I just recently got The Matrix Binary Watch, a great combination of different and inexpensive. This has been mentioned here before here and here.
The ordering process is now pretty normal. The watch itself is a bit chunky, and appears to be well made. The included band is leather, but not quite big enough for me (not unusual). The watch band is a less common size (19mm) so it took several different stores/malls, but eventually I found an extra long band.
It's relatively inexpensive (~$1500 w/GPS antenna and cable last time I checked) and seems to work well.
The best way to set this particular box up is to have ntp running on a regular server, and have that server get it's updates from this box while other clients get their update from the server. This unit doesn't seem to scale well to provide a time source for a lot of clients.
If you're really paranoid, you could simultaneously set up a system to check NIST signals for time and then let GPS and NIST duke it out in your ntp server.
Reads from Raid 5 spread over lots of disks tend to be pretty quick (quicker than Raid 0+1 / 10) since you've got more spindles active. There very well may be some kind of highly optimized raid 0+1 style controllers that utilize all spindles for reads -- that would negate the raid 5 advantage there, but I haven't seen such a controller yet.
Writes are what give you the 4x performance hit. For a read-intensive database (e.g.: datamart, datawarehouse) it might make sense to trade off some write performance for read performance.
As far as the question goes, it doesn't make a lot of sense. If you're worried about raw, sustainable transfer rates (e.g.: ftp downloads) then your tuning options are very different than if you are worried about how many transactions per second your DB is able to handle.
Instead of trying to get a tape system that will back up in 2 hours, why not get a 'SAN' system that supports "snapshots" and then back up the "snapshot" on a tape medium that could take many more hours?
Well, cost I suppose, but the 2 hour window will probably shrink over time, while the 200GB will turn into 250 and 300GB, so you always end up chasing the most expensive tape technologies to keep up.
As far as the DLT vs. LTO question, I don't know much about LTO in production use, but DLT is so widely used and trusted it would seem to have more of an incumbent status as the technology of choice. I ended up using AIT 50/100GB tapes -- cheaper media, faster than DLT, smaller tapes (easier to carry lots off-site), larger 'jukebox' designs available, and higher capacity (at the time I made that decision). It's worked well and I've never regretted it.
Although I agree with a previous poster that there is probably some gross margin shenanigans going on with the pricing of this individual item, this does raise an interesting issue.
As far as I can tell a technical e-book's only appeal is that you could (in theory) dump a bunch on your laptop hard drive, on cdr's, or into your palmpilot, and carry them around as reference material. Much better than lugging around 80lbs of books to client sites. Of course with all of the digital rights junk floating around this may or may not be practical, and your ability to read the book 5 years from now under such a system is highly questionable.
Without such DRM type controls, the publishers fears that the books will get freely distributed over the Internet are well justified -- but it's happening anyway as people scan in books and OCR them, or simply crack the DRM schemes.
Reading a book on a laptop or palmpilot screen is a fundamentally different experience than curling up with a paper book. Although my preference is paper, it is nice to have my Palmpilot loaded up with a good Heinlein novel during boring meetings.
E-books seem to be stopgap measure driven by short-term profit potential. IMO the publishing industry faces fundamental changes over the next few years as their primary value-add transitions to making nice printed versions of works already freely (although perhaps illegally) available.
There are four categories of books that seem to be under-represented:
1: XYZ to ABC guides, ala Perl to Python Migration or Java for Cobol programmers. Would be nice to see more in this vein -- e.g.: Microsoft ASP to PHP/Zope/mod_perl/whatever, Perl to Ruby, mySQL for MS SQL server admins, etc.
2: 'Cookbook' style programming guides. The Perl Cookbook should be the prototype -- just get someone to translate it all to Python, Scheme, whatever. Same idea for website development (or at least html).
3: Computer Science books for non-computer scientists. The Perl Journal used to have lots of nifty articles that talked about CS subjects, but applied them to the real world and made them accessible to a relatively wide audience. Dr. Dobbs has some similar articles, although they tend to get a bit more CS'ish. Algorithms in Perl is one take on this but still too textbookish, something with more narrative perhaps describing a specific project rather than a laundry list of all possibilities.
4: Software "craftsmanship" books -- ala The Pragmatic Programmer, or Programming Pearls. Cover more subjects than making string searches _really_ fast -- how about books like this that deal with "best practices" for setting up a datacenter, migrating to/from and/or co-exisiting with Unix and Mainframes, setting up scalable web sites, system administration. These are different from cookbooks in that they don't prescribe, they simply describe common circumstances and provide heuristics for dealing with them.
NAS, or "Network attached Storage" is often better for maintaing large collections of data to be accessed by multiple computers. You can simulate NAS by exporting some filesystems via NFS (Unix) or CIFS (Windows). Network Appliance "Filers" are said to be very good. On the lower end are the Maxtor MaxAttach and Quantum Snap! devices.
