I had the same thought when going through the same class. I was hoping to simply test out of it -- ended up compromising with the professor and was able to get all of the class material ahead of time (finished it within two days) but had to take the tests on test days (which I assume is understandable to minimize cheating).
A prior learning assessment would have been nice but atleast it got me out of having to attend every class session.
For your goal on attempting to change the course contents -- yah, good luck.. I'm assuming Microsoft is still pouring a LOT of $$$ into the colleges to make sure their software is being used -- definitely very annoying as other software choices are more than adequate and would require learning concepts vs following step-by-step directions (nothing more frustrating than watching someone pull out a step-by-step instruction sheet and have absolutely NO clue what is actually happening within the program to give them their results).
A forum and some custom LAMP pages... soooo why not focus on that and leave the server/hosting/etc to someone else? There are a lot of LAMP hosting providers out there that you could use for a few $$ per month.. Go with a company that can give you SSH access so you can get familiar with some CLI management (or concurrently attempting to run the same stuff on a virtualbox linux setup) and perhaps down the road once your more comfortable with the LAMP config, then migrate the site to your own server.
Google is Microsoft's #1 competition right? Of course Microsoft wants Google to continue to use Windows.. not using Windows puts Google at an even further advantage.. its not like Microsoft can drop using Windows for its internal systems.
I think open source will benefit. There will be a certain number of people who are laid off that it will be extremely difficult to find a similar job. I'm thinking of people in industries and service sector jobs that were contracting before the downturn and have become much worse (the printing industry is one in particular I'm very familiar with).
These individuals will need to "reinvent" themselves to an extent. Getting involved in an open source project will give them some experience in a new field. In addition, others may consider this being a good time to start a business (I know many people who have the desire to start their own businesses but have been reluctant due to the imaginary security blanket of a corporate job). If these people want to be successful in this economy with a new venture, they will need to be very smart with their expenses which does generally equate to utilizing open source.
actually you would have 1/4 the resolution.. cyan, magenta, yellow and black and at that, you would need quite a high resolution to make it blend properly to provide a good color gamut -- much more difficult than a simple black and white display.
Quick question. I have a Windows 2003 server and the partition where my major file shares & user's roaming profiles are stored is getting quite full.
On a Unix machine, I can add additional storage, run rsync to sync all of the files & permissions to the new location and then I run rsync one last time and remount that new storage space at the same location as the old space so all the network sharing configuration stays the same. From the users perspective, access to those shares is interuppted for less than 1 minute (keep from them updating files while the final rsync makes sure everythings up-to-date).
On Windows, I can't seem to figure out exactly how to do this. The default copy does not preserve permissions, there is no rsync type tool that comes with Windows. There is the recommendation by certain websites to use the CLI based xcopy (which doesn't seem to offer any file sync capability and most likely will puke on copying files that are currently in use).. Microsoft's KB recommends doing modifications to the registry to offer this functionality (yikes!!) and the end result is just *copying* the files, which I don't think allows me to keep all of my network mappings configured (for the record, if I changed the position of a share on a unix box, I can modify one file and do a search replace, restart the sharing and I'm done.. let it be one network share or thousands..).
According to the article, Dell says that lining up certification, support, and training will 'take a lot of work.'
Indeed. They can't even get their Vista loads correct. The recent Dell's I bought had Vista.. On boot the system had an error message about the Roxio/Sonic driver not being compatible!! hah.. On top of this, their default Outlook integration completely screws with Outlook (ie you can't use Outlook due to a significant decrease in speed and stability).
I spent way too much time troubleshooting these issues and after having the computers for over two weeks, finally got them deployed yesterday after wasting countless hours with a (very bad assumption) that a computer sold by Dell Small Business and running Vista Business would actually have been tested and work flawlessly with probably the most popular email client on Windows (Outlook 2003).
Its still new, give it time. Firefox vs IE6 was a no brainer.. there were major security issues with IE6, non-standards compliant, feature poor (ie no tabs, pop-up blocker, etc..) and these outweighed the issues with Firefox (separate download, no official MSI or group policy settings).
As IE7 gets more established and issues with IE6 are determined to no longer be an issue, there is less motivation for both home and corporate users to continue to invest time into Firefox (downloading, making MSIs, maintaining two browsers (IE + Firefox), etc..) when IE7 integrates smoothly into the "Windows experience" (comes with the OS, automatic updates, programs using MSHTML use it by default, etc..)
Hopefully I am wrong and Firefox continues to grow or atleast maintain marketshare. If anything, it should limit browser-specific sites (Which I absolutley hate).
There have *always* be viable, functionally equivalent alternatives to Windows (and all other pieces of Microsoft software).
Not when you take applications into account. While perhaps some mainstream apps have cross platform alternatives (office, databases, mail clients, web browsers, gui interface, etc..) there is still LOTS of niche/market specific software that is Windows only. Unfortunately this ties many companies to Windows. It would be great if third party developers started porting their apps to other platforms (or make them platform agnostic). Please don't mention wine/crossover or windows emulation -- just TRY getting support for that configuration.
I imagine a world where Windows is banned and replaced with Ubuntu (for the sake of argument). Imagine your family installing and updating software from CLI or giving up your favorite software or games.
