So how do you actually convince your company and/or cohorts to make this conversion? Or even do some trials or testing?
I went to the COO with a white paper showing potential areas of costs savings. He brought in the CFO and they gave me resources and billing codes (necessary, we are a systems integrator, all techie folk have to produce billable work) so I could run research and analysis into the viability of IPTel and VoIP for our company. He also held a meeting with the telecom specialists and the network team (I was, at the time, the senior architect for the operations group) and between the COO and myself we gained consensus that we needed to investigate the technologies.
Then I went off and worked with several vendors (Avaya, Cisco, Nortel, etc.) to define what we would need to migrate and how soon we could expect an ROI (about 20 months) if we migrated our PBX system to IPTel. Then I brought in a project manager, we put a project charter and cost model together and took that back to the COO as a proposal. We also gained agreement from the call center and the various IT groups (technical delivery, operations, telecom, network operations) by taking them on tours of existing IPTel installations and meeting with organizations that had already done the migration. We included all of those details in the proposal as well. The COO and CFO liked it and chartered a project to implement an IPTel system to replace our PBX.
Re:Big Blue vs. The Banna Republic Phone Company
on
IBM To Run VoIP On Linux
·
· Score: 2, Informative
If a significant proportion of the profitable lines are abandoned in favour of voip, the teleco gets upset.
The poster obviously doesn't understand what enterprise VOIP is all about. Actually, there are two different things going on, and people often confuse the two.
VOIP - send voice protocols over IP. So, instead of paying for a tie line between two of my campus locations, I route voice from my PBX across my wide area data circuits to my other location and terminate onto another PBX. This is all about cost savings by reducing the number of PSTN circuits I need. I'm still using PBX and TDM technology for my phone system.
IPTEL - IP Telephony, no PBX or TDM at all. My PSTN circuits terminate on a DSP based gateway and are converted to TCP/IP packets. Those packets are routed by a server based call management system to IP hosts (telephones, computers with software, analog converters, etc.). This is an internal strategy to converge my voice and data networks and eliminate costly and proprietary PBX and TDM systems.
In the not too distant past I was the architect for an IPTEL project. We eliminated 3 PBX'es in our 5 building campus and replaced it with a single IPTEL system. We have no PBX, our phones run on the same network as our data. Our data network is extremely redundant and high speed. Switched gigabit to the servers, switched 100 megabit to the end points, collapsed backbone layer 3 switching throughout. All core switches are redundant, all call manager servers are redundant, all voice mail servers are redundant.
Our PBX system was about 12 years old and needed to be replaced. Migrating to a converged solution cost us less than replacing and we moved to modern technology. Going forward over the next 5 to 10 years most business telephony will migrate to IPTel. We also use VOIP to route phonecalls within our campus. External calls, both in and out, come in on PSTN circuits, hit our gateway and then are pure IP from that point onwards. With QSIG and QoS our quality is just as good as any PBX system. That's the short term advantage. Also, short term, we have integrated our voice mail and our email. Employees can now receive and listen to voice mail in their groupware client (Outlook in our case) or they can listen to email via the voicemail system while on the road. We publish system alerts and the like directly to the phones via an XML browser capability on the telephones. We can converge voice and data into any application we want. Our CRM system will be completely integrated (voice and data) in about two more months.
This is what IBM and other enterprises want VOIP (more properly IPTEL) for, not for the PSTN reductions. They can already achieve that by routing phone calls on their global private network from PBX to PBX using VOIP technology. And they probably already do that.
While all of these ideas are "really cool", let's operate on the KISS principle here. With the low cost of IDE RAID these days, why not just create a RAID 1 mirror set and NFS export it. You could do 200 GB for probably $300 or so, based on a quick check of the prices on Comp USA. And if you really want distributed redundancy, set up a second system with another RAID 1 array. Then rsync the two with cron.
Using nbd or afs is pretty cool, technically speaking, buy way overkill for a home network and way more trouble, both to set up and to maintain, than it's worth. Instead, for the cost of one more PC you can set up a very redundant system. And in the extremely unlikely case that you lost both hard drives supporting your primary nfs simultaneously you could redirect yourself to your secondary and keep right on working. In a more likely scenario, where you lose just one drive, you immediately rsync to the secondary, repoint all clients to the secondary, and keep going. Replace the failed drive in your primary, then you could either fail back, or demote it to be the new secondary.
Pieroxy wrote: Dude: 7 minutes vs 28 seconds. That's more than 1:14!!!
Dude: This comparison leaves so much out that it is completely meaningless.
No mention is made of data layout, I would say that the "benchmark" author didn't bother to ensure that he had the exact same data layout and filesystem layout between the two disks.
Were the files clean, or fragmented? Was one disk clean and the other fragmented?
Were these the only disks in each system? Or was some of the other I/O mentioned hitting against a different spindle and/or controller?
More spindles, smaller disks on the spindles, will give better performance than one spindle (or a few) with a larger disk. I'm betting that he added the SCSI drive to his server and much of the I/O activity on the system was hitting a different disk, possibly a different controller as well.
What drivers is he using and what operating system? Is it exactly the same, down to the kernel version on both systems? Did he use a generic IDE driver on one and the vendor's highly optimized and tuned SCSI driver on the other?
Are both systems basic PC in nature? Or is one a laptop and the other a low end server?
I have actually been part of benchmarks that had such a wide disparity in performance. But we could back up our results with rigorous benchmark standards and results. The benchmarks involved several different database engines on the exact same server hardware, storage subsystem and operating system. All system parameters, data load processes, operating system optimizations, DBMS tuning and queries were carefully documented and reviewed by independent SME's. Benchmark results were documented and reviewed after being repeated multiple times for each system. At the end of it all we had a benchmark that was acceptable to our customers, the engineers and the PHB's.
This so called benchmark would be laughed out the door. SCSI generally performs better for I/O intensive scenarios and that can be proven with appropriate benchmarking. But this "benchmark" was not rigorous or thorough and proved nothing.
The problem is that you can't take all of those things in isolation. When you put it all together, Teradata is a closed system. If you want to move from Teradata to something else, Oracle or DB2, for example, it will be quite costly and difficult. In my current scenario I can, fairly easily, change hardware and OS. Both Oracle and DB2 run on a wide variety of platforms, including wintel, lintel, and RISC/Unix solutions. If my Sun platforms, as they age, can be better replaced with HP platforms, for example, that is not a huge issue. The SA's will gripe, but they will survive.
Otherwise, given it's performance and capabilities, I agree that Teradata is a great data warehousing platform. Unfortunately with small market share and a closed system I don't think it will survive.
None taken, your opinion of me doesn't really bother me. I just restarted it and don't know if I'll continue it, or not. I deleted all the old entries because I had not been active on/. for over a year and didn't see any reason to keep the old stuff. The comment was made after our discussion, but it is not a realization just because of our discussion.
