Also, having worked in the arse end of the film/TV industry, I'd estimate that 80-90% of the average budget is sheer waste.
Witness: I worked on one show where everything came out of the producer's own pocket, rather than being financed by a studio, hence he wouldn't stand for any waste. That show cost $80k per episode -- when similar studio-backed productions were costing around $500k per episode (and SF was around $1M/ep.)
I agree with you completely about the need to test, test, refine, and test some more. However, that's a matter of content, not of distribution.
What I was getting at, is that given X content (and an assumption that SOMEONE has tested it, because untested patches/updates are tantamount to network suicide), anyone with half a clue could DEPLOY that content with ZenWorks -- selectively, if necessary.
This sort of "bigger thinking" (ZenWorks for everyone, not just Platform N) is why I always find Novell's seminars so interesting, and mind you I'm not even a networking dude:)
"With ZenWorks setup in your environment, you can have your Windows helpdesk people install or repair applications on Linux systems without any real Linux knowledge."
I am not a sysadmin or even a real networking dude; my network-grok is at best crude. However... when I went to Novell's seminar and watched the ZenWorks live demo, my immediate thought was, "That is SO simple, even *I* could do that". How to accomplish desired tasks was flamingly evident, and did not assume any serious knowledge of networks beyond "this is connected to that". Very much designed for real people in the real world, not just for toplevel experts.
System history goes back thru Win98/95/3.1, online an average of 140 hrs/month since 1996. NO patches except for the con/con bug. I run an old version of ZoneAlarm set variously to medium or high (high doesn't always play nice). I visit a LOT of, um, dark corners, and download tons of shit (which I scan later with FProt for DOS; I don't run a resident AV), but I don't use P2P, and I tend not to install much noncommercial software, because frankly too much of it sucks (clean or not). Once in a while I run AdAware or Spybot, but they never find anything.
But I don't use IE/Outlook, WSH is disabled, and I don't visit iffy places with javascript active. I use a mail client that doesn't autoexecute anything. And I don't open spam attachments.
And that's all the precautions that are truly necessary.
I've taught my clients to follow these simple rules (primarily "beware of attachments and downloads of unknown provenance" and "always run your firewall"), and my regulars likewise have ZERO infections. The only ones I ever have to clean up anymore are the first-time clients.
Speaking as a mostly-Republican and fiscal conservative, with a social agenda of "Stay the fuck out of my life" (as contrasted to the liberal viewpoint of "We know what's best for you"), *I* am baffled by the phrase "family values". I haven't the vaguest idea what it refers to, and I've yet to meet anyone who does.
But buzzwords tend to be like that. And just wait til some politician uses a phrase like "leverage our traditions". Figure THAT one out.:/
Except that there are very few truly "American brands" left. The majority are now either foreign-owned or foreign-manufactured (when did you last see a consumer product that wasn't "Made in China"??)
So a boycott wouldn't actually hurt *American* interests.
But the issue you refer to isn't whether America is being a bully. It's whether Australia still has the balls to stand up and say "NO".
That was my thought as well. If you're going to go to this much trouble to design a mainboard that swings both ways, and given that the market for "need to *switch* between AMD and Intel CPUs" is probably limited to game programmers......why not have it all on one board? Have both CPUs and their necessary chipsets present, and switch between them in the BIOS, so it would require nothing more onerous than a reboot and a brief trip into CMOS setup.
Otherwise -- by about the 3rd time a person has to open up the box to change the CPU jumper, they'd be ready to fling up their hands and just buy a second box.
[blink] Oh, I get it... this is a marketing gimmick to get everyone to buy a SECOND motherboard!;)
I agree... DRM alone is enough to prevent me from buying downloadable music. I don't want what I paid for to be beholden to a company or tied to hardware either of which could go tits-up at any time.
Here's my own pricing scheme, what I would be willing to go for:
MP3, low-bitrate (64k mono would be fine) -- no charge, as "free samples" so I can taste-test without worrying about dropping cash on Crud (this would function exactly the same as radio exposure, at about the same audio quality).
128k to 360k stereo -- 5 to 10 cents each. Not worth chasing 'em around the net at that price.