I do a lot of support for *nix, Mac OS X and Windows. Even if the end-user doesn't frequent the CLI often, it does provide a very quick and easy way to support them. With Windows, doing phone support is horribly ineffective. I spend a significant amount of time navigating the user around the interface and then trying to tell them how to uncover hidden options or wade through dialogs with dozens of tabs and hundreds of options. With a good CLI, I can tell them to launch it, run a few commands, read me some info from the screen and generally fix an issue much quicker. With Windows, use of remote desktop tools is almost mandatory for all but the simpliest tasks.
If the world banned Windows, hmm.. your favorite software and games would have been ported to the dominate operating system.. you think those companies would simply stop making their software?
Imagine relearning all they know about their desktop in a Linux environment.
I really don't buy this one AT ALL nor have to imagine anything. I helped migrate an elementary school to a Solaris based network (Gnome desktops). This included teachers, administrators, students, computer labs.. everything. Besides having to learn some new icons and a few new locations for items, the "massive learning curve" issue we were anticipating never happened. Even friends, family and co-workers that use my system (KDE) don't seem to have any major issues being productive.. they ask how to access the web, write a letter, whatever, I tell them what app(s) they can try and thats about it. Sure doing adminsitrative tasks are different, but day to day operations for *most* end users is such a small learning curve as to be a non-issue.
Windows also has a lot of software not offered on other platforms, such as Photoshop, Flash (the IDE), Dreamweaver, 3DSMax and so on.
That is true and the #1 reason why people stick with Windows. I think ultimately these apps need to be ported. The amount of development time and effort that has gone into each of those apps is astronomical and it is unrealistic to think that, at this time, these types of niche apps would attract the necessary development community to make a highly successful, competitive open source alternative.
they would need to change their procedures and way of running
That is true. Admins that only know the Windows/Microsoft way of doing things would need to be removed and unix/linux admins would need to take their place.
The second reply was saying that they couldn't roll out updates through out the network with free software in the same way that they could MS and they couldn't control it with the same level of precision that they could with MS products
As others have said, they are looking at rolling out a Unix infrastructure that resembles a Windows network. Ideally most systems would be running as thin clients with a central server cluster. You update the software at the master server, it populates the cluster and all of the thin clients run software from there. It is a much less intensive administrative overhead, reduces the need for admins/IT to go computer-to-computer (Thin clients are more reliable, if one dies, unplug it and plug in another one..) and all administrative tasks are centralized.
I provide technical consulting to a school that rolled out a thin client configuration across 150 computers (elementary school). There are 4 backend servers (for future expansion). The network just umm.. works. Upgrading software only happens at the master server, and everything rolls out nice. We can have multiple versions of software (if necessary) without having to deal with conflicts as on Windows. We don't have to ghost or create disk images. If a user has a problem, because they are ultimately running on the server, it is a piece-of-cake to see exactly what is on their screen and assist them from anywhere (remotely, from the front office, where-ever). There are no emergencies. The backend servers provide redunancy so if one happens to fail -- no problem. If a software upgrade fails, we only need to rollback at the master -- no need to jump around to various computers.
Furthermore -- centralization makes backups a breeze.. there is no need for antivirus/spyware software on each of the computers (simplified), user accounts are true "underprivleged" without needing to do registry hacks to make software run.. information is centralized so if updates need to be made, a simple shell script usually suffices -- not to mention significantly less equipment to monitor and have spare parts on-hand in the event of hardware failure. Then there is the upgrade factor -- when we need more power -- we can add another server or upgrade the existing few servers we have.. no need to buy hundreds of standalone computers, ghost, migrate desktops, etc.
The only major issue is legacy (windows) software that doesn't have a *nix port or equivilant. Those end up being run via terminal server which is adequate but not nearly as integrated as we would like.
The only way for Linux to succeed in these kind of settings is to make Wine work flawlessly.
Not necessarily. There *IS* the possibility of a thin client configuration (most office workers do not need a fat client at their desk) and run terminal services for legacy Windows apps. Granted this depends on the needs of the user/organization but this does provide Windows compatibility to the desktop without abandoning support (your running our Windows app in Wine?!? no support for you!!). In many situations, this type of setup makes a LOT of sense -- much lower TCO, centralized backups/redundancy, centralized support, portable office workers (hop on any thin client and get access to everything.. infact, your desktop including all running apps can follow you from one thin client to the next...).
I'll agree.. IF your using XP/Vista and IE7 is default -- there needs to be a compelling reason to install and maintain another browser. With IE6 this was simple as you had many choices: security, interface features, standards compliance, etc.
IE7 *seems* to address many of those issues.. In addition, if your a company that has a Microsoft based network, IE7 fits nicer in the network than Firefox -- Official Active Directory/Group Policy support, updates via a centralized method alongside other Windows updates, etc. I think the lack of official AD/MSI features for Firefox was a HUGE mistake that made many companies stay away from Firefox.
ISP's block all outbound port 25 traffic from customer's computers (except to their own smtp server); and then unblock anyone who complains - the rationale is if you know enough to complain, you know enough to keep your own system reasonably safe
Mobile individuals HATE the port 25 limits. That rates VERY high as a complaint. If it was easy & free to get this unblocked, I think a LOT more than just "geeks" would do this for the convienance of using a non-ISP SMTP server. I think "geeks" and IT types favor using port 587 w/SMTP_AUTH to a remote SMTP server (which is what I end up doing).