I guess one would argue that a movie star is 100x more skilled that a firefighter (due to their differeing compensations and higher competition) but I don't see any difference. I suspect you dont' agree with my views:)
I think there's a difference here between what we are calling skilled labor. I have been both a skilled laborer (carpenter) and a skilled white collar worked (programmer, as well as a member of management (security officer) and a couple of things in between, like an army sergeant and a helpdesk technician. Skilled labor means skilled in what it is that you do. A movie star and a firefighter are skilled in different ways. I wouldn't ask a movie star to go in a burning building and rescue a small child any more than I would ask a firefighter to play the lead role in "Casablanca" (unless he/she could act of course). Your skills in a specific area add value to the material you work with. This is obvious if you consider carpentry, or computer programming for that matter. A skilled carpenter can contribute to a well built house, an unskilled day laborer can't. This seems obvious, and which house will have more value I wonder? I don't see how you can fail to distinguish between skilled and unskilled labor.
I think you are distinguishing from a class competition perspective and I'm making the distinction from a "what value or lack of value" does the labor add to the material. Then again, Americans tend not to make class distinctions. I personally find it absurd to pay sports and movie stars the salaries they get, but it's also a question of what value the society places on the skill. I don't agree with some of the values of my society, but I'm not in the majority:-).
Most liberatarians who run for office actually end up running or supporting the Republicans. In essence, Americans do not consider anything related to anarchism to be liberatarian, which IMO is wrong...
Well, if my choice is between a platform whose aim is to try and reduce government and a platform whose aim is to try and increase government (and I realize that is a huge simplification) then I choose the first alternative. I believe in the least government possible, preferably none except that little needed to keep the strong from taking advantage of the weak. I believe in personal responsibility and choice. I believe that I should be absolutely free to do as I please up to the point that my choices infringe on someone else's freedom. At that point, "the system", whether that is a group of individuals or a government, should prevent my actions, or correct them after the fact. Now, that is only a short step up the scale that runs from anarchy to totalitarianism. I think there is major confusion here. There are two different political sprectrums involved. One is what I discussed above, anarchy on one end and totalitarianism on the other. The other is the more traditional one with fascism (i.e. the right) on one end and communism/socialism (i.e the left) on the other end. The problem with the second spectrum is that our choice is simply between two different forms of totalitarianism: the dictatorship of the proletariat or the elite, or some middle road that tries to balance between the two. I don't like those choices, and I don't think they represent reality either. I think reality is that Libertarianism and Anarchy are closely related, the prime difference being that Libertarianism argues for a minimal level of rules and structure and Anarchy argues for none.
See, you can't always tell just from a few sentences and where he comes from what a man believes. Hopefully I've posed some food for thought, or at least corrected what I believe are some misperceptions. If you want to continue the conversation, try my journal. Have a good one.
Yes, but Teradata has too narrow a market base, too expensive, too proprietary, not enough people who actually know the product. And yes, it runs on Intel, but it has the same major reason to dislike it as to dislike Microsoft: It's a closed, proprietary system that doesn't conform to open standards. Unix, java and DB2, on the other hand, conform to published standards that are, more or less, open and other products that conform to the standards are interchangeable at need.
$DEITY only knows why. Intel have demonstrated their ability to make the fastest x86 CPU on the market. If the fact they do this by lengthening the pipeline and ramping up the clockspeed is an affront to whatever biases you hold, then feel free to buy one of the slower, lower clocked CPUs on the market - intel even make a couple of them for you to pick from if you want.
Actually, my one bias is for reliable, scalable, stable systems. Intel, especially running Windows, doesn't fit into that bias very well.:-)
That is what I don';t like. Liberatarians in USA have their own definition which is somewhat incompatible with the rest of the world.
Just one last thing that bothered me a bit. Why is it your (i.e. the non-US definition> definition that I have to achieve compatibility with? Perhaps it's the other way around. Why not just say that there are multiple definitions of Libertarianism and yours is different from mine? This reminds me of the Soviet and ChiCom quibbling over who's communism was purer. On the other hand, if you have some valid reasons for that beyond disliking it as different because it is from the U.S. let's discuss it.
what i want to know is how much you pay your admin people.. you bought more CPU's even though you were 60% iowait ?
Actually, we knew we needed both CPU's and disk. Here's why. The system was IO bound, that was clear from a simple reading of top. But we also intended to add more users, and even if we removed the factors contributing to being IO bound, we still would need more CPU cycles for the user queries. Because of the way the project was built and interconnected with some other projects, we first added the CPU's, and then moved the data to new disk. So, we were able to measure performance in both states.
in general, 64bit computing is a waste of time and performance, unless you need a 64 bit address space. you can fit half the instructions in cache, half the pointers in your data structures, load half as many addresses per cycle, etc. We've got a couple of 8 and 16GB SQL server boxes so when Win64 and SQL64 have baked a bit longer we may migrate those databases to 64bit platforms..
Sorry, but your beloved wintel doesn't support the heavy lifting needed. I have yet to see a true multi-terabyte data warehouse run on wintel and sql server. Although you just might be able to do it reasonably using Windows, Intel and Sybase IQ Multiplex. The application I was talking about was a 350 GB fraud datamart.... not even the data warehouse. This datamart, on technology that is not "interesting or novel", manages to support that much data, real time data loads, real time interactive queries and so forth. And, as you might imagine, a fraud datamart gets very heavy ad-hoc analytical queries. We aren't really worried about the technology being interesting or novel, we are worried about it doing what we need it to do. I suspect that a 4 or 8 CPU system based on Intel technology would be CPU bound in this situation, not IO bound. Although I'm not really willing to try it and waste the money.
We have consistently hit performance and scalability ceilings with Intel, especially when running Windows. Intel processors seem to scale a bit better with Solaris x86 or Linux than with Windows, although not much. By contrast, the main limiting factor we have found with our Sun and IBM technology is our disk farm. Then again, the organization I work for deals with nearly 200 million transactions annually, a data warehouse that contains about 2.5 terabytes of data, an imaging system with more than one billion images available in either real time or as little as 10 minutes (for the offline images), about 5,000 total users and more than 5,000,000 customers who all have to interact with the system in some fashion or another. So, not only is the performance and scalability important, but so is the reliability, availability and stability. When the downtime costs more than entire wintel server, you find that the ROI of those Sun and IBM servers you scoffed at makes a lot of sense. The data center doesn't run on those platforms (not to mention HP Superdome and Alpha) because of some sort of hegemony, but rather because the systems are proven, reliable, stable, scalable and perform well with enormous user and data loads on them. On top of all of that, they aren't vulnerable to the worm of the week because the OS vendor can't manage to separate the user from the OS.