WAV (or a lossless format that can be backconverted to a WAV *identical* with the original CD rip, so I can burn my own for-really audio CDs playable in ANY device) -- $1.00 each.
Chances are with that scenario, I'd dabble a lot in the freebies, go with the 10 centers for artists I already know I like well enough for casual listening, and often skip straight to the dollar files when I *know* I'm going to want a good copy (after all, from there I can make my own MP3s).
Side note: the only times I've bought lots of CDs are when 1) I had a good connection and access to lots of MP3s, and 2) back when I was DJing, and could tape any LP in the station's library. The freebie serves as advertising -- and best of all from a marketer's standpoint, advertising at my own expense.
Hmm... it's not just control over the distribution channels that the RIAA is afraid of losing. If the CD market as we know it goes away, there are an awful lot of ADVERTISING AGENCIES (and their major customers such as Clear Channel) that will be going hungry, who now get rich selling our eyes and ears to the RIAA cartel.
That was my impression of MacOS 7 too. Horrible memory management, on a par with Win3.0 (not even 3.1!), no crash protection, no true multitasking, and way slower than the GEM desktop it still so greatly resembled. -- The OS was tied to the hardware because part of it was in ROM, my friend said because otherwise boot time was unacceptably long by anyone's standards. He also told me that *the* reason Macs used SCSI HDs is because that was the only way they could get acceptable data throughput. (Macs still had a 16bit data bus 10 years after PCs had gone to a 32bit bus.)
In my observation MacOS 8.5 was better, but it still didn't truly multitask, had driv^H^H^H^H extension issues, and was far too easy to crash with what looked like an OS resource leak. (I remember trying to view a particular rather large but plain web page on OS 8.5 and finding that it froze solid at about 200k, no matter what browser was used. On Win3.1x or 9x, the *lowest* single plain-page limit I've found is 2.1 megs.)
I've been told that the reason Apple shitcanned the old OS was because it was such a mess of kludges that there was no upgrading it any further.
I haven't messed with OS X itself, but I did look at Darwin for x86, and liked that -- at least as a CLI (which was all it would give me) it didn't seem to have any obvious issues, leaks, or sluggish points (other than the fact that the installer was easy to break), and gave me good feelings toward BSD. Of course, that doesn't say anything about the desktop!!
I picked up OS X at a yard sale for a buck, but don't have any Apple hardware to try it on (and after having had my hands inside a few Macs, nor would I buy any!)
Someone mentioned elsewhere that there is one problem with the traceback system: if it's tied to the app in question, and the app's failure also crashes the traceback component, you'll never see a report on that fatal bug.
Likewise, if the traceback component itself crashes, you won't get your bug reports.
So it's important that traceback systems be robust and able to operate independently of the app they are supposed to crash-monitor.
Kindof like method, which boils down to "If I can do this with the software, why can't I do that?" until I hit something that falls over. Not so much random as "but it LOOKS like I should be able to...." in ways the coder frequently didn't anticipate.
Hence I am widely feared as "the beta tester who can break anything":)
Damn, so THAT'S why sometimes stuff goes pear-shaped!!
Next time I will be sure to keep those oranges around.;)
Re:Shouldn't that be too bloated to test?
on
Too Darned Big to Test?
·
· Score: 2, Interesting
I knew a programmer who worked for Apple as a member of their core OS development team, back around MacOS7. He told horror stories about how poorly managed it was. One problem he specifically ranted about was that some manager would decide that YOU were DONE with a given project, and physically remove your work machine from your desk, give it to some other coder, and give YOU someone else's half-finished work (which you'd then have to figure out before you could work on it). So no one ever got to actually FINISH their coding, hence there was a lot of half-baked code, kludges, and workarounds. And management *forbid* them from publishing a patch to fix a particular broken firmware, because management wanted people to buy their next machine (with fixed firmware), not just fix the old one!!
Anyway, my point is that just because what you see on the surface looks polished, doesn't necessarily mean the QA or development process is any better.