While I agree that ISPs should block outgoing 25 for residential users (according to RFC2476, port 25 should be used by MTAs only -- generally running a server is against the ISP's TOS for residential/personal accounts), unfortunately, it is not an Internet law enforcing this for all ISPs. As an ISP, there is no managable way to only accept mail from "authorized" mail servers (ie other ISP run systems).
ISP's attempt to block spam before the spam arrives in their network. If they can block it (ie a specific mail server is a known spam source, so block the IP via a realtime blacklist) this reduces the bandwidth to receive the message, the cpu cycles to do a spam/virus scan and the resources to store the message.
For my private company mail servers, they end up averaging about 60%-80% of all incoming mail is SPAM. I'd expect with larger ISPs, such as AOL and Comcast, this ratio is even worse -- perhaps 4 spams or more for every 1 legitimate mail (or greater) due to being a much larger target for things like distributed mail campaigns, dictionary-based mailings, etc.
So this is a HUGE problem. Unfortunately it is getting worse with no real tangable solution available. As a result, spam filtering is getting more agressive and false positives are more common.
Anyone who has a machine of that generation is going to leave it as it is. Linux is not an option.
Not necessarily. If I am a school with lets say two computer labs (30 computers each), I could buy a server (or two) and utilize all of those systems as thin clients. This would be a huge upgrade as far as performance and capabilities from a Win98 config (in most cases) and significantly reduce the administrative overhead (frequent re-imaging of systems, BSODs, etc..). In addition, the upgrade would be probably 1/10th the cost of the upgrade to WinXP (with hardware) and offer much more out of the box (desktop publishing, office suite, graphic editing, full internet suite, programming tools, educational titles, etc..) with the added bonus of minimal cost of future upgrades (just upgrade the server).
Yeah, but let's not pretend that everyone else is able to meet specs and standards perfectly either.
Yah but I can develop a XHTML/CSS website to the standard, using one of the mentioned browsers (firefox, mozilla, konqueror, safari, opera, etc..) and when I view that website in any other standards compliant browser on various operating systems, it looks VERY close if not the same. Then I view it in Internet Explorer and absolutely cringe.
As a developer, you get the following options:
1. Develop exclusively for Internet Explorer and don't care about any other browser (fortunately this mentality is dying due to the marketshare of alternative browsers). Other browsers tend to display these sites fairly well as long as there is no IE-specific crap (ie: active x)
2. Develop a tables based design with limited CSS.. basically, throw out lots of accessibility related formatting, but limit overall development time. Seems like most sites favor this method.
3. Develop two separate sites.. do browser detection (yuck) and serve up (at minimum) a separate CSS doc for Internet Explorer than for all other current browsers. Works ok until you realize the incompatibilities between different versions of IE and end up having to do version checks and maintain many separate layouts.
4. Develop to the standard and if IE can't display it properly, oh well.. (not terribly useful for most sites).
Its absolutely aggervating as a web developer to not only learn a standard and code to the standard, but end up having to learn the "IE" way and all the various hacks and workarounds (I believe Microsoft refers to this type of crap as "shims").. when working on a new layout, its not unrealistic to end up having to spend twice as long just to make a standards compliant XHTML/CSS design work in IE.. Its a shame because *most* developers simply can't devote the time ($$) and as a result, webpages are not standards compliant, less accessible and harder to maintain.
Microsoft already has competitors, in the form of Apple, Linux, Google, and web app vendors who want to kill the desktop altogether. One more competitor, loaded with cash, unencumbered by a requirement to maintain backward compatibility with Windows, and given a well-articulated mission might be able to come up with something radically new and better than anything currently available.
That is true. But building something radically new and better at this point in time is a FAR larger task than building the Mac OS over 22 years earlier.
Over the past 22 years, the GUI has largely remained unchanged.. we ar still using mice, icons, and windows to interact with the system. Alternative systems such as 3D interfaces, voice activated computing, stylus based computing, etc have largely been relegated to small niche markets as they don't offer the balance of ease of use and productivity for *most* people.
I definitely like the idea of building out a new platform that is designed for todays computing environment. I think the use of hardware virtualization to provide Win32 compatibility could be a definite option to ease transition. However it is a HUGE gamble. If it forces developers to largely rewrite their applications -- who is to say they will end up sticking with the Windows lock-in? It is definitely an interesting thought and I'd actually like to see Microsoft take a stab at it.. out of any company, they are in a position to realize this massive undertaking.
Obviously it sounds like we both find value in both systems (we are using both.. though it sounds like I lean slightly more heavily toward the *nix usage).
IMHO, adding users by running 'adduser blah' or even by manually editing/etc/passwd and friends is not meaningfully less abstracted than clicking "Add User" in Windows.
It is very clear how the system functions in *nix. When reading books on system administration regarding *nix, it does go into detail on boot process, configurations, how tools/applications interact with one another, etc.. Couple this with generally very good verbose output (either to the console or the syslog), it is very easy to pin-point problems when they exist. As a result, I feel more confident in resolving an issue due to this openness compared to Windows (which troubleshooting tends to take the form of running various 3rd party tools or diagnosing vague error messages or symptoms). While I think abstraction level is a good term for this, perhaps I am mistaken.
Windows is always going to fundamentally be a black box - I don't dispute that. What I do argue is that a) the average unix system is just as much a black box to the average unix administrator, even the more old-skool ones; and b) this doesn't have a meaningfully negative impact in the real world.
