I don't yet see any evidence of even a mild decline.
More and more people are becoming skeptical about Microsoft's motives and the true value of their new products (e.g., Win XP really isn't all that great once the novelty wears off). Also, I've seen people at all-Microsoft shops become defensive about why they use the software they do (in the back of their minds they are beginning to realize where their eggs are). And, in general, the adoptions rates for new MS products (Win XP, XBOX, etc.) aren't stellar.
I really think Microsoft has peaked. However, it'll be several years, still, until they really decline, because the installed base is just so huge. And, in those years, genuine much-more-open competition, such as OpenOffice, Mac OS, Linux, etc., will mature further and make Microsoft look more obviously like the proprietary pit of technology that they are.
A better idea. Make this road tax, a tax on Gasoline.
Agreed. Gas tax is much much fairer than GPS or car property tax. I'm really sick of property taxes that make people want to drive old dirty cars, and putting a GPS reciever into cars is unconstitutional. I say unconstitutional, because the government could track the activities of protected groups of people and strategically interfere in favor of the government's agenda. GPS data would really be ripe for abuse of all types. It could be a new era of witch hunting (e.g., why was political-enemy XYZ's car at motel ABC on Tuesday morning...)
This is quite bogus. A single drive can easily exceed 40MB/s sequential transfer and your hard drive is the slowest storage device on most pcs.
I guess it's been a while since I looked at disk specs. At Seagate's website, the newer Barracudas and Cheetas all do about 30 to 60 MB/sec. It feels like it wasn't long ago when hard drives really were doing 20 to 30 MB/sec.
So, the conclusion is that the 80MB/sec and 160MB/sec SCSI busses would be a good fit for the newer drives.
The performance of the IDE drives are almost the same as their SCSI counterparts.
I'd like to see real non-single-user benchmarks. Multi-user UNIX environments and/or RAID are where SCSI shines. I trust SCSI's ability to aggregate the drives to truly utilize the bus' bandwidth better than I would trust IDE. IDE has always been designed from the single-user PC point of view.
I remember seeing a review of IDE RAID controllers a while back. The aggregate performance shown on the benchmarks was disappointing (gaining only a couple percent performance gain from a striped or mirrored array)--I'd think much better should be possible.
Ever since 40MB/sec SCSI came out...there really is no need for anything faster in a workstation...until hard drives become dramatically faster. Most workstations have no more than two hard drives (get it? 2 X 20MB/sec = 40MB/sec).
Only servers and workstations with massive external storage arrays benefit from multiple high-bandwidth SCSI controllers, such as FibreChannel, Ultra160 or Ultra320. Those bus speeds handle the aggregate bandwidths of the hard drives.
...faster, but less dependable IDE drives...
I still don't see 10,000 or 15,000RPM IDE drives, do you?
How fast do they propogate? Since the Sun is younger than the universe, could it be that distant stars have yet to be influenced by the Sun's magnetic, gravitational, or electric fields?
It's a serious disappointment that society has arrived (not recently) at a state where truly worthwhile information is rightfully withheld because we, as humans, can't treat things with respect.
This is only devil's advocate: the nostalgic feeling we have towards these shipwrecks is of arguable value, but the monetary value of the ships' materials can provide a direct injection of wealth into the economy.
Alas, choosing when to be sentimental is often hard (especially when spring cleaning comes around...)
When not using Passport, this is not true. I have the same login/password for several websites, but I always manually log in. If one of those websites was cracked, how would the others be compromised? Answer: they won't.
it's better to trust a single organisation than many.
We are talking about Microsoft, here. I'd rather have individual logins for each website. Each one is totally self-contained, even if the same username and password is used (see above).
No-one can evaluate a method until they've done a few non-trivial projects with it, and that takes years. If all the people who jumped on the RUP bandwagon then the XP bandwagon jump on this, the industry's track record for delivering on time and within budget will only get worse.
Thus the importance of not adopting RUP, XP, etc. for real projects. These methodologies can be informative, but it is better to create a simplified custom process for each project. It isn't very hard, and the development team can establish the tool chain, conventions, and documentation methods that suits them and the project's requirements best. Note that simplifying the process is critical, because no one can seriously keep track of developing real software while trying to learn some baroque process. Also, it is always critical to avoid proprietary documentation formats (e.g., basically anything by Microsoft), trendy IDEs, acronyms of the month, and other neat but immature development toys.