Not to turn this into a holy war or something, but Intel CPU's may increase in computing power each generation, but if you plot a curve using something objective like SPEC you see that the increase is a parabolic curve along the X axis, that is performance is not increasing as fast as it did in the prior generation. Put processor generation/clock speed on the X axis and SPEC benchmark baseline on the Y axis. Now, hopefully, Intel will figure a technology path out of their dead end. However, if you take the increase in performance of "boring" 64bit RISC processors, interestingly enough the curve is parabolic along the Y axis. Admittedly the improvement is gentle, but still there.
drmsithy said, "What on earth makes you think that ? Clockspeed != performance."
Did you read my post? Or just pick out one line that you thought was the essence of the post? The point I was making was that clockspeed does not equal performance. I suppose, to make it much clearer for you, I should have written "If clockspeed equals performance as Intel would have us believe, then theoretically the PE2650 should outperform the PE2550 and 280R by about 3 times....".
Geeez, read the whole post the next time and then you would see my closing line where I take issue with how Intel is building CPU's.
For instance, you claim Jefferson is a liberatarian (partly he was). But he is mostly a (classical) liberal. He was mostly interested in empowering people (and eliminating the monarchy) more so than the liberatarian notion of "self-rule".
Actually, there was no such thing as a "Libertarian" then, so it's kind of tough to argue. But, I think if we look at the U.S. Constitution and especially the Bill of Rights we see a recognition that government of some sort is necessary (to keep other governments from taking up the power vacuum primarily) while attempting to limit so that each individual man (and they really meant man at the time, but they were, after all, a product of their times) would be able to rule themselves to the greatest extent possible. At this point I'm trying to read the minds of men like Jefferson, Hamilton and Franklin based on scraps of their writings and my own political bent. So maybe I'm off a bit. But at least I'm not totally rewriting what they intended as the politicians have been doing basically since the days of Lincoln.
I think you are wrong about Marx's views.
Notice how I worded it. In Marx's writings he made it clear that labor added value to material, but he did not distinguish between skilled and unskilled labor.
I guess we can agree on one thing: Free Software is not capitalism--or do you disagree?
I don't think the Free Software movement is necessarily for or against capitalism. In that sense I'm pretty sure we agree. The worst excess of capitalism, monopoly, is what Free Software is about. Or rather, preventing the monopolization of a workers ideas, contributions and skills by an employer. It is, more over, a fairly natural reaction to the desire by the owners to control IP. So, the workers say "no, we own the IP, but you can use it". This is a natural outcome of an open economic system. That is, when one force becomes too power, then another arises to counterbalance it. Occasionally an unnatural condition exists that we call a monopoly where one side of the balance, or the other, has too much power, preventing the natural balancing of the open system. But even that can only last so long, as is already becoming obvious with Microsoft, or with IBM two decades ago or, for that matter, with the downfall of Leninist/Stalinist communism in eastern Europe.
Free Software is a reaction to the attempt to monopolize ideas, creativity and work.
I can't speak for SCSI, Firewire, SIDE, or any other drive techs 'cause I'm a cheap S.O.B. and won't pay the big bucks for them.
We moved an application from 2 UltraSPARC III 750 MHz CPU's to 6 UltraSPARC III Cu 900 MHz CPU's and saw very little improvement in performance. Then we moved the disk for the application from 9 internal drives to 20 external SCSI over FC drives, and voila our IO wait dropped from 60% or so to 10% +/-. Our query response times dropped by a factor of three or more. Faster, and even more, CPU's are not the answer to data intensive problems, I/O is. Slower (clock speed wise) 64bit CPU's, with better efficiency, more memory addressing, etc. are the norm in the data center for just this reason. IF you can take advantage of your L1/L2 cache then faster clock speed on the CPU will improve performance. The reason most Intel PC's benchmark better than an older box is because the disk, memory and video sub-systems have improved, not because the CPU is making a huge difference.
As proof, search SPEC's benchmark results using Dell and then Sun as your search criteria. Notice the following:
A Dell PE2550 with a PIII 1.13 GHz CPU has a CINT baseline of 561.
A Dell PE2650 with a Xeon 3.06 GHz CPU has a CINT baseline of 1014
A Sun 280R with an US III 1.2 Ghz CPU has a CINT baseline of 637
Theoretically the PE2650 should outperform the PE2550 and 280R by about 3 times, all other factors being equal (i.e. same benchmark). The SPEC benchmark does its absolute best to eliminate I/O systems and network interfaces as a factor, so if we are just talking CPU, cache and memory, the Xeon should have had a CINT baseline of about 1600 or so.
Things get even worse when you start looking at the SMP capabilities and scalability. In a truly linearly scalable SMP system you should be able to go from 1 CPU to 2 CPU's and have the benchmark double. Even the best SMP systems (Sun UltraSPARC and IBM Power) can't quite achieve that. But Itanium really has trouble. Search on Dell and look at the CINT and CFP rates benchmarks. Look at 1, 2 and 4 CPU scores for the Dell 7150.
Bottom line? If you are doing heavy lifting on a server, go SMP with 64bit RISC, or, in some cases, use a cluster of 2 CPU x86 servers. If you are a PC user, you are unlikely to see a significant performance increase with new Intel CPU's unless you upgrade the whole system, not just the CPU.
This whole thing of adding clock cycles and deepening the pipeline is not working out well.
Of course the sort of Libertarianism I talk about is U.S. centric, that's where I'm from after all. A man named Thomas Jefferson almost slipped a Libertarian system over on us, but not quite. And these days the values that Jefferson built into the U.S. Constitution seem to be dead anyhow.
On a side note, your description of anarchism sounds like a subset of it, just as my idea of libertarianism is a subset of the whole.
That said, not all people who work together for the common good (i.e. communal or community effort) are socialists. In fact, the idea existed long before the idea of communism or socialism. There are some good examples of communal type societies and groups in the United States in the early 1800's. Japanese society of the middle ages, while feudal in nature, contained many communal type drivers and a disdain of money as a means for compensation. So did the early Christian groups (pre-Catholic church). So, while I will agree that open-source is about communal effort, I don't really agree that it is socialism. Socialism is as much about trying to level the playing field by government intervention as it is about communal effort. Communism, as defined by Marx, Friederich and Engels, is closer to what open-source is about. And even there they seem to have reversed the basics. In Marxist communism an item's value is based on its material worth and skilled workmanship/labor contributes no more added value than unskilled workmanship/labor does. Open-source says the opposite, code in and of itself has no value until a skilled individual makes it useful.
Anyhow, I agree with much of what you say, but not entirely. Thanks for an interesting discussion.
And some years back I used Netscape's server and never found anything wrong with it.
I ran Netscape Server in the 96 time frame and it was a good product. These days we are running a pretty large intranet website on Sun ONE, the newest incarnation of Netscape. I've also tried out IBM's httpd, which is basically, but not quite, Apache. I've never run AOLServer, so I can't comment on it. But, the point is that there are other alternatives besides Apache. In fact, if you license Solaris 9 for x86 you get the basic version of the Sun ONE webserver and Sun ONE Directory Server with 200,000 user licenses for free. The webserver is as capable as Apache and the Directory Server is one of the best around.
