I still play Quake, which according to wikipedia was released June 22, 1996... I can always find people to play against, and thanks to ID open sourcing the quake code i have a version which is better than the original, and runs on modern hardware.
Add to that, games which are 15 years old require a fraction of today's hardware and run really well under emulation or virtualization which is widely available now. Dosbox for example, and there's nothing to stop you running a real copy of dos or windows 9x inside of something like vmware.
Lots of people also still have older machines sitting around, which would quite happily play older games... I have a real Amiga in a closet here for instance.
Some people play games on laptops, there are even laptops specifically designed for gaming (the big heavy ones), sure they have poor battery life but they will survive 1-2 seconds without power just fine. But more importantly, what about people who want to play on laptops like these in places where connections aren't available?
Or how about places where connections are limited in some way (ie you're stuck behind an http proxy or such)?
I have a desktop with 8gb ram, and a dec tulip 10/100 network card (designed for use in 64bit alpha systems)... Windows has no 64bit drivers for this card at all, and the 32bit version won't let me use all of the memory in the system.
A disturbing number of mac users actually install adobe reader and let it set itself as their default pdf viewer, despite that OSX already comes with a much better PDF viewer, people are conditioned to think that PDF files require adobe acrobat.
The difference is that windows is the only platform which doesn't come with a PDF reader by default...
And to make matters worse, many users aren't aware that alternative pdf readers exist at all, how many mac users do you think install adobe's viewer because they don't realise preview.app can handle PDF files very well. Users have the mindset that file formats are proprietary and belong to specific programs.
They target Adobe's PDF reader because it is extremely widespread, most users don't even realise PDF is a standard and that other readers exist... They think it's a proprietary format only supported by a single program. As a consequence, virtually every potential victim will be running exactly the same code, or a small subset of possible versions making them a very easy target. Also Adobe's software hasn't been attacked much before, and therefore is likely to have many more undiscovered bugs.
This is also the reason IE is generally targeted less, now that other browsers are taking significant market share away, except in corporate deployments (where the recent attacks on google proved that targeting IE is still an effective strategy).
Also, most malware filters permit PDF files through..
For people like your described "dad", Linux would sell very well if appropriately marketed... It will let them do what people want, and be available very cheaply including a large set of apps, on the cheapest of hardware. The key however, is marketing and training the sales droids...
Apple might be very well known in the US, but in other countries they are less so... They are also considered, rightly or wrongly, to be expensive and many people don't realise apple even make computers...
For linux on a desktop/laptop not that i know of, someone did mention roadnav but i haven't looked at it...
But a lot of dedicated gps devices (eg tomtom) actually run on linux, since those boxes run arm processors i wonder if it would be feasible to make their software run on an arm netbook...
When i was in school (age 6), we had a single computer in the whole school which ran a selection of very simple programs, one of which simulated a snooker table and calculated how many times a ball would bounce before falling down a corner pocket. You had to enter the width/height of the table and guess how many bounces... I entered a size of 0 for the table, and the program promptly crashed.. The teacher saw, called my actions stupid and sent me to the headmaster, who promptly banned me from ever touching a computer again so long as i was at that school.
What happens when MS stop supporting windows 7 and turn the activation servers off? Does that mean it will become useless 90 days afterwards?
What about for machines which aren't networked, or are on isolated networks which can't or aren't allowed to access the internet?
If they provide a corporate version which doesn't need to phone home, then pirates will simply pirate that instead, just like they did with the corporate versions of xp that didn't need activation.
They didn't say playing catchup in marketshare, they mean catchup in features, stability performance and appearance etc..
People don't buy windows because they've assessed multiple competing options (including osx and linux) and found windows to be technically superior or better value for money, they buy it because they don't realise anything else exists, because they're afraid to learn something new or because theyre forced to by having their data held to ransom by proprietary applications and their formats.
For bigger businesses (and governments), open source makes even more sense. Just think economies of scale, a million installs of a single open source application is the same cost as one install.
Support costs are variable, but with proprietary software they will be *ON TOP* of the cost of the software itself... Proprietary software support can really only come from one place, and is unlikely to be provided indefinitely... Sooner or later support for the versions you use will end and you will be forced to upgrade (if an upgrade path exists - no guarantee it will or if the vendor will even still exist). By contrast open source code can be supported by multiple parties, and if its important enough you always have the option to support it yourself. How many companies are lumbered with proprietary legacy systems that noone supports at all anymore?
And frankly there are only a few OSS software vendors that can provide quality enterprise level support.
