A mail client is one thing I never find myself wanting for on any platform
Wish I could say that.
if you don't like Mozilla's bundled client
What if you just don't want to use Mozilla, but would like to use the mail part? Oh... you can't...
Windows users have The Bat!, Eudora, and Mulberry
All of which are commercial and/or ad-ware, or (in the case of Eudora) repeatedly associated with allegations of malware.
I considered using The Bat! at one point... even downloaded it to try. And then my computer's clock got screwed up and The Bat! went into expired mode. And wouldn't come out. No way for me to access the email it had already retrieved (and it happily went off and tried to retrieve more email... and THEN wouldn't let me access it). Emails to the company were ignored as well.
I used to use Eudora, but the recent versions are bloated, filled with ads (unless you register), and they keep having allegations of spyware/malware. Even without the ads or spy/malware the bloat stinks.
I even heard Microsoft makes a mail client or two
Which are repeatedly filled with security problems... and lack features like not displaying embedded HTML graphics unless requested - essential with modern spam methods.
There are, of course, a host of other MUA's for Windows. I've tried most of them. Most stink. The one I'm currently using is ok, but it's dead - no future development. And there are some areas that could certainly use improvement.
Nifty... I like it. And answered my own questions about it after reading the FAQ (yes, you can increase the count... yes, you can set whitelist addresses that don't decrease the count, etc).
I may setup a new clean email address and try this out. My hotmail account is a bed of spam at this point.
Uh, because being a subscribed user has its perks?
It's not a scam. If you don't want to pay, then don't. If you want to pay, then you'll not only get things a whole week ahead of time, you'll also get support.
What on earth does this have to do with "the cult of Linux"?
Well, it does change the truth... 14 is not "fully half of" 30.
Your point is well made though - there are lies, damned lies, and statistics. If you don't know how the survey was made (including the questions asked), the survey population, and (possibly) some of the raw data, then the statistics coming out of it is nearly useless.
Although I'd say the main cause of poor data here was the/. blurb. The article had issues too, but the/. summary was, well, crap.
If only you could get a PVR that just worked, and was programmable like a VCR, with Showview or some other listings, and could pause live TV?
Sure. Are you willing to pay $500 for it? That's how much the hardware costs. Ooops... forgot. You're not spreading the software development costs out over a monthly subscription... up that to $750 then.
and allow you to interface it to a PC for archiving of old shows
Oh... up it to $1000. We're gonna get our pants sued off.
Wait... what am I saying? Why not just buy a PC with an ATI All-in-wonder card?
Because the interface sucks rocks. Actually, sucking rocks would be an improvement on the interface and recording quality.
Frankly, the monthly fees aren't really for the guide data - yeah, some of the costs go to that and to the dialup/web servers to support the customer base, but the vast majority covers development costs and day-to-day operations. If you're willing to have a PVR that will never have a software update, never have new features, and has to rely on the amazingly crappy show data that is broadcast in sideband (ala VCRPlus+) then you could get a standalone PVR. But it's going to be around $500 to cover the hardware and development.
Or I can sell you one for $400, then charge you ~$10/month or a largish lifetime fee, provide software updates, new features, oh, and better guide data. As business models go, I'll take the second one.
You're right about the giants coming though -- most cable operators are looking to provide PVR style functionality soon. But wow are they crippled. Of course, it may not matter - the boxes are cheap (usually free) and the monthly fee just becomes another line item on a $100/mo cable bill. I keep hoping that TiVo, at least, will be able to fight a lot of these guys on a patent basis, but it seems unlikely.
If it was filed in October 2002 then Amazon is claiming that they didn't use the method prior to October 2001... which I find doubtful. I also question that nobody else was using it prior to October 2001. Which is what needs to be proven to invalidate it.
And while the patent is somewhat novel, I don't think it's sufficiently different from other advertising models (magazine publishing, television, radio) that select what ads to play during which shows to be considered inobvious. But, hey, neither of us are patent clerks. Thank God.
Yes, and AOL was doing it prior to Prodigy on their proprietary service.
I (or, more likely, my parents) somehow got solicited as a beta tester for AOL when they were first releasing their PC client. This was in the late 80's. It ran on GeoWorks and sucked... badly. Didn't handle anything faster than 2400 baud (I believe I had a USR 9600 HST at the time) and apparantly used xmodem as the transfer protocol.