Re:Enough of those double standards!
on
HyperSCSI Examined
·
· Score: 3, Interesting
Fiber Channel maximum cable length: 10,000 m
Add the appropriate routers and switches and you can easily go 90 km on dark fiber. Add some appropriate routers onto a fast network (T3, ATM, what have you) and you can go 500 km. With fast FC connected storage at each end. Of course, this sort of solution is used by data centers, not home users. But Fiber is the obvious solution to data storage problems. And there is enough mass in the server storage market now that prices are starting to come down. Of course, if you need fast, redundant, capable storage you won't blink at the cost.
SCSI may try to continue living, but it got only one direction to go: history's garbage pail.
You do realize that the serious data center hasn't run the SCSI protocol on a SCSI cable for years, right? We run FC everywhere. Even for our internal disks on some servers (Sun Vx80 series). Serial ATA doesn't even come close to SCSI over FC. My Hitachi 9570V gives me 19.5 TB raw storage and more than 100,000 IOPS. Let's see serial ATA and USB2 do that.
One more time, HyperSCSI, like FC and iSCSI is NOT for a desktop PC. It's for the data center. Although HyperSCSI provides some promise for extending centralized data storage to the desktop.
SCSI is dead? For the home computer perhaps, but not in the data center. I have yet to see a serial ATA enclosure perform on anything like the level of enclosures using SCSI over FC. Why don't you go look around a large office environment or a data center and see how many storage devices are SCSI based. This stuff (iSCSI, HyperSCSI, Ultra320, etc.) are aimed at servers and storage networks, not your desktop PC.
In 1983 I had a Bondwell "portable" computer. It weighed about 15 pounds or so, had two double density 5.25" floppies, a 4" CRT, the keyboard snapped onto the end to make it a portable unit. It ran CPM and had either 64K or 128K of RAM, I don't recall which. My dad's office had some sort of mini, probably a DEC. Between the two, plus the TRS-80's in the lab at school, I got really interested in computers. It was all downhill from there. Now I play with SunFire 6800's and Linux is my hobby at home.
So Free Software is 70% anarchist and 30% socialist.
I would actually categorize Free Software as Libertarian. Check this web site for more information on what Libertarianism is all about. This quote is the essence of it: "everyone should be free to do as they choose, so long as they don't infringe upon the equal freedom of others."
What Stallman rebelled against was the fact that his employer had freedom regarding his code, but he did not. GNU, FSF and OSS is about ensuring that we all have equal freedom to the greatest extent possible with software. The Libertarian is not an anarchist, he realizes that some sort of controls are necessary to protect the freedom desired. He is not a communist who believes in the dictatorship of the proletariat that will "wither away", nor is he a socialist who believes in leveling the playfield, economically speaking. RMS believes in protecting IP, that keeps me from infringing on your freedom, but he doesn't believe that you should be able stop me from modifying what you have done, as long as I don't steal your work. That's the essence of the GPL, and is a very Libertarian idea.
There are a lot of people out there, in jobs similar to mine. We don't all work in situations where efficiency of large tasks is paramount- but ease of use for many different small tasks. A GUI is perfect for us.
And fortunately for you a couple of things about Linux/Unix are true:
Modern Unix/Linux GUI's give you a fairly good degree of control for system administration tasks, without locking you to the GUI when the CLI is a better choice
Unix/Linux servers, especially in a small office, will require far less in the way of system administration than your Windows servers do.
Your hardware life cycle will lengthen, reducing the number of work cycles you have to put into maintaining your older systems and implementing your newer systems.
An appropriately implemented Unix network can work just fine in a small office too. I've worked in a small IT shop, the whole company had 10 employees, 1 database guy, 2 programmers, an engineer (me), a system administrator, a sales guy, a receptionist, an accountant and the owner. We ran Unix and Linux just fine. Our *nix servers (web, mail, database) took about the same amount of administration as our one and only Windows server (file server). That was in 1995 or so, well before the advent of the user friendly *nix.
However it's taken 4 generations of windows operating systems (depending on how you count) to make things buisness-friendly as they are in NT/XP
Okay, I've been a system administrator, network engineer, ISP webmaster, security administrator, C programmer and, most recently, a security officer. The height of Microsoft's "business friendliness" was the Win95/NT 4 combination, and that was really on "friendly" to the small/medium business market. Today, with the advent of weekly patch cycles, proprietary document format lockin, Software Assurance, continuous virus/worm threats, lack of appropriate security in the software, and more I would hardly characterize Microsoft as "business friendly". They are universally used because they are a monopoly with a stranglehold on the PC and desktop software market. As for ease of use, Mac has always been easier to use, as was OS/2. Windows didn't win on the desktop because it was better, it won because Microsoft used the VHS vs. Beta approach. DOS and Windows cost less and had more software available for it, and it was ubiquitious as more and more hardware vendors took advantage of OEM pricing.
Unfortunately, most Microsoft based organizations that try to migrate to F/OSS will implement things in the way you are discussing, where the user can "get under the hood". If you deployed Unix (Linux is just another incarnation of Unix) appropriately, with the OS and apps residing on a server and the users connecting via smart terminals this becomes a non-problem. The terminal session is set up to deliver the GUI and apps the user needs to do their job. For office automation workers this makes perfect sense and is extremely efficient from a resource, dollars and cents and people perspective. For more information, read the portion of this article that deals with how Unix in a distributed environment is set up see this story.
There was also recent/. coverage of this approach. Of course most Microsoft shops will try to migrate to Unix in a way that will allow "tinkering under the hood" and all sorts of issues. However, one of my colleagues implemented a Unix/thin client solution in a large data center. Not only did it cost about 60% of a PC solution, but the desktop support dropped to nearly zero. The sys admins, network admins and engineers don't have the lost productivity involved in updating, maintaining and patching desktop PC's. Their "workstation" has the resiliency of a server and they can connect to their desktop from anywhere and have the exact settings they want every time.
But, unfortunately, your scenario is probably more likely after all. Guys who have grown up with the concept of stand alone Windows PC's will try to clone that with Unix/Linux.
Ok, that was a bad example. But moving to "open source" is only the first step. Next step is user education, and thats going to be a big step to climb.
What are you talking about? The "typical user" needs an email client, office automation suite and a web browser. They need to be able to connect to network locations to access shared data. They need to run third party applications like Business Objects, Crystal Reports, custom stuff created in PeopleSoft or SAP or Oracle, etc. They don't need education in using the operating system for the most part.
They are going to get educated in those things, and need training in those things, regardless of the operating system. And a lot of those things they already know how to use. The differences between OpenOffice and MS Office are pretty small, as is the difference between Evolution and Outlook. Since most of the ISV's are moving towards thin applications they pretty much work the same on Linux, UNIX and Windows.
