer people that computing in practice is too much like magic - it is a collection of arcance sequences and irrational invocations, memorised by the practitioners as "spells that work", without them neccesarily knowing how or why.
;^)) is still as weak as the X-Files, new age hippies, down-shifting and the odd eastern-inspired religious cult phenomena.
I'm looking forward to the day when I don't get told "you must follow these exact steps, if you get any two in the wrong order it won't work but we don't know why".
Any but the most specialist computing niches are now too large and too genereal for deep understanding to be feasible. So we rely on too many arbitrary black boxes, hence the popular success of OpenSource (yeah, preaching to the converted I know) so that instead of having to remember "that's the way it is" we can read the source and replace some of the magic with a bit of science.
I think this idea was inspired if not blatantly stolen from Weinberg's "The Psychology of Computer Programming" (1973 and still true).
As for why the magic/mysticism imagery is so popular right now ?
Fin-de-siecle always produces a swing towards mysticism, this coming at the peak of quite a sustained period of vicious technological acceleration, and I'm surprised the "backlash" (I think "whiplash" may be more accurate, but probably that word is banned for our friends in Oz
>> have warred in a very public way against hackers...
... hackers tend to have a deep understanding of the systems they work on, crackers tend to just know surface details and incantations. Crackers (in general, obvious exceptions apply) feed off the teachings of hackers, they're rarely the source of new information themselves.
[snip]
>> hackers don't steal, vandalize or damage. They are most often freedom-loving and generous problem solvers and information sharers
Hackers may be, but they're not who're being chased.
The problem is crackers, and "real crackers" are not [gush, gush], they're anti-social malicious self-righteous spolit brats.
I know the article goes on to draw a difference, but it implies that all crackers are hackers "crossed over to the dark side" (my quote, not JK), I'd disagree
Once you've started down the path of computing, the choice is hacking vs. cracking, not one as a specialisation of the other.
Tim
>> Game developer that's overclocking?
>> Pretty dumb I would say
Why so, if it runs on my overclocked CPU, then a "normal" CPU should be no problem...
As a (non-games) developer I run loads of Beta OS code etc.
I test under a "normal" platform too, but I find it less frustrating to develop under a less tolerant environment (debugging kernels, heap libraries etc. anyone) and test under a more tolerant env. than vice versa.
YMMV
Tim
>> why do these things seem to be centred around the US?
Because the US news gets reported everywhere else (CNN anyone), whereas australian etc. news is rarely reported overseas.
Australia has ahuge number of "berzerker" incidents when considered per head of population (16 million = less than 10% of US population) - Hoddle Street anyone ? Backpackers murders ? The postal workers in Melbourne ? The Tasmanian incident that is so rarely mentioned that I forget it's name (not to mention the determined extermination of aboriginals in Tasmania).
Australia suffers the same "beat the crap out of anyone different" mentality as the US, but because so many kids are immigrants they learn pretty quick to abandon their own personality and join in with the mob-think.
Tim (yeah, 2nd anti-Aus posting in 2 days...)
The author of the parent comment raises exactly the points taht the scientists in question are rasing and examining.
Most of the other comments here are as uninformed as the source article. I heard an interview with the head of the department on Today this morniing (BBC Radio 4 for US readers) and yes, the correct figures are 6,3 and 9 years, NOT 6, 3 and 6.
The bloke said (as memory serves) "this is one measure of cell age (out of many), and we don't know how important it is over many generatiosn, but in a single case it is not important BUT WORTH INVESTIGATING, and the test size is only one, but this is the point of a test case".
The more I read news on the web the more I despair for anything even approaching accurate reporting.
Tim
Australians are under the illusion that they're living in the free-est of democracies, whereas they are one of the most dominated of the western nations. The Aus government continues to treat its citizens as "naughty children", or dare I say convicts.
... treat your citizens as idiots and you end up with idiots for citizens.
Guns wasn't too bad, the dangerous signs were compulsory tax-file numbers, compulsory voting, compulsory carrying of drivers licenses (with photos)
Read John Pilger's A Secret Country for background...
Tim (trying hard not to go back there...)