My 1992 Saturn was stolen about 3 months ago. There was a 2001 Jeep Grand Cherokee with leather sitting next to it. The cops said they picked my car because it had no alarm system and so it was easier to steal.
Dark green '94 Grand Cherokee with light brown leather? That'd be mine.:-)
Mind you, if you'd have looked at the passenger door you'd see where some dumbass with a slide hammer knocked the ring off the door lock and used a screwdriver to gain entry. At 2am in my driveway in a sleepy little rural Ontario town. The alarm went off and they left empty-handed.
Though this doesn't address what happens when a transmission gets missed (from interference or distance), but another reply covered that one pretty well...
It looks for codes that are n codes away from the next desired, as well. So if you hit the button n+1 times away from the receiver, you're buggered unless you can reset them both to code 0, but I can personally attest to this working. (I have a button-happy 18-month-old son).
In most cases you're talking about $99 for a factory installed alarm. And a good unit from Viper (500ESP+) that will have the code hopping feature is gonna cost you at least $300. The other advantages of a GOOD alarm system is that you can install modules like power window/sunroof and remote start (for automatics).
You're full of it. Code hopping is probably cheaper to implement these days than non. There are chipsets (Microchip's HCS series, for example) which do all the hard work for you. My CDN$90 (US$40) remote starter uses code hopping, for Christ's sake.
Maybe that was not what they said out loud, but that was pretty much what they were thinking.
? The article said they didn't go against the owner's wishes on purpose (re the hay on the walls) -- I agree that the designers should be held liable and pay for the repair -- give the fuckers some accountability -- but it's not like the homeowners said "I don't want hay on the walls" -- they didn't know not to ask for that. I wouldn't have known, either, that someone would be so detached from reality that they would do such a thing.
Now I did read in a different article that the designer went against the protection clauses, and that is plain wrong.
'These people are going to hate hay on the wall - let's do it.'
Reread the article. They didn't do that:
"You can undertake a plan and not really know how it's going to turn out. That's different from saying, 'These people are going to hate hay on the wall - let's do it.' "
You might want to give XWT a shot. You write all your business logic in whatever you need (C++, Java, Perl, Fortran, whatever), exporting the functions you need to use as XMLRPC or SOAP connections. Then you write your user interfaces in XWT. Native clients for Linux and Win32, with everything else supported by a Java 1.2 RE. It's fast, scalable, configurable and truly multiplatform.
One really neat use of this would be to bundle VMWare into the CD dist so that you could actually drop the CD into a machine you don't trust (maybe your employer's?) to be free of keystroke loggers, etc.
Two problems immediately come to mind
A hardware dongle that logs keys isn't circumvented
Now you need a "CD#2" for your windows image
Let's face it -- it's a company computer. If you don't want them reading your private shit, don't use the company computer for it.
Anything that compromises security (anti-piracy, DRM, anti-virus, licensing, encryption, authentication).
Great, that covers oh... everything except for skinning UIs.
Filesystem specs? No, that would compromise DRM and antivirus. Authentication? Sorry, no Domain Master or AD internals for you. This doesn't seem like a very happy judgement to me.
f CF-to-IDE adapters are so cheap, why are the CF based solid state IDE hard disks so expensive!?! I've been waiting for Sandisk's 1GB 2.5" ide drive to come down under $500 so I can replace my laptop drive. It doesn't look like that will happen any time soon.
Because Flash is expensive, that's why.
And you won't be replacing an IDE HDD with a CF card for very long. Flash has a limited number of write cycles. And it's dog-slow.
What motherboard+CPU did you see there for under $250? I couldn't find a motherboard for a fanless CPU, other than Crusoe, Pentium or 486DX4, and even they were $250-$380 without CPU.
It's rather telling that the first thing you think of in connection with coding discipline is concrete syntax and Python and not, say, strong, powerful type systems, enforced modularity, declarative clarity and so forth.
I'm not sure how you're piecing together my view of programming dos and donts and then attacking Python with them, but let me try to clarify:
I like programming
over the years, I've developed a certain style
Python makes me break that style through forced formatting, so
I am not a fan of Python
I never said Python was a bad language.
As for your comment about hiring disciplined programmers, well, it's banal. And yet you present it as though it supported some point against strongly typed languages.