Well, not exactly. With a traditional auction, the auctioneer can indeed check whether the goods exist (something eBay cannot reasonable do), but it's impractical for him to verify whether the goods are stolen property, or broken in some non-obvious way, etc. All he can really confirm is that they are available to be delivered to the buyer, and look more or less as described.
eBay fraud is more akin to if the seller sneaked back and stole or broke the items in the display area after the buyer has paid but before the buyer can collect his new property.
It would make more sense for *eBay* to require that sellers post a bond.. but there again, that just keeps honest people honest, and would be prohibitively expensive for most small shops. Dishonest sellers already change eBay identities regularly (or have multiple IDs), to avoid getting kicked out of the system.
I did RTFA, and what I see is that some technologically-challenged type mistakes eBay for a meatspace auction -- the requirements as stated are exactly those I'd expect for licensing of a meatspace auctioneer.
What on earth does bid-calling have to do with selling stuff on eBay, where you never see or hear the buyers' spoken or gestured responses, but only a final high bid as determined by a computer?? That alone tells me that whoever thinks this applies to eBay sellers is weak on the concept. In fact, eBay ITSELF is the "auctioneer" here, and the seller is essentially the same as someone who is *consigning* items to a meatspace auction.
I agree that it smells strongly of "let's find another point to extract money from our constituents' wallets". It won't impact scammers one bit.
It may be even simpler. I think those people who dislike violent video games, actively select for testing scenarios to support the idea that such games increase violence -- whether they realise they're skewing the data or not.
Considering how many monsters and Former Humans I've killed, if such gaming induced violence, California would long since have become uninhabited!!;)
Moreover, there are so many different varieties of shitty parenting, some of which look perfectly good from the outside, that it's hard to identify shitty parenting such that any idiot can grok the concept.
It's a lot easier to sue something you can ID in no uncertain terms (rightly or wrongly) and then point an accusing finger at.
Besides, no shitty parent wants to ADMIT that they fucked up their kid. So they have to find someone else to blame.
OTOH, from what I've read the World Bank is such a muddle of fraud, waste, graft, and bribery, that putting Carly in charge may be a ploy to kill the wretched thing so it can be rebuilt from scratch;)
When I was following the XP-beta newsgroups, there was a hoorah over some serious bugs that had been reported and documented until everyone was tired of hearing about them, yet nothing was done to fix them. One of the M$ MVPs told me that he'd submitted these bugs several times, to the most official of internal channels, to no result, and he was finally told in so many words that they would not be fixed. -- At the time M$ management was having one of their periodic spasms of "Everyone aboard THIS train!" which in this case was focused on the cutesy new interface, and to hell with the OS under it.
It's the same sort of stupidity as observed in FuturePower's initial comment, but exercised at the corporate-management level.:(
[disclaimer] My XP is well-mannered overall, but I agree the CLI sucks in ways that were *not* broken in Win2K. However, you can run other command.com interfaces, from other Windows -- I've even used the one from DRDOS7:)
I have seen this kind of thing over and over. Here's an example: This particular program was designed to run on DOS, because 99% of its market were DOS users. This particular type of program normally includes a shell-to-DOS function (indeed, it is required for full functionality, as a sort of poor-man's multitasking).
However, the current maintainer REMOVED the shell-to-DOS function. When people complained, his response was "If you want to run more than one program at once, get a REAL operating system." Why? Because he's a linux bigot, and in his opinion anyone who runs DOS is a luser. This from someone maintaining an app used primarily on DOS... WTF??!
It boils down to "*I* want it that way, and if users don't like it, fuck 'em." Yet these same coders are the worst about rejecting user complaints or bug reports, or taking them as personal attacks.
Exactly. -- When I report a bug, it's because I like or care enough about the program in question to want it to work right (much as the parent post says). Thus I'm willing to put in my time to document where it went wrong, so the coder *has another tool with which to [hopefully] locate and fix the bug*.
But an awful lot of coders can't tolerate being told they screwed up, and as you say, ALL they get out of any bug report is the idea that they're being personally attacked. They can't admit to making an honest mistake, so they react as if the USER made the mistake, or like the user is out to get them.
Slashdot Mirror
Also, having worked in the arse end of the film/TV industry, I'd estimate that 80-90% of the average budget is sheer waste.