Once the users get past the "oh my god, it's a new OS" reaction, they'll get right back to work.
It's really the developers, engineers and administrators where the issue is at. In most government organizations Microsoft is deeply entrenched and these folks are going to have to make large changes in how they think and work. If they try to implement Linux and F/OSS using the same methodologies and thought processes they have used for Windows they will have issues, to say the least.
Just getting the sys admins to understand that they can administer a Linux server far more effectively, efficiently and quickly from a command line than they could in the Windows GUI will take some doing, let alone convincing them that they can, with only a bit more effort, replicate that work across a large number of servers/workstations simultaneously, all from the command line. And wait til the first time they have to compile something because there's no pretty install package available!
All through the early to mid 90's the Army was using Windows for the computers that commanders and operations staff used in the field. I suspect that a lot of the computer gear that the individual soldiers carry now is based on Windows CE, although I can't speak from personal experience on that. My first comment is based on direct, first hand experience.
Slashdot Mirror
So how do you actually convince your company and/or cohorts to make this conversion? Or even do some trials or testing?
I went to the COO with a white paper showing potential areas of costs savings. He brought in the CFO and they gave me resources and billing codes (necessary, we are a systems integrator, all techie folk have to produce billable work) so I could run research and analysis into the viability of IPTel and VoIP for our company. He also held a meeting with the telecom specialists and the network team (I was, at the time, the senior architect for the operations group) and between the COO and myself we gained consensus that we needed to investigate the technologies.
Then I went off and worked with several vendors (Avaya, Cisco, Nortel, etc.) to define what we would need to migrate and how soon we could expect an ROI (about 20 months) if we migrated our PBX system to IPTel. Then I brought in a project manager, we put a project charter and cost model together and took that back to the COO as a proposal. We also gained agreement from the call center and the various IT groups (technical delivery, operations, telecom, network operations) by taking them on tours of existing IPTel installations and meeting with organizations that had already done the migration. We included all of those details in the proposal as well. The COO and CFO liked it and chartered a project to implement an IPTel system to replace our PBX.
If a significant proportion of the profitable lines are abandoned in favour of voip, the teleco gets upset.
The poster obviously doesn't understand what enterprise VOIP is all about. Actually, there are two different things going on, and people often confuse the two.
In the not too distant past I was the architect for an IPTEL project. We eliminated 3 PBX'es in our 5 building campus and replaced it with a single IPTEL system. We have no PBX, our phones run on the same network as our data. Our data network is extremely redundant and high speed. Switched gigabit to the servers, switched 100 megabit to the end points, collapsed backbone layer 3 switching throughout. All core switches are redundant, all call manager servers are redundant, all voice mail servers are redundant.
Our PBX system was about 12 years old and needed to be replaced. Migrating to a converged solution cost us less than replacing and we moved to modern technology. Going forward over the next 5 to 10 years most business telephony will migrate to IPTel. We also use VOIP to route phonecalls within our campus. External calls, both in and out, come in on PSTN circuits, hit our gateway and then are pure IP from that point onwards. With QSIG and QoS our quality is just as good as any PBX system. That's the short term advantage. Also, short term, we have integrated our voice mail and our email. Employees can now receive and listen to voice mail in their groupware client (Outlook in our case) or they can listen to email via the voicemail system while on the road. We publish system alerts and the like directly to the phones via an XML browser capability on the telephones. We can converge voice and data into any application we want. Our CRM system will be completely integrated (voice and data) in about two more months.
This is what IBM and other enterprises want VOIP (more properly IPTEL) for, not for the PSTN reductions. They can already achieve that by routing phone calls on their global private network from PBX to PBX using VOIP technology. And they probably already do that.
While all of these ideas are "really cool", let's operate on the KISS principle here. With the low cost of IDE RAID these days, why not just create a RAID 1 mirror set and NFS export it. You could do 200 GB for probably $300 or so, based on a quick check of the prices on Comp USA. And if you really want distributed redundancy, set up a second system with another RAID 1 array. Then rsync the two with cron.
Using nbd or afs is pretty cool, technically speaking, buy way overkill for a home network and way more trouble, both to set up and to maintain, than it's worth. Instead, for the cost of one more PC you can set up a very redundant system. And in the extremely unlikely case that you lost both hard drives supporting your primary nfs simultaneously you could redirect yourself to your secondary and keep right on working. In a more likely scenario, where you lose just one drive, you immediately rsync to the secondary, repoint all clients to the secondary, and keep going. Replace the failed drive in your primary, then you could either fail back, or demote it to be the new secondary.
Why make it so hard?
Pieroxy wrote: Dude: 7 minutes vs 28 seconds. That's more than 1:14!!!
Dude: This comparison leaves so much out that it is completely meaningless.
I have actually been part of benchmarks that had such a wide disparity in performance. But we could back up our results with rigorous benchmark standards and results. The benchmarks involved several different database engines on the exact same server hardware, storage subsystem and operating system. All system parameters, data load processes, operating system optimizations, DBMS tuning and queries were carefully documented and reviewed by independent SME's. Benchmark results were documented and reviewed after being repeated multiple times for each system. At the end of it all we had a benchmark that was acceptable to our customers, the engineers and the PHB's.
This so called benchmark would be laughed out the door. SCSI generally performs better for I/O intensive scenarios and that can be proven with appropriate benchmarking. But this "benchmark" was not rigorous or thorough and proved nothing.
The problem is that you can't take all of those things in isolation. When you put it all together, Teradata is a closed system. If you want to move from Teradata to something else, Oracle or DB2, for example, it will be quite costly and difficult. In my current scenario I can, fairly easily, change hardware and OS. Both Oracle and DB2 run on a wide variety of platforms, including wintel, lintel, and RISC/Unix solutions. If my Sun platforms, as they age, can be better replaced with HP platforms, for example, that is not a huge issue. The SA's will gripe, but they will survive.
Otherwise, given it's performance and capabilities, I agree that Teradata is a great data warehousing platform. Unfortunately with small market share and a closed system I don't think it will survive.
Your journal sucks... no offense :)
None taken, your opinion of me doesn't really bother me. I just restarted it and don't know if I'll continue it, or not. I deleted all the old entries because I had not been active on /. for over a year and didn't see any reason to keep the old stuff. The comment was made after our discussion, but it is not a realization just because of our discussion.
Here's my response. Take it or leave it.I guess one would argue that a movie star is 100x more skilled that a firefighter (due to their differeing compensations and higher competition) but I don't see any difference. I suspect you dont' agree with my views :)
I think there's a difference here between what we are calling skilled labor. I have been both a skilled laborer (carpenter) and a skilled white collar worked (programmer, as well as a member of management (security officer) and a couple of things in between, like an army sergeant and a helpdesk technician. Skilled labor means skilled in what it is that you do. A movie star and a firefighter are skilled in different ways. I wouldn't ask a movie star to go in a burning building and rescue a small child any more than I would ask a firefighter to play the lead role in "Casablanca" (unless he/she could act of course). Your skills in a specific area add value to the material you work with. This is obvious if you consider carpentry, or computer programming for that matter. A skilled carpenter can contribute to a well built house, an unskilled day laborer can't. This seems obvious, and which house will have more value I wonder? I don't see how you can fail to distinguish between skilled and unskilled labor.