#1 Windows is a blackbox -- limits capability. #2 open source *nix is not. It has a meaningful impact on the real world. Open source *nix runs on more hardware. It fosters "3rd party" development throughout the entire scope of the system. It allows scaling and fine-tuning that is not possible in Windows (which I might add in certain circumstances has provided a huge performance boost). Openness and open standards are the fundamental underpinnings of the Internet, sites like google (given the sheer amount of low-level tweaks Google has made to Linux, I *highly* doubt Windows could have provided a platform for them to build their empire) and even new "killer apps" such as Asterisk and all it provides (Which btw is a GREAT example of openness vs traditional blackbox pbx implimentations). I think it has meaningful impact.
Well, there's little you can do about broken software on any platform. Unfortunately Windows has suffered a plague of the stuff, largely (IMHO) due to the unmatched low barrier to entry into developing for the platform.
Unmatched? A person can download Linux for free, boot it up on a computer for free and get a full suite of software development tools, for free. They can develop graphical apps (GTK or QT), standard C/C++ apps, web apps, database apps.. all for free. I can take any of this software developed for Linux, FreeBSD, whatever and package it up in a manner that can be distributed to hundreds of thousands of systems in a consistent, reproducable method. Look at FreeBSD.. I can install over 14,000 apps using "make install".. create a custom binary package and push it out to hundreds of systems with a full method of reporting if the update failed/etc. I don't BUY that it is due to the "low barrier of entry". It truly is aggervating. I guess Windows has much more broken software as there is VERY FEW apps that I can throw into a group policy and distribute without issues (oddly, most are open source packages such as PDFCreator.. funny).
Another problem with Windows, being more designed and coherent rather than evolved and fragmented like Unix is/was, it tends to rely on a similarly "designed" and "coherent" infrastructure to work well. If you don't have - and can't provide - that, you're probably never going to get it working at its peak potential.
Give me a break. The only reason why it gets its "peak potential" is due to Microsoft's reluctance to open up its protocols and APIs. period. It has nothing to do with the design. Microsoft breaks compatiability consistently.
1. Built under their "security is top priority" and "trustworthy computing" iniatives.
2. Microsoft built security focused tools such as.NET.. I'm sure its used extensively in their flagship operating system and applications.
3. Given the long development cycle, I'd have to imagine they recoded most of the system and not based it off of their previous code which all has major critical security issues.
4. I'd have to imagine in the effort to keep the system secure, backwards compatibility is largely sandboxed to not allow this insecure code to infect the integrity of the system.
5. With the knowledge that most home users (And small business users) ARE THE administrator, I'm sure they are taking special precautions to provide resources to enhance their knowledge of security and maintaining a secure system. With the 10+ gigabyte default install and modern day video capabilities, I'd imagine they have lots of video to get this knowledge out to people.
6. They have stated it is not only the most secure WINDOWS release ever, but the most secure OPERATING SYSTEM ever. I don't recall this being the case with previous releases. They even attended a blackhat conference (or something) to prove this! It must be true.
7. For extra precaution, they have high system requirements and excessive annoyances (such as making the simple task of deleting a desktop icon into a 6+ step procedure) to provide a barrier so just not everyone buys it the day it is released. Seems like they have structured it so most people won't get it until atleast SP1 or later which should be great to provide extra time to make it even more secure then the most secure OS ever.
Based on all of this. I am positive that Microsoft is right and you are wrong. a'Yup..
While you certainly don't get the Windows source as a matter of course, the documentation - if you can be bothered to look (and, I'll agree, most cannot) - is quite comprehensive.
Can you recommend some reading? I have a myriad of books on NT, 2000, XP and Windows 2003 Server (yes I do admin these systems) but even after reading extensive documentation, I still very much feel like there is a black box level that is inhibiting me from truly groking the system. Do realize that I have used Windows for far longer (~16 years) than *nix (about 6.5 years).
because I don't know any other unix admins who consider looking at the source code to their OS as a normal part of their system maintenance.
It is useful. There have been many times I have found having access to source code to be quite useful in adding a feature here, adjusting a database, etc..etc.. Even checking out diffs for updates can be enlightening -- espeically (As you have mentioned) when stuff goes wrong. As mentioned, with Windows you are stuck with relying on others (generally phone support to someone who doesn't care) to resolve even very simple issues.
serious flaw in the platform
Obviously a difference of opinion. Having the option to dig down, debug and fix issues in the field is very powerful. While I don't have to do it often, when the need arises, it is really great to truly understand why an issue happens versus largely guessing, reinstalling or otherwise waiting for a support person to attempt to recreate the scenario and develop a solution. Sure, not all administrators have the skill set to do these things, but its nice not to be limited when the need arises.
I was being a bit cowboyish
Indeed. I'm guessing these were fairly critical servers to be having a cluster/HA config. Thats one of the features of Linux.. you can get into that bleeding edge software (CVS code patches?!?) and end up finding yourself part of the development cycle. I really don't think thats a recommend approach for managing production level servers -- especially without testing on a dev server first. Seems like an enterprise grade distribution and only using approved updates would have been a smarter thing to do.. even if it is using a freebie version (ie CentOS). Besides, once you got in over your head, you should have rolled back the system (you did save a copy of the old package/source tree.. yes?).
