A fatty diet causes a body missing a pancreas, or much of a pancreas, much grief. It's hard to digest, and can make your need for bathroom breaks in the middle of a meeting increase quite profoundly.
That's a very fair point. It's why I disliked the claim that 'Linux has $0 cost', because switching infrastructure tools is a very real cost, even when the software itself is free and high quality, and ignoring that switchover cost and risk makes a Linux advocate look foolish.
It's often well _worth_ the risk. Linux NIS, for example, is obviously superior to Solaris NIS because of its correct handling of low and high-numbered UID accounts and very superior handling of configurations files outside of/etc/passwd,/etc/group, etc.
b) Linux, overall, is not inferior. Many developer grade completely free Linuxes, like Fedora and Gentoo, would require a serious investment in skilled personnel to maintain or in an expensive external contract to maintain to the support quality you get with a Solaris license. But for most uses, even industrial, the lesser licenses of such Linuxes is quite enough for modest industrial use.
One of the big uses of Solaris right now is satisfying bean-counters that you have a supported architecture that they will accept, irrelevant of the technical feasibility of an appropriate Linux distribution. That's a political issue, not a technical one, and it's a compelling reason to use Solaris.
c) Cheap disk is almost always what you want these days. While some applications do require 15,000 RPM drives and the higher bandwidth of SAS or even fiber channel, most modern systems are quite happy with a reasonable amount of RAM, RAID configured 7200 RPM drives with hot spares at 7200 RPM, and 4 times the capacity or more. That kind of capacity enables the complete retirement of old equipment before the drives reach their MTBF, with the drives released for ligher load use in QA systems or selling the equipment off on Ebay.
Why? For the same overall reliability as Sun, you can buy a modest Linux box and install twice as much cheap disk on it for double the capacity. It won't be as fast without some of the complex file systems, but it's quite enough for personal and most industrial use.
Unfortunately, that's only for printout. There's still not a good open source PDF editor, and Word documents and their ilk are useful for editing and adjusting.
How many Solaris file servers have you run? I've had noticeable issues with them, most often fixed by buying twice as many Linux servers for the same price and gaining reedundancy.
Oh, I do that! It's an effective approach to reduce wasteful external bandwidth usage. But it's still not _free_. You have to pay for the server space and manpower to maintain the local repository, and the bandwidth for the one-time downloads. And internal LAN bandiwdth is not free, either. 200 Windows desktops, all hitting the server at the same time, can saturate some classes of links and interfere with other operations such as backups.
Using open source is like eating at home. It's normally a lot cheaper, and healthier, than eating out. But you still have to buy or grow the food, and even growing your own food costs time and resources, so it's not 'free as in beer'.
I've admittedly not worked with Google's calendar. But the alleged 'drop-in replacements' for Exchange aren't, or at least not yet: they just don't work well with real Outlook clients. Replacing the tight integration of Outlook with its calendar tools is very difficult. All of the Exchange replacements seem to require an unstable and unreliable 'Connector' to inter-operate with Outlook, and it costs a lot more to support in IT resources and wasted user time than simply buying an Exchange server.
A mail client that uses an open source calendar source, and integrates it well with email, would be a great Outlook replacement, but I've not seen this either. Evolution, for example, behaves very poorly with its Exchange service, at least the last time I tried it. And don't even think of suggesting 'Horde': I spent a lot of time trying to get it integrated and working in a production kenvironment, and it was awful.
Replaceing both Outlook and Exchange together is theoretically possible, but it would have to have very tight, effective calendar and email integration. Is Gmail and Google Calendar working well? Even if it is, it raises the problem that your calendar and email are off-site and you are vulnerable to data theft and loss of services if your external connection is cut.
Yes, bandwidth does cost money for Windows. That's partly why the retail price can be so misleading.
Some commercial support is included with a bare Windows license, including the nominal $50 OEM prices with new machines. Mind you, those OEM prices are a bit misleading, but they do include someone to call and rant to when you can't figure out what happened and you need help getting your machine to boot. That these calls are likely to be more frequent and more devastatingly unhelpful with Windows is a separate matter, often not factored in properly.