And while AOL still sucks, it doesn't suck nearly as much as it once did. Scary thought.
There's AcceleratedX, which I used long, long ago in a galaxy far, far away because it was one of the only X Servers that supported the Imagine128 chipset back when it was new.
Yes. It's commercial.
I believe there are others as well, but don't feel like Googleing hard enough to find them.
Does he do this while waving his hand in a dismissing motion?
Gotta wonder if it work on the IRS auditors though...
Finance Director: "These are not the balance books you are looking for. Everything is fine." Auditor: "These aren't the balance books we're looking for. Everything is fine."
One question is, how binding is this? If a member of the W3C patents a process then starts telling people to pay up (a few years down the line, maybe), is this really any protection?
Yes it is. If you are part of a standard's committee, and part of the committee's rules of participation state that you must disclose (and/or license at little or no cost) any intellectual property rights that are relevant to a standard you're involved in, then you can't submarine IP and expect to collect on it. It's been ruled illegal several times - the clearest case being Dell attempting to claim patent rights on the "VL-bus" VESA standard. See this for more details.
Rambus is in pretty much the same situation now - they're trying to claim patent rights on DDRAM, SDRAM, and pretty much every other kind of memory process... even though they were part of JEDEC at the time the standards were being decided on. And it's looking highly unlikely that they'll win at this point. Darn.
Why don't you go include some Microsoft code in your project and then try and pick your own license. You'll find Microsoft on your back so quick that your head will spin
Depends on what code you're using. Most of the Microsoft libraries are not under any sort of restrictive licensing. You can link to them to your heart's content and then distribute your program under whatever license you care to.
The same is not true for GPL licensed libraries. Link to them at all and there goes your freedom as a developer - you no longer have a choice as to what license your code is under. It's under GPL. Hope you like it (and if you don't, then, yes, choose a different library or write your own).
By merely using a predefined API the GPL has taken over your own intellectual property. That is what is meant by "viral". And it's why any reasonable developer licenses libraries under the LGPL and not the GPL -- because if I interface with your library using a predefined API and don't muck around with the internals then there's no legitimate reason you should be able to claim my code.
So the GPL is only viral if you concede that ALL copyright is viral
Uh... no. All copyright does affect derivative works (which you didn't state correctly in your post), but most derivative works would not include things like making references to other texts, using the pre-defined APIs, etc. GPL does claim that merely linking code makes it a derivative work.
We use several BSD, MIT, and LGPL libraries in our product here. We've even modified some GPL code to provide services we need. And we'll happily provide the modifications we've done to all of the libraries back to the community -- even though none of it is distributed -- because none of it contains our own core business logic. But we damn well won't link a GPL library (even though we're not distributing) because it could cause too many problems down the line.
GPL for programs is fine and dandy by me -- it makes sense that any derivative work from a stand-alone program should be considered GPL (or BSD, or whatever). On libraries, however, it's freaking stupid and removes the freedom from the developer. No thanks.
Well, of the things you listed - non-verbal alliances, vendettas, and "brainless voice comments" -- only one of those is really social interaction. And a rather negative one at that. Most computer players wouldn't bother with a vendetta - it's not very efficient, especially if the opponent is much better than you and not merely lucky.
The others are essentially poking holes in the AI. Human opponents are better than bots because they're much more "interesting" -- humans can learn and adapt far better than bots do, and they don't use ungodly accuracy to be "better" (although I do know humans who have accuracies that rate up there with the bots... scary really. One of my good friends was one of the top rated Quake players once-upon-a-time, and I know to avoid the hell out of him if playing him in a game of CTF. I don't suck, but he's that much better than me.)
I wouldn't really call FPS or RTS games "social interaction". If you want to see what real social interaction is, then try one of the MMORPGs. I will never play one again because of the time sucking capability they have. But there's much more realistic social interaction there, albeit primarily with people with very low social interaction abilities in the first place.
I think the recall has something to do with recognising an ad that youve seen previously
Perhaps. But both my wife and myself have been FF'ing through commercials (at 20x - 2x is too slow, 60x usually too fast) and will see a commercial that either looks interesting, is for some product we want to watch commercials for (generally her employer or former employer), or have heard about and want to watch (like the Nike soccer streaker commercial).
