Apparently you do, because you're complaining about it.
You keep claiming that Linux is the de facto standard, and that's simply not true. Maybe for your projects it is, but globally it's Linux that is the aberration.
Secondly, de facto standards are only so valid when there are real, official standards out there which are followed by all but a single player.
"I could write portable scripts, but this is not one of the goals of the projects I work in."
Well then, you should be developing on Linux. If you're never going to be exposed to anything else, then use the tools you want. If you're playing with multiple platforms, then using non-standard tools for non-portable results is your own fault.
"Just because it's a new kernel doesn't mean the syscall interface changes."
Absolutely true. On the other hand, there's no guarantee that they won't in Linux, and sometimes they do. That's the problem.
Sun Studio 11 (compiler/development IDE) is available for Linux, for free. Give it a try. On Solaris, it typically produces code that runs 10-15% faster than gcc. On Linux, the difference isn't quite so pronounced, but it's worth trying.
You've hit on a few of the key points in your post. However, let me address your complaint about sysadmins not liking people trying new things. In general, we admins LOVE new things! Take a look at all the gadgets around me or software installed on my workstations, and you'll see the truth in that.
However, there's also a supportability issue. If I have five users I'm responsible for, then I'll happily accept five different machines. If I have 30 users, then I don't want 30 different builds and application bundles. If I have 500 users (or even 100), then I cannot AFFORD to have variance between machines, if I'm expected to support them.
You want a program installed? If I'm going to install it, then I will have to make sure it won't interfere with the existing software, and then I have to keep track of the fact that your machine is different than anyone else's. If someone else wants a different program installed, same problem, squared. Alternatively, I can give you admin access to your workstation or laptop, but then I can't guarantee anything about that machine anymore, and can't support it.
The third alternative is to put in a formal request to have the software added to the official bundle, or at least put on an 'allowed/approved' list. That's the best solution, but also the most onerous, bureaucracy-laden, time-intensive one, as you well know.
Mostly, it's a matter of (a) scale, (b) supportability, and (c) accountability. If your system is strange and nonstandard then when it breaks it's easier to say, "it's " than explain the reasoning behind, "because you have installed, I can't help you."
I feel your pain, but there is some valid reason behind it.
Heh. At 4:30 in the morning, my sarcasm detector was offline. Partly from exhaustion, and partly because I've run into too many people who actually embrace obscurity as job protection.
You're right about the rates being driven partly by the sector. Solaris is huge in oil and gas, and is used extensively in the commodities trading side of O&G where the demands are almost the same as the financials. ISPs use a lot of Solaris and HPUX as well, where the perceived (but not necessarily actual) uptime requirements are also very high.
As for me, I'm quite happy with my new admin job, and don't see myself leaving anytime soon. Sorry.:-)
Actually, the biggest reason that Sun has been in the news so much lately is that they're actually doing a lot of new stuff. They _are_ getting better at talking to the media about it, but they do have stuff to talk about.
I do NOT want a default shell that is Bourne-like, but can't actually parse Bourne shell syntax properly. zsh, ksh, fine. bash, never. Not until they fix it at least.
Userland Solaris...isn't perfect. Not by a long shot.
But that said, gnu/linux is hardly standard Unix! Your tools like grep, tar, awk, etc., ARE standard! The GNU extensions to them are what breaks from the standard, and if you depend on the extensions, then you're the one who's leaving the standard behind, not Sun. Sun, HP-UX, AIX, *BSD, IRIX, and the rest all have standard tools. Linux tools are just Linux tools.
Put another way: Solaris tools don't break your scripts, Linux tools prevent your scripts from being portable. 99+% of the time I've seen Linux scripts break on another platform, it's because the author was too ignorant or lazy to write it properly and portably.
OK, Sun has deliberately skipped explicit backwards compatibility THREE TIMES in their entire history!!! The last time was about fifteen YEARS ago, and you're still complaining about it. Linux, on the other hand, does this routinely. A program compiled under a 2.2 kernel often wouldn't run under a 2.4 kernel, ditto for the 2.4-2.6 transition. Worse, the actual toolsets and configuration keep changing.
