MIPS is only one measure of computer load, anyway, and in a mainframe context processor power isn't necessarily the main performance determinant -- folks are often more interested in the raw amount of data being processed, or the raw number of transactions per second, not the speed at which each discrete transaction is being executed.
Uncontrolled change is not always desireable on a production system, especialy one which is involved in finance or has either a safety or a critical operational impact on a given company.
The mainframe shops I've worked in have never had issues with change itself, only with changes which are not well thought out and which don't follow due process. Those processes were put in place for a reason -- so person A's cure idea doesn't trash stuff needed by the rest of the company!!
I ran OS/2 2.0 and 2.1 on an 8MB box between 1992 and 1994, and it was just fine even with DOS support, WinOS2, and the WPS were all enabled. The key was to configure the swap file to a large size initially instead of letting it dymanically manage the filesize (which used both disk access time and CPU cycles).
You could get OS/2's resource requirements down to 4MB by dropping the WPS entirely and going with a text alternative shell like TSHELL, and OS/2 Warp 3 actually *REDUCED* the required footprint in 1994. That's why it was codenamed "Warp" in beta -- it was a performance update to OS/2 2.1, and in many cases it actually performed better on the same hardware in low RAM situations than previous versions.
The Pentium Pro was available in 1996, roughly the same time that OS/2 Warp 4 was released by IBM -- I timed my first PPro purchase to coincide with the release of Warp 4 so I could use that on my new box (a Micron Millenia Pro2 Plus, which is the very same box and OS I'm using to type in this message here on Slashdot today).
* was smaller and faster than Microsoft's equivalent product at the time (Windows NT 3.1)
and not THAT much larger than Windows 95,
* was far more flexible in terms of connectivity and application APIs supported than either
Windows 3.1 or NT,
* was arguably superior (on purely technical grounds) to both Windows 3.1 and Windows NT 3.1
and yet it could not obtain preloads with major vendors, it had a hard time obtaining device driver support from several major hardware manufacturers, and it eventually lost to a product (Windows 95) which was a relative joke (technically speaking).
Microsoft's Windows Vista
* is larger and slower than anything else on the market (including eComStation, Linux, Solaris,
and MacOSX),
* is less flexible than any of those in terms of connectivity and no better in terms of APIs
* is also arguably inferior (on purely technical grounds) to Linux, Solaris, MacOS X, and even
eCS in many respects,
and yet it will probably be preloaded on almost every system purchased over the next few years, it will be the first priority for device driver support from just about every hardware vendor under the sun, and it will probably win over the rest of the pack in spite of its resource requirements, performance problems, and reduced multimedia functionality.
I'd say those two products are about as dissimilar as two products could possibly be.
The first 32-bit OS/2 was OS/2 2.0 released in the spring of 1992 (just after Windows 3.1), and IBM's standard upgrade pricing at the time was US$49 for DOS users and US$99 for Windows 3.x users.
I'm not really a gadget collector, and I have four palms (two m105's, one IIIc, and an Abacus watch). Why would I purchase any more? Featurewise, I have what I want.
I hate phone PDAs. Too many functions in a phone detracts from its usefulness, IMO. I want a phone with a phone book, period. If I want more, I'll use my PDA.:-)
Any MP3 file, once created, is copyrighted by its creator at the moment of its creation, at least of the country the creator resides in is a signatory to the Berne Conventions. This includes Australia.
That means that the statement "copyrighted mp3s" is meaningless. All MP3 files are copyrighted. Some MP3 files may be freely copied, while others may not, but that is unrelated to their copyright status.
Prudential used to have several large data centers scattered throughout the US (my father used to work at one), but I don't know how many are left. They had a large IT department, though.
1. Don't confuse "Computer Science" with commercial programming. They are NOT the same thing.
Especially these days. When I received my degree, all IT-related degrees were CS degrees at a fair number of schools, and one simply chose a specialized track (systems, scientific, business) after finishing the CS core, but that's not the approach used at many schools today.
I liked the mix of practical and theoretical classes I took in the program I went through, though, since I think I've derived a lot of benefit from both types of classes over the years.
2. You will soon realize that coding is a far smaller portion of your job then you expect. The coding portion decreases as you move up the food chain.
Yes, unless you're a dedicated code monkey (something I've never personally encountered), you will be expected to do design work, create specifications, do support, talk to customers, help to coordinate tasks on complex projects, etc.
3. Do not ignore the business/finance side of your job. The business side keeps you employed.
Probably sound advice. In a large IT shop, you won't necessarily USE that type of knowledge in an overt manner, but it never hurts to be able to understand the business process and how it relates to your current position, and in future positions it could be tremendously helpful.