Personally, I think taking the time to actually implement the dogma of RUP, XP, etc. is a waste of time, when 1) no one really understands them, anyway and 2) they are like fashion: here today, gone tomorrow, possibly reborn in 20 years, but who knows.
I've got access to the Windows source at work, and it's not like you just get the source tree as files.
So, can you compile a full Windows distribution (or at least a kernel), if you wanted to? If not, then there is no guarantee that you actually have the Windows source!
Are you claiming that PCs running Windows are used for heart monitoring equipment? In a situation where a crash would be fatal??
I don't know about heart monitoring equipment, but I have seen non-trivial lab equipment run off PCs, where errors could certainly lead to misdiagnoses or loss of patient data. The other things I was thinking about were military command and control applications and air traffic control information displays. Windows really has no place in these environments (not even tangentially). Additionally, I'm not sure most UNIX systems would even be suitable.
No, it is just offensive (meaning smelly and repulsive).
TogetherSoft and Rational have the two most popular round-trip UML tools out there. Imagine MS getting their dirty hands into it. That would possibly alientate lots of people who choose those tools, because MS already spoiled Visio.
They will probably try to force rewriting it for.NET. Imagine the cultural collision. I wouldn't be surprised if hundreds of Borland employees suddenly update their resumes and post them to Monster, et. al.
I think the quote, "Equipment is built with the dumpster in mind, not the repair shop." is particularly telling, especially coming from a tech.
I often wonder when we will have to mine our landfills for raw materials. Everything will have to be recycled eventually, because mining the landfills will, one day, be cheaper than digging 1000 miles into the Earth for metal or oil.
If a heart monitoring computer crashes, people die. If a missile control system fails, people die. When the computer in your car fails, you might die. A corporate LAN goes down and your company looses millions. Computers are very important, not just comodities anymore.
Given how often PCs and Windows are used in these applications...I guess our lives aren't that important after all.
You do NOT want the Empire to go to a CAD/CAE-tool style, probably FLEXlm based, licensing model! PTC _may_ work with their customers. The monopolist has demonstrated that _they_ don't have to.
I guess the difference is that PTC doesn't generally assume its customers are moronic criminals from the start.
I haven't administered FLEXlm, but from an end user's point of view, it seems pretty fair about dealing out the floating licenses and it is easy to see exactly what has been purchased.
Well, I didn't mean to be obvious about it, but I have worked with Pro/E quite a bit and was really impressed by one of PTC's sales people when we had a licensing problem. Although I do have reservations about some of the proprietary aspects of Pro/E, I've never felt that PTC was out to screw me over. There's enough competition remaining in the CAD market to keep them mostly on target (they have streamlined some of their licensing, and a decent Pro/E package really isn't all that expensive anymore).
Personally, I think it would be much easier to move to a service model -- but the cost can't come at purchase time.
It seems that, in practice, service models for software are a mix between initial purchace price and on-going costs (though I agree totally about the activation-time cost).
It isn't uncommon in high-end software, such as CAD/CAM, for every add-on module to be shipped on the CD-ROM media. However, the customer pays only for the parts used, where a licensing server unlocks the purchased modules. Unlocking the modules is covered in the up-front cost of the software. The licensing server is kept local, so there really isn't any "big brother" spying on the customer (yes, Microsoft, it is possible to do things without a centralized licensing system).
Additionally, high-end software typcially comes with "maintenance" contracts, too, where the user can pay over time for support and upgrades. Without these contracts, the customer is typically on their own when problems occur (the customer should know this up front). These contracts help cover the on-going costs of developing the software.
One thing I've noticed about service-contract oriented software is that the companies are usually still working for the customer. Their salespeople will often jump through all sorts of hoops to get a sale, often willing to research what the customer needs to make them happy. This is most likely possible due to the high costs and risks involved when spending tens of thousands of dollars.
Thus back to Microsoft--are the costs and risks for a single sale of Windows sufficient for Microsoft to actually give a rat's ass about making the customer happy? Or, do they just try to milk the customer base for everything they're worth? Microsoft isn't in a position to have to work hard for the small-time customer to get a sale and a support contract. Where's the profit in that?
And all the old unix clinets (~800) running solaris are super slow (5+ minutes to log in!).
Me thinks they are connected to a badly architected network...or you are still using the SPARC lunchboxes from over a decade ago. How much of that login time is just the little old processor struggling against the transition from dtlogin (or equivalent) to CDE (or openwin) while working against a wee old NIS server over a 10Mb Ethernet?
Explain to me why a city, with offices here,there, and everywhere, manages to run a linux-based thin-client network, while a university with a huge IT budget runs one that's too slow to use!