The big advantage to NAS is that dozens of web servers can mount the NAS volume and all serve up the same content. Developers, Administrators, etc. can also mount the NAS volume and do updates etc. Compared to a SAN and buying a fibre channel card, cabling, switch ports, etc. for anything but non-essential components gets very expensive very quickly. Although a previous poster indicated that multiple computers can mount the same SAN volume, It's much more difficult than with NAS since you're essentially operating at the same level as a SCSI bus, wheras with NAS you're operating via TCP/IP.
A Fibre Channel SAN is good for multiple computers running I/O intensive processes, e.g. a SQL database. It's also good as a foundation for clusters since (usually) LUNs can be re-mapped w/out a reboot. SANs really shine for fully redundant storage as well -- multiple loops, switches, controllers, etc.
Many products in both categories suffer in support for backup -- the typical low-end devices require you to mount the data on a server then use a server-attached tape device. Some products feature built-in tape drives or offer ways to back up the entire storage unit to a fibre channel attached tape drive, however this option tends to get very expensive very quickly.
One major bonus in the backup arena is the "snapshot" feature many products have (SAN or NAS). This lets you freeze 'the drive' so that no updates happen to the drive for your backup, but the system still stays up and allows updates. See vendor propaganda for more details.
Although Gigabit Ethernet is 1000Mbps in theory, in practice you don't usually get that kind of throughput -- so Firewire might not be all _that_ disadvantaged.
The basic reality is that you can _get_ cameras, hard drives, etc. with firewire ports while gigabit ports aren't readily available (if at all) on these sorts of 'consumer' devices. Will Gigabit supplant firewire? Maybe -- but why deprive yourself of the advantages of firewire for the next few years until it does (or doesn't) happen?
I've got this book, and actually read quite a bit of it. My only complaint is that it's too big to effectively balance on your desk and work with next to a computer. Maybe I just need a bigger desk.
On the other hand, it sure would be nice to have this in a 3 or 4-volume boxed set. I'd pay a few bucks more for that format -- smaller (300-400) page books are a lot easier to handle.
"Mainframes have the largest (fastest) CPUs except for the class of supercomputers."
I have to disagree. Mainframe CPUs follow the same laws of physics as other computers. Yes they are build with good I/O in mind, and yes they have good memory bandwidth.
I disassembled a 5-MIPS machine last summer and discovered the CPU unit had a 60MHz clock chip (and a 55 and a 72 and a few others in that range) This would put a 60MIPS machine (the size a medium-size grocery store chain might use) at about 720MHz. Using a Pentium whatever at one end of the scale and your choice of High-end CPU at the other, and factoring in the good I/O, etc. it's hard to say that machine (about $180k + 20k/month in licensing fees) is going to be substantially faster than say a quad-900MHz 8GB RAM box from Sun for essentially the same price, or big box from Compaq (alpha), HP, and perhaps even a high end Intel-based machine.
What it _does_ do is run legacy business applications very well, and it runs reliably. So go ahead and use it -- just don't claim it's almost as fast as a supercomputer.
Going to the opposite extreme in Mainframe land -- multiple zSeries tied together in a Sysplex cluster running at thousands of MIPS -- sure that's a fast machine, but you just dropped $10million or more on the hardware and license fees would be well over a $million a year. You're then forced to compare once again to a (for example) 72-way Sparc system from SUN or any of the other high-end Unix systems out there, usually at about 1/5 to 1/10th the cost, and usually with minimal on-going licensing fees.
Given that (for better or worse) RedHat is one of the cornerstone linux distros out there, forming the basis of Mandrake and many others, and Given the 'pay per view' mentality of cable combined with the 'enslave the idiots' mentality of AOL, do we have a potentially explosive mix coming together?
Just suppose that this transaction went through -- given the millions if not billions that AOLTW could piss away on legal fees, would this pose a serious challenge to the GPL? I don't doubt that the FSF, EFF, RMS, and a whole bunch of people would get ticked off about it, file suit, and generally raise a lot of hell. But when push comes to shove and RedHat becomes AOL 8.5, closed source, $xyz per copy (or per view) -- what are we going to do about it? Heck, they could just stall long enough to buy politicians, not unlike how MS has been behaving lately.
On the other hand, perhaps it would just cause RedHat to simply stagnate, too busy integrating corporate systems and dealing with lost employees to do much of anything else. Certainly the Netscape buyout hasn't exactly set the world on fire yet.
And lest I be branded an eternal pessimist, maybe they will instead piss away the budgeted fund earmarked for legal fees related to destroying linux on Free software development and contribution back to the community. To their credit the Mozilla project is still going.
But why should I have to create custom install CDs? I already dedicate several hours a day to keep the blasted things running, now I have to spend the rest of my time creating new install CDs?