And whether or not you've 'gotten beyond what other proprietary offerings can get you', you've still entered the world where MySQL itself is not $0 cost. So please, let's remember to count the cost of these other components of open source, to make the comparisons to closed source products more fair and complete, and not get tricked by someone who misunderstands things like the support requirements of Windows.
I cost a lot more than the average Windows guy, as a case in point. On the other hand, I replace about 4 Windows guys in personal productivity, and tend to provide a lot more services on the same amount of hardware, so it's a good investment.
Note also, that $0/box is misleading. Updates cost bandwidth, commercial support costs license money, and some Linux compatible software is licensed in ways requiring payment for commercial use. (The MySQL licenses and their interesting clauses come to mind.) Nevertheless, the ability to do very low-cost or free prototype and testing systems is invaluable in industrial work.
OpenOffice will not take much more corporate share until two things happen:
1: A workable replacement for Outlook, Exchange, and its calendar service is released. Too many office personnel use it, and it's not a bad calendar service.
2: Word document compatibility improves quite a lot. Swapping back and forth between OpenOffice and Word still causes nasty layout and compatibility problems, especially for graphics intensive documents and templates made by Microsoft Office users.
I get quite a lot of money for integrating open source tools to specific client needs. It's much cheaper, and more flexible, to pay my salary to provide access to a huge range of tools than it is to specify, purchase, and implement one closed source project that turns out to be complete vaporware or where the company goes belly up and the software can no longer be supported.
It is a great joy of an open source developer's career when, during the time that a closed source company or consultant is writing their bids and release schedules and Gant charts and Powerpoint presentations, an open source developer has already found a compatible tool, tested it, modified it for local use, and put it in production. This has happened to me repeatedly throughout my career. On occasion, I've been overruled and the closed source tools used for 'business support' reasons, and on several of those occasions I've seen the closed source toolkit thrown out a year later and the entire system rebuilt from my notes.
That is an event to warm one's heart on those long, cold nights sitting in the server room repairing a mess.
While that's true, it's a separate matter. There's just no need for Microsoft to _plan_ or deliberately leak the software. It's far too likely to be stolen as a matter of course, with no corporate malfeisance necessary.
"A few hundred people", including off-site partners, is more broadly distributed than many commercial and most private software projects. It could, possibly, have been a malicious behavior by a Microsoft employee or partner, but there is no compelling reason to think that it was anything other than a simple theft of the code.
That's not a "dodgy business practice issue", that's a "it's very difficult to secure several hundred DVD's in several hundred people's hands", coupled with the understandable popularity of such a pre-release of one of the world's most popular OS.
This is complete nonsense. Please look into the history of the theft of Half-Life 2 from Steam, and of the various internal corporate documents that crackers like Kevin Mitnick used to steal with considerable success and considerable frequency. Such documents have considerable appeal to crackers to try and steal, and they will certainly attempt to do so via any means they find available, whether that is using an undetected rootkit on a laptop inside Microsoft's networks, or stealing a DVD or DVD image from a partner. Tight security for anything as broadly deployed as a Windows operating system at this stage of testing is basically a fond desire, but unlikely in the extreme to be successful.
Well, you do have a point. It's possible, in theory, to extract the resources needed to support a corporation in a community by other means, such as forcing them to build low-cost housing, forcing them to pay for new parking facilities, etc. But I'm not aware of any cases where such deals, themselves, covered the overall cost of the corporation to the community. As I understand the economics, that comes from property taxes, income taxes on the employees, sales taxes on the employee and corporate purchases, etc.
One can't wave a magic wand, say "no taxes" and not provide some alternative means of providing the necessary community resources for most businesses, especially brick & mortar based businesses. Too much density of such new, opportunistic businesses can destroy the existing community, by driving out residents, by raising the costs of food and water and power that have to be increasingly imported beyond their means, by hiring only people from outside the local community and eliminating companies that provided work for local residents, etc. These phenomena are particularly common when large complexes move near to small towns, especially when Wal-mart shows up.
This is the same rule as 'lower your price to gain more customers'. Corporations cost money to taxpayers to support. Roads, hospitals, schools, utilities, emergency services, military protection, etc. are all expected by corporations as part of local services. Lower the taxes too far, and those huge business moving to your nation will bleed you dry.