The prevailing culture in the university's bureaucracy is living in the stone ages. I.e., it is probably made up of professors and admins who grew up, academically speaking, in the 70's and 80's and haven't learned a new thing since.
This is also why Microsoft is so successful - they constantly innovate, rather than sitting on their market dominance.
Sure they do. Microsoft is extremely good at seeing what other people are doing and, then, copying, buying, or squashing it to meet their agenda. In the case of C#, they first inhibited Java with their VM "embrace and extend" scheme, and then took Java, copied it, renamed it, and added a few things.
I guess you could call standing on other peoples shoulders innovation, but that is a relatively weak use of the word. Microsoft is more like the business partner that is your friend right up to the point of stabbing you in the back and taking credit for your work.
As for real innovation, I often wonder what "killer apps" were stifled over the years by Microsoft. Like weeds in a garden, Microsoft is always there...
OpenOffice is horrendously slow compared to Office 97/2K/XP on all my hardware from the trusty P-120 with 48MB up to P4-1.6 w/512MB.
In Word 2K, I can type faster than the characters display (especially in tables)--on a 400MHz CPU, and I'm not a fast typist by any measure.
Word is a messy ugly kludge. What kind of crappy software requires faster than a 400MHz CPU for just text entry? Word is crap.
One serious advantage of OpenOffice over MS Office is that OpenOffice will always be making progress, and the GPL ensures that progress will never be lost. MS Office seems to take two steps back for each step forward.
So, if OpenOffice is too slow for your needs, there is a good chance that won't be true forever.
Now, I'm no genius, but isn't something like this wrong?
Go read a Sociology textbook. There are decades of tales about cult leaders, population control, con artists, etc. Tales about power over other people.
Now go read a Management or Marketing textbook...same thing, but different jargon.
Except now, their tatics will be even more potent, as they manipulate our core humanity against us. Don't be suprised when the hopeless flocks grow even greater than before.
Slashdot Mirror
I don't yet see any evidence of even a mild decline.
More and more people are becoming skeptical about Microsoft's motives and the true value of their new products (e.g., Win XP really isn't all that great once the novelty wears off). Also, I've seen people at all-Microsoft shops become defensive about why they use the software they do (in the back of their minds they are beginning to realize where their eggs are). And, in general, the adoptions rates for new MS products (Win XP, XBOX, etc.) aren't stellar.
I really think Microsoft has peaked. However, it'll be several years, still, until they really decline, because the installed base is just so huge. And, in those years, genuine much-more-open competition, such as OpenOffice, Mac OS, Linux, etc., will mature further and make Microsoft look more obviously like the proprietary pit of technology that they are.
A better idea. Make this road tax, a tax on Gasoline.
Agreed. Gas tax is much much fairer than GPS or car property tax. I'm really sick of property taxes that make people want to drive old dirty cars, and putting a GPS reciever into cars is unconstitutional. I say unconstitutional, because the government could track the activities of protected groups of people and strategically interfere in favor of the government's agenda. GPS data would really be ripe for abuse of all types. It could be a new era of witch hunting (e.g., why was political-enemy XYZ's car at motel ABC on Tuesday morning...)
This is quite bogus. A single drive can easily exceed 40MB/s sequential transfer and your hard drive is the slowest storage device on most pcs.
I guess it's been a while since I looked at disk specs. At Seagate's website, the newer Barracudas and Cheetas all do about 30 to 60 MB/sec. It feels like it wasn't long ago when hard drives really were doing 20 to 30 MB/sec.
So, the conclusion is that the 80MB/sec and 160MB/sec SCSI busses would be a good fit for the newer drives.
The performance of the IDE drives are almost the same as their SCSI counterparts.
I'd like to see real non-single-user benchmarks. Multi-user UNIX environments and/or RAID are where SCSI shines. I trust SCSI's ability to aggregate the drives to truly utilize the bus' bandwidth better than I would trust IDE. IDE has always been designed from the single-user PC point of view.
I remember seeing a review of IDE RAID controllers a while back. The aggregate performance shown on the benchmarks was disappointing (gaining only a couple percent performance gain from a striped or mirrored array)--I'd think much better should be possible.
...slower SCSI bus.
...faster, but less dependable IDE drives...
Troll.
Ever since 40MB/sec SCSI came out...there really is no need for anything faster in a workstation...until hard drives become dramatically faster. Most workstations have no more than two hard drives (get it? 2 X 20MB/sec = 40MB/sec).