A few is better than just one, there is only one vendor that can provide a proper level of support for windows, only one vendor that can support oracle etc.
Large companies typically have some applications developed in house, open source can help a lot here since your developers can now customise general applications rather than having to put up with what a vendor sells you, or code from scratch.
Don't forget other costs, like license management, this can become very expensive especially when not all of your software is licensed for every machine so you need to keep track of how many licenses you have, where it's deployed, and deal with deploying different software to different locations, not to mention technical license enforcement measures which must be maintained, managed and supported. With open source license management becomes pretty much irrelevant, even GPL code only has restrictions once you start distributing it to third parties, for your own internal use you can install it as many times as you want and use it for anything you want.
Cost is only a small part of the overall picture, open source offers you a level of flexibility not available with proprietary software, you are freed from the burdens of license management, you have no lock-in, you have multiple support options, you have guaranteed future availability of at least the versions you are currently using, you always have
And frankly, the code quality for most OSS solutions (even Apache) doesn't compare with good commercial code.
Having seen commercial code, i have to say a lot of it is complete garbage and extremely poorly written. Just look at formerly commercial code which has been open sourced for some examples, netscape, staroffice etc. Commercial code is typically churned out by bored programmers who are forced to rush to meet deadlines which causes them to cut corners and make some pretty nasty kludges. I personally find that i write better code when i'm doing it for my own benefit and at my own pace, rather than being rushed to meet a deadline and waiting for the clock to strike home-time. The biggest difference is that you can see the open source code, whereas you can only imagine just how shoddy the closed source code was before it got compiled.
I don't live very far from work, but to get there by bus requires taking 2 buses (have to travel quite a distance in the wrong direction before the bus meets up with the other bus route that goes past my workplace)... So a total of 4 bus rides a day for a relatively short distance. I can walk it in the same time it takes to catch the first bus, then wait for the second one to arrive (altogether about 1.5 hours), or i could drive which takes about 5 minutes and costs less in fuel than i would be paying for the bus tickets (i take a direct route and don't stop until i get there, unlike the bus).
Video editing is an interesting comparison, but video editing software generally doesn't lock you in... The input formats are standard, the output formats are standard, so while you may buy proprietary applications now, in a few years time you might find that replacing them with open source replacements becomes a viable and cheaper option, so you switch.
For anything where there aren't standards like this, you have to consider the very serious dangers of proprietary lockin. While your company is small and only employs 20 editors today, some vendor may give you a very cheap deal and you buy in to their software... A couple of years down the line, you now have 200 editors instead of 20 and this same vendor no longer offers you a cheap deal and is trying to milk you for everything you've got, meanwhile their software has largely stagnated and their software is now inferior to everyone else's. However you can't switch easily, because all your data is in formats only supported by that program, and all the rest of your workflow depends on a sequence of proprietary tools that don't work with anything else so you would need to replace lots of other components too. So now you're stuck with a choice between inferior overpriced tools or spending a huge amount of money to free yourself from them.
While developing your own software, or modifying an existing open source package to suit your needs probably doesn't make much sense for a company with 20 users, when you have an organization the size of the australian government you will get economies of scale. Software only needs to be written once and then you can distribute it to all your users at no additional cost. Most organizations of this size already employ developers to write in house bespoke applications anyway so you can split their time between projects, reuse developers for other projects once one goes into maintenance mode etc.
As to support, when buying off the shelf software support is usually extra (what you get included in the price is typically nothing, sometimes you might get 90 days installation help or something), especially if you want the top tier support packages, and even then there's no guarantee you'd get to speak to the actual developers or have changes made.. You end up having to adapt your business to the way the software works. By contrast, when you employ your own developers they will code the application to your exact requirements.
Consider your video editing again for a moment, of that $12k you pay to sony or apple, how much of that goes towards development of the product you're buying? Very little i imagine, most will go to profit, paying for staff not related to development, paying for marketing etc... Consider on a grander scale, all the thousands of other companies that need video editing software, splitting all the money they spend between a handful of companies... If we assume that 10% (a very high estimate) of the purchase price goes directly towards development/testing... That 10% is split between say 3-4 companies who are all competing with each other and therefore develop equivalent features independently.. And some of that development effort will go to things which don't benefit the end user - like license enforcement schemes... If all the companies that bought proprietary software instead of buying, spent 10% of the money to sponsor open source development you would have a massively superior product, with far less duplication of effort and far less wastage.
In order to connect the system to a network, it would need to be booted and thus the key must already have been entered. That's the problem, storing the key in the TPM system makes you vulnerable to attack since the thief now has the key and all the time in the world to extract it.