I think you are distinguishing from a class competition perspective and I'm making the distinction from a "what value or lack of value" does the labor add to the material. Then again, Americans tend not to make class distinctions. I personally find it absurd to pay sports and movie stars the salaries they get, but it's also a question of what value the society places on the skill. I don't agree with some of the values of my society, but I'm not in the majority :-).
Most liberatarians who run for office actually end up running or supporting the Republicans. In essence, Americans do not consider anything related to anarchism to be liberatarian, which IMO is wrong...
Well, if my choice is between a platform whose aim is to try and reduce government and a platform whose aim is to try and increase government (and I realize that is a huge simplification) then I choose the first alternative. I believe in the least government possible, preferably none except that little needed to keep the strong from taking advantage of the weak. I believe in personal responsibility and choice. I believe that I should be absolutely free to do as I please up to the point that my choices infringe on someone else's freedom. At that point, "the system", whether that is a group of individuals or a government, should prevent my actions, or correct them after the fact. Now, that is only a short step up the scale that runs from anarchy to totalitarianism. I think there is major confusion here. There are two different political sprectrums involved. One is what I discussed above, anarchy on one end and totalitarianism on the other. The other is the more traditional one with fascism (i.e. the right) on one end and communism/socialism (i.e the left) on the other end. The problem with the second spectrum is that our choice is simply between two different forms of totalitarianism: the dictatorship of the proletariat or the elite, or some middle road that tries to balance between the two. I don't like those choices, and I don't think they represent reality either. I think reality is that Libertarianism and Anarchy are closely related, the prime difference being that Libertarianism argues for a minimal level of rules and structure and Anarchy argues for none.
See, you can't always tell just from a few sentences and where he comes from what a man believes. Hopefully I've posed some food for thought, or at least corrected what I believe are some misperceptions. If you want to continue the conversation, try my journal. Have a good one.
Yes, but Teradata has too narrow a market base, too expensive, too proprietary, not enough people who actually know the product. And yes, it runs on Intel, but it has the same major reason to dislike it as to dislike Microsoft: It's a closed, proprietary system that doesn't conform to open standards. Unix, java and DB2, on the other hand, conform to published standards that are, more or less, open and other products that conform to the standards are interchangeable at need.
$DEITY only knows why. Intel have demonstrated their ability to make the fastest x86 CPU on the market. If the fact they do this by lengthening the pipeline and ramping up the clockspeed is an affront to whatever biases you hold, then feel free to buy one of the slower, lower clocked CPUs on the market - intel even make a couple of them for you to pick from if you want.
Actually, my one bias is for reliable, scalable, stable systems. Intel, especially running Windows, doesn't fit into that bias very well. :-)
That is what I don';t like. Liberatarians in USA have their own definition which is somewhat incompatible with the rest of the world.
Just one last thing that bothered me a bit. Why is it your (i.e. the non-US definition> definition that I have to achieve compatibility with? Perhaps it's the other way around. Why not just say that there are multiple definitions of Libertarianism and yours is different from mine? This reminds me of the Soviet and ChiCom quibbling over who's communism was purer. On the other hand, if you have some valid reasons for that beyond disliking it as different because it is from the U.S. let's discuss it.
what i want to know is how much you pay your admin people.. you bought more CPU's even though you were 60% iowait ?
Actually, we knew we needed both CPU's and disk. Here's why. The system was IO bound, that was clear from a simple reading of top. But we also intended to add more users, and even if we removed the factors contributing to being IO bound, we still would need more CPU cycles for the user queries. Because of the way the project was built and interconnected with some other projects, we first added the CPU's, and then moved the data to new disk. So, we were able to measure performance in both states.
in general, 64bit computing is a waste of time and performance, unless you need a 64 bit address space. you can fit half the instructions in cache, half the pointers in your data structures, load half as many addresses per cycle, etc. We've got a couple of 8 and 16GB SQL server boxes so when Win64 and SQL64 have baked a bit longer we may migrate those databases to 64bit platforms..
Sorry, but your beloved wintel doesn't support the heavy lifting needed. I have yet to see a true multi-terabyte data warehouse run on wintel and sql server. Although you just might be able to do it reasonably using Windows, Intel and Sybase IQ Multiplex. The application I was talking about was a 350 GB fraud datamart .... not even the data warehouse. This datamart, on technology that is not "interesting or novel", manages to support that much data, real time data loads, real time interactive queries and so forth. And, as you might imagine, a fraud datamart gets very heavy ad-hoc analytical queries. We aren't really worried about the technology being interesting or novel, we are worried about it doing what we need it to do. I suspect that a 4 or 8 CPU system based on Intel technology would be CPU bound in this situation, not IO bound. Although I'm not really willing to try it and waste the money.
We have consistently hit performance and scalability ceilings with Intel, especially when running Windows. Intel processors seem to scale a bit better with Solaris x86 or Linux than with Windows, although not much. By contrast, the main limiting factor we have found with our Sun and IBM technology is our disk farm. Then again, the organization I work for deals with nearly 200 million transactions annually, a data warehouse that contains about 2.5 terabytes of data, an imaging system with more than one billion images available in either real time or as little as 10 minutes (for the offline images), about 5,000 total users and more than 5,000,000 customers who all have to interact with the system in some fashion or another. So, not only is the performance and scalability important, but so is the reliability, availability and stability. When the downtime costs more than entire wintel server, you find that the ROI of those Sun and IBM servers you scoffed at makes a lot of sense. The data center doesn't run on those platforms (not to mention HP Superdome and Alpha) because of some sort of hegemony, but rather because the systems are proven, reliable, stable, scalable and perform well with enormous user and data loads on them. On top of all of that, they aren't vulnerable to the worm of the week because the OS vendor can't manage to separate the user from the OS.
Not to turn this into a holy war or something, but Intel CPU's may increase in computing power each generation, but if you plot a curve using something objective like SPEC you see that the increase is a parabolic curve along the X axis, that is performance is not increasing as fast as it did in the prior generation. Put processor generation/clock speed on the X axis and SPEC benchmark baseline on the Y axis. Now, hopefully, Intel will figure a technology path out of their dead end. However, if you take the increase in performance of "boring" 64bit RISC processors, interestingly enough the curve is parabolic along the Y axis. Admittedly the improvement is gentle, but still there.
So, back to my original point, usin
drmsithy said, "What on earth makes you think that ? Clockspeed != performance."