Witness: I worked on one show where everything came out of the producer's own pocket, rather than being financed by a studio, hence he wouldn't stand for any waste. That show cost $80k per episode -- when similar studio-backed productions were costing around $500k per episode (and SF was around $1M/ep.)
I agree with you completely about the need to test, test, refine, and test some more. However, that's a matter of content, not of distribution.
What I was getting at, is that given X content (and an assumption that SOMEONE has tested it, because untested patches/updates are tantamount to network suicide), anyone with half a clue could DEPLOY that content with ZenWorks -- selectively, if necessary.
This sort of "bigger thinking" (ZenWorks for everyone, not just Platform N) is why I always find Novell's seminars so interesting, and mind you I'm not even a networking dude :)
"With ZenWorks setup in your environment, you can have your Windows helpdesk people install or repair applications on Linux systems without any real Linux knowledge."
I am not a sysadmin or even a real networking dude; my network-grok is at best crude. However... when I went to Novell's seminar and watched the ZenWorks live demo, my immediate thought was, "That is SO simple, even *I* could do that". How to accomplish desired tasks was flamingly evident, and did not assume any serious knowledge of networks beyond "this is connected to that". Very much designed for real people in the real world, not just for toplevel experts.
"So... back to the auto shops with the bums!"
;)
I'm not sure if you misspelled "chop shop" or "Lincoln Park Pirates"
ZERO infections here as well:
System history goes back thru Win98/95/3.1, online an average of 140 hrs/month since 1996. NO patches except for the con/con bug. I run an old version of ZoneAlarm set variously to medium or high (high doesn't always play nice). I visit a LOT of, um, dark corners, and download tons of shit (which I scan later with FProt for DOS; I don't run a resident AV), but I don't use P2P, and I tend not to install much noncommercial software, because frankly too much of it sucks (clean or not). Once in a while I run AdAware or Spybot, but they never find anything.
But I don't use IE/Outlook, WSH is disabled, and I don't visit iffy places with javascript active. I use a mail client that doesn't autoexecute anything. And I don't open spam attachments.
And that's all the precautions that are truly necessary.
I've taught my clients to follow these simple rules (primarily "beware of attachments and downloads of unknown provenance" and "always run your firewall"), and my regulars likewise have ZERO infections. The only ones I ever have to clean up anymore are the first-time clients.
Speaking as a mostly-Republican and fiscal conservative, with a social agenda of "Stay the fuck out of my life" (as contrasted to the liberal viewpoint of "We know what's best for you"), *I* am baffled by the phrase "family values". I haven't the vaguest idea what it refers to, and I've yet to meet anyone who does.
:/
But buzzwords tend to be like that. And just wait til some politician uses a phrase like "leverage our traditions". Figure THAT one out.
Except that there are very few truly "American brands" left. The majority are now either foreign-owned or foreign-manufactured (when did you last see a consumer product that wasn't "Made in China"??)
So a boycott wouldn't actually hurt *American* interests.
But the issue you refer to isn't whether America is being a bully. It's whether Australia still has the balls to stand up and say "NO".
That was my thought as well. If you're going to go to this much trouble to design a mainboard that swings both ways, and given that the market for "need to *switch* between AMD and Intel CPUs" is probably limited to game programmers... ...why not have it all on one board? Have both CPUs and their necessary chipsets present, and switch between them in the BIOS, so it would require nothing more onerous than a reboot and a brief trip into CMOS setup.
;)
Otherwise -- by about the 3rd time a person has to open up the box to change the CPU jumper, they'd be ready to fling up their hands and just buy a second box.
[blink] Oh, I get it... this is a marketing gimmick to get everyone to buy a SECOND motherboard!
I agree... DRM alone is enough to prevent me from buying downloadable music. I don't want what I paid for to be beholden to a company or tied to hardware either of which could go tits-up at any time.
Here's my own pricing scheme, what I would be willing to go for:
MP3, low-bitrate (64k mono would be fine) -- no charge, as "free samples" so I can taste-test without worrying about dropping cash on Crud (this would function exactly the same as radio exposure, at about the same audio quality).
128k to 360k stereo -- 5 to 10 cents each. Not worth chasing 'em around the net at that price.