Amen! I've spent quite a bit of time in group environments trying to convince managers and developers that they _must not_ rely on heavily personalized local systems, because the time spent cleaning up the mess and trying to 'save' their working environment when they have a major hardware or software problem is hideously expensive to both them and the IT group.
I support them by helping them have good software management and configuration tools to be able to re-establish the necessary parts of a custom environment from the standard image. It's incredibly helpful to build environments and testing efforts as well, because you know what tweaks were done to allow the software to build and to perform well or to avoid failing in the development environment, because those tweaks must be replicated to operate in the testing environment rather than simply imaging nsome developer's mess of a build environment.
Gentoo seems to be fabulous at this, and its software build components are often a good model of how to do this. The RedHat style 'mock' utility is also a good teaching tool for this approach. Lessons learned there are very useful in even Windows environments.
I got "Nation" for Christmas, and am looking forward to reading it. But it's a collaborative work: if my claim is correct, it's exactly how Sir Pratchett could extend his useful time as a beloved author.
Oh, my yes, Cheery Littlebottom is a character to prize. So was Vimes's reaction to her name: that was priceless, and the sort of thing that does not get taught in a normal management class because very few people could pull it off.
But my point was not that Pratchett did not collaborate before, but that he is collaborating a lot more _now_. I think it's a wonderful defense against losing focus and against loss of creativity as he fights Alzheimer's: I understand that Alzheimer's can lead to a serious flattening of creativity in its early stages, one that the victim may be aware of but which may not be apparent to others. Collaboration with sharp, witty people such as those who helped Sir Pratchett with his 'Science of Discworld' books is a great way to continue creative work when you're feeling a bit rusty, as almost any gifted professor with grad students doing research can testify.
Oh, please. I'd pay _money_ to see that. Terry Pratchett is very short, and seeing the breastplate haning below his hips and the hose all doubled up would be....
Well, it would be like the Librarian wearing a ruff.
Slashdot Mirror
A fatty diet causes a body missing a pancreas, or much of a pancreas, much grief. It's hard to digest, and can make your need for bathroom breaks in the middle of a meeting increase quite profoundly.
That's a very fair point. It's why I disliked the claim that 'Linux has $0 cost', because switching infrastructure tools is a very real cost, even when the software itself is free and high quality, and ignoring that switchover cost and risk makes a Linux advocate look foolish.
It's often well _worth_ the risk. Linux NIS, for example, is obviously superior to Solaris NIS because of its correct handling of low and high-numbered UID accounts and very superior handling of configurations files outside of /etc/passwd, /etc/group, etc.
No, I _haven't_ tried that. I'll do so ASAP, thank you for the pointer.
a) Me, too.
b) Linux, overall, is not inferior. Many developer grade completely free Linuxes, like Fedora and Gentoo, would require a serious investment in skilled personnel to maintain or in an expensive external contract to maintain to the support quality you get with a Solaris license. But for most uses, even industrial, the lesser licenses of such Linuxes is quite enough for modest industrial use.
One of the big uses of Solaris right now is satisfying bean-counters that you have a supported architecture that they will accept, irrelevant of the technical feasibility of an appropriate Linux distribution. That's a political issue, not a technical one, and it's a compelling reason to use Solaris.
c) Cheap disk is almost always what you want these days. While some applications do require 15,000 RPM drives and the higher bandwidth of SAS or even fiber channel, most modern systems are quite happy with a reasonable amount of RAM, RAID configured 7200 RPM drives with hot spares at 7200 RPM, and 4 times the capacity or more. That kind of capacity enables the complete retirement of old equipment before the drives reach their MTBF, with the drives released for ligher load use in QA systems or selling the equipment off on Ebay.
Why? For the same overall reliability as Sun, you can buy a modest Linux box and install twice as much cheap disk on it for double the capacity. It won't be as fast without some of the complex file systems, but it's quite enough for personal and most industrial use.
Unfortunately, that's only for printout. There's still not a good open source PDF editor, and Word documents and their ilk are useful for editing and adjusting.
How many Solaris file servers have you run? I've had noticeable issues with them, most often fixed by buying twice as many Linux servers for the same price and gaining reedundancy.