The capabilities in such a system for security, auditing, concurrent access, consistency and automated backup/revision control - to name a few - are primitive at best.
Throw it into a revision control system and your off and running. Pretty standard stuff.. I'm curious, talking about concurrent access, consistency, revision control, versioning, etc.. how are these things handled in a Windows environment? Things like active directory/group policy are great for centralization, but I've haven't really seen where I can get revision/version control.. granted, the Windows networks I help maintain are relatively small (under 100 client systems, one or two servers) so administrative access is largely restricted, but to understand config changes over time would be useful.
error-prone humans are continually forced to do work that computers can do faster, more reliably and more accurately.
*shrug* I dunno.. Even when setting up Mac OS X servers which do have a lot of semi-decent GUI tools (or even Webmin which does a good job for gui based configuration) I still found myself regularly editing configs directly due to the fact the GUI was lacking in fully exploiting the configuration capacity of an application. The Windows method has largely been to develop overly complex options/customization dialogs with tiers of tabs and hundreds of check boxes, pull downs and other widgets.. I guess after a while, I find doing a text
Is this not the clearest evidence yet that Microsoft is "dying" and on its way out?
No. This taken by itself is just the life cycle of the business. This could result in a shift in upper management which could rejuvinate MS and make it more competetive in the marketplace. Its hard to say. However, there have been many business to have their founders leave/retire/fired and the company continues to survive and be successful.
While I have enjoyed this friendly banter, I believe our viewpoints and perspectives on the issue differ too much to enlighten each other.. and it is a shame to put this amount of effort into these very long posting to not atleast get some karma for our efforts.:)
Having said that, it would be discourteous of me not to respond to your lengthy reply. So here it comes..
1. As far as "more abstracted".. It has nothing to do with the GUI, it has to do with the fact that many aspects of Windows are "off limits". I have the full source to my *nix machines and basic introductory text tend to fully explain how the entire system functions. This empowers me as an adminsitrator to quickly diagnose and correct issues. Windows troubleshooting tends to be largely curing symptoms of issues due to the fact that core problems are less easily uncovered (such as a bug in an application, I have no way to personally go in and analyze the code or have tools at my disposal to easily debug an application to find the root cause.. generaly this requires support calls and waiting around for a solution).
2. There is more commonality in different variations of *nix than differnet releases of Windows (Windows 3.x vs Windows Vista for example). While commands may different, underlying philosophies are largely the same and while it does take some time to become acquanted with a different variety of *nix, a large set of core knowledge about the system is applicable from one to the next (ie hopping between FreeBSD and RedHat or Solaris). Given that there is no equivilent in the Windows world, I think this is largely a moot point.
3. I'll agree with Linux being a patchwork... its definitely getting better, but I migrated from Linux to FreeBSD about 3 years ago and while I do like to tinker with Linux on occasion, I definitely enjoy the consistency of FreeBSD. When discussion *nix vs Windows, I tend to overlook the patchwork-esqe qualities of Linux due to my far more familiarity with FreeBSD, and for this, I am wrong.
4. When I originally started using Linux after using Windows for close to a decade, I was put off by the lack of GUI config tools.. while the situation has improved, I have found the use of these GUI interfaces to be lacking when compared to editing text based configs. I find the power to script against the configs, configuring versioning and test different configs very useful. The fact that Microsoft is developing monad and greatly enhancing its CLI interface (along with XML based configs, etc..) tend to lend creedence that perhaps configs *should* remain in the textual relam.
5. "capable of meeting todays computing needs".. there are many areas that Windows falls short. It does not scale well either up or down. Microsoft had to release the WinCE which besides having the Windows name and being developed by Microsoft is a completely separate code base. Linux has been scaled to work on cell phones, PDAs, set-top boxes (Tivo), embedded applications, networking boxes (ie Linksys Routers) and so forth. Even at the high end, Linux has been scaled to work on the fastest of the supercomputers. It has proven itself as a capable server system at a variety of levels. Besides perhaps a lack of 3rd party support, it is even a very respectable desktop OS that continues to gain traction and marketshare. Linux and BSD's are able to run on a variety of hardware platforms and CPUs. It is generally one of the first operating systems to be ported to a new platform. Take a look at how long 64-bit Windows took to debut. This is what I am talking about. Staying with Windows limits your choice in hardware and your choice in how to deploy your technical solution. While it may work adequately for tasks Microsoft has deemed within its focus, it has not proven to scale well outside of that realm.
Perhaps after typing all of this, looking back, everything is largely "market driven" and perhaps not "technically better" (in a very broad sense). Ultimately the issue en
Slashdot Mirror
I had the same thought when going through the same class. I was hoping to simply test out of it -- ended up compromising with the professor and was able to get all of the class material ahead of time (finished it within two days) but had to take the tests on test days (which I assume is understandable to minimize cheating).
A prior learning assessment would have been nice but atleast it got me out of having to attend every class session.
For your goal on attempting to change the course contents -- yah, good luck .. I'm assuming Microsoft is still pouring a LOT of $$$ into the colleges to make sure their software is being used -- definitely very annoying as other software choices are more than adequate and would require learning concepts vs following step-by-step directions (nothing more frustrating than watching someone pull out a step-by-step instruction sheet and have absolutely NO clue what is actually happening within the program to give them their results).