A lot of these "overly cautious" security people are actually being extremely detrimental to security... They prefer to stick to what they know, even when that is known to be far less secure than something else they could switch to. Also consider the various accreditation schemes, getting your product accredited is a costly and time consuming process, so quite often you will find that only old versions are accredited and new versions are not (and might not ever be)...
People erroneously think the accredited version is more secure when in fact it isn't and may well have known flaws that people could exploit. You end up with extremely ridiculous situations where a lot of effort is expended to work around the known flaws, additional software is purchased to try and mitigate the known flaws (and by having additional software there you increase the attack surface anyway), and time/money is even expended cleaning up the mess from worm/virus infections with such infections being considered an acceptable cost of doing business.
The one positive from this whole mess, is that you should be able to use this as evidence of why being tied to a proprietary system is bad. Even Microsoft strongly recommend that you upgrade from IE6 these days. Use this as a lesson going forward, any new apps you procure must comply with standards rather than being tied to a proprietary client. Make it a requirement on any companies you buy from, if your company is relatively large (and 100+ apps suggests it is) then vendors will fall over themselves to comply with your demands and if other companies do the same pretty soon standards compliant will become the norm and everyone will be much better off.
Standards compliant pages support lynx pretty well, lynx will simply ignore things it doesn't support and render the alt-tags instead of images etc. So long as you chose sensible names for the alt tags the textual page content is perfectly viewable. I use lynx quite often and find it great for getting to information quickly on well designed sites, without being distracted by any of the fancy graphics. IE6 doesn't degrade gracefully, it doesn't ignore unsupported features like lynx does, it tries to render them and botches the job totally.
Perhaps if some of those people making the decisions find themselves unable to view the sites they want to, they will realise how stupid those decisions were and make efforts to change them. With any luck, this will teach them the dangers of using proprietary apps which are tied to a single platform and they will take the dangers seriously when making any future decisions. So long as sites accommodate IE6 it will be considered acceptable to run it, which just continues the headache for everyone involved.
Slashdot Mirror
I still play Quake, which according to wikipedia was released June 22, 1996...
I can always find people to play against, and thanks to ID open sourcing the quake code i have a version which is better than the original, and runs on modern hardware.
Just think how much bandwidth they will be paying for, not only to handle all the legit players, but also to mitigate the risk of being attacked...
I agree entirely...
Add to that, games which are 15 years old require a fraction of today's hardware and run really well under emulation or virtualization which is widely available now. Dosbox for example, and there's nothing to stop you running a real copy of dos or windows 9x inside of something like vmware.
Lots of people also still have older machines sitting around, which would quite happily play older games... I have a real Amiga in a closet here for instance.
Some people play games on laptops, there are even laptops specifically designed for gaming (the big heavy ones), sure they have poor battery life but they will survive 1-2 seconds without power just fine.
But more importantly, what about people who want to play on laptops like these in places where connections aren't available?
Or how about places where connections are limited in some way (ie you're stuck behind an http proxy or such)?
I would rather pirate and crack, if i purchased i would feel i had bought a defective product and had to jump through extra steps to make it work.
I have a desktop with 8gb ram, and a dec tulip 10/100 network card (designed for use in 64bit alpha systems)...
Windows has no 64bit drivers for this card at all, and the 32bit version won't let me use all of the memory in the system.
A disturbing number of mac users actually install adobe reader and let it set itself as their default pdf viewer, despite that OSX already comes with a much better PDF viewer, people are conditioned to think that PDF files require adobe acrobat.
The difference is that windows is the only platform which doesn't come with a PDF reader by default...
And to make matters worse, many users aren't aware that alternative pdf readers exist at all, how many mac users do you think install adobe's viewer because they don't realise preview.app can handle PDF files very well. Users have the mindset that file formats are proprietary and belong to specific programs.
They target Adobe's PDF reader because it is extremely widespread, most users don't even realise PDF is a standard and that other readers exist... They think it's a proprietary format only supported by a single program.
As a consequence, virtually every potential victim will be running exactly the same code, or a small subset of possible versions making them a very easy target.
Also Adobe's software hasn't been attacked much before, and therefore is likely to have many more undiscovered bugs.
This is also the reason IE is generally targeted less, now that other browsers are taking significant market share away, except in corporate deployments (where the recent attacks on google proved that targeting IE is still an effective strategy).
Also, most malware filters permit PDF files through..