Did you read my post? Or just pick out one line that you thought was the essence of the post? The point I was making was that clockspeed does not equal performance. I suppose, to make it much clearer for you, I should have written "If clockspeed equals performance as Intel would have us believe, then theoretically the PE2650 should outperform the PE2550 and 280R by about 3 times ....".
Geeez, read the whole post the next time and then you would see my closing line where I take issue with how Intel is building CPU's.
For instance, you claim Jefferson is a liberatarian (partly he was). But he is mostly a (classical) liberal. He was mostly interested in empowering people (and eliminating the monarchy) more so than the liberatarian notion of "self-rule".
Actually, there was no such thing as a "Libertarian" then, so it's kind of tough to argue. But, I think if we look at the U.S. Constitution and especially the Bill of Rights we see a recognition that government of some sort is necessary (to keep other governments from taking up the power vacuum primarily) while attempting to limit so that each individual man (and they really meant man at the time, but they were, after all, a product of their times) would be able to rule themselves to the greatest extent possible. At this point I'm trying to read the minds of men like Jefferson, Hamilton and Franklin based on scraps of their writings and my own political bent. So maybe I'm off a bit. But at least I'm not totally rewriting what they intended as the politicians have been doing basically since the days of Lincoln.
I think you are wrong about Marx's views.
Notice how I worded it. In Marx's writings he made it clear that labor added value to material, but he did not distinguish between skilled and unskilled labor.
I guess we can agree on one thing: Free Software is not capitalism--or do you disagree?
I don't think the Free Software movement is necessarily for or against capitalism. In that sense I'm pretty sure we agree. The worst excess of capitalism, monopoly, is what Free Software is about. Or rather, preventing the monopolization of a workers ideas, contributions and skills by an employer. It is, more over, a fairly natural reaction to the desire by the owners to control IP. So, the workers say "no, we own the IP, but you can use it". This is a natural outcome of an open economic system. That is, when one force becomes too power, then another arises to counterbalance it. Occasionally an unnatural condition exists that we call a monopoly where one side of the balance, or the other, has too much power, preventing the natural balancing of the open system. But even that can only last so long, as is already becoming obvious with Microsoft, or with IBM two decades ago or, for that matter, with the downfall of Leninist/Stalinist communism in eastern Europe.
Free Software is a reaction to the attempt to monopolize ideas, creativity and work.
I can't speak for SCSI, Firewire, SIDE, or any other drive techs 'cause I'm a cheap S.O.B. and won't pay the big bucks for them.
We moved an application from 2 UltraSPARC III 750 MHz CPU's to 6 UltraSPARC III Cu 900 MHz CPU's and saw very little improvement in performance. Then we moved the disk for the application from 9 internal drives to 20 external SCSI over FC drives, and voila our IO wait dropped from 60% or so to 10% +/-. Our query response times dropped by a factor of three or more. Faster, and even more, CPU's are not the answer to data intensive problems, I/O is. Slower (clock speed wise) 64bit CPU's, with better efficiency, more memory addressing, etc. are the norm in the data center for just this reason. IF you can take advantage of your L1/L2 cache then faster clock speed on the CPU will improve performance. The reason most Intel PC's benchmark better than an older box is because the disk, memory and video sub-systems have improved, not because the CPU is making a huge difference.
As proof, search SPEC's benchmark results using Dell and then Sun as your search criteria. Notice the following:
Theoretically the PE2650 should outperform the PE2550 and 280R by about 3 times, all other factors being equal (i.e. same benchmark). The SPEC benchmark does its absolute best to eliminate I/O systems and network interfaces as a factor, so if we are just talking CPU, cache and memory, the Xeon should have had a CINT baseline of about 1600 or so.
Things get even worse when you start looking at the SMP capabilities and scalability. In a truly linearly scalable SMP system you should be able to go from 1 CPU to 2 CPU's and have the benchmark double. Even the best SMP systems (Sun UltraSPARC and IBM Power) can't quite achieve that. But Itanium really has trouble. Search on Dell and look at the CINT and CFP rates benchmarks. Look at 1, 2 and 4 CPU scores for the Dell 7150.
Bottom line? If you are doing heavy lifting on a server, go SMP with 64bit RISC, or, in some cases, use a cluster of 2 CPU x86 servers. If you are a PC user, you are unlikely to see a significant performance increase with new Intel CPU's unless you upgrade the whole system, not just the CPU.
This whole thing of adding clock cycles and deepening the pipeline is not working out well.
Of course the sort of Libertarianism I talk about is U.S. centric, that's where I'm from after all. A man named Thomas Jefferson almost slipped a Libertarian system over on us, but not quite. And these days the values that Jefferson built into the U.S. Constitution seem to be dead anyhow.
On a side note, your description of anarchism sounds like a subset of it, just as my idea of libertarianism is a subset of the whole.
That said, not all people who work together for the common good (i.e. communal or community effort) are socialists. In fact, the idea existed long before the idea of communism or socialism. There are some good examples of communal type societies and groups in the United States in the early 1800's. Japanese society of the middle ages, while feudal in nature, contained many communal type drivers and a disdain of money as a means for compensation. So did the early Christian groups (pre-Catholic church). So, while I will agree that open-source is about communal effort, I don't really agree that it is socialism. Socialism is as much about trying to level the playing field by government intervention as it is about communal effort. Communism, as defined by Marx, Friederich and Engels, is closer to what open-source is about. And even there they seem to have reversed the basics. In Marxist communism an item's value is based on its material worth and skilled workmanship/labor contributes no more added value than unskilled workmanship/labor does. Open-source says the opposite, code in and of itself has no value until a skilled individual makes it useful.
Anyhow, I agree with much of what you say, but not entirely. Thanks for an interesting discussion.
And some years back I used Netscape's server and never found anything wrong with it.
I ran Netscape Server in the 96 time frame and it was a good product. These days we are running a pretty large intranet website on Sun ONE, the newest incarnation of Netscape. I've also tried out IBM's httpd, which is basically, but not quite, Apache. I've never run AOLServer, so I can't comment on it. But, the point is that there are other alternatives besides Apache. In fact, if you license Solaris 9 for x86 you get the basic version of the Sun ONE webserver and Sun ONE Directory Server with 200,000 user licenses for free. The webserver is as capable as Apache and the Directory Server is one of the best around.
Fiber Channel maximum cable length: 10,000 m
Add the appropriate routers and switches and you can easily go 90 km on dark fiber. Add some appropriate routers onto a fast network (T3, ATM, what have you) and you can go 500 km. With fast FC connected storage at each end. Of course, this sort of solution is used by data centers, not home users. But Fiber is the obvious solution to data storage problems. And there is enough mass in the server storage market now that prices are starting to come down. Of course, if you need fast, redundant, capable storage you won't blink at the cost.
SCSI may try to continue living, but it got only one direction to go: history's garbage pail.