WAV (or a lossless format that can be backconverted to a WAV *identical* with the original CD rip, so I can burn my own for-really audio CDs playable in ANY device) -- $1.00 each.
Chances are with that scenario, I'd dabble a lot in the freebies, go with the 10 centers for artists I already know I like well enough for casual listening, and often skip straight to the dollar files when I *know* I'm going to want a good copy (after all, from there I can make my own MP3s).
Side note: the only times I've bought lots of CDs are when 1) I had a good connection and access to lots of MP3s, and 2) back when I was DJing, and could tape any LP in the station's library. The freebie serves as advertising -- and best of all from a marketer's standpoint, advertising at my own expense.
Hmm... it's not just control over the distribution channels that the RIAA is afraid of losing. If the CD market as we know it goes away, there are an awful lot of ADVERTISING AGENCIES (and their major customers such as Clear Channel) that will be going hungry, who now get rich selling our eyes and ears to the RIAA cartel.
That was my impression of MacOS 7 too. Horrible memory management, on a par with Win3.0 (not even 3.1!), no crash protection, no true multitasking, and way slower than the GEM desktop it still so greatly resembled. -- The OS was tied to the hardware because part of it was in ROM, my friend said because otherwise boot time was unacceptably long by anyone's standards. He also told me that *the* reason Macs used SCSI HDs is because that was the only way they could get acceptable data throughput. (Macs still had a 16bit data bus 10 years after PCs had gone to a 32bit bus.)
In my observation MacOS 8.5 was better, but it still didn't truly multitask, had driv^H^H^H^H extension issues, and was far too easy to crash with what looked like an OS resource leak. (I remember trying to view a particular rather large but plain web page on OS 8.5 and finding that it froze solid at about 200k, no matter what browser was used. On Win3.1x or 9x, the *lowest* single plain-page limit I've found is 2.1 megs.)
I've been told that the reason Apple shitcanned the old OS was because it was such a mess of kludges that there was no upgrading it any further.
I haven't messed with OS X itself, but I did look at Darwin for x86, and liked that -- at least as a CLI (which was all it would give me) it didn't seem to have any obvious issues, leaks, or sluggish points (other than the fact that the installer was easy to break), and gave me good feelings toward BSD. Of course, that doesn't say anything about the desktop!!
I picked up OS X at a yard sale for a buck, but don't have any Apple hardware to try it on (and after having had my hands inside a few Macs, nor would I buy any!)
Someone mentioned elsewhere that there is one problem with the traceback system: if it's tied to the app in question, and the app's failure also crashes the traceback component, you'll never see a report on that fatal bug.
Likewise, if the traceback component itself crashes, you won't get your bug reports.
So it's important that traceback systems be robust and able to operate independently of the app they are supposed to crash-monitor.
Kindof like method, which boils down to "If I can do this with the software, why can't I do that?" until I hit something that falls over. Not so much random as "but it LOOKS like I should be able to...." in ways the coder frequently didn't anticipate.
:)
Hence I am widely feared as "the beta tester who can break anything"
Damn, so THAT'S why sometimes stuff goes pear-shaped!!
;)
Next time I will be sure to keep those oranges around.
I knew a programmer who worked for Apple as a member of their core OS development team, back around MacOS7. He told horror stories about how poorly managed it was. One problem he specifically ranted about was that some manager would decide that YOU were DONE with a given project, and physically remove your work machine from your desk, give it to some other coder, and give YOU someone else's half-finished work (which you'd then have to figure out before you could work on it). So no one ever got to actually FINISH their coding, hence there was a lot of half-baked code, kludges, and workarounds. And management *forbid* them from publishing a patch to fix a particular broken firmware, because management wanted people to buy their next machine (with fixed firmware), not just fix the old one!!
Anyway, my point is that just because what you see on the surface looks polished, doesn't necessarily mean the QA or development process is any better.
Bug free, cheap, on time, works. Pick two.
(BTW thanks for the informative viewpoint.)
Well, not exactly. With a traditional auction, the auctioneer can indeed check whether the goods exist (something eBay cannot reasonable do), but it's impractical for him to verify whether the goods are stolen property, or broken in some non-obvious way, etc. All he can really confirm is that they are available to be delivered to the buyer, and look more or less as described.
eBay fraud is more akin to if the seller sneaked back and stole or broke the items in the display area after the buyer has paid but before the buyer can collect his new property.