I'm not trying to be unique. I have a higher standard for hired programmers than many others it seems. I'm not sure what your point is.
As far as disciplined programmers not needing strongly typed languages: it's true. It doesn't mean that disciplined programmers won't use strongly typed languages or won't like them. I don't ever recall saying anything to the effect of "I use disciplined programmers so I don't need strongly typed languages," but rather "I don't like strongly typed languages, so I hire disciplined programmers." It's a significant difference.
Having written that, I don't know why I'm bothering to reply to you.
I smile as I read this, because I don't think I'd ever fit in to an organization where I was to follow a command chain and (potentially) not know why. I've never been on an army base, although the RCF often stop by at the McDonald's in my town on their way to Borden, Ontario.
I don't know -- perhaps I'm a self-riteous pissant who takes "my way or the highway" too seriously, but I put a lot of forethought and research into a decision and become very rigid on the implementation. It doesn't mean I won't move from it, but rather that there has to be a damned good reason to move away from what I'd set out in a project. If I'm proven wrong, I'll correct it and my belief in the new method will be no less strong than my belief in the old method was (until it was shown incorrect, that is.):-)
If there were more people like us during the dot-com boom, it might still be going.
heh, I don't know if I'd go that far... There were a lot of things wrong with the dot-com boom.
When coding Perl as part of a large team, there's always one that hacks out his code so poorly that his code works only on Fridays, and when the Environment table is deleted, and with glibc-0.0.12 because he calls the C API directly which buffer overflows somewhere. -wT and use strict were forgotten long ago.
The key is not to let it get "long ago". Bi-weekly (mini) code reviews and constant interaction are important to prevent bad code from affecting a codebase in any language.
This is all, of course, horses for courses. We can hand-wave and hypothesize (sp?) forever and not get anywhere.
So do you turn on use strict and -wT, or do you decide not to grovel to your boss and install the unstable script? Afterwards, you can blame Perl and recommend switching to PHP. Explains a lot, eh?
I've been lucky; I have been given the power to eliminate lousy coders who won't at least try to maintain some semblance of good coding methods, and I've been given the power to delay some things in order to make sure they are done right so that the future won't take twice as long. Others aren't so lucky.
The use of -wT only works in the smallest scripts.
Bullshit. Complete and utter bullshit. You don't inherit 9M-LOC of Perl and THEN put -wT on it, you start with 10 LOC and use -wT from the start, eliminating the warnings as you go. I've got perl scripts thousands of lines long, across multiple files and modules, all with use strict and -wT. Not a single warning. Discipline, it's what I stated from the start.
Now your comment about "run until error" -- Isn't that what testing with the "million monkeys" data input method is all about? I don't care what language you use, you can't catch runtime errors at compile time.
Now as far as catching syntactical errors at compile-time: you're absolutely correct. Without having a way for Perl to run over every single line beforehand, there is no way to do this. It is a downfall, IMO.
Just because you don't like compile errors doesn't mean that you should ignore them.
I don't follow. All my perl code executes with -wT and strict mode. When I have to do strange things like call subroutine by reference, I turn off the specific strict-mode bit. So my perl is less good because...?
Curious, why are you "hacking" your ICQ client? Why doesn't it do what you want in the first place?
? I'm not hacking my ICQ client, I wrote a couple of small patches for Psi which add functionality I want (and which I hope make it in to the main tree, as I can't compile for win32 and all my win32 contacts seem to want the same features).
That's the beauty of OSS -- Things aren't just the way I want them, but I have carte blanche to modify it, or to find someone and pay them to do so. Try getting rid of all the crap that the latest ICQ clients have.
There is not a computer in existence that is powerful enough to translate "stupid politician" into "English" or "French", or even any other human language. I simply cannot be done.
We can't go that way, but how about an English to Chretien translator instead?:-)
Slashdot Mirror
My 1992 Saturn was stolen about 3 months ago. There was a 2001 Jeep Grand Cherokee with leather sitting next to it. The cops said they picked my car because it had no alarm system and so it was easier to steal.
Dark green '94 Grand Cherokee with light brown leather? That'd be mine. :-)
Mind you, if you'd have looked at the passenger door you'd see where some dumbass with a slide hammer knocked the ring off the door lock and used a screwdriver to gain entry. At 2am in my driveway in a sleepy little rural Ontario town. The alarm went off and they left empty-handed.