Nevertheless, I actually do do this to some extent -- I have a couple of "NT 4" cds where I've whacked some of the useless stuff (e.g.: Alpha support and other languages), and used the space for useful things (e.g.: Option pack, Service Packs, IE5 install etc.). But this doesn't address the core problem -- Microsoft should stop distributing broken versions of it's operating system. Maybe have a quarterly release with a subscription (now there's a way for MS to suck the lifeblood out of corporate America).
Ghost is OK, but only for identical machines, and then there's the whole SID problem. This can be made to work, but it tends to be extremely time consuming.
I recently had to rebuild a web server after a machine crashed, and getting NT4, IIS Option pack, etc. up and running with all patches was a _very_ long task.
It's not enough that Microsoft patches their products -- they are still shipping CDs of NT4 and win2k with the original 'release' of the product, so installing it means the original install plus a dozen or more service packs, hotfixes, etc. This makes it very tempting for internal corporate PC usage to just skip most of the patches to save time, and makes the process of securing Microsoft software that much more difficult.
They should just release new 'point' versions of the OS with every service pack, and stop selling the out of date CDs! Maybe this would cut down on the useless churn of moving from NT4 to 2K to XP to whatever -- and that would have to be good.
Has he ever had to _support_ a big MS server installation?
Sure, the "Mainframe is dead", except for the tens of thousands of businesses that rely on fast, efficient, reliable, and comparatively cheap processing provided by mainframes and the relatively inexpensive cobol programmers that man them.
Sure, Unix is a 'niche market', except for the millions of users who use it every day for tasks ranging from mainframe replacement to destop applications, not to mention the countless academic, engineering, and other uses Unix is put to. For example, running most of the infrastructure on the Internet.
Yeah, Java runs slow. Boo hoo. So does a windows machine, even when you ignore downtime due to reboots and system crashes.
When this bozo is ready to bet his business on a technology, and is ready to assume full responsibility for the consequences of his decision, and is able to execute on his strategy, then and only then is he qualified to write a credible version of the article referenced.
Perhaps he would have an easier time selling himself as a software architect rather than a software engineer. i.e.: He's good at relating business requirements to system requirements and making sure the pieces fit together in a cohesive way.
This way his actual programming experience is _much_ less important than his people skills, ability to withstand long meetings in a single bound, and general political savvy in a large corporation.
Sysadmins probably shouldn't write the equivalent of shell scripts in assembler, but having a firm grounding in it helps to:
1: Analyze core dumps, blue screens, etc. (regardless of platform)
2: Understand tricky things like big and little endian, binary, octal, and hex notation, etc.
4: Have a firm fundamental understanding of what object code is, what linkers and loaders do, why it's important, etc. In a world of DLL hell, at least understand what's going on "behind the scenes"
5: Have a theoretical framework to hang concepts like processes, threads, priority, scheduling, interrupts, etc. -- it might not be _the_ correct model for a particular platform, but it will give a much better starting frame of reference.
On the other hand, if by "sysadmin" you mean "village idiot", and the only requirement is to know how to walk people through rebooting their PC and knowing how to call 1-800-dial-a-vendor and sit on hold, then NO, you don't need to learn assembly. In fact, you could probably get away with not knowing math, history, economics, or much of anythig else. Better yet, how about totally untrained sysadmins? I mean, c'mon, NT/2k/XP is virtually self-administering and it's all "point and click", you could entrust all of your corporate systems to a high school dropout for a fraction of the cost of even an incompetent sysadmin!
Lets face it: Linux is not a clear choice for joe idiot consumer looking to buy a PC to put under the xmas tree.
At the same time, it's come a _long_ way in just a few years. I'd bet my job (well, actually I just about have) that Linux is a better business desktop than Windows. For a business, Linux makes a lot of sense. It gets you off the Microsoft-upgrade-churn cycle, most everything you'd want is freely available, and the simplicity of administration and the excellent security make it a great choice.
Yes, KDE/Koffice, Gnome/'Gnome office', and StarOffice are not MS-Office. So what? As more businesses adopt Linux as their desktop, manufacturers will take note and start offering it, ISVs will take note and start selling more software, and consumers will take note and start buying linux for home since they want to be compatible with what's at the office. Same sort of cycle that made the IBM PC more popular than the Mac back in the late 80's/early 90's.
This might not be the year of linux for the consumer, but it's getting close for linux on corporate desktops.
"Enterprise" has two components, first is the relatively straightforward "Core Application" of a company, which might be some sort of ERP system like PeopleSoft, or some other commercial produce. For many companies, this will be an amalgamation of custom programs written over many years. These applications are typically based on some form of transaction processing system (e.g. CICS on the Mainframe, Tuxedo on Unix, or even database-driven transactions ala Oracle, DB2, etc.)