Only servers and workstations with massive external storage arrays benefit from multiple high-bandwidth SCSI controllers, such as FibreChannel, Ultra160 or Ultra320. Those bus speeds handle the aggregate bandwidths of the hard drives.
I still don't see 10,000 or 15,000RPM IDE drives, do you?
Magenetic fields never end.
How fast do they propogate? Since the Sun is younger than the universe, could it be that distant stars have yet to be influenced by the Sun's magnetic, gravitational, or electric fields?
It's a serious disappointment that society has arrived (not recently) at a state where truly worthwhile information is rightfully withheld because we, as humans, can't treat things with respect.
This is only devil's advocate: the nostalgic feeling we have towards these shipwrecks is of arguable value, but the monetary value of the ships' materials can provide a direct injection of wealth into the economy.
Alas, choosing when to be sentimental is often hard (especially when spring cleaning comes around...)
Crack one, and it opens up access to the rest.
When not using Passport, this is not true. I have the same login/password for several websites, but I always manually log in. If one of those websites was cracked, how would the others be compromised? Answer: they won't.
it's better to trust a single organisation than many.
We are talking about Microsoft, here. I'd rather have individual logins for each website. Each one is totally self-contained, even if the same username and password is used (see above).
No-one can evaluate a method until they've done a few non-trivial projects with it, and that takes years. If all the people who jumped on the RUP bandwagon then the XP bandwagon jump on this, the industry's track record for delivering on time and within budget will only get worse.
Thus the importance of not adopting RUP, XP, etc. for real projects. These methodologies can be informative, but it is better to create a simplified custom process for each project. It isn't very hard, and the development team can establish the tool chain, conventions, and documentation methods that suits them and the project's requirements best. Note that simplifying the process is critical, because no one can seriously keep track of developing real software while trying to learn some baroque process. Also, it is always critical to avoid proprietary documentation formats (e.g., basically anything by Microsoft), trendy IDEs, acronyms of the month, and other neat but immature development toys.
Personally, I think taking the time to actually implement the dogma of RUP, XP, etc. is a waste of time, when 1) no one really understands them, anyway and 2) they are like fashion: here today, gone tomorrow, possibly reborn in 20 years, but who knows.
I've got access to the Windows source at work, and it's not like you just get the source tree as files.
So, can you compile a full Windows distribution (or at least a kernel), if you wanted to? If not, then there is no guarantee that you actually have the Windows source!
Ahh, the smell of burning cow entrails!
You sound nostalgic...
Regardless, I'll concede that I'd rather live near a paper plant than an entrail furnace.
Are you claiming that PCs running Windows are used for heart monitoring equipment? In a situation where a crash would be fatal??
I don't know about heart monitoring equipment, but I have seen non-trivial lab equipment run off PCs, where errors could certainly lead to misdiagnoses or loss of patient data. The other things I was thinking about were military command and control applications and air traffic control information displays. Windows really has no place in these environments (not even tangentially). Additionally, I'm not sure most UNIX systems would even be suitable.
... they already do that. :)
As homeless scavengers or as genuine mining for new manufacturing?
This move is both offensive and defensive.
No, it is just offensive (meaning smelly and repulsive).
TogetherSoft and Rational have the two most popular round-trip UML tools out there. Imagine MS getting their dirty hands into it. That would possibly alientate lots of people who choose those tools, because MS already spoiled Visio.
What will they do with it... convert it to C#?
.NET. Imagine the cultural collision. I wouldn't be surprised if hundreds of Borland employees suddenly update their resumes and post them to Monster, et. al.
They will probably try to force rewriting it for
I think the quote, "Equipment is built with the dumpster in mind, not the repair shop." is particularly telling, especially coming from a tech.
I often wonder when we will have to mine our landfills for raw materials. Everything will have to be recycled eventually, because mining the landfills will, one day, be cheaper than digging 1000 miles into the Earth for metal or oil.
If a heart monitoring computer crashes, people die. If a missile control system fails, people die. When the computer in your car fails, you might die. A corporate LAN goes down and your company looses millions. Computers are very important, not just comodities anymore.
Given how often PCs and Windows are used in these applications...I guess our lives aren't that important after all.
Those people in the article should try living in a paper mill town sometime. Yech!
You do NOT want the Empire to go to a CAD/CAE-tool style, probably FLEXlm based, licensing model! PTC _may_ work with their customers. The monopolist has demonstrated that _they_ don't have to.