As for some of their innovations, Sun has been putting their money where their core is: Good technology. ZFS is quietly transforming small-to-medium data centres worldwide. dtrace is shaking out bad code for good in places people hadn't thought to look before. You seem to be equating user-facing tweaks (changes to OpenView, etc.), an area Sun has never been entirely successful in, with serious internal design development. The two are vastly different.
I'm a full time Solaris admin, and getting paid well for it. If supply and demand changes to get idiots like you out of the field, then I'd be happy.
Good, professional admins will always command a premium. There are very VERY few good, professional Linux or Windows admins, but they're paid roughly as well as Solaris (or HP-UX, etc.) admins.
"any monkey thats run linux for a bit has the same skills?"
I suggest you upgrade your skills beyond those of a monkey. And your attitude, while you're at it.
Interesting take, but let us not forget the one claim that has yet to be proven: camcorder copies hurt revenue.
Honestly, unless a movie is so promising but also so bad that everyone who wants to see it either does so in the first two days or downloads a camcorder version without reading any reviews, it won't lose any appreciable revenue. When Lord of the Rings came out, I had a copy burned almost a week before theatrical release, but I went to see it in the theatres (twice!) anyways, because it was a good movie and the effect of seeing it on the big screen was totally different than watching a blurry camcorder version. Same thing with buying it, when the DVDs came out.
Endless examples are out there. One that comes to mind is Tripwire. OS X is another one. Commercial Sendmail must have made a dollar or two for someone. Also, depending on how you view things, the entire set of commercial Linux companies might qualify.
There's a lot more philosophy at stake here than people may realise. Ask yourself this: When you use a search engine, what results do you want to see? That is a complex question, partly because the answer changes from search to search.
If I want to buy some fly fishing gear, I might search for "fly fishing equipment." Pretty straightforward, but the search engine has to decide whether I want to learn about the equipment, read reviews of specific items, or find retailers to buy from. If I then search for "Berkley fly rods," the engine has to make the same decision, and also has to throw in the possibility of the manufacturer's website. The trick is that I'm more likely to be looking for retailers with the second search than the first, so they should be given more prominence in the results.
All well and good, but (a) trying to build this logic is tricky, and (b) companies benefit greatly by landing high on the list for any and every remotely relevant (and in some cases, even totally irrelevant) search. Therefore, companies try hard to get their name up on the list as often as possible, and google (and other search engines) try to present a useful set of results.
The question comes down to this: Who is the search engine company beholden to? They're making money by selling advertising to companies, so they don't want to deliberately censure them; however, advertising is only as effective as the number of potential customers, so they want to maximise exposure--by providing the best results to the customer. Ultimately, companies and consumers are at odds about what constitutes the "best" results, and google has to sit in the middle, acting as gatekeeper.
Having a neutral algorithm that tries to minimise companies' attempts at gaming the system is a good system. They can use it to back their 'useful results' ideal, and avoid having to beat down companies directly, risking revenue.
In short, this guy paid too much money to a scammer masquerading as a consultant, and is paying the penalty for it.
Lots of very good points there. I'm not sure I'd agree with the final point about Office, though.
First of all, exit cost is absolutely key. However, exit cost is a one-time event. If you turf an existing product for a new one, you have to retrain everyone to use the new product, and then you're done. At least, for your existing staff. If new people come in, then they need to be trained on the new product if they don't already know it. That's an area where the monopoly can drive (and in fact, force) changes. Office 2007 DOES break the continuity, and forces a company to change their product from "old office" to "something new." Now this leads to two scenarios: A company can decide to avoid excess effort and retraining, and stick with their "old office;" or they can migrate to "something new." In the first case, no one will buy Office 2007, so MS will eliminate that possibility by cutting off support for previous versions. Now the company is forced to migrate to something new, and could theoretically reevaluate the field evenly, since all products are equally new. (In fact, most competing products will have been working hard to lower the entry cost by behaving as much like old office as possible, so they're likely much easier to retrain towards. Regardless...)