The second, and more critical part of 'Enterprise' is the nature of the computing service. Generally any outage is measured in dollars per minute or hour. It's not unusual for a large company to face severe monetary losses for even slight outages. Think millions of dollars an hour (or even per minute). This measure tends to be a little slippery, but with some analysis a pretty solid figure can usually be determined.
For some enterprises, Linux might make complete sense (e.g. Google). For others, the potential of saving a few thousand or hundred thousand in licensing costs pales in comparison to the probable re-training, new hardware, and "potential" instability of moving to Linux. If you've got something that works, why fix it?
Given the above, even if all of the big 'Enterprise' vendors port their software to Linux, you're not done. Linux clustering in a business context such as Solaris, AIX, and (in the good old days) VMS provide would be one stumbling block. The lack of high-end hardware is another -- and yes I know that Linux runs on anything from a PC to a SPARC server to a S/390 mainframe. In reality, you're unlikely to drop $2million on a big Sun box then load Linux -- you'll want to take advantage of Solaris's dynamic partitioning and other proprietary hooks.
Loading Linux on diverse old hardware makes business sense -- turn that old Sun box into something useful. It doesn't make nearly as much business sense when buying a new non-intel server, since the license fee of the OS (if any) is negligable compared to the overall value of the system in the 'Enterprise'.
Over time this is likely to change, since Linux represents a constantly improving and freely available system, vendors will start adopting it as 'their' OS. IBM is an early starter here, but the process will take time. And like a battle of attrition Linux has the advantage over time, since it is constantly improving (for free), while commercial vendors have to dump $millions into R&D to bring out each new version of their OS.
I seem to recall all of these (or their predecessors) offering things for sale. I remember, in particular, that one of these late-80's early 90's online services featured an on-line shopping mall. Unfortunately I don't recall which since I never used it. Thought it was too gimmicky.
What about CompuServe? They had a lot of 'features' that would instantly charge your account for access, e.g.: I took an IQ test that cost not just connect time but an actual fee. After choosing the option the test was presented. (unfortunately I didn't score high enough to catch on to the e-commerce boom of the mid-late 90's).
All of these services also allowed you to dial in, enter a credit card, and have (more or less) instant access, which smells a lot like e-commerce. For a while compuserve even sent out a kit with a manual and some other stuff. This back in 1986-1987.
I haven't seen a trackball on a laptop in years, although the side-car style that old TI and Toshiba laptops used was pretty nice.
If you can find one of the old sidecar style ones, or even a serial or USB trackball that is suitably sized to bolt onto the side of a notebook you could probably rig something up:
Most laptops have removable a floppy/cd drive bay, or PCMCIA slot or 2nd battery slot, usually one of these would be located on the "correct" side of the computer.
Whichever slot is free, find a plastic blank or construct one out of plastic/wood/whatever, then mount the trackball to that.
Not as elegant as having one purpose-built, but with a little craftsmanship and a bit of luck it would probably turn out looking just dandy.
Cable Modem is alive and well in upstate New York. DSL however has always been much more difficult to get. Not surprising, when you look at the equation:
Old copper + recalcitrant phone company / severe technical limitations + high cost == bad business.
Lets face it, just getting DSL to work is virtually a miracle, and getting it to work on every copper line going to every home is simply unrealistic.
DSL seems to be a good onesy-twosey kind of thing to implement, but I don't envy the people trying to make it work at thousands of subscriber sites.
1: Free software (& 'open source') is not about killing Microsoft. Microsoft is bent on screwing the user over for as much cash as they can take. Free software is about freeing the user from that domination, and giving the user control of their own software destiny.
2: Leader? We don't need to steenking leader! This is a grassroots movement in what is probably the most egalitarian forum ever devised. If you can write good code, people respect that. What's a leader going to do? Enforce project timelines? Talk to the press? We've already got lots of people doing that e.g.: RMS & ESR.
3: Left to themselves, the people writing the code will go through their own darwinian selection process. Some projects will gain at the expense of others, some will merge, some will die, some will co-exist. This process takes time, (sometimes _too_long_), but so what? If you want something to move faster, contribute to it! With so many good coders contributing to the community, the richness and quality of Free Software accretes over time. At some point, the sheer mass of high-quality free software will overwhelm the ability of proprietary software to compete. This is already starting to happen with Linux vs. proprietary Unix, and will likely happen in other areas in the next few years.
Slashdot Mirror
I just recently got The Matrix Binary Watch, a great combination of different and inexpensive. This has been mentioned here before here and here.
The ordering process is now pretty normal. The watch itself is a bit chunky, and appears to be well made. The included band is leather, but not quite big enough for me (not unusual). The watch band is a less common size (19mm) so it took several different stores/malls, but eventually I found an extra long band.
I use the Lantronix CoBox E1.