A forum and some custom LAMP pages ... soooo why not focus on that and leave the server/hosting/etc to someone else? There are a lot of LAMP hosting providers out there that you could use for a few $$ per month.. Go with a company that can give you SSH access so you can get familiar with some CLI management (or concurrently attempting to run the same stuff on a virtualbox linux setup) and perhaps down the road once your more comfortable with the LAMP config, then migrate the site to your own server.
Google is Microsoft's #1 competition right? Of course Microsoft wants Google to continue to use Windows.. not using Windows puts Google at an even further advantage.. its not like Microsoft can drop using Windows for its internal systems.
I think open source will benefit. There will be a certain number of people who are laid off that it will be extremely difficult to find a similar job. I'm thinking of people in industries and service sector jobs that were contracting before the downturn and have become much worse (the printing industry is one in particular I'm very familiar with).
These individuals will need to "reinvent" themselves to an extent. Getting involved in an open source project will give them some experience in a new field. In addition, others may consider this being a good time to start a business (I know many people who have the desire to start their own businesses but have been reluctant due to the imaginary security blanket of a corporate job). If these people want to be successful in this economy with a new venture, they will need to be very smart with their expenses which does generally equate to utilizing open source.
actually you would have 1/4 the resolution .. cyan, magenta, yellow and black and at that, you would need quite a high resolution to make it blend properly to provide a good color gamut -- much more difficult than a simple black and white display.
Quick question. I have a Windows 2003 server and the partition where my major file shares & user's roaming profiles are stored is getting quite full.
.. Microsoft's KB recommends doing modifications to the registry to offer this functionality (yikes!!) and the end result is just *copying* the files, which I don't think allows me to keep all of my network mappings configured (for the record, if I changed the position of a share on a unix box, I can modify one file and do a search replace, restart the sharing and I'm done .. let it be one network share or thousands..).
On a Unix machine, I can add additional storage, run rsync to sync all of the files & permissions to the new location and then I run rsync one last time and remount that new storage space at the same location as the old space so all the network sharing configuration stays the same. From the users perspective, access to those shares is interuppted for less than 1 minute (keep from them updating files while the final rsync makes sure everythings up-to-date).
On Windows, I can't seem to figure out exactly how to do this. The default copy does not preserve permissions, there is no rsync type tool that comes with Windows. There is the recommendation by certain websites to use the CLI based xcopy (which doesn't seem to offer any file sync capability and most likely will puke on copying files that are currently in use)
Indeed. They can't even get their Vista loads correct. The recent Dell's I bought had Vista.. On boot the system had an error message about the Roxio/Sonic driver not being compatible!! hah.. On top of this, their default Outlook integration completely screws with Outlook (ie you can't use Outlook due to a significant decrease in speed and stability).
I spent way too much time troubleshooting these issues and after having the computers for over two weeks, finally got them deployed yesterday after wasting countless hours with a (very bad assumption) that a computer sold by Dell Small Business and running Vista Business would actually have been tested and work flawlessly with probably the most popular email client on Windows (Outlook 2003).
Its still new, give it time. Firefox vs IE6 was a no brainer .. there were major security issues with IE6, non-standards compliant, feature poor (ie no tabs, pop-up blocker, etc..) and these outweighed the issues with Firefox (separate download, no official MSI or group policy settings).
As IE7 gets more established and issues with IE6 are determined to no longer be an issue, there is less motivation for both home and corporate users to continue to invest time into Firefox (downloading, making MSIs, maintaining two browsers (IE + Firefox), etc..) when IE7 integrates smoothly into the "Windows experience" (comes with the OS, automatic updates, programs using MSHTML use it by default, etc..)
Hopefully I am wrong and Firefox continues to grow or atleast maintain marketshare. If anything, it should limit browser-specific sites (Which I absolutley hate).
I read that headline and thought you were referring to an army of this guy .. AHH!!!
If the world banned Windows, hmm.. your favorite software and games would have been ported to the dominate operating system.. you think those companies would simply stop making their software?
I really don't buy this one AT ALL nor have to imagine anything. I helped migrate an elementary school to a Solaris based network (Gnome desktops). This included teachers, administrators, students, computer labs.. everything. Besides having to learn some new icons and a few new locations for items, the "massive learning curve" issue we were anticipating never happened. Even friends, family and co-workers that use my system (KDE) don't seem to have any major issues being productive.. they ask how to access the web, write a letter, whatever, I tell them what app(s) they can try and thats about it. Sure doing adminsitrative tasks are different, but day to day operations for *most* end users is such a small learning curve as to be a non-issue.
That is true and the #1 reason why people stick with Windows. I think ultimately these apps need to be ported. The amount of development time and effort that has gone into each of those apps is astronomical and it is unrealistic to think that, at this time, these types of niche apps would attract the necessary development community to make a highly successful, competitive open source alternative.
That is true. Admins that only know the Windows/Microsoft way of doing things would need to be removed and unix/linux admins would need to take their place.
As others have said, they are looking at rolling out a Unix infrastructure that resembles a Windows network. Ideally most systems would be running as thin clients with a central server cluster. You update the software at the master server, it populates the cluster and all of the thin clients run software from there. It is a much less intensive administrative overhead, reduces the need for admins/IT to go computer-to-computer (Thin clients are more reliable, if one dies, unplug it and plug in another one..) and all administrative tasks are centralized.