However, here's the thing: Everyone "knows" that Office is the standard. Everyone knows that Office 2007 will become the new standard eventually. If you have to switch to a new product, will you go with the one that is most likely to exploit the largest knowledge and user base after a year or so, or do you use a fringe product (relatively)?
Microsoft knows that they won't lose many customers by releasing a completely revamped piece of turd, so they're taking advantage of that fact to recreate the interface. It doesn't matter how good or bad it is, market forces will force it on a big enough segment that the competition will be left in the dust again. StarOffice (for instance) is a better product in many ways that MS Office 2003 and very similar to use; but now that the paradigm of an office suite has changed, they'll have to retool. In the meantime, SO will be less compatible from a usage perspective.
Or put another way, Microsoft changed the interface to break the end-user training compatibility that other products have been building. By increasing the entry costs to their own product for one generation, they've bumped the entry cost to other suites even more.
Well I was going to point out what was wrong in your post, but then I realised that it would leave nothing left. Still, I can't resist making a few points:
First of all, this has nothing to do with Open Source whatsoever. It has everything to do with monopolies and monoculture. The competition could be OSS or commercial software, and there would be no difference.
Microsoft and other large companies will routinely sell things that don't make money, in order to increase market share. Once they have a hard lock on the market, they will increase their prices to make a profit. Also, it's more accurate to say that open source people will keep devoting their time to something that (perhaps) hardly anyone uses, UNTIL THEY GET BORED OF IT. Then it languishes as another abandoned bit of unusable code. Try to count how many abandoned projects there are on freshmeat sometime when you have a spare week.
I'm not going to touch that car dealership analogy. It's...fascinating just as it is.
I think we're both looking at a fringe part of the article's central argument, and different fringes at that.
The/. article started by saying, "Why does Microsoft win the development environment war so often, when we all know it's a lifetime lock-in to Windows?"
But the question begs the answer: "Microsoft wins the development environment war so often because it's a lifetime lock-in to Windows." Being 'locked-in' in this manner means that this is potentially the best way to exploit all of the non-standard but friendly Windows development features. (And even languages, for that matter - C# comes to mind.) Furthermore, developers are happy and willing to be locked into Windows in this manner because they're already there.
Part of my original point is that monopolies are self-supporting. An end-user monopoly (i.e. Windows) will drive progress towards sustaining that monopoly (development tools designed for Windows, writing Windows software with Windows quirks, for Windows users). That's about it.
What the monopoly says, goes. They define a standard. Because they're MS, they define a standard that's different and incompatible with official standards. You either go with the market, or you swim upstream. This is about as clever as saying, "the reason red is red is because it's not yellow."
Nothing to see here. Market forces and ease of use win over features, stability, or quality.
Microsoft is in the almost unique position (in the computer industry, that is) of not having to care about timing.
Release Vista this year or next, it doesn't matter. People adopt now or in three years, it doesn't matter. If you're an accountant you'll have to change the forecasts and projections, but ultimately it's not that big of a deal. If the 'buy Vista' bubble is huge and short or moderate and sustained, people WILL buy it, when you can force it on them.
Here's what I'm afraid of. I'm afraid that a public for-profit corporation can get away with MORE abuse of the general public because their stupid marketing slogan has convinced people to let them do anything they want.
You said, "If Google can reign in doubleclick's ads..."
Why would they want to do that? They're a profit-making company, and they bought doubleclick to increase their profit, on some timescale. While it makes sense to buy-and-gut competitors, it makes sense in this case to buy-and-grow partners.
Google needs more evangelists like you, but their veneer is starting to crack. Pity that nobody has come up with a decent search engine since them.
Awwww...isn't that nice. Someone believes that Google still behaves nicely.
They're a public company. Their existence is now based on growing profits faster than the cost of living. That is NOT POSSIBLE without eventually screwing over the customers.
(and in fact, it's not possible at all, as a sustainable model. Sooner or later, the whole thing collapses.)