Oh, I do that! It's an effective approach to reduce wasteful external bandwidth usage. But it's still not _free_. You have to pay for the server space and manpower to maintain the local repository, and the bandwidth for the one-time downloads. And internal LAN bandiwdth is not free, either. 200 Windows desktops, all hitting the server at the same time, can saturate some classes of links and interfere with other operations such as backups.
Using open source is like eating at home. It's normally a lot cheaper, and healthier, than eating out. But you still have to buy or grow the food, and even growing your own food costs time and resources, so it's not 'free as in beer'.
I've admittedly not worked with Google's calendar. But the alleged 'drop-in replacements' for Exchange aren't, or at least not yet: they just don't work well with real Outlook clients. Replacing the tight integration of Outlook with its calendar tools is very difficult. All of the Exchange replacements seem to require an unstable and unreliable 'Connector' to inter-operate with Outlook, and it costs a lot more to support in IT resources and wasted user time than simply buying an Exchange server.
A mail client that uses an open source calendar source, and integrates it well with email, would be a great Outlook replacement, but I've not seen this either. Evolution, for example, behaves very poorly with its Exchange service, at least the last time I tried it. And don't even think of suggesting 'Horde': I spent a lot of time trying to get it integrated and working in a production kenvironment, and it was awful.
Replaceing both Outlook and Exchange together is theoretically possible, but it would have to have very tight, effective calendar and email integration. Is Gmail and Google Calendar working well? Even if it is, it raises the problem that your calendar and email are off-site and you are vulnerable to data theft and loss of services if your external connection is cut.
Yes, bandwidth does cost money for Windows. That's partly why the retail price can be so misleading.
Some commercial support is included with a bare Windows license, including the nominal $50 OEM prices with new machines. Mind you, those OEM prices are a bit misleading, but they do include someone to call and rant to when you can't figure out what happened and you need help getting your machine to boot. That these calls are likely to be more frequent and more devastatingly unhelpful with Windows is a separate matter, often not factored in properly.
And whether or not you've 'gotten beyond what other proprietary offerings can get you', you've still entered the world where MySQL itself is not $0 cost. So please, let's remember to count the cost of these other components of open source, to make the comparisons to closed source products more fair and complete, and not get tricked by someone who misunderstands things like the support requirements of Windows.
Right, like those God botherers have brought peace to the Middle East with their latest Crusade?
I cost a lot more than the average Windows guy, as a case in point. On the other hand, I replace about 4 Windows guys in personal productivity, and tend to provide a lot more services on the same amount of hardware, so it's a good investment.
Note also, that $0/box is misleading. Updates cost bandwidth, commercial support costs license money, and some Linux compatible software is licensed in ways requiring payment for commercial use. (The MySQL licenses and their interesting clauses come to mind.) Nevertheless, the ability to do very low-cost or free prototype and testing systems is invaluable in industrial work.
OpenOffice will not take much more corporate share until two things happen:
1: A workable replacement for Outlook, Exchange, and its calendar service is released. Too many office personnel use it, and it's not a bad calendar service.
2: Word document compatibility improves quite a lot. Swapping back and forth between OpenOffice and Word still causes nasty layout and compatibility problems, especially for graphics intensive documents and templates made by Microsoft Office users.
No need. They had beards and headscarves.
I actually how Richard Stallman manages to fly anywhere with his raggedy beard and that, non-conformist look in his eye.
I get quite a lot of money for integrating open source tools to specific client needs. It's much cheaper, and more flexible, to pay my salary to provide access to a huge range of tools than it is to specify, purchase, and implement one closed source project that turns out to be complete vaporware or where the company goes belly up and the software can no longer be supported.
It is a great joy of an open source developer's career when, during the time that a closed source company or consultant is writing their bids and release schedules and Gant charts and Powerpoint presentations, an open source developer has already found a compatible tool, tested it, modified it for local use, and put it in production. This has happened to me repeatedly throughout my career. On occasion, I've been overruled and the closed source tools used for 'business support' reasons, and on several of those occasions I've seen the closed source toolkit thrown out a year later and the entire system rebuilt from my notes.
That is an event to warm one's heart on those long, cold nights sitting in the server room repairing a mess.
While that's true, it's a separate matter. There's just no need for Microsoft to _plan_ or deliberately leak the software. It's far too likely to be stolen as a matter of course, with no corporate malfeisance necessary.