I provide technical consulting to a school that rolled out a thin client configuration across 150 computers (elementary school). There are 4 backend servers (for future expansion). The network just umm.. works. Upgrading software only happens at the master server, and everything rolls out nice. We can have multiple versions of software (if necessary) without having to deal with conflicts as on Windows. We don't have to ghost or create disk images. If a user has a problem, because they are ultimately running on the server, it is a piece-of-cake to see exactly what is on their screen and assist them from anywhere (remotely, from the front office, where-ever). There are no emergencies. The backend servers provide redunancy so if one happens to fail -- no problem. If a software upgrade fails, we only need to rollback at the master -- no need to jump around to various computers.
Furthermore -- centralization makes backups a breeze
The only major issue is legacy (windows) software that doesn't have a *nix port or equivilant. Those end up being run via terminal server which is adequate but not nearly as integrated as we would like.
I'll agree.. IF your using XP/Vista and IE7 is default -- there needs to be a compelling reason to install and maintain another browser. With IE6 this was simple as you had many choices: security, interface features, standards compliance, etc.
.. In addition, if your a company that has a Microsoft based network, IE7 fits nicer in the network than Firefox -- Official Active Directory/Group Policy support, updates via a centralized method alongside other Windows updates, etc. I think the lack of official AD/MSI features for Firefox was a HUGE mistake that made many companies stay away from Firefox.
IE7 *seems* to address many of those issues
While I agree that ISPs should block outgoing 25 for residential users (according to RFC2476, port 25 should be used by MTAs only -- generally running a server is against the ISP's TOS for residential/personal accounts), unfortunately, it is not an Internet law enforcing this for all ISPs. As an ISP, there is no managable way to only accept mail from "authorized" mail servers (ie other ISP run systems).
ISP's attempt to block spam before the spam arrives in their network. If they can block it (ie a specific mail server is a known spam source, so block the IP via a realtime blacklist) this reduces the bandwidth to receive the message, the cpu cycles to do a spam/virus scan and the resources to store the message.
For my private company mail servers, they end up averaging about 60%-80% of all incoming mail is SPAM. I'd expect with larger ISPs, such as AOL and Comcast, this ratio is even worse -- perhaps 4 spams or more for every 1 legitimate mail (or greater) due to being a much larger target for things like distributed mail campaigns, dictionary-based mailings, etc.
So this is a HUGE problem. Unfortunately it is getting worse with no real tangable solution available. As a result, spam filtering is getting more agressive and false positives are more common.
As a developer, you get the following options:
1. Develop exclusively for Internet Explorer and don't care about any other browser (fortunately this mentality is dying due to the marketshare of alternative browsers). Other browsers tend to display these sites fairly well as long as there is no IE-specific crap (ie: active x)
2. Develop a tables based design with limited CSS
3. Develop two separate sites
4. Develop to the standard and if IE can't display it properly, oh well.. (not terribly useful for most sites).
Its absolutely aggervating as a web developer to not only learn a standard and code to the standard, but end up having to learn the "IE" way and all the various hacks and workarounds (I believe Microsoft refers to this type of crap as "shims").. when working on a new layout, its not unrealistic to end up having to spend twice as long just to make a standards compliant XHTML/CSS design work in IE.. Its a shame because *most* developers simply can't devote the time ($$) and as a result, webpages are not standards compliant, less accessible and harder to maintain.
That is true. But building something radically new and better at this point in time is a FAR larger task than building the Mac OS over 22 years earlier.
Over the past 22 years, the GUI has largely remained unchanged
I definitely like the idea of building out a new platform that is designed for todays computing environment. I think the use of hardware virtualization to provide Win32 compatibility could be a definite option to ease transition. However it is a HUGE gamble. If it forces developers to largely rewrite their applications -- who is to say they will end up sticking with the Windows lock-in? It is definitely an interesting thought and I'd actually like to see Microsoft take a stab at it
It is very clear how the system functions in *nix. When reading books on system administration regarding *nix, it does go into detail on boot process, configurations, how tools/applications interact with one another, etc.. Couple this with generally very good verbose output (either to the console or the syslog), it is very easy to pin-point problems when they exist. As a result, I feel more confident in resolving an issue due to this openness compared to Windows (which troubleshooting tends to take the form of running various 3rd party tools or diagnosing vague error messages or symptoms). While I think abstraction level is a good term for this, perhaps I am mistaken.
#1 Windows is a blackbox -- limits capability. #2 open source *nix is not. It has a meaningful impact on the real world. Open source *nix runs on more hardware. It fosters "3rd party" development throughout the entire scope of the system. It allows scaling and fine-tuning that is not possible in Windows (which I might add in certain circumstances has provided a huge performance boost). Openness and open standards are the fundamental underpinnings of the Internet, sites like google (given the sheer amount of low-level tweaks Google has made to Linux, I *highly* doubt Windows could have provided a platform for them to build their empire) and even new "killer apps" such as Asterisk and all it provides (Which btw is a GREAT example of openness vs traditional blackbox pbx implimentations). I think it has meaningful impact.
Unmatched? A person can download Linux for free, boot it up on a computer for free and get a full suite of software development tools, for free. They can develop graphical apps (GTK or QT), standard C/C++ apps, web apps, database apps
Give me a break. The only reason why it gets its "peak potential" is due to Microsoft's reluctance to open up its protocols and APIs. period. It has nothing to do with the design. Microsoft breaks compatiability consistently.