Obviously it's only the first one that would be of interest to advertisers - the second is a non-sell and the third requires someone to have watched it in the first place. And while the first one doesn't happen very often, neither does our watching commercials at all. I'd say it's about equal in occurance to the others.
Fact of the matter is, however, most commercials are even crappier than most shows. I only watch the shows I like. I've been stuck watching commercials at friends' and relatives' houses and... wow... no wonder so many people think TV sucks.
Re:Gartner Group is at it again
on
CIOs Looking At OSS
·
· Score: 4, Insightful
It has to go somewhere because going nowhere is called dying. If you don't move forward you will be left behind by your competition as they continue to innovate, refine, and improve.
And while I disagre with what Mr. Weiss said (there are several companies or groups that can tell you where Linux is going -- but probably only one or two (Redhat and IBM) can express it in a way that a C?O would grok), it is a valid question to ask.
Constant as expected? So you're saying there are absolutely zero bugs or security holes in Linux right now? And that it has everything you could possibly want, so there's no reason to continue development on it, right? Guess we better tell the Kernel devs, GCC devs, KDE/Gnome peeps, and XFree86 group that they need to find something else to do now because it's "perfect".
Get real. Companies want a stable, well supported platform but they also want to know where it's going -- because it's freaking expensive to change platforms. Companies want to know that the platform they chose right now will be supported for a decade, and that it'll be the platform they chose again in a decade. And don't say "well, you have the source! You can make it do whatever you want!". Yeah. Right. It's a smart business plan to move away from your core competencies and spend resources on something like that.
You may want to try here for Calvin and Hobbes. Far Side doesn't appear to be online, at least not officially - but if you Google for it, you'll get to read one of those nifty DMCA complaints.
There are tons of commercial apps. Which you'll never see. Because you're looking for consumer apps, while most developers don't write for the consumer market - they write for the business market.
I've been developing on Unix (Solaris, HP-UX, and AIX) for over a decade. Not one of the applications will ever be seen by a consumer because it's business logic and backbone server stuff. Heck, most of this stuff isn't even seen by anyone outside the company -- I think my current position is the closest to even that occuring, but realistically all a customer will see is a spec on how to interface to our system.
As far as Nick goes - you're pretty much spot on. I was a huge OS/2 advocate and Petreley was one of the few columnists that praised OS/2. But even then he made crap up and promised things that only the most devout zealots would actually believe. Frankly, having him on the Linux bandwagon isn't exactly a great thing - he's pretty well discounted by anyone with a clue. (Of course, most of those people don't read Infoworld/eWeek/whatever anymore either -- but execs do, so he should play well in that audience).
Oh... and we're desperately hoping to move our current app off AIX and onto Linux. AIX's linker sucks rocks with C++, and gives us problems debugging whether we're using xlC or g++.
Which is pretty much the same stresses you go through around here - we may not get snow that often, but most of December and January are spent near or below freezing with the relative humidity right at about 0%.
Of course, steel strut construction, which is becoming more popular, eliminates these problems too.
Once again, this isn't a problem that hasn't been solved. Another poster probably nailed it though - the land is so valuable it doesn't matter much what you put on top of it. It's going to sell.
That's what I thought and that's what common sense is. But not here in Japan. Summer is humid and long (not just 1 week humid hot like it is usually in Germany once in a year). 3 Monthes humidty of easily 70% up to 90% and temperatures of 30+ centigrades. At night it often stays between 25 and 30 centigrades
Uh... this is a problem? What, have they totally ignored other areas in the world where that kind of weather would be refreshing? Heck, a 30 centigrade day here during summer is cool. Here being the American South. The worst areas are the coastal ones - try Houston, TX; Baton Rouge, LA; Mobile, AL; or most of Florida during the summer. Atlanta and Dallas are also hot and humid (Atlanta being more humid, Dallas being hotter). In a normal year the months of July and August are above 90 deg. F (32 C) all day, with streaks above 100 (38 C).
And, yet, somehow we manage to insulate both for the heat of summer and the cold of winter. And it's done with fiberglass insulation, vapor barriers, double paned windows, and lots of ventilation in the roof (soffit vents, wind-driven turbines, and ridge vents). There are some who claim that reflective barriers do a better job than fiberglass, but I've yet to see anyone but salespeople give data proving it.