You do realize that the serious data center hasn't run the SCSI protocol on a SCSI cable for years, right? We run FC everywhere. Even for our internal disks on some servers (Sun Vx80 series). Serial ATA doesn't even come close to SCSI over FC. My Hitachi 9570V gives me 19.5 TB raw storage and more than 100,000 IOPS. Let's see serial ATA and USB2 do that.
One more time, HyperSCSI, like FC and iSCSI is NOT for a desktop PC. It's for the data center. Although HyperSCSI provides some promise for extending centralized data storage to the desktop.
SCSI is dead? For the home computer perhaps, but not in the data center. I have yet to see a serial ATA enclosure perform on anything like the level of enclosures using SCSI over FC. Why don't you go look around a large office environment or a data center and see how many storage devices are SCSI based. This stuff (iSCSI, HyperSCSI, Ultra320, etc.) are aimed at servers and storage networks, not your desktop PC.
In 1983 I had a Bondwell "portable" computer. It weighed about 15 pounds or so, had two double density 5.25" floppies, a 4" CRT, the keyboard snapped onto the end to make it a portable unit. It ran CPM and had either 64K or 128K of RAM, I don't recall which. My dad's office had some sort of mini, probably a DEC. Between the two, plus the TRS-80's in the lab at school, I got really interested in computers. It was all downhill from there. Now I play with SunFire 6800's and Linux is my hobby at home.
So Free Software is 70% anarchist and 30% socialist.
I would actually categorize Free Software as Libertarian. Check this web site for more information on what Libertarianism is all about. This quote is the essence of it: "everyone should be free to do as they choose, so long as they don't infringe upon the equal freedom of others."
What Stallman rebelled against was the fact that his employer had freedom regarding his code, but he did not. GNU, FSF and OSS is about ensuring that we all have equal freedom to the greatest extent possible with software. The Libertarian is not an anarchist, he realizes that some sort of controls are necessary to protect the freedom desired. He is not a communist who believes in the dictatorship of the proletariat that will "wither away", nor is he a socialist who believes in leveling the playfield, economically speaking. RMS believes in protecting IP, that keeps me from infringing on your freedom, but he doesn't believe that you should be able stop me from modifying what you have done, as long as I don't steal your work. That's the essence of the GPL, and is a very Libertarian idea.
There are a lot of people out there, in jobs similar to mine. We don't all work in situations where efficiency of large tasks is paramount- but ease of use for many different small tasks. A GUI is perfect for us.
And fortunately for you a couple of things about Linux/Unix are true:
- Modern Unix/Linux GUI's give you a fairly good degree of control for system administration tasks, without locking you to the GUI when the CLI is a better choice
- Unix/Linux servers, especially in a small office, will require far less in the way of system administration than your Windows servers do.
- Your hardware life cycle will lengthen, reducing the number of work cycles you have to put into maintaining your older systems and implementing your newer systems.
An appropriately implemented Unix network can work just fine in a small office too. I've worked in a small IT shop, the whole company had 10 employees, 1 database guy, 2 programmers, an engineer (me), a system administrator, a sales guy, a receptionist, an accountant and the owner. We ran Unix and Linux just fine. Our *nix servers (web, mail, database) took about the same amount of administration as our one and only Windows server (file server). That was in 1995 or so, well before the advent of the user friendly *nix.However it's taken 4 generations of windows operating systems (depending on how you count) to make things buisness-friendly as they are in NT/XP
Okay, I've been a system administrator, network engineer, ISP webmaster, security administrator, C programmer and, most recently, a security officer. The height of Microsoft's "business friendliness" was the Win95/NT 4 combination, and that was really on "friendly" to the small/medium business market. Today, with the advent of weekly patch cycles, proprietary document format lockin, Software Assurance, continuous virus/worm threats, lack of appropriate security in the software, and more I would hardly characterize Microsoft as "business friendly". They are universally used because they are a monopoly with a stranglehold on the PC and desktop software market. As for ease of use, Mac has always been easier to use, as was OS/2. Windows didn't win on the desktop because it was better, it won because Microsoft used the VHS vs. Beta approach. DOS and Windows cost less and had more software available for it, and it was ubiquitious as more and more hardware vendors took advantage of OEM pricing.
Unfortunately, most Microsoft based organizations that try to migrate to F/OSS will implement things in the way you are discussing, where the user can "get under the hood". If you deployed Unix (Linux is just another incarnation of Unix) appropriately, with the OS and apps residing on a server and the users connecting via smart terminals this becomes a non-problem. The terminal session is set up to deliver the GUI and apps the user needs to do their job. For office automation workers this makes perfect sense and is extremely efficient from a resource, dollars and cents and people perspective. For more information, read the portion of this article that deals with how Unix in a distributed environment is set up see this story.
There was also recent /. coverage of this approach. Of course most Microsoft shops will try to migrate to Unix in a way that will allow "tinkering under the hood" and all sorts of issues. However, one of my colleagues implemented a Unix/thin client solution in a large data center. Not only did it cost about 60% of a PC solution, but the desktop support dropped to nearly zero. The sys admins, network admins and engineers don't have the lost productivity involved in updating, maintaining and patching desktop PC's. Their "workstation" has the resiliency of a server and they can connect to their desktop from anywhere and have the exact settings they want every time.
But, unfortunately, your scenario is probably more likely after all. Guys who have grown up with the concept of stand alone Windows PC's will try to clone that with Unix/Linux.
Ok, that was a bad example. But moving to "open source" is only the first step. Next step is user education, and thats going to be a big step to climb.
What are you talking about? The "typical user" needs an email client, office automation suite and a web browser. They need to be able to connect to network locations to access shared data. They need to run third party applications like Business Objects, Crystal Reports, custom stuff created in PeopleSoft or SAP or Oracle, etc. They don't need education in using the operating system for the most part.
They are going to get educated in those things, and need training in those things, regardless of the operating system. And a lot of those things they already know how to use. The differences between OpenOffice and MS Office are pretty small, as is the difference between Evolution and Outlook. Since most of the ISV's are moving towards thin applications they pretty much work the same on Linux, UNIX and Windows.
Once the users get past the "oh my god, it's a new OS" reaction, they'll get right back to work.
It's really the developers, engineers and administrators where the issue is at. In most government organizations Microsoft is deeply entrenched and these folks are going to have to make large changes in how they think and work. If they try to implement Linux and F/OSS using the same methodologies and thought processes they have used for Windows they will have issues, to say the least.
Just getting the sys admins to understand that they can administer a Linux server far more effectively, efficiently and quickly from a command line than they could in the Windows GUI will take some doing, let alone convincing them that they can, with only a bit more effort, replicate that work across a large number of servers/workstations simultaneously, all from the command line. And wait til the first time they have to compile something because there's no pretty install package available!
All through the early to mid 90's the Army was using Windows for the computers that commanders and operations staff used in the field. I suspect that a lot of the computer gear that the individual soldiers carry now is based on Windows CE, although I can't speak from personal experience on that. My first comment is based on direct, first hand experience.