It's relatively inexpensive (~$1500 w/GPS antenna and cable last time I checked) and seems to work well.
The best way to set this particular box up is to have ntp running on a regular server, and have that server get it's updates from this box while other clients get their update from the server. This unit doesn't seem to scale well to provide a time source for a lot of clients.
If you're really paranoid, you could simultaneously set up a system to check NIST signals for time and then let GPS and NIST duke it out in your ntp server.
Reads from Raid 5 spread over lots of disks tend to be pretty quick (quicker than Raid 0+1 / 10) since you've got more spindles active. There very well may be some kind of highly optimized raid 0+1 style controllers that utilize all spindles for reads -- that would negate the raid 5 advantage there, but I haven't seen such a controller yet.
Writes are what give you the 4x performance hit. For a read-intensive database (e.g.: datamart, datawarehouse) it might make sense to trade off some write performance for read performance.
As far as the question goes, it doesn't make a lot of sense. If you're worried about raw, sustainable transfer rates (e.g.: ftp downloads) then your tuning options are very different than if you are worried about how many transactions per second your DB is able to handle.
Instead of trying to get a tape system that will back up in 2 hours, why not get a 'SAN' system that supports "snapshots" and then back up the "snapshot" on a tape medium that could take many more hours?
Well, cost I suppose, but the 2 hour window will probably shrink over time, while the 200GB will turn into 250 and 300GB, so you always end up chasing the most expensive tape technologies to keep up.
As far as the DLT vs. LTO question, I don't know much about LTO in production use, but DLT is so widely used and trusted it would seem to have more of an incumbent status as the technology of choice. I ended up using AIT 50/100GB tapes -- cheaper media, faster than DLT, smaller tapes (easier to carry lots off-site), larger 'jukebox' designs available, and higher capacity (at the time I made that decision). It's worked well and I've never regretted it.
Although I agree with a previous poster that there is probably some gross margin shenanigans going on with the pricing of this individual item, this does raise an interesting issue.
As far as I can tell a technical e-book's only appeal is that you could (in theory) dump a bunch on your laptop hard drive, on cdr's, or into your palmpilot, and carry them around as reference material. Much better than lugging around 80lbs of books to client sites. Of course with all of the digital rights junk floating around this may or may not be practical, and your ability to read the book 5 years from now under such a system is highly questionable.
Without such DRM type controls, the publishers fears that the books will get freely distributed over the Internet are well justified -- but it's happening anyway as people scan in books and OCR them, or simply crack the DRM schemes.
Reading a book on a laptop or palmpilot screen is a fundamentally different experience than curling up with a paper book. Although my preference is paper, it is nice to have my Palmpilot loaded up with a good Heinlein novel during boring meetings.
E-books seem to be stopgap measure driven by short-term profit potential. IMO the publishing industry faces fundamental changes over the next few years as their primary value-add transitions to making nice printed versions of works already freely (although perhaps illegally) available.
There are four categories of books that seem to be under-represented:
1: XYZ to ABC guides, ala Perl to Python Migration or Java for Cobol programmers. Would be nice to see more in this vein -- e.g.: Microsoft ASP to PHP/Zope/mod_perl/whatever, Perl to Ruby, mySQL for MS SQL server admins, etc.
2: 'Cookbook' style programming guides. The Perl Cookbook should be the prototype -- just get someone to translate it all to Python, Scheme, whatever. Same idea for website development (or at least html).
3: Computer Science books for non-computer scientists. The Perl Journal used to have lots of nifty articles that talked about CS subjects, but applied them to the real world and made them accessible to a relatively wide audience. Dr. Dobbs has some similar articles, although they tend to get a bit more CS'ish. Algorithms in Perl is one take on this but still too textbookish, something with more narrative perhaps describing a specific project rather than a laundry list of all possibilities.
4: Software "craftsmanship" books -- ala The Pragmatic Programmer, or Programming Pearls. Cover more subjects than making string searches _really_ fast -- how about books like this that deal with "best practices" for setting up a datacenter, migrating to/from and/or co-exisiting with Unix and Mainframes, setting up scalable web sites, system administration. These are different from cookbooks in that they don't prescribe, they simply describe common circumstances and provide heuristics for dealing with them.
NAS, or "Network attached Storage" is often better for maintaing large collections of data to be accessed by multiple computers. You can simulate NAS by exporting some filesystems via NFS (Unix) or CIFS (Windows). Network Appliance "Filers" are said to be very good. On the lower end are the Maxtor MaxAttach and Quantum Snap! devices.
The big advantage to NAS is that dozens of web servers can mount the NAS volume and all serve up the same content. Developers, Administrators, etc. can also mount the NAS volume and do updates etc. Compared to a SAN and buying a fibre channel card, cabling, switch ports, etc. for anything but non-essential components gets very expensive very quickly. Although a previous poster indicated that multiple computers can mount the same SAN volume, It's much more difficult than with NAS since you're essentially operating at the same level as a SCSI bus, wheras with NAS you're operating via TCP/IP.