I guess the difference is that PTC doesn't generally assume its customers are moronic criminals from the start.
I haven't administered FLEXlm, but from an end user's point of view, it seems pretty fair about dealing out the floating licenses and it is easy to see exactly what has been purchased.
I see that you have worked with/for PTC. :)
Well, I didn't mean to be obvious about it, but I have worked with Pro/E quite a bit and was really impressed by one of PTC's sales people when we had a licensing problem. Although I do have reservations about some of the proprietary aspects of Pro/E, I've never felt that PTC was out to screw me over. There's enough competition remaining in the CAD market to keep them mostly on target (they have streamlined some of their licensing, and a decent Pro/E package really isn't all that expensive anymore).
Personally, I think it would be much easier to move to a service model -- but the cost can't come at purchase time.
It seems that, in practice, service models for software are a mix between initial purchace price and on-going costs (though I agree totally about the activation-time cost).
It isn't uncommon in high-end software, such as CAD/CAM, for every add-on module to be shipped on the CD-ROM media. However, the customer pays only for the parts used, where a licensing server unlocks the purchased modules. Unlocking the modules is covered in the up-front cost of the software. The licensing server is kept local, so there really isn't any "big brother" spying on the customer (yes, Microsoft, it is possible to do things without a centralized licensing system).
Additionally, high-end software typcially comes with "maintenance" contracts, too, where the user can pay over time for support and upgrades. Without these contracts, the customer is typically on their own when problems occur (the customer should know this up front). These contracts help cover the on-going costs of developing the software.
One thing I've noticed about service-contract oriented software is that the companies are usually still working for the customer. Their salespeople will often jump through all sorts of hoops to get a sale, often willing to research what the customer needs to make them happy. This is most likely possible due to the high costs and risks involved when spending tens of thousands of dollars.
Thus back to Microsoft--are the costs and risks for a single sale of Windows sufficient for Microsoft to actually give a rat's ass about making the customer happy? Or, do they just try to milk the customer base for everything they're worth? Microsoft isn't in a position to have to work hard for the small-time customer to get a sale and a support contract. Where's the profit in that?
And all the old unix clinets (~800) running solaris are super slow (5+ minutes to log in!).
Me thinks they are connected to a badly architected network...or you are still using the SPARC lunchboxes from over a decade ago. How much of that login time is just the little old processor struggling against the transition from dtlogin (or equivalent) to CDE (or openwin) while working against a wee old NIS server over a 10Mb Ethernet?
Explain to me why a city, with offices here,there, and everywhere, manages to run a linux-based thin-client network, while a university with a huge IT budget runs one that's too slow to use!
The prevailing culture in the university's bureaucracy is living in the stone ages. I.e., it is probably made up of professors and admins who grew up, academically speaking, in the 70's and 80's and haven't learned a new thing since.
This is also why Microsoft is so successful - they constantly innovate, rather than sitting on their market dominance.
Sure they do. Microsoft is extremely good at seeing what other people are doing and, then, copying, buying, or squashing it to meet their agenda. In the case of C#, they first inhibited Java with their VM "embrace and extend" scheme, and then took Java, copied it, renamed it, and added a few things.
I guess you could call standing on other peoples shoulders innovation, but that is a relatively weak use of the word. Microsoft is more like the business partner that is your friend right up to the point of stabbing you in the back and taking credit for your work.
As for real innovation, I often wonder what "killer apps" were stifled over the years by Microsoft. Like weeds in a garden, Microsoft is always there...
OpenOffice is horrendously slow compared to Office 97/2K/XP on all my hardware from the trusty P-120 with 48MB up to P4-1.6 w/512MB.
In Word 2K, I can type faster than the characters display (especially in tables)--on a 400MHz CPU, and I'm not a fast typist by any measure.
Word is a messy ugly kludge. What kind of crappy software requires faster than a 400MHz CPU for just text entry? Word is crap.
One serious advantage of OpenOffice over MS Office is that OpenOffice will always be making progress, and the GPL ensures that progress will never be lost. MS Office seems to take two steps back for each step forward.
So, if OpenOffice is too slow for your needs, there is a good chance that won't be true forever.
Now, I'm no genius, but isn't something like this wrong?
Go read a Sociology textbook. There are decades of tales about cult leaders, population control, con artists, etc. Tales about power over other people.
Now go read a Management or Marketing textbook...same thing, but different jargon.
Except now, their tatics will be even more potent, as they manipulate our core humanity against us. Don't be suprised when the hopeless flocks grow even greater than before.