Women are refusing to burn themselves out. That might appear to be a very short-term disadvantage, but if there's any difference, it's only that the men in IT are a few years behind. By 2009, the men who aren't quitting now will either be dead (or disabled), will have left the industry, or will have demanded proper treatment.
I quit my last job two months ago, when I realised that charging more money (doubletime? Tripletime? Travel expenses?) for overtime wasn't making it go away, and I hadn't had a weekend in three months. I'm now getting paid about $15k less for more complex work, and am much happier. And I'm still in IT.
Bruce is a rare guy who is deeply knowledgeable in his field of expertise, and yet can see the rest of the world around him. His books and his articles constantly reiterate the point that computer security is no different from physical security in most cases, and security products are no different from any other products in most cases. In this article, he reminds us that the details of whether you're talking about a secure USB stick or a used car or a bathroom sink don't change the base economics of the matter, in general.
Fundamentally, it's cheaper and faster to sell shit to people than it is to sell quality. Making quality products is more expensive, more involved, and more time consuming--that means that I have to charge more to the customer, who generally won't know the difference. In the rare cases (maybe 10%?) where the consumer knows better, he will make a value decision on whether or not it's worth paying the premium, and will probably decide against it.
As a maker of quality products, I not only have higher costs and lower turnover, but my potential market share is probably only 5% of the market. That means I need to make a significant profit on each unit sold. My product which may be 10% better than the average will probably have to sell for twice as much.
I can't imagine when I was ever that optimistic. Here's my interpretation:
"DoubleClick is making money and everyone hates them. If we buy them and scale back the degree of shit, people will think we're a good company, our profits will go up, and we'll have lowered the bar for 'not evil' even further."
Google is a shitty company with a great service, who just bought out a profitable and even shittier company with a shitty service. The only rationale for any of this is to make cash for the owners.
Slashdot Mirror
"But I do not care about that."
Apparently you do, because you're complaining about it.
You keep claiming that Linux is the de facto standard, and that's simply not true. Maybe for your projects it is, but globally it's Linux that is the aberration.
Secondly, de facto standards are only so valid when there are real, official standards out there which are followed by all but a single player.
"I could write portable scripts, but this is not one of the goals of the projects I work in."
Well then, you should be developing on Linux. If you're never going to be exposed to anything else, then use the tools you want. If you're playing with multiple platforms, then using non-standard tools for non-portable results is your own fault.
"Just because it's a new kernel doesn't mean the syscall interface changes."
Absolutely true. On the other hand, there's no guarantee that they won't in Linux, and sometimes they do. That's the problem.
Sun Studio 11 (compiler/development IDE) is available for Linux, for free. Give it a try. On Solaris, it typically produces code that runs 10-15% faster than gcc. On Linux, the difference isn't quite so pronounced, but it's worth trying.
You've hit on a few of the key points in your post. However, let me address your complaint about sysadmins not liking people trying new things. In general, we admins LOVE new things! Take a look at all the gadgets around me or software installed on my workstations, and you'll see the truth in that.
However, there's also a supportability issue. If I have five users I'm responsible for, then I'll happily accept five different machines. If I have 30 users, then I don't want 30 different builds and application bundles. If I have 500 users (or even 100), then I cannot AFFORD to have variance between machines, if I'm expected to support them.
You want a program installed? If I'm going to install it, then I will have to make sure it won't interfere with the existing software, and then I have to keep track of the fact that your machine is different than anyone else's. If someone else wants a different program installed, same problem, squared. Alternatively, I can give you admin access to your workstation or laptop, but then I can't guarantee anything about that machine anymore, and can't support it.
The third alternative is to put in a formal request to have the software added to the official bundle, or at least put on an 'allowed/approved' list. That's the best solution, but also the most onerous, bureaucracy-laden, time-intensive one, as you well know.
Mostly, it's a matter of (a) scale, (b) supportability, and (c) accountability. If your system is strange and nonstandard then when it breaks it's easier to say, "it's " than explain the reasoning behind, "because you have installed, I can't help you."
I feel your pain, but there is some valid reason behind it.