A Fibre Channel SAN is good for multiple computers running I/O intensive processes, e.g. a SQL database. It's also good as a foundation for clusters since (usually) LUNs can be re-mapped w/out a reboot. SANs really shine for fully redundant storage as well -- multiple loops, switches, controllers, etc.
Many products in both categories suffer in support for backup -- the typical low-end devices require you to mount the data on a server then use a server-attached tape device. Some products feature built-in tape drives or offer ways to back up the entire storage unit to a fibre channel attached tape drive, however this option tends to get very expensive very quickly.
One major bonus in the backup arena is the "snapshot" feature many products have (SAN or NAS). This lets you freeze 'the drive' so that no updates happen to the drive for your backup, but the system still stays up and allows updates. See vendor propaganda for more details.
Although Gigabit Ethernet is 1000Mbps in theory, in practice you don't usually get that kind of throughput -- so Firewire might not be all _that_ disadvantaged.
The basic reality is that you can _get_ cameras, hard drives, etc. with firewire ports while gigabit ports aren't readily available (if at all) on these sorts of 'consumer' devices. Will Gigabit supplant firewire? Maybe -- but why deprive yourself of the advantages of firewire for the next few years until it does (or doesn't) happen?
I've got this book, and actually read quite a bit of it. My only complaint is that it's too big to effectively balance on your desk and work with next to a computer. Maybe I just need a bigger desk.
On the other hand, it sure would be nice to have this in a 3 or 4-volume boxed set. I'd pay a few bucks more for that format -- smaller (300-400) page books are a lot easier to handle.
"Mainframes have the largest (fastest) CPUs except for the class of supercomputers."
I have to disagree. Mainframe CPUs follow the same laws of physics as other computers. Yes they are build with good I/O in mind, and yes they have good memory bandwidth.
I disassembled a 5-MIPS machine last summer and discovered the CPU unit had a 60MHz clock chip (and a 55 and a 72 and a few others in that range) This would put a 60MIPS machine (the size a medium-size grocery store chain might use) at about 720MHz. Using a Pentium whatever at one end of the scale and your choice of High-end CPU at the other, and factoring in the good I/O, etc. it's hard to say that machine (about $180k + 20k/month in licensing fees) is going to be substantially faster than say a quad-900MHz 8GB RAM box from Sun for essentially the same price, or big box from Compaq (alpha), HP, and perhaps even a high end Intel-based machine.
What it _does_ do is run legacy business applications very well, and it runs reliably. So go ahead and use it -- just don't claim it's almost as fast as a supercomputer.
Going to the opposite extreme in Mainframe land -- multiple zSeries tied together in a Sysplex cluster running at thousands of MIPS -- sure that's a fast machine, but you just dropped $10million or more on the hardware and license fees would be well over a $million a year. You're then forced to compare once again to a (for example) 72-way Sparc system from SUN or any of the other high-end Unix systems out there, usually at about 1/5 to 1/10th the cost, and usually with minimal on-going licensing fees.
Given that (for better or worse) RedHat is one of the cornerstone linux distros out there, forming the basis of Mandrake and many others, and Given the 'pay per view' mentality of cable combined with the 'enslave the idiots' mentality of AOL, do we have a potentially explosive mix coming together?
Just suppose that this transaction went through -- given the millions if not billions that AOLTW could piss away on legal fees, would this pose a serious challenge to the GPL? I don't doubt that the FSF, EFF, RMS, and a whole bunch of people would get ticked off about it, file suit, and generally raise a lot of hell. But when push comes to shove and RedHat becomes AOL 8.5, closed source, $xyz per copy (or per view) -- what are we going to do about it? Heck, they could just stall long enough to buy politicians, not unlike how MS has been behaving lately.
On the other hand, perhaps it would just cause RedHat to simply stagnate, too busy integrating corporate systems and dealing with lost employees to do much of anything else. Certainly the Netscape buyout hasn't exactly set the world on fire yet.
And lest I be branded an eternal pessimist, maybe they will instead piss away the budgeted fund earmarked for legal fees related to destroying linux on Free software development and contribution back to the community. To their credit the Mozilla project is still going.
But why should I have to create custom install CDs? I already dedicate several hours a day to keep the blasted things running, now I have to spend the rest of my time creating new install CDs?
Nevertheless, I actually do do this to some extent -- I have a couple of "NT 4" cds where I've whacked some of the useless stuff (e.g.: Alpha support and other languages), and used the space for useful things (e.g.: Option pack, Service Packs, IE5 install etc.). But this doesn't address the core problem -- Microsoft should stop distributing broken versions of it's operating system. Maybe have a quarterly release with a subscription (now there's a way for MS to suck the lifeblood out of corporate America).