For people like your described "dad", Linux would sell very well if appropriately marketed... It will let them do what people want, and be available very cheaply including a large set of apps, on the cheapest of hardware.
The key however, is marketing and training the sales droids...
Apple might be very well known in the US, but in other countries they are less so...
They are also considered, rightly or wrongly, to be expensive and many people don't realise apple even make computers...
For linux on a desktop/laptop not that i know of, someone did mention roadnav but i haven't looked at it...
But a lot of dedicated gps devices (eg tomtom) actually run on linux, since those boxes run arm processors i wonder if it would be feasible to make their software run on an arm netbook...
When i was in school (age 6), we had a single computer in the whole school which ran a selection of very simple programs, one of which simulated a snooker table and calculated how many times a ball would bounce before falling down a corner pocket. You had to enter the width/height of the table and guess how many bounces...
I entered a size of 0 for the table, and the program promptly crashed.. The teacher saw, called my actions stupid and sent me to the headmaster, who promptly banned me from ever touching a computer again so long as i was at that school.
What happens when MS stop supporting windows 7 and turn the activation servers off?
Does that mean it will become useless 90 days afterwards?
What about for machines which aren't networked, or are on isolated networks which can't or aren't allowed to access the internet?
If they provide a corporate version which doesn't need to phone home, then pirates will simply pirate that instead, just like they did with the corporate versions of xp that didn't need activation.
They didn't say playing catchup in marketshare, they mean catchup in features, stability performance and appearance etc..
People don't buy windows because they've assessed multiple competing options (including osx and linux) and found windows to be technically superior or better value for money, they buy it because they don't realise anything else exists, because they're afraid to learn something new or because theyre forced to by having their data held to ransom by proprietary applications and their formats.
For bigger businesses (and governments), open source makes even more sense.
Just think economies of scale, a million installs of a single open source application is the same cost as one install.
Support costs are variable, but with proprietary software they will be *ON TOP* of the cost of the software itself... Proprietary software support can really only come from one place, and is unlikely to be provided indefinitely... Sooner or later support for the versions you use will end and you will be forced to upgrade (if an upgrade path exists - no guarantee it will or if the vendor will even still exist). By contrast open source code can be supported by multiple parties, and if its important enough you always have the option to support it yourself. How many companies are lumbered with proprietary legacy systems that noone supports at all anymore?
And frankly there are only a few OSS software vendors that can provide quality enterprise level support.
A few is better than just one, there is only one vendor that can provide a proper level of support for windows, only one vendor that can support oracle etc.
Large companies typically have some applications developed in house, open source can help a lot here since your developers can now customise general applications rather than having to put up with what a vendor sells you, or code from scratch.
Don't forget other costs, like license management, this can become very expensive especially when not all of your software is licensed for every machine so you need to keep track of how many licenses you have, where it's deployed, and deal with deploying different software to different locations, not to mention technical license enforcement measures which must be maintained, managed and supported. With open source license management becomes pretty much irrelevant, even GPL code only has restrictions once you start distributing it to third parties, for your own internal use you can install it as many times as you want and use it for anything you want.
Cost is only a small part of the overall picture, open source offers you a level of flexibility not available with proprietary software, you are freed from the burdens of license management, you have no lock-in, you have multiple support options, you have guaranteed future availability of at least the versions you are currently using, you always have
And frankly, the code quality for most OSS solutions (even Apache) doesn't compare with good commercial code.
Having seen commercial code, i have to say a lot of it is complete garbage and extremely poorly written. Just look at formerly commercial code which has been open sourced for some examples, netscape, staroffice etc. Commercial code is typically churned out by bored programmers who are forced to rush to meet deadlines which causes them to cut corners and make some pretty nasty kludges. I personally find that i write better code when i'm doing it for my own benefit and at my own pace, rather than being rushed to meet a deadline and waiting for the clock to strike home-time. The biggest difference is that you can see the open source code, whereas you can only imagine just how shoddy the closed source code was before it got compiled.
I don't live very far from work, but to get there by bus requires taking 2 buses (have to travel quite a distance in the wrong direction before the bus meets up with the other bus route that goes past my workplace)... So a total of 4 bus rides a day for a relatively short distance. I can walk it in the same time it takes to catch the first bus, then wait for the second one to arrive (altogether about 1.5 hours), or i could drive which takes about 5 minutes and costs less in fuel than i would be paying for the bus tickets (i take a direct route and don't stop until i get there, unlike the bus).
Video editing is an interesting comparison, but video editing software generally doesn't lock you in... The input formats are standard, the output formats are standard, so while you may buy proprietary applications now, in a few years time you might find that replacing them with open source replacements becomes a viable and cheaper option, so you switch.