Heh. At 4:30 in the morning, my sarcasm detector was offline. Partly from exhaustion, and partly because I've run into too many people who actually embrace obscurity as job protection.
:-)
You're right about the rates being driven partly by the sector. Solaris is huge in oil and gas, and is used extensively in the commodities trading side of O&G where the demands are almost the same as the financials. ISPs use a lot of Solaris and HPUX as well, where the perceived (but not necessarily actual) uptime requirements are also very high.
As for me, I'm quite happy with my new admin job, and don't see myself leaving anytime soon. Sorry.
Pardon me?
/bin/sh. Has been for ages, if not always. (what standards are set by companies using Solaris is another story.)
The default shell in Solaris is
bash is a horrible choice. It looks like it's sh compatible, but isn't. At least csh is clearly different.
Actually, the biggest reason that Sun has been in the news so much lately is that they're actually doing a lot of new stuff. They _are_ getting better at talking to the media about it, but they do have stuff to talk about.
Ummm...Solaris is free too.
And Linux very much is trendy.
FUCK no!!!
I do NOT want a default shell that is Bourne-like, but can't actually parse Bourne shell syntax properly. zsh, ksh, fine. bash, never. Not until they fix it at least.
Userland Solaris...isn't perfect. Not by a long shot.
But that said, gnu/linux is hardly standard Unix! Your tools like grep, tar, awk, etc., ARE standard! The GNU extensions to them are what breaks from the standard, and if you depend on the extensions, then you're the one who's leaving the standard behind, not Sun. Sun, HP-UX, AIX, *BSD, IRIX, and the rest all have standard tools. Linux tools are just Linux tools.
Put another way: Solaris tools don't break your scripts, Linux tools prevent your scripts from being portable. 99+% of the time I've seen Linux scripts break on another platform, it's because the author was too ignorant or lazy to write it properly and portably.
OK, Sun has deliberately skipped explicit backwards compatibility THREE TIMES in their entire history!!! The last time was about fifteen YEARS ago, and you're still complaining about it. Linux, on the other hand, does this routinely. A program compiled under a 2.2 kernel often wouldn't run under a 2.4 kernel, ditto for the 2.4-2.6 transition. Worse, the actual toolsets and configuration keep changing.
As for some of their innovations, Sun has been putting their money where their core is: Good technology. ZFS is quietly transforming small-to-medium data centres worldwide. dtrace is shaking out bad code for good in places people hadn't thought to look before. You seem to be equating user-facing tweaks (changes to OpenView, etc.), an area Sun has never been entirely successful in, with serious internal design development. The two are vastly different.
I'm a full time Solaris admin, and getting paid well for it. If supply and demand changes to get idiots like you out of the field, then I'd be happy.
Good, professional admins will always command a premium. There are very VERY few good, professional Linux or Windows admins, but they're paid roughly as well as Solaris (or HP-UX, etc.) admins.
"any monkey thats run linux for a bit has the same skills?"
I suggest you upgrade your skills beyond those of a monkey. And your attitude, while you're at it.
Interesting take, but let us not forget the one claim that has yet to be proven: camcorder copies hurt revenue.
Honestly, unless a movie is so promising but also so bad that everyone who wants to see it either does so in the first two days or downloads a camcorder version without reading any reviews, it won't lose any appreciable revenue. When Lord of the Rings came out, I had a copy burned almost a week before theatrical release, but I went to see it in the theatres (twice!) anyways, because it was a good movie and the effect of seeing it on the big screen was totally different than watching a blurry camcorder version. Same thing with buying it, when the DVDs came out.
Endless examples are out there. One that comes to mind is Tripwire. OS X is another one. Commercial Sendmail must have made a dollar or two for someone. Also, depending on how you view things, the entire set of commercial Linux companies might qualify.
There's a lot more philosophy at stake here than people may realise. Ask yourself this: When you use a search engine, what results do you want to see? That is a complex question, partly because the answer changes from search to search.