Ghost is OK, but only for identical machines, and then there's the whole SID problem. This can be made to work, but it tends to be extremely time consuming.
I recently had to rebuild a web server after a machine crashed, and getting NT4, IIS Option pack, etc. up and running with all patches was a _very_ long task.
It's not enough that Microsoft patches their products -- they are still shipping CDs of NT4 and win2k with the original 'release' of the product, so installing it means the original install plus a dozen or more service packs, hotfixes, etc. This makes it very tempting for internal corporate PC usage to just skip most of the patches to save time, and makes the process of securing Microsoft software that much more difficult.
They should just release new 'point' versions of the OS with every service pack, and stop selling the out of date CDs! Maybe this would cut down on the useless churn of moving from NT4 to 2K to XP to whatever -- and that would have to be good.
Where in the h*ll is this guy coming from?
Has he ever had to _support_ a big MS server installation?
Sure, the "Mainframe is dead", except for the tens of thousands of businesses that rely on fast, efficient, reliable, and comparatively cheap processing provided by mainframes and the relatively inexpensive cobol programmers that man them.
Sure, Unix is a 'niche market', except for the millions of users who use it every day for tasks ranging from mainframe replacement to destop applications, not to mention the countless academic, engineering, and other uses Unix is put to. For example, running most of the infrastructure on the Internet.
Yeah, Java runs slow. Boo hoo. So does a windows machine, even when you ignore downtime due to reboots and system crashes.
When this bozo is ready to bet his business on a technology, and is ready to assume full responsibility for the consequences of his decision, and is able to execute on his strategy, then and only then is he qualified to write a credible version of the article referenced.
Perhaps he would have an easier time selling himself as a software architect rather than a software engineer. i.e.: He's good at relating business requirements to system requirements and making sure the pieces fit together in a cohesive way.
This way his actual programming experience is _much_ less important than his people skills, ability to withstand long meetings in a single bound, and general political savvy in a large corporation.
Just my $.02...
Sysadmins probably shouldn't write the equivalent of shell scripts in assembler, but having a firm grounding in it helps to:
1: Analyze core dumps, blue screens, etc. (regardless of platform)
2: Understand tricky things like big and little endian, binary, octal, and hex notation, etc.
4: Have a firm fundamental understanding of what object code is, what linkers and loaders do, why it's important, etc. In a world of DLL hell, at least understand what's going on "behind the scenes"
5: Have a theoretical framework to hang concepts like processes, threads, priority, scheduling, interrupts, etc. -- it might not be _the_ correct model for a particular platform, but it will give a much better starting frame of reference.
On the other hand, if by "sysadmin" you mean "village idiot", and the only requirement is to know how to walk people through rebooting their PC and knowing how to call 1-800-dial-a-vendor and sit on hold, then NO, you don't need to learn assembly. In fact, you could probably get away with not knowing math, history, economics, or much of anythig else. Better yet, how about totally untrained sysadmins? I mean, c'mon, NT/2k/XP is virtually self-administering and it's all "point and click", you could entrust all of your corporate systems to a high school dropout for a fraction of the cost of even an incompetent sysadmin!
Lets face it: Linux is not a clear choice for joe idiot consumer looking to buy a PC to put under the xmas tree.
At the same time, it's come a _long_ way in just a few years. I'd bet my job (well, actually I just about have) that Linux is a better business desktop than Windows. For a business, Linux makes a lot of sense. It gets you off the Microsoft-upgrade-churn cycle, most everything you'd want is freely available, and the simplicity of administration and the excellent security make it a great choice.
Yes, KDE/Koffice, Gnome/'Gnome office', and StarOffice are not MS-Office. So what? As more businesses adopt Linux as their desktop, manufacturers will take note and start offering it, ISVs will take note and start selling more software, and consumers will take note and start buying linux for home since they want to be compatible with what's at the office. Same sort of cycle that made the IBM PC more popular than the Mac back in the late 80's/early 90's.
This might not be the year of linux for the consumer, but it's getting close for linux on corporate desktops.
"A few good men from Univac" is a great book (good luck finding it though, as it's out of print...)
"Eniac" is good.
"Nerds" is a good history of the genesis of the Internet.
"We were burning" is a good book about the japanese semiconductor industry.
"A history of modern computing" is good.
"The invention that changed the world", actually about radar, but nice lead-ins from 1940s technology providing the genesis for the computer industry.
"Enterprise" has two components, first is the relatively straightforward "Core Application" of a company, which might be some sort of ERP system like PeopleSoft, or some other commercial produce. For many companies, this will be an amalgamation of custom programs written over many years. These applications are typically based on some form of transaction processing system (e.g. CICS on the Mainframe, Tuxedo on Unix, or even database-driven transactions ala Oracle, DB2, etc.)