And air conditioners completely changed the way of cooling, introducing the humidity problem
Uh... air conditioners remove humidity. And rather well.
Frankly it sounds like they're just built like crap, and simply because Japanese architects and foremen are unwilling to look at problems that were already solved a half century ago elsewhere.
Well, if you look purely at CPU speed/price then the best buy is the Athlon 1800 for $58. The 1900 costs $63, which is a 5.5% performance improvement for a 8.6% price increase.
Except that that's rather silly. I'd recommend buying an Athlon 2200 ($97) or 2400 ($123) - the performance improvements are marginal, but so are the price increases. It's not like it used to be where CPU speed steps were several hundred dollars apart. I certainly wouldn't recommend anything higher unless you had a good reason for it - and good reasons for faster CPU are few and far between when the bottleneck is more likely to be RAM, HD, or (most likely) the user.
Other than that, I very much agree with your point, and it's how I've been building systems for sometime now. The product cycles are too short and speed improvements too large to buy top of the line constantly - you'll be purchasing a new $1500-2000 system every year for a 20% speed increase. That or buy a $750-1000 system every year for the same increase and minimal realistic differences in performance.
Clock rate means nothing - a 2.0 GHz Athlon will outperform a 2.0 GHz P4 by a considerable margin.
The PR is a much better indicator, although not always accurate. And when you compare Athlon XP PR's to Intel MHz ratings then the Athlon's beat out Intel in nearly every case - the top of the line is unusual in that the P4 2.8 and 3.06 is actually (slightly) cheaper than the Athlon 2800 and 3000. But for every other processor speed the Athlon is cheaper - often much cheaper. Hell, for the price of a P4 1.8 ($103) I can buy an Athlon 2200 ($97) (and this is the most favorable comparison at the low end -- I could also buy a P4 1.3 for $103, but that would be silly).
AMD is certainly behind, but not by much. And as far as $/performance is concerned, they're considerably ahead of Intel.
Did the guy receive spam the day after he opted-in? The day after that? What if it didn't start for a week? Then why should he expect the opt-out to be any faster than that?
If you bother looking at the various state opt-out programs you'll see that they're handled quarterly - if you sign up for it on Dec 31 then you'll be opted-out starting Jan 1. If you sign up on Jan 1, however, you'll need to wait until the next time the list is distributed - usually Apr 1. The dates may vary, but it's almost always a quarterly schedule.
Now, admittedly, email is a bit of a different thing from traditional telemarketing or snail mail - you don't have the lead time involved in either of those and there's no way you can claim that your computer system doesn't have connectivity. I think that for spam it wouldn't be unreasonable for opt-in/out lists to have a maximum turn-around time of 24 hours (to allow distribution to partners - even connected systems aren't going to be instantaneous).
I plan on taking a stand against this personally by breaking off business relationships with companies that insist on sharing my data with their "exclusive marketing partners" and crap like that
So who are you going to bank with? What about insurance? Home loan? Car loan?
I can help you on one front at least - State Farm does not share it's data with 3rd parties. Or at least the branch of State Farm that I still have a policy with - I don't recall which one that is. It was pleasant to read through their notice to discover that, for once, I didn't have to send in the form to opt out.
But other than that? Forget it. Especially the home loan - refi my mortgage just to hope that it doesn't end up getting sold to someone that doesn't share data? That's a laugh.
Oh, there is one way to beat the telemarketers at their own game. It works best with credit card companies and their slimy offers. When they call you up to offer you some discount club membership sign up for it. Then excercise your right of return (usually 60 days). Do this a couple times and you'll be amazed at how quickly they stop offering you these fantastic deals -- it costs too much to sign someone up and then deal with the accounting on undoing it. They'll be sure to flag your data as "do not share" afterwards.
Why are you comparing Oracle's pricing to Redhat's? They have nothing to do with one another.
As it happens, $2500/yr for support is considerably better than the $10k+/year offered by most Unix vendors. That's for a mid-level box... have more CPUs or faster CPUs and it's usually more expensive.
Slashdot Mirror
A mail client is one thing I never find myself wanting for on any platform
Wish I could say that.
if you don't like Mozilla's bundled client
What if you just don't want to use Mozilla, but would like to use the mail part? Oh... you can't...