It would make more sense for *eBay* to require that sellers post a bond.. but there again, that just keeps honest people honest, and would be prohibitively expensive for most small shops. Dishonest sellers already change eBay identities regularly (or have multiple IDs), to avoid getting kicked out of the system.
I did RTFA, and what I see is that some technologically-challenged type mistakes eBay for a meatspace auction -- the requirements as stated are exactly those I'd expect for licensing of a meatspace auctioneer.
What on earth does bid-calling have to do with selling stuff on eBay, where you never see or hear the buyers' spoken or gestured responses, but only a final high bid as determined by a computer?? That alone tells me that whoever thinks this applies to eBay sellers is weak on the concept. In fact, eBay ITSELF is the "auctioneer" here, and the seller is essentially the same as someone who is *consigning* items to a meatspace auction.
I agree that it smells strongly of "let's find another point to extract money from our constituents' wallets". It won't impact scammers one bit.
"If video games could so drastically affect behavior, where are all the Pac-Man addicts
;)
who should be running around eating everything in sight?"
Nonsense. It's quite clear that the current "epidemic of obesity" is due to a generation that played too much Pac-Man.
It may be even simpler. I think those people who dislike violent video games, actively select for testing scenarios to support the idea that such games increase violence -- whether they realise they're skewing the data or not.
;)
Considering how many monsters and Former Humans I've killed, if such gaming induced violence, California would long since have become uninhabited!!
Moreover, there are so many different varieties of shitty parenting, some of which look perfectly good from the outside, that it's hard to identify shitty parenting such that any idiot can grok the concept.
It's a lot easier to sue something you can ID in no uncertain terms (rightly or wrongly) and then point an accusing finger at.
Besides, no shitty parent wants to ADMIT that they fucked up their kid. So they have to find someone else to blame.
OTOH, from what I've read the World Bank is such a muddle of fraud, waste, graft, and bribery, that putting Carly in charge may be a ploy to kill the wretched thing so it can be rebuilt from scratch ;)
When I was following the XP-beta newsgroups, there was a hoorah over some serious bugs that had been reported and documented until everyone was tired of hearing about them, yet nothing was done to fix them. One of the M$ MVPs told me that he'd submitted these bugs several times, to the most official of internal channels, to no result, and he was finally told in so many words that they would not be fixed. -- At the time M$ management was having one of their periodic spasms of "Everyone aboard THIS train!" which in this case was focused on the cutesy new interface, and to hell with the OS under it.
:(
:)
It's the same sort of stupidity as observed in FuturePower's initial comment, but exercised at the corporate-management level.
[disclaimer] My XP is well-mannered overall, but I agree the CLI sucks in ways that were *not* broken in Win2K. However, you can run other command.com interfaces, from other Windows -- I've even used the one from DRDOS7
I have seen this kind of thing over and over. Here's an example: This particular program was designed to run on DOS, because 99% of its market were DOS users. This particular type of program normally includes a shell-to-DOS function (indeed, it is required for full functionality, as a sort of poor-man's multitasking).
However, the current maintainer REMOVED the shell-to-DOS function. When people complained, his response was "If you want to run more than one program at once, get a REAL operating system." Why? Because he's a linux bigot, and in his opinion anyone who runs DOS is a luser. This from someone maintaining an app used primarily on DOS... WTF??!
It boils down to "*I* want it that way, and if users don't like it, fuck 'em." Yet these same coders are the worst about rejecting user complaints or bug reports, or taking them as personal attacks.
Exactly. -- When I report a bug, it's because I like or care enough about the program in question to want it to work right (much as the parent post says). Thus I'm willing to put in my time to document where it went wrong, so the coder *has another tool with which to [hopefully] locate and fix the bug*.
But an awful lot of coders can't tolerate being told they screwed up, and as you say, ALL they get out of any bug report is the idea that they're being personally attacked. They can't admit to making an honest mistake, so they react as if the USER made the mistake, or like the user is out to get them.