If I want to buy some fly fishing gear, I might search for "fly fishing equipment." Pretty straightforward, but the search engine has to decide whether I want to learn about the equipment, read reviews of specific items, or find retailers to buy from. If I then search for "Berkley fly rods," the engine has to make the same decision, and also has to throw in the possibility of the manufacturer's website. The trick is that I'm more likely to be looking for retailers with the second search than the first, so they should be given more prominence in the results.
All well and good, but (a) trying to build this logic is tricky, and (b) companies benefit greatly by landing high on the list for any and every remotely relevant (and in some cases, even totally irrelevant) search. Therefore, companies try hard to get their name up on the list as often as possible, and google (and other search engines) try to present a useful set of results.
The question comes down to this: Who is the search engine company beholden to? They're making money by selling advertising to companies, so they don't want to deliberately censure them; however, advertising is only as effective as the number of potential customers, so they want to maximise exposure--by providing the best results to the customer. Ultimately, companies and consumers are at odds about what constitutes the "best" results, and google has to sit in the middle, acting as gatekeeper.
Having a neutral algorithm that tries to minimise companies' attempts at gaming the system is a good system. They can use it to back their 'useful results' ideal, and avoid having to beat down companies directly, risking revenue.
In short, this guy paid too much money to a scammer masquerading as a consultant, and is paying the penalty for it.
Lots of very good points there. I'm not sure I'd agree with the final point about Office, though.
First of all, exit cost is absolutely key. However, exit cost is a one-time event. If you turf an existing product for a new one, you have to retrain everyone to use the new product, and then you're done. At least, for your existing staff. If new people come in, then they need to be trained on the new product if they don't already know it. That's an area where the monopoly can drive (and in fact, force) changes. Office 2007 DOES break the continuity, and forces a company to change their product from "old office" to "something new." Now this leads to two scenarios: A company can decide to avoid excess effort and retraining, and stick with their "old office;" or they can migrate to "something new." In the first case, no one will buy Office 2007, so MS will eliminate that possibility by cutting off support for previous versions. Now the company is forced to migrate to something new, and could theoretically reevaluate the field evenly, since all products are equally new. (In fact, most competing products will have been working hard to lower the entry cost by behaving as much like old office as possible, so they're likely much easier to retrain towards. Regardless...)
However, here's the thing: Everyone "knows" that Office is the standard. Everyone knows that Office 2007 will become the new standard eventually. If you have to switch to a new product, will you go with the one that is most likely to exploit the largest knowledge and user base after a year or so, or do you use a fringe product (relatively)?
Microsoft knows that they won't lose many customers by releasing a completely revamped piece of turd, so they're taking advantage of that fact to recreate the interface. It doesn't matter how good or bad it is, market forces will force it on a big enough segment that the competition will be left in the dust again. StarOffice (for instance) is a better product in many ways that MS Office 2003 and very similar to use; but now that the paradigm of an office suite has changed, they'll have to retool. In the meantime, SO will be less compatible from a usage perspective.
Or put another way, Microsoft changed the interface to break the end-user training compatibility that other products have been building. By increasing the entry costs to their own product for one generation, they've bumped the entry cost to other suites even more.
Well I was going to point out what was wrong in your post, but then I realised that it would leave nothing left. Still, I can't resist making a few points:
First of all, this has nothing to do with Open Source whatsoever. It has everything to do with monopolies and monoculture. The competition could be OSS or commercial software, and there would be no difference.
Microsoft and other large companies will routinely sell things that don't make money, in order to increase market share. Once they have a hard lock on the market, they will increase their prices to make a profit. Also, it's more accurate to say that open source people will keep devoting their time to something that (perhaps) hardly anyone uses, UNTIL THEY GET BORED OF IT. Then it languishes as another abandoned bit of unusable code. Try to count how many abandoned projects there are on freshmeat sometime when you have a spare week.
I'm not going to touch that car dealership analogy. It's...fascinating just as it is.
I think we're both looking at a fringe part of the article's central argument, and different fringes at that.
/. article started by saying, "Why does Microsoft win the development environment war so often, when we all know it's a lifetime lock-in to Windows?"