Windows users have The Bat!, Eudora, and Mulberry
All of which are commercial and/or ad-ware, or (in the case of Eudora) repeatedly associated with allegations of malware.
I considered using The Bat! at one point... even downloaded it to try. And then my computer's clock got screwed up and The Bat! went into expired mode. And wouldn't come out. No way for me to access the email it had already retrieved (and it happily went off and tried to retrieve more email... and THEN wouldn't let me access it). Emails to the company were ignored as well.
I used to use Eudora, but the recent versions are bloated, filled with ads (unless you register), and they keep having allegations of spyware/malware. Even without the ads or spy/malware the bloat stinks.
I even heard Microsoft makes a mail client or two
Which are repeatedly filled with security problems... and lack features like not displaying embedded HTML graphics unless requested - essential with modern spam methods.
There are, of course, a host of other MUA's for Windows. I've tried most of them. Most stink. The one I'm currently using is ok, but it's dead - no future development. And there are some areas that could certainly use improvement.
Doesn't help when someone "trusted" gets pissed off because they're a lying, stealing asshole and signs up your account for every spam list out there.
:)
Yeah... it happened to me. Was still worth exposing the lying, stealing asshole though
Nifty... I like it. And answered my own questions about it after reading the FAQ (yes, you can increase the count... yes, you can set whitelist addresses that don't decrease the count, etc).
I may setup a new clean email address and try this out. My hotmail account is a bed of spam at this point.
Uh, because being a subscribed user has its perks?
It's not a scam. If you don't want to pay, then don't. If you want to pay, then you'll not only get things a whole week ahead of time, you'll also get support.
What on earth does this have to do with "the cult of Linux"?
Well, it does change the truth... 14 is not "fully half of" 30.
/. blurb. The article had issues too, but the /. summary was, well, crap.
Your point is well made though - there are lies, damned lies, and statistics. If you don't know how the survey was made (including the questions asked), the survey population, and (possibly) some of the raw data, then the statistics coming out of it is nearly useless.
Although I'd say the main cause of poor data here was the
If only you could get a PVR that just worked, and was programmable like a VCR, with Showview or some other listings, and could pause live TV?
Sure. Are you willing to pay $500 for it? That's how much the hardware costs. Ooops... forgot. You're not spreading the software development costs out over a monthly subscription... up that to $750 then.
and allow you to interface it to a PC for archiving of old shows
Oh... up it to $1000. We're gonna get our pants sued off.
Wait... what am I saying? Why not just buy a PC with an ATI All-in-wonder card?
Because the interface sucks rocks. Actually, sucking rocks would be an improvement on the interface and recording quality.
Frankly, the monthly fees aren't really for the guide data - yeah, some of the costs go to that and to the dialup/web servers to support the customer base, but the vast majority covers development costs and day-to-day operations. If you're willing to have a PVR that will never have a software update, never have new features, and has to rely on the amazingly crappy show data that is broadcast in sideband (ala VCRPlus+) then you could get a standalone PVR. But it's going to be around $500 to cover the hardware and development.
Or I can sell you one for $400, then charge you ~$10/month or a largish lifetime fee, provide software updates, new features, oh, and better guide data. As business models go, I'll take the second one.
You're right about the giants coming though -- most cable operators are looking to provide PVR style functionality soon. But wow are they crippled. Of course, it may not matter - the boxes are cheap (usually free) and the monthly fee just becomes another line item on a $100/mo cable bill. I keep hoping that TiVo, at least, will be able to fight a lot of these guys on a patent basis, but it seems unlikely.
Well that's not likely, but people with pacemakers are likely to fall over dead if they get too close to one of these mats.
Pacemaker + powerful EM field = powerful EM field + corpse.
I have no idea how close too close would be though...
If it was filed in October 2002 then Amazon is claiming that they didn't use the method prior to October 2001... which I find doubtful. I also question that nobody else was using it prior to October 2001. Which is what needs to be proven to invalidate it.
And while the patent is somewhat novel, I don't think it's sufficiently different from other advertising models (magazine publishing, television, radio) that select what ads to play during which shows to be considered inobvious. But, hey, neither of us are patent clerks. Thank God.
Yes, and AOL was doing it prior to Prodigy on their proprietary service.