For anything where there aren't standards like this, you have to consider the very serious dangers of proprietary lockin. While your company is small and only employs 20 editors today, some vendor may give you a very cheap deal and you buy in to their software... A couple of years down the line, you now have 200 editors instead of 20 and this same vendor no longer offers you a cheap deal and is trying to milk you for everything you've got, meanwhile their software has largely stagnated and their software is now inferior to everyone else's. However you can't switch easily, because all your data is in formats only supported by that program, and all the rest of your workflow depends on a sequence of proprietary tools that don't work with anything else so you would need to replace lots of other components too. So now you're stuck with a choice between inferior overpriced tools or spending a huge amount of money to free yourself from them.
While developing your own software, or modifying an existing open source package to suit your needs probably doesn't make much sense for a company with 20 users, when you have an organization the size of the australian government you will get economies of scale. Software only needs to be written once and then you can distribute it to all your users at no additional cost.
Most organizations of this size already employ developers to write in house bespoke applications anyway so you can split their time between projects, reuse developers for other projects once one goes into maintenance mode etc.
As to support, when buying off the shelf software support is usually extra (what you get included in the price is typically nothing, sometimes you might get 90 days installation help or something), especially if you want the top tier support packages, and even then there's no guarantee you'd get to speak to the actual developers or have changes made.. You end up having to adapt your business to the way the software works. By contrast, when you employ your own developers they will code the application to your exact requirements.
Consider your video editing again for a moment, of that $12k you pay to sony or apple, how much of that goes towards development of the product you're buying? Very little i imagine, most will go to profit, paying for staff not related to development, paying for marketing etc...
Consider on a grander scale, all the thousands of other companies that need video editing software, splitting all the money they spend between a handful of companies... If we assume that 10% (a very high estimate) of the purchase price goes directly towards development/testing... That 10% is split between say 3-4 companies who are all competing with each other and therefore develop equivalent features independently.. And some of that development effort will go to things which don't benefit the end user - like license enforcement schemes...
If all the companies that bought proprietary software instead of buying, spent 10% of the money to sponsor open source development you would have a massively superior product, with far less duplication of effort and far less wastage.
The cert also expired in 2008...
In order to connect the system to a network, it would need to be booted and thus the key must already have been entered.
That's the problem, storing the key in the TPM system makes you vulnerable to attack since the thief now has the key and all the time in the world to extract it.
I'm sure all the evil blackhat hackers will be delighted to support users of such old software...
A lot of these "overly cautious" security people are actually being extremely detrimental to security...
They prefer to stick to what they know, even when that is known to be far less secure than something else they could switch to.
Also consider the various accreditation schemes, getting your product accredited is a costly and time consuming process, so quite often you will find that only old versions are accredited and new versions are not (and might not ever be)...
People erroneously think the accredited version is more secure when in fact it isn't and may well have known flaws that people could exploit. You end up with extremely ridiculous situations where a lot of effort is expended to work around the known flaws, additional software is purchased to try and mitigate the known flaws (and by having additional software there you increase the attack surface anyway), and time/money is even expended cleaning up the mess from worm/virus infections with such infections being considered an acceptable cost of doing business.
The one positive from this whole mess, is that you should be able to use this as evidence of why being tied to a proprietary system is bad. Even Microsoft strongly recommend that you upgrade from IE6 these days.
Use this as a lesson going forward, any new apps you procure must comply with standards rather than being tied to a proprietary client. Make it a requirement on any companies you buy from, if your company is relatively large (and 100+ apps suggests it is) then vendors will fall over themselves to comply with your demands and if other companies do the same pretty soon standards compliant will become the norm and everyone will be much better off.
Standards compliant pages support lynx pretty well, lynx will simply ignore things it doesn't support and render the alt-tags instead of images etc. So long as you chose sensible names for the alt tags the textual page content is perfectly viewable. I use lynx quite often and find it great for getting to information quickly on well designed sites, without being distracted by any of the fancy graphics.
IE6 doesn't degrade gracefully, it doesn't ignore unsupported features like lynx does, it tries to render them and botches the job totally.
Perhaps if some of those people making the decisions find themselves unable to view the sites they want to, they will realise how stupid those decisions were and make efforts to change them. With any luck, this will teach them the dangers of using proprietary apps which are tied to a single platform and they will take the dangers seriously when making any future decisions.
So long as sites accommodate IE6 it will be considered acceptable to run it, which just continues the headache for everyone involved.