4. As you learn more, you will realize how little you actually know.
There's always someone else out there who's been doing it longer or better than you have. Or both.:-)
Pay attention to them -- such people are valuable teachers and resources, and I've learned a lot from people like that myself. Some programming tricks might be as old as YOU are.:-)
5. Your current position is nothing more than a software assembly line job. All of those "cool" technologies are being developed by more experienced engineers.
In all the shops I've worked in over the years, we NEVER had folks who did software in an assembly-line manner. Even folks like me right out of school were doing (mentored) design work for the live system. Other shops may be different, obviously, but even the folks I've seen who were writing software from a func spec that someone else created had a certain amount of latitude in terms of its actual implementation (even if screens and inputs/outputs were all predefined, the internal structure was often left up to the coder).
Don't be afraid of trying to create things on your own. I've seen folks right out of school make a huge difference by writing a little utility or by applying something they learned from another platform, and sometimes even something small can make a large difference. Experienced people are often very smart, but their tred-and-true experience (while often relevant) can also blind them to new approaches at times. I'm guilty of that as much as anyone at times.:-(
6. "Engineering" software and "programming" are more different than you realize.
Both should involve a formal process (although not all processes which people have in place are constructive or even useful). However, real "engineering" seems to rarely apply to software development. I still haven't decided if that's a good thing or a bad thing overall.
7. Coding is the easy part. You can teach a cat to bang out code. It takes an artist to design good software.
Absolutely. The top priority should be readable code that is easy for someone unfamiliar with the gory details to maintain. That means relevant comments in the source and (hopefully) a good set of programmer support documents written in parallel with the software. I've had the privilege to work in two shops where that was done quite well, but that's the exception, not the rule.
...and it's been a learning experience ever since. I've been "lucky" enough to actually be able to write software on a platform similar to what we used in college for much of my career, but the languages used and the scale/complexity of development I encountered in the "real world" (in my case writing custom software for a major airline) made me realize right away that college only gave me some basic tools -- the rest was up to me to learn on my own.
Much of what I've learned over the years has come in a work environment. I've been given some OJT flexibility and job function flexibility at work where I could work on pet projects to a certain extent, and that has allowed me to learn some non-mainstream languages and techniques on the main platform I work on. I've aso gone out of my way to ask questions of the folks I've run into who knew a lot in various areas, and that's been helpful.
However, I've also done a lot of learning at home (started playing with Linux in 1993, for example, mainly because we had no extended access to UNIX at the university I went to), and that has helped me tremendously as I've transitioned from mainframe programming to UNIX programming professionally. I've done both environments in parallel now for several years.
You can't keep up with it all. Once you get one trendy language or envirionment mastered, another three will materialize somewhere else. Just pick and choose things you think you might actually use and focus on them -- otherwise, you'll quickly find yourself overwhelmed.
The Palm is still powered up and retaining its memory state when you turn it "off", and turning it on only restores the power to the display. When you do a reset, that actually reboots the Palm OS, and it takes MUCH longer than one second.
The Nokia 5180i I received from Tracfone was an awesome phone. Featurewise it was fairly vanilla, but it was built like a tank and its multi-technology analog/digital reception made sure it would connect just about anywhere. I used it up in the Twin Cities and then down here in Atlanta for a little while, and it worked just fine.
I also bought the yearly cards. Well worth it if you don't use the phone except for emergency stuff and maybe a few other calls per month.
This, incidentally, cuts to the heart of the problem: Whichever way you cut it, the patent system, as it is, is a barrier to innovation in some industries.
Worse, it is a barrier to many *routine* activities in some industries.
I also have a custom DOS boot CD with a NeoBook menu that contains around 300MB of various DOS tools (Partition Magic, DriveImage, Ghost, Paragon's MountEverything, etc).
However, when someone has done quite as much as Bush and the Neocons have, supposedly in your names, mere apathetic inaction isn't enough. The American people have to either swiftly and pro-actively either make it clear that you disapprove of his actions, or be condemned to history as supporting him.
Except at election time, where the American people have spoken quite loudly about their opposition to the actions of the Republican party, how would you propose that the American people take action? Protests against the war have been happening for years, for example. Do those go ignored outside the US?
The OS/2 kernel (at least in my experience from v2.0 and onwards) was never an issue in terms of performance or stability, so if they had internal issues, those issues were not reflected in the resulting product shipped to customers.
OS/2's failure was not technical in nature. It was, if anything, a marketing failure.
It took many months and tremendous expense for European goods to make it to the American colonies back in the 17th-19th centuries, so recreating European products was often the only viable way to obtain such products locally. I'm not convinced that the same is true for China today.
From everything I've seen, their development processes and product lines both could use a serious ground-up redesign, and both Microsoft and their customer base would benefit from such a move.
According to the Berne convention, no such submission is required (a work is copyrighted the moment it is embodied in a tangible medium), but it's a good step to take if you want to defend your copyright in court. I don't think the complete source code has to be submitted, however.