Though this doesn't address what happens when a transmission gets missed (from interference or distance), but another reply covered that one pretty well...
It looks for codes that are n codes away from the next desired, as well. So if you hit the button n+1 times away from the receiver, you're buggered unless you can reset them both to code 0, but I can personally attest to this working. (I have a button-happy 18-month-old son).
In most cases you're talking about $99 for a factory installed alarm. And a good unit from Viper (500ESP+) that will have the code hopping feature is gonna cost you at least $300. The other advantages of a GOOD alarm system is that you can install modules like power window/sunroof and remote start (for automatics).
You're full of it. Code hopping is probably cheaper to implement these days than non. There are chipsets (Microchip's HCS series, for example) which do all the hard work for you. My CDN$90 (US$40) remote starter uses code hopping, for Christ's sake.
Maybe that was not what they said out loud, but that was pretty much what they were thinking.
? The article said they didn't go against the owner's wishes on purpose (re the hay on the walls) -- I agree that the designers should be held liable and pay for the repair -- give the fuckers some accountability -- but it's not like the homeowners said "I don't want hay on the walls" -- they didn't know not to ask for that. I wouldn't have known, either, that someone would be so detached from reality that they would do such a thing.
Now I did read in a different article that the designer went against the protection clauses, and that is plain wrong.
Anyone who signs up to participate in that show has no room to complain--bout anything at all!
That's untrue.
'These people are going to hate hay on the wall - let's do it.'
Reread the article. They didn't do that:
I showed my wife an article about the behind the sceens of one of those episodes and how much damage/money it costs to repair the "renovation".
LINKS, DAMMIT, LINKS!
Ok, so what was teh word that sounded like "Gunnerstchaffel!"?
You might want to give XWT a shot. You write all your business logic in whatever you need (C++, Java, Perl, Fortran, whatever), exporting the functions you need to use as XMLRPC or SOAP connections. Then you write your user interfaces in XWT. Native clients for Linux and Win32, with everything else supported by a Java 1.2 RE. It's fast, scalable, configurable and truly multiplatform.
One really neat use of this would be to bundle VMWare into the CD dist so that you could actually drop the CD into a machine you don't trust (maybe your employer's?) to be free of keystroke loggers, etc.
Two problems immediately come to mind
Let's face it -- it's a company computer. If you don't want them reading your private shit, don't use the company computer for it.
Anything that compromises security (anti-piracy, DRM, anti-virus, licensing, encryption, authentication).
Great, that covers oh... everything except for skinning UIs.
Filesystem specs? No, that would compromise DRM and antivirus. Authentication? Sorry, no Domain Master or AD internals for you. This doesn't seem like a very happy judgement to me.
Isn't that like rp_filter on Linux?
The Via Eden 533mhz x86 based platform goes for about $90.
Most excellent. Thank you for the pointer. However the Via Eden isn't on the site mentioned. :-)
f CF-to-IDE adapters are so cheap, why are the CF based solid state IDE hard disks so expensive!?! I've been waiting for Sandisk's 1GB 2.5" ide drive to come down under $500 so I can replace my laptop drive. It doesn't look like that will happen any time soon.
Because Flash is expensive, that's why.
And you won't be replacing an IDE HDD with a CF card for very long. Flash has a limited number of write cycles. And it's dog-slow.
What motherboard+CPU did you see there for under $250? I couldn't find a motherboard for a fanless CPU, other than Crusoe, Pentium or 486DX4, and even they were $250-$380 without CPU.
Statically-typed languages (e.g., Eiffel) are easier to maintain than dynamically-typed ones (e.g., Perl, Python), for they are self-documenting.
Depends on what you want from your documentation, I suppose.
It's rather telling that the first thing you think of in connection with coding discipline is concrete syntax and Python and not, say, strong, powerful type systems, enforced modularity, declarative clarity and so forth.
I'm not sure how you're piecing together my view of programming dos and donts and then attacking Python with them, but let me try to clarify:
- I like programming
- over the years, I've developed a certain style
- Python makes me break that style through forced formatting, so
- I am not a fan of Python
I never said Python was a bad language.As for your comment about hiring disciplined programmers, well, it's banal. And yet you present it as though it supported some point against strongly typed languages.