The second, and more critical part of 'Enterprise' is the nature of the computing service. Generally any outage is measured in dollars per minute or hour. It's not unusual for a large company to face severe monetary losses for even slight outages. Think millions of dollars an hour (or even per minute). This measure tends to be a little slippery, but with some analysis a pretty solid figure can usually be determined.
For some enterprises, Linux might make complete sense (e.g. Google). For others, the potential of saving a few thousand or hundred thousand in licensing costs pales in comparison to the probable re-training, new hardware, and "potential" instability of moving to Linux. If you've got something that works, why fix it?
Given the above, even if all of the big 'Enterprise' vendors port their software to Linux, you're not done. Linux clustering in a business context such as Solaris, AIX, and (in the good old days) VMS provide would be one stumbling block. The lack of high-end hardware is another -- and yes I know that Linux runs on anything from a PC to a SPARC server to a S/390 mainframe. In reality, you're unlikely to drop $2million on a big Sun box then load Linux -- you'll want to take advantage of Solaris's dynamic partitioning and other proprietary hooks.
Loading Linux on diverse old hardware makes business sense -- turn that old Sun box into something useful. It doesn't make nearly as much business sense when buying a new non-intel server, since the license fee of the OS (if any) is negligable compared to the overall value of the system in the 'Enterprise'.
Over time this is likely to change, since Linux represents a constantly improving and freely available system, vendors will start adopting it as 'their' OS. IBM is an early starter here, but the process will take time. And like a battle of attrition Linux has the advantage over time, since it is constantly improving (for free), while commercial vendors have to dump $millions into R&D to bring out each new version of their OS.
I seem to recall all of these (or their predecessors) offering things for sale. I remember, in particular, that one of these late-80's early 90's online services featured an on-line shopping mall. Unfortunately I don't recall which since I never used it. Thought it was too gimmicky.
What about CompuServe? They had a lot of 'features' that would instantly charge your account for access, e.g.: I took an IQ test that cost not just connect time but an actual fee. After choosing the option the test was presented. (unfortunately I didn't score high enough to catch on to the e-commerce boom of the mid-late 90's).
All of these services also allowed you to dial in, enter a credit card, and have (more or less) instant access, which smells a lot like e-commerce. For a while compuserve even sent out a kit with a manual and some other stuff. This back in 1986-1987.
I haven't seen a trackball on a laptop in years, although the side-car style that old TI and Toshiba laptops used was pretty nice.
If you can find one of the old sidecar style ones, or even a serial or USB trackball that is suitably sized to bolt onto the side of a notebook you could probably rig something up:
Most laptops have removable a floppy/cd drive bay, or PCMCIA slot or 2nd battery slot, usually one of these would be located on the "correct" side of the computer.
Whichever slot is free, find a plastic blank or construct one out of plastic/wood/whatever, then mount the trackball to that.
Not as elegant as having one purpose-built, but with a little craftsmanship and a bit of luck it would probably turn out looking just dandy.
Cable Modem is alive and well in upstate New York. DSL however has always been much more difficult to get. Not surprising, when you look at the equation:
Old copper + recalcitrant phone company / severe technical limitations + high cost == bad business.
Lets face it, just getting DSL to work is virtually a miracle, and getting it to work on every copper line going to every home is simply unrealistic.
DSL seems to be a good onesy-twosey kind of thing to implement, but I don't envy the people trying to make it work at thousands of subscriber sites.
Subject says it all...
Best freakin laptop I've ever used. Great keyboard. Great screen. Great size. Not too heavy. I love this thing and it runs Linux really well.
I don't use the winmodem, as I generally don't use a modem at all. If it came down to it though, I'd get a PCMCIA modem before I'd give up my T20.
Looks like this guy just doesn't get it.
1: Free software (& 'open source') is not about killing Microsoft. Microsoft is bent on screwing the user over for as much cash as they can take. Free software is about freeing the user from that domination, and giving the user control of their own software destiny.
2: Leader? We don't need to steenking leader! This is a grassroots movement in what is probably the most egalitarian forum ever devised. If you can write good code, people respect that. What's a leader going to do? Enforce project timelines? Talk to the press? We've already got lots of people doing that e.g.: RMS & ESR.
3: Left to themselves, the people writing the code will go through their own darwinian selection process. Some projects will gain at the expense of others, some will merge, some will die, some will co-exist. This process takes time, (sometimes _too_long_), but so what? If you want something to move faster, contribute to it! With so many good coders contributing to the community, the richness and quality of Free Software accretes over time. At some point, the sheer mass of high-quality free software will overwhelm the ability of proprietary software to compete. This is already starting to happen with Linux vs. proprietary Unix, and will likely happen in other areas in the next few years.