I (or, more likely, my parents) somehow got solicited as a beta tester for AOL when they were first releasing their PC client. This was in the late 80's. It ran on GeoWorks and sucked... badly. Didn't handle anything faster than 2400 baud (I believe I had a USR 9600 HST at the time) and apparantly used xmodem as the transfer protocol.
And while AOL still sucks, it doesn't suck nearly as much as it once did. Scary thought.
There's AcceleratedX, which I used long, long ago in a galaxy far, far away because it was one of the only X Servers that supported the Imagine128 chipset back when it was new.
Yes. It's commercial.
I believe there are others as well, but don't feel like Googleing hard enough to find them.
Does he do this while waving his hand in a dismissing motion?
Gotta wonder if it work on the IRS auditors though...
Finance Director: "These are not the balance books you are looking for. Everything is fine."
Auditor: "These aren't the balance books we're looking for. Everything is fine."
One question is, how binding is this? If a member of the W3C patents a process then starts telling people to pay up (a few years down the line, maybe), is this really any protection?
Yes it is. If you are part of a standard's committee, and part of the committee's rules of participation state that you must disclose (and/or license at little or no cost) any intellectual property rights that are relevant to a standard you're involved in, then you can't submarine IP and expect to collect on it. It's been ruled illegal several times - the clearest case being Dell attempting to claim patent rights on the "VL-bus" VESA standard. See this for more details.
Rambus is in pretty much the same situation now - they're trying to claim patent rights on DDRAM, SDRAM, and pretty much every other kind of memory process... even though they were part of JEDEC at the time the standards were being decided on. And it's looking highly unlikely that they'll win at this point. Darn.
Why don't you go include some Microsoft code in your project and then try and pick your own license. You'll find Microsoft on your back so quick that your head will spin
Depends on what code you're using. Most of the Microsoft libraries are not under any sort of restrictive licensing. You can link to them to your heart's content and then distribute your program under whatever license you care to.
The same is not true for GPL licensed libraries. Link to them at all and there goes your freedom as a developer - you no longer have a choice as to what license your code is under. It's under GPL. Hope you like it (and if you don't, then, yes, choose a different library or write your own).
By merely using a predefined API the GPL has taken over your own intellectual property. That is what is meant by "viral". And it's why any reasonable developer licenses libraries under the LGPL and not the GPL -- because if I interface with your library using a predefined API and don't muck around with the internals then there's no legitimate reason you should be able to claim my code.
So the GPL is only viral if you concede that ALL copyright is viral
Uh... no. All copyright does affect derivative works (which you didn't state correctly in your post), but most derivative works would not include things like making references to other texts, using the pre-defined APIs, etc. GPL does claim that merely linking code makes it a derivative work.
We use several BSD, MIT, and LGPL libraries in our product here. We've even modified some GPL code to provide services we need. And we'll happily provide the modifications we've done to all of the libraries back to the community -- even though none of it is distributed -- because none of it contains our own core business logic. But we damn well won't link a GPL library (even though we're not distributing) because it could cause too many problems down the line.
GPL for programs is fine and dandy by me -- it makes sense that any derivative work from a stand-alone program should be considered GPL (or BSD, or whatever). On libraries, however, it's freaking stupid and removes the freedom from the developer. No thanks.
Well, of the things you listed - non-verbal alliances, vendettas, and "brainless voice comments" -- only one of those is really social interaction. And a rather negative one at that. Most computer players wouldn't bother with a vendetta - it's not very efficient, especially if the opponent is much better than you and not merely lucky.
The others are essentially poking holes in the AI. Human opponents are better than bots because they're much more "interesting" -- humans can learn and adapt far better than bots do, and they don't use ungodly accuracy to be "better" (although I do know humans who have accuracies that rate up there with the bots... scary really. One of my good friends was one of the top rated Quake players once-upon-a-time, and I know to avoid the hell out of him if playing him in a game of CTF. I don't suck, but he's that much better than me.)
I wouldn't really call FPS or RTS games "social interaction". If you want to see what real social interaction is, then try one of the MMORPGs. I will never play one again because of the time sucking capability they have. But there's much more realistic social interaction there, albeit primarily with people with very low social interaction abilities in the first place.