But Vista is the one! Just think about it..
.NET .. I'm sure its used extensively in their flagship operating system and applications.
1. Built under their "security is top priority" and "trustworthy computing" iniatives.
2. Microsoft built security focused tools such as
3. Given the long development cycle, I'd have to imagine they recoded most of the system and not based it off of their previous code which all has major critical security issues.
4. I'd have to imagine in the effort to keep the system secure, backwards compatibility is largely sandboxed to not allow this insecure code to infect the integrity of the system.
5. With the knowledge that most home users (And small business users) ARE THE administrator, I'm sure they are taking special precautions to provide resources to enhance their knowledge of security and maintaining a secure system. With the 10+ gigabyte default install and modern day video capabilities, I'd imagine they have lots of video to get this knowledge out to people.
6. They have stated it is not only the most secure WINDOWS release ever, but the most secure OPERATING SYSTEM ever. I don't recall this being the case with previous releases. They even attended a blackhat conference (or something) to prove this! It must be true.
7. For extra precaution, they have high system requirements and excessive annoyances (such as making the simple task of deleting a desktop icon into a 6+ step procedure) to provide a barrier so just not everyone buys it the day it is released. Seems like they have structured it so most people won't get it until atleast SP1 or later which should be great to provide extra time to make it even more secure then the most secure OS ever.
Based on all of this. I am positive that Microsoft is right and you are wrong. a'Yup..
Can you recommend some reading? I have a myriad of books on NT, 2000, XP and Windows 2003 Server (yes I do admin these systems) but even after reading extensive documentation, I still very much feel like there is a black box level that is inhibiting me from truly groking the system. Do realize that I have used Windows for far longer (~16 years) than *nix (about 6.5 years).
It is useful. There have been many times I have found having access to source code to be quite useful in adding a feature here, adjusting a database, etc..etc.. Even checking out diffs for updates can be enlightening -- espeically (As you have mentioned) when stuff goes wrong. As mentioned, with Windows you are stuck with relying on others (generally phone support to someone who doesn't care) to resolve even very simple issues.
Obviously a difference of opinion. Having the option to dig down, debug and fix issues in the field is very powerful. While I don't have to do it often, when the need arises, it is really great to truly understand why an issue happens versus largely guessing, reinstalling or otherwise waiting for a support person to attempt to recreate the scenario and develop a solution. Sure, not all administrators have the skill set to do these things, but its nice not to be limited when the need arises.
Indeed. I'm guessing these were fairly critical servers to be having a cluster/HA config. Thats one of the features of Linux .. you can get into that bleeding edge software (CVS code patches?!?) and end up finding yourself part of the development cycle. I really don't think thats a recommend approach for managing production level servers -- especially without testing on a dev server first. Seems like an enterprise grade distribution and only using approved updates would have been a smarter thing to do.. even if it is using a freebie version (ie CentOS). Besides, once you got in over your head, you should have rolled back the system (you did save a copy of the old package/source tree.. yes?).
Throw it into a revision control system and your off and running. Pretty standard stuff.. I'm curious, talking about concurrent access, consistency, revision control, versioning, etc.. how are these things handled in a Windows environment? Things like active directory/group policy are great for centralization, but I've haven't really seen where I can get revision/version control .. granted, the Windows networks I help maintain are relatively small (under 100 client systems, one or two servers) so administrative access is largely restricted, but to understand config changes over time would be useful.
*shrug* I dunno.. Even when setting up Mac OS X servers which do have a lot of semi-decent GUI tools (or even Webmin which does a good job for gui based configuration) I still found myself regularly editing configs directly due to the fact the GUI was lacking in fully exploiting the configuration capacity of an application. The Windows method has largely been to develop overly complex options/customization dialogs with tiers of tabs and hundreds of check boxes, pull downs and other widgets.. I guess after a while, I find doing a text
No. This taken by itself is just the life cycle of the business. This could result in a shift in upper management which could rejuvinate MS and make it more competetive in the marketplace. Its hard to say. However, there have been many business to have their founders leave/retire/fired and the company continues to survive and be successful.
While I have enjoyed this friendly banter, I believe our viewpoints and perspectives on the issue differ too much to enlighten each other.. and it is a shame to put this amount of effort into these very long posting to not atleast get some karma for our efforts. :)
.. It has nothing to do with the GUI, it has to do with the fact that many aspects of Windows are "off limits". I have the full source to my *nix machines and basic introductory text tend to fully explain how the entire system functions. This empowers me as an adminsitrator to quickly diagnose and correct issues. Windows troubleshooting tends to be largely curing symptoms of issues due to the fact that core problems are less easily uncovered (such as a bug in an application, I have no way to personally go in and analyze the code or have tools at my disposal to easily debug an application to find the root cause.. generaly this requires support calls and waiting around for a solution).
.. there are many areas that Windows falls short. It does not scale well either up or down. Microsoft had to release the WinCE which besides having the Windows name and being developed by Microsoft is a completely separate code base. Linux has been scaled to work on cell phones, PDAs, set-top boxes (Tivo), embedded applications, networking boxes (ie Linksys Routers) and so forth. Even at the high end, Linux