Even it terms of UI, one user might find a certain type of desktop to be much more production than another, or might prefer to eschew the desktop altogether in favor of a command line (as I often do under OS/2, Linux, and Solaris).
I can see having a generalized default, perhaps, but the removal of choices will almost always mean a removal of functionality (and perhaps also productivity) at some level. IMO.
Slashdot Mirror
MIPS is only one measure of computer load, anyway, and in a mainframe context processor power isn't necessarily the main performance determinant -- folks are often more interested in the raw amount of data being processed, or the raw number of transactions per second, not the speed at which each discrete transaction is being executed.
Uncontrolled change is not always desireable on a production system, especialy one which is involved in finance or has either a safety or a critical operational impact on a given company.
The mainframe shops I've worked in have never had issues with change itself, only with changes which are not well thought out and which don't follow due process. Those processes were put in place for a reason -- so person A's cure idea doesn't trash stuff needed by the rest of the company!!
Hey, we're just following the behavior patterns established by our European ancestors. What else do you expect?
I ran OS/2 2.0 and 2.1 on an 8MB box between 1992 and 1994, and it was just fine even with DOS support, WinOS2, and the WPS were all enabled. The key was to configure the swap file to a large size initially instead of letting it dymanically manage the filesize (which used both disk access time and CPU cycles).
You could get OS/2's resource requirements down to 4MB by dropping the WPS entirely and going with a text alternative shell like TSHELL, and OS/2 Warp 3 actually *REDUCED* the required footprint in 1994. That's why it was codenamed "Warp" in beta -- it was a performance update to OS/2 2.1, and in many cases it actually performed better on the same hardware in low RAM situations than previous versions.
The Pentium Pro was available in 1996, roughly the same time that OS/2 Warp 4 was released by IBM -- I timed my first PPro purchase to coincide with the release of Warp 4 so I could use that on my new box (a Micron Millenia Pro2 Plus, which is the very same box and OS I'm using to type in this message here on Slashdot today).
IBM's 32-bit version of OS/2
* was smaller and faster than Microsoft's equivalent product at the time (Windows NT 3.1)
and not THAT much larger than Windows 95,
* was far more flexible in terms of connectivity and application APIs supported than either
Windows 3.1 or NT,
* was arguably superior (on purely technical grounds) to both Windows 3.1 and Windows NT 3.1
and yet it could not obtain preloads with major vendors, it had a hard time obtaining device driver support from several major hardware manufacturers, and it eventually lost to a product (Windows 95) which was a relative joke (technically speaking).
Microsoft's Windows Vista
* is larger and slower than anything else on the market (including eComStation, Linux, Solaris,
and MacOSX),
* is less flexible than any of those in terms of connectivity and no better in terms of APIs
* is also arguably inferior (on purely technical grounds) to Linux, Solaris, MacOS X, and even
eCS in many respects,
and yet it will probably be preloaded on almost every system purchased over the next few years, it will be the first priority for device driver support from just about every hardware vendor under the sun, and it will probably win over the rest of the pack in spite of its resource requirements, performance problems, and reduced multimedia functionality.
I'd say those two products are about as dissimilar as two products could possibly be.
The first 32-bit OS/2 was OS/2 2.0 released in the spring of 1992 (just after Windows 3.1), and IBM's standard upgrade pricing at the time was US$49 for DOS users and US$99 for Windows 3.x users.
Price was *NOT* an issue.
I'm not really a gadget collector, and I have four palms (two m105's, one IIIc, and an Abacus watch). Why would I purchase any more? Featurewise, I have what I want.
:-)
I hate phone PDAs. Too many functions in a phone detracts from its usefulness, IMO. I want a phone with a phone book, period. If I want more, I'll use my PDA.
Any MP3 file, once created, is copyrighted by its creator at the moment of its creation, at least of the country the creator resides in is a signatory to the Berne Conventions. This includes Australia.
That means that the statement "copyrighted mp3s" is meaningless. All MP3 files are copyrighted. Some MP3 files may be freely copied, while others may not, but that is unrelated to their copyright status.
Music died after the 70's. :-)
:-) :-)
But what does "music" have to do with rap?
They switched the design on the back (reverse) side of the US penny from the classic "wheat penny" to the newer Lincoln Memorial design in 1959.
So what do folks like me with dual-tuner analog TVs do? One external converter doesn't cut it.
Prudential used to have several large data centers scattered throughout the US (my father used to work at one), but I don't know how many are left. They had a large IT department, though.
Especially these days. When I received my degree, all IT-related degrees were CS degrees at a fair number of schools, and one simply chose a specialized track (systems, scientific, business) after finishing the CS core, but that's not the approach used at many schools today.
I liked the mix of practical and theoretical classes I took in the program I went through, though, since I think I've derived a lot of benefit from both types of classes over the years.