I'm not trying to be unique. I have a higher standard for hired programmers than many others it seems. I'm not sure what your point is.
As far as disciplined programmers not needing strongly typed languages: it's true. It doesn't mean that disciplined programmers won't use strongly typed languages or won't like them. I don't ever recall saying anything to the effect of "I use disciplined programmers so I don't need strongly typed languages," but rather "I don't like strongly typed languages, so I hire disciplined programmers." It's a significant difference.
Having written that, I don't know why I'm bothering to reply to you.
Indeed.
Sometimes he has to tell me which language he's coding in, it's that bad.
Any way to relegate him to less important tasks to build his skill level up?
Guess what - due to the recession the software company I work for is diversifying into catering
You're joking -- where do you work?
Were you ever in the army?
I smile as I read this, because I don't think I'd ever fit in to an organization where I was to follow a command chain and (potentially) not know why. I've never been on an army base, although the RCF often stop by at the McDonald's in my town on their way to Borden, Ontario.
I don't know -- perhaps I'm a self-riteous pissant who takes "my way or the highway" too seriously, but I put a lot of forethought and research into a decision and become very rigid on the implementation. It doesn't mean I won't move from it, but rather that there has to be a damned good reason to move away from what I'd set out in a project. If I'm proven wrong, I'll correct it and my belief in the new method will be no less strong than my belief in the old method was (until it was shown incorrect, that is.) :-)
If there were more people like us during the dot-com boom, it might still be going.
heh, I don't know if I'd go that far... There were a lot of things wrong with the dot-com boom.
When coding Perl as part of a large team, there's always one that hacks out his code so poorly that his code works only on Fridays, and when the Environment table is deleted, and with glibc-0.0.12 because he calls the C API directly which buffer overflows somewhere. -wT and use strict were forgotten long ago.
The key is not to let it get "long ago". Bi-weekly (mini) code reviews and constant interaction are important to prevent bad code from affecting a codebase in any language.
This is all, of course, horses for courses. We can hand-wave and hypothesize (sp?) forever and not get anywhere.
So do you turn on use strict and -wT, or do you decide not to grovel to your boss and install the unstable script? Afterwards, you can blame Perl and recommend switching to PHP. Explains a lot, eh?
I've been lucky; I have been given the power to eliminate lousy coders who won't at least try to maintain some semblance of good coding methods, and I've been given the power to delay some things in order to make sure they are done right so that the future won't take twice as long. Others aren't so lucky.
The use of -wT only works in the smallest scripts.
Bullshit. Complete and utter bullshit. You don't inherit 9M-LOC of Perl and THEN put -wT on it, you start with 10 LOC and use -wT from the start, eliminating the warnings as you go. I've got perl scripts thousands of lines long, across multiple files and modules, all with use strict and -wT. Not a single warning. Discipline, it's what I stated from the start.
Now your comment about "run until error" -- Isn't that what testing with the "million monkeys" data input method is all about? I don't care what language you use, you can't catch runtime errors at compile time.
Now as far as catching syntactical errors at compile-time: you're absolutely correct. Without having a way for Perl to run over every single line beforehand, there is no way to do this. It is a downfall, IMO.
Just because you don't like compile errors doesn't mean that you should ignore them.
I don't follow. All my perl code executes with -wT and strict mode. When I have to do strange things like call subroutine by reference, I turn off the specific strict-mode bit. So my perl is less good because...?
Curious, why are you "hacking" your ICQ client? Why doesn't it do what you want in the first place?
? I'm not hacking my ICQ client, I wrote a couple of small patches for Psi which add functionality I want (and which I hope make it in to the main tree, as I can't compile for win32 and all my win32 contacts seem to want the same features).
That's the beauty of OSS -- Things aren't just the way I want them, but I have carte blanche to modify it, or to find someone and pay them to do so. Try getting rid of all the crap that the latest ICQ clients have.
There is not a computer in existence that is powerful enough to translate "stupid politician" into "English" or "French", or even any other human language. I simply cannot be done.
We can't go that way, but how about an English to Chretien translator instead? :-)
wait, bilingual?? whoa dude! add some languages on top of it's voice recognition,, and it'd be like a mini interpreter!! that'd be sweet!
Yeah, we could finally understand what the fuck Chretien is trying to say.