The
But the question begs the answer: "Microsoft wins the development environment war so often because it's a lifetime lock-in to Windows." Being 'locked-in' in this manner means that this is potentially the best way to exploit all of the non-standard but friendly Windows development features. (And even languages, for that matter - C# comes to mind.) Furthermore, developers are happy and willing to be locked into Windows in this manner because they're already there.
Part of my original point is that monopolies are self-supporting. An end-user monopoly (i.e. Windows) will drive progress towards sustaining that monopoly (development tools designed for Windows, writing Windows software with Windows quirks, for Windows users). That's about it.
What the monopoly says, goes. They define a standard. Because they're MS, they define a standard that's different and incompatible with official standards. You either go with the market, or you swim upstream. This is about as clever as saying, "the reason red is red is because it's not yellow."
Nothing to see here. Market forces and ease of use win over features, stability, or quality.
Microsoft is in the almost unique position (in the computer industry, that is) of not having to care about timing.
Release Vista this year or next, it doesn't matter. People adopt now or in three years, it doesn't matter. If you're an accountant you'll have to change the forecasts and projections, but ultimately it's not that big of a deal. If the 'buy Vista' bubble is huge and short or moderate and sustained, people WILL buy it, when you can force it on them.
Here's what I'm afraid of. I'm afraid that a public for-profit corporation can get away with MORE abuse of the general public because their stupid marketing slogan has convinced people to let them do anything they want.
You said, "If Google can reign in doubleclick's ads..."
Why would they want to do that? They're a profit-making company, and they bought doubleclick to increase their profit, on some timescale. While it makes sense to buy-and-gut competitors, it makes sense in this case to buy-and-grow partners.
Google needs more evangelists like you, but their veneer is starting to crack. Pity that nobody has come up with a decent search engine since them.
Awwww...isn't that nice. Someone believes that Google still behaves nicely.
They're a public company. Their existence is now based on growing profits faster than the cost of living. That is NOT POSSIBLE without eventually screwing over the customers.
(and in fact, it's not possible at all, as a sustainable model. Sooner or later, the whole thing collapses.)
Women are refusing to burn themselves out. That might appear to be a very short-term disadvantage, but if there's any difference, it's only that the men in IT are a few years behind. By 2009, the men who aren't quitting now will either be dead (or disabled), will have left the industry, or will have demanded proper treatment.
I quit my last job two months ago, when I realised that charging more money (doubletime? Tripletime? Travel expenses?) for overtime wasn't making it go away, and I hadn't had a weekend in three months. I'm now getting paid about $15k less for more complex work, and am much happier. And I'm still in IT.
Even better would be "minority party in power, but not really government."
Harper's government is an embarassment. The sooner the liberals get back in the game, the better.
But I digress.
Bruce is a rare guy who is deeply knowledgeable in his field of expertise, and yet can see the rest of the world around him. His books and his articles constantly reiterate the point that computer security is no different from physical security in most cases, and security products are no different from any other products in most cases. In this article, he reminds us that the details of whether you're talking about a secure USB stick or a used car or a bathroom sink don't change the base economics of the matter, in general.
Fundamentally, it's cheaper and faster to sell shit to people than it is to sell quality. Making quality products is more expensive, more involved, and more time consuming--that means that I have to charge more to the customer, who generally won't know the difference. In the rare cases (maybe 10%?) where the consumer knows better, he will make a value decision on whether or not it's worth paying the premium, and will probably decide against it.
As a maker of quality products, I not only have higher costs and lower turnover, but my potential market share is probably only 5% of the market. That means I need to make a significant profit on each unit sold. My product which may be 10% better than the average will probably have to sell for twice as much.
I can't imagine when I was ever that optimistic. Here's my interpretation:
"DoubleClick is making money and everyone hates them. If we buy them and scale back the degree of shit, people will think we're a good company, our profits will go up, and we'll have lowered the bar for 'not evil' even further."
Google is a shitty company with a great service, who just bought out a profitable and even shittier company with a shitty service. The only rationale for any of this is to make cash for the owners.