I think the recall has something to do with recognising an ad that youve seen previously
Perhaps. But both my wife and myself have been FF'ing through commercials (at 20x - 2x is too slow, 60x usually too fast) and will see a commercial that either looks interesting, is for some product we want to watch commercials for (generally her employer or former employer), or have heard about and want to watch (like the Nike soccer streaker commercial).
Obviously it's only the first one that would be of interest to advertisers - the second is a non-sell and the third requires someone to have watched it in the first place. And while the first one doesn't happen very often, neither does our watching commercials at all. I'd say it's about equal in occurance to the others.
Fact of the matter is, however, most commercials are even crappier than most shows. I only watch the shows I like. I've been stuck watching commercials at friends' and relatives' houses and... wow... no wonder so many people think TV sucks.
It has to go somewhere because going nowhere is called dying. If you don't move forward you will be left behind by your competition as they continue to innovate, refine, and improve.
And while I disagre with what Mr. Weiss said (there are several companies or groups that can tell you where Linux is going -- but probably only one or two (Redhat and IBM) can express it in a way that a C?O would grok), it is a valid question to ask.
Constant as expected? So you're saying there are absolutely zero bugs or security holes in Linux right now? And that it has everything you could possibly want, so there's no reason to continue development on it, right? Guess we better tell the Kernel devs, GCC devs, KDE/Gnome peeps, and XFree86 group that they need to find something else to do now because it's "perfect".
Get real. Companies want a stable, well supported platform but they also want to know where it's going -- because it's freaking expensive to change platforms. Companies want to know that the platform they chose right now will be supported for a decade, and that it'll be the platform they chose again in a decade. And don't say "well, you have the source! You can make it do whatever you want!". Yeah. Right. It's a smart business plan to move away from your core competencies and spend resources on something like that.
You may want to try here for Calvin and Hobbes. Far Side doesn't appear to be online, at least not officially - but if you Google for it, you'll get to read one of those nifty DMCA complaints.
There are tons of commercial apps. Which you'll never see. Because you're looking for consumer apps, while most developers don't write for the consumer market - they write for the business market.
I've been developing on Unix (Solaris, HP-UX, and AIX) for over a decade. Not one of the applications will ever be seen by a consumer because it's business logic and backbone server stuff. Heck, most of this stuff isn't even seen by anyone outside the company -- I think my current position is the closest to even that occuring, but realistically all a customer will see is a spec on how to interface to our system.
As far as Nick goes - you're pretty much spot on. I was a huge OS/2 advocate and Petreley was one of the few columnists that praised OS/2. But even then he made crap up and promised things that only the most devout zealots would actually believe. Frankly, having him on the Linux bandwagon isn't exactly a great thing - he's pretty well discounted by anyone with a clue. (Of course, most of those people don't read Infoworld/eWeek/whatever anymore either -- but execs do, so he should play well in that audience).
Oh... and we're desperately hoping to move our current app off AIX and onto Linux. AIX's linker sucks rocks with C++, and gives us problems debugging whether we're using xlC or g++.
Which is pretty much the same stresses you go through around here - we may not get snow that often, but most of December and January are spent near or below freezing with the relative humidity right at about 0%.
Of course, steel strut construction, which is becoming more popular, eliminates these problems too.
Once again, this isn't a problem that hasn't been solved. Another poster probably nailed it though - the land is so valuable it doesn't matter much what you put on top of it. It's going to sell.
That's what I thought and that's what common sense is. But not here in Japan. Summer is humid and long (not just 1 week humid hot like it is usually in Germany once in a year). 3 Monthes humidty of easily 70% up to 90% and temperatures of 30+ centigrades. At night it often stays between 25 and 30 centigrades
Uh... this is a problem? What, have they totally ignored other areas in the world where that kind of weather would be refreshing? Heck, a 30 centigrade day here during summer is cool. Here being the American South. The worst areas are the coastal ones - try Houston, TX; Baton Rouge, LA; Mobile, AL; or most of Florida during the summer. Atlanta and Dallas are also hot and humid (Atlanta being more humid, Dallas being hotter). In a normal year the months of July and August are above 90 deg. F (32 C) all day, with streaks above 100 (38 C).