Yes, unless you're a dedicated code monkey (something I've never personally encountered), you will be expected to do design work, create specifications, do support, talk to customers, help to coordinate tasks on complex projects, etc.
Probably sound advice. In a large IT shop, you won't necessarily USE that type of knowledge in an overt manner, but it never hurts to be able to understand the business process and how it relates to your current position, and in future positions it could be tremendously helpful.
There's always someone else out there who's been doing it longer or better than you have. Or both. :-)
Pay attention to them -- such people are valuable teachers and resources, and I've learned a lot from people like that myself. Some programming tricks might be as old as YOU are. :-)
In all the shops I've worked in over the years, we NEVER had folks who did software in an assembly-line manner. Even folks like me right out of school were doing (mentored) design work for the live system. Other shops may be different, obviously, but even the folks I've seen who were writing software from a func spec that someone else created had a certain amount of latitude in terms of its actual implementation (even if screens and inputs/outputs were all predefined, the internal structure was often left up to the coder).
Don't be afraid of trying to create things on your own. I've seen folks right out of school make a huge difference by writing a little utility or by applying something they learned from another platform, and sometimes even something small can make a large difference. Experienced people are often very smart, but their tred-and-true experience (while often relevant) can also blind them to new approaches at times. I'm guilty of that as much as anyone at times. :-(
Both should involve a formal process (although not all processes which people have in place are constructive or even useful). However, real "engineering" seems to rarely apply to software development. I still haven't decided if that's a good thing or a bad thing overall.
Absolutely. The top priority should be readable code that is easy for someone unfamiliar with the gory details to maintain. That means relevant comments in the source and (hopefully) a good set of programmer support documents written in parallel with the software. I've had the privilege to work in two shops where that was done quite well, but that's the exception, not the rule.
My appro
...and it's been a learning experience ever since. I've been "lucky" enough to actually be able to write software on a platform similar to what we used in college for much of my career, but the languages used and the scale/complexity of development I encountered in the "real world" (in my case writing custom software for a major airline) made me realize right away that college only gave me some basic tools -- the rest was up to me to learn on my own.
Much of what I've learned over the years has come in a work environment. I've been given some OJT flexibility and job function flexibility at work where I could work on pet projects to a certain extent, and that has allowed me to learn some non-mainstream languages and techniques on the main platform I work on. I've aso gone out of my way to ask questions of the folks I've run into who knew a lot in various areas, and that's been helpful.
However, I've also done a lot of learning at home (started playing with Linux in 1993, for example, mainly because we had no extended access to UNIX at the university I went to), and that has helped me tremendously as I've transitioned from mainframe programming to UNIX programming professionally. I've done both environments in parallel now for several years.
You can't keep up with it all. Once you get one trendy language or envirionment mastered, another three will materialize somewhere else. Just pick and choose things you think you might actually use and focus on them -- otherwise, you'll quickly find yourself overwhelmed.
So says _Not The Bible_, anyway... :-)
The Palm is still powered up and retaining its memory state when you turn it "off", and turning it on only restores the power to the display. When you do a reset, that actually reboots the Palm OS, and it takes MUCH longer than one second.
The Nokia 5180i I received from Tracfone was an awesome phone. Featurewise it was fairly vanilla, but it was built like a tank and its multi-technology analog/digital reception made sure it would connect just about anywhere. I used it up in the Twin Cities and then down here in Atlanta for a little while, and it worked just fine.
I also bought the yearly cards. Well worth it if you don't use the phone except for emergency stuff and maybe a few other calls per month.
Worse, it is a barrier to many *routine* activities in some industries.
I also have a custom DOS boot CD with a NeoBook menu that contains around 300MB of various DOS tools (Partition Magic, DriveImage, Ghost, Paragon's MountEverything, etc).
The OS/2 kernel (at least in my experience from v2.0 and onwards) was never an issue in terms of performance or stability, so if they had internal issues, those issues were not reflected in the resulting product shipped to customers.
OS/2's failure was not technical in nature. It was, if anything, a marketing failure.
It took many months and tremendous expense for European goods to make it to the American colonies back in the 17th-19th centuries, so recreating European products was often the only viable way to obtain such products locally. I'm not convinced that the same is true for China today.
From everything I've seen, their development processes and product lines both could use a serious ground-up redesign, and both Microsoft and their customer base would benefit from such a move.
According to the Berne convention, no such submission is required (a work is copyrighted the moment it is embodied in a tangible medium), but it's a good step to take if you want to defend your copyright in court. I don't think the complete source code has to be submitted, however.
Even it terms of UI, one user might find a certain type of desktop to be much more production than another, or might prefer to eschew the desktop altogether in favor of a command line (as I often do under OS/2, Linux, and Solaris).
I can see having a generalized default, perhaps, but the removal of choices will almost always mean a removal of functionality (and perhaps also productivity) at some level. IMO.