That's because it's _cold_. I'm looking at the weathermap of Ontario right now, and parts of Ontario are -20 degrees, Fahrenheit.
"A few hundred people", including off-site partners, is more broadly distributed than many commercial and most private software projects. It could, possibly, have been a malicious behavior by a Microsoft employee or partner, but there is no compelling reason to think that it was anything other than a simple theft of the code. That's not a "dodgy business practice issue", that's a "it's very difficult to secure several hundred DVD's in several hundred people's hands", coupled with the understandable popularity of such a pre-release of one of the world's most popular OS.
This is complete nonsense. Please look into the history of the theft of Half-Life 2 from Steam, and of the various internal corporate documents that crackers like Kevin Mitnick used to steal with considerable success and considerable frequency. Such documents have considerable appeal to crackers to try and steal, and they will certainly attempt to do so via any means they find available, whether that is using an undetected rootkit on a laptop inside Microsoft's networks, or stealing a DVD or DVD image from a partner. Tight security for anything as broadly deployed as a Windows operating system at this stage of testing is basically a fond desire, but unlikely in the extreme to be successful.
Well, you do have a point. It's possible, in theory, to extract the resources needed to support a corporation in a community by other means, such as forcing them to build low-cost housing, forcing them to pay for new parking facilities, etc. But I'm not aware of any cases where such deals, themselves, covered the overall cost of the corporation to the community. As I understand the economics, that comes from property taxes, income taxes on the employees, sales taxes on the employee and corporate purchases, etc.
One can't wave a magic wand, say "no taxes" and not provide some alternative means of providing the necessary community resources for most businesses, especially brick & mortar based businesses. Too much density of such new, opportunistic businesses can destroy the existing community, by driving out residents, by raising the costs of food and water and power that have to be increasingly imported beyond their means, by hiring only people from outside the local community and eliminating companies that provided work for local residents, etc. These phenomena are particularly common when large complexes move near to small towns, especially when Wal-mart shows up.
This is the same rule as 'lower your price to gain more customers'. Corporations cost money to taxpayers to support. Roads, hospitals, schools, utilities, emergency services, military protection, etc. are all expected by corporations as part of local services. Lower the taxes too far, and those huge business moving to your nation will bleed you dry.
Amen! I've spent quite a bit of time in group environments trying to convince managers and developers that they _must not_ rely on heavily personalized local systems, because the time spent cleaning up the mess and trying to 'save' their working environment when they have a major hardware or software problem is hideously expensive to both them and the IT group.
I support them by helping them have good software management and configuration tools to be able to re-establish the necessary parts of a custom environment from the standard image. It's incredibly helpful to build environments and testing efforts as well, because you know what tweaks were done to allow the software to build and to perform well or to avoid failing in the development environment, because those tweaks must be replicated to operate in the testing environment rather than simply imaging nsome developer's mess of a build environment.
Gentoo seems to be fabulous at this, and its software build components are often a good model of how to do this. The RedHat style 'mock' utility is also a good teaching tool for this approach. Lessons learned there are very useful in even Windows environments.
I got "Nation" for Christmas, and am looking forward to reading it. But it's a collaborative work: if my claim is correct, it's exactly how Sir Pratchett could extend his useful time as a beloved author.
Oh, my yes, Cheery Littlebottom is a character to prize. So was Vimes's reaction to her name: that was priceless, and the sort of thing that does not get taught in a normal management class because very few people could pull it off.
But my point was not that Pratchett did not collaborate before, but that he is collaborating a lot more _now_. I think it's a wonderful defense against losing focus and against loss of creativity as he fights Alzheimer's: I understand that Alzheimer's can lead to a serious flattening of creativity in its early stages, one that the victim may be aware of but which may not be apparent to others. Collaboration with sharp, witty people such as those who helped Sir Pratchett with his 'Science of Discworld' books is a great way to continue creative work when you're feeling a bit rusty, as almost any gifted professor with grad students doing research can testify.
Oh, please. I'd pay _money_ to see that. Terry Pratchett is very short, and seeing the breastplate haning below his hips and the hose all doubled up would be....
Well, it would be like the Librarian wearing a ruff.