And, yet, somehow we manage to insulate both for the heat of summer and the cold of winter. And it's done with fiberglass insulation, vapor barriers, double paned windows, and lots of ventilation in the roof (soffit vents, wind-driven turbines, and ridge vents). There are some who claim that reflective barriers do a better job than fiberglass, but I've yet to see anyone but salespeople give data proving it.
And air conditioners completely changed the way of cooling, introducing the humidity problem
Uh... air conditioners remove humidity. And rather well.
Frankly it sounds like they're just built like crap, and simply because Japanese architects and foremen are unwilling to look at problems that were already solved a half century ago elsewhere.
Well, if you look purely at CPU speed/price then the best buy is the Athlon 1800 for $58. The 1900 costs $63, which is a 5.5% performance improvement for a 8.6% price increase.
Except that that's rather silly. I'd recommend buying an Athlon 2200 ($97) or 2400 ($123) - the performance improvements are marginal, but so are the price increases. It's not like it used to be where CPU speed steps were several hundred dollars apart. I certainly wouldn't recommend anything higher unless you had a good reason for it - and good reasons for faster CPU are few and far between when the bottleneck is more likely to be RAM, HD, or (most likely) the user.
Other than that, I very much agree with your point, and it's how I've been building systems for sometime now. The product cycles are too short and speed improvements too large to buy top of the line constantly - you'll be purchasing a new $1500-2000 system every year for a 20% speed increase. That or buy a $750-1000 system every year for the same increase and minimal realistic differences in performance.
Clock rate means nothing - a 2.0 GHz Athlon will outperform a 2.0 GHz P4 by a considerable margin.
The PR is a much better indicator, although not always accurate. And when you compare Athlon XP PR's to Intel MHz ratings then the Athlon's beat out Intel in nearly every case - the top of the line is unusual in that the P4 2.8 and 3.06 is actually (slightly) cheaper than the Athlon 2800 and 3000. But for every other processor speed the Athlon is cheaper - often much cheaper. Hell, for the price of a P4 1.8 ($103) I can buy an Athlon 2200 ($97) (and this is the most favorable comparison at the low end -- I could also buy a P4 1.3 for $103, but that would be silly).
AMD is certainly behind, but not by much. And as far as $/performance is concerned, they're considerably ahead of Intel.
What a load of junk.
Did the guy receive spam the day after he opted-in? The day after that? What if it didn't start for a week? Then why should he expect the opt-out to be any faster than that?
If you bother looking at the various state opt-out programs you'll see that they're handled quarterly - if you sign up for it on Dec 31 then you'll be opted-out starting Jan 1. If you sign up on Jan 1, however, you'll need to wait until the next time the list is distributed - usually Apr 1. The dates may vary, but it's almost always a quarterly schedule.
Now, admittedly, email is a bit of a different thing from traditional telemarketing or snail mail - you don't have the lead time involved in either of those and there's no way you can claim that your computer system doesn't have connectivity. I think that for spam it wouldn't be unreasonable for opt-in/out lists to have a maximum turn-around time of 24 hours (to allow distribution to partners - even connected systems aren't going to be instantaneous).
I plan on taking a stand against this personally by breaking off business relationships with companies that insist on sharing my data with their "exclusive marketing partners" and crap like that
So who are you going to bank with? What about insurance? Home loan? Car loan?
I can help you on one front at least - State Farm does not share it's data with 3rd parties. Or at least the branch of State Farm that I still have a policy with - I don't recall which one that is. It was pleasant to read through their notice to discover that, for once, I didn't have to send in the form to opt out.
But other than that? Forget it. Especially the home loan - refi my mortgage just to hope that it doesn't end up getting sold to someone that doesn't share data? That's a laugh.
Oh, there is one way to beat the telemarketers at their own game. It works best with credit card companies and their slimy offers. When they call you up to offer you some discount club membership sign up for it. Then excercise your right of return (usually 60 days). Do this a couple times and you'll be amazed at how quickly they stop offering you these fantastic deals -- it costs too much to sign someone up and then deal with the accounting on undoing it. They'll be sure to flag your data as "do not share" afterwards.
Why are you comparing Oracle's pricing to Redhat's? They have nothing to do with one another.
As it happens, $2500/yr for support is considerably better than the $10k+/year offered by most Unix vendors. That's for a mid-level box... have more CPUs or faster CPUs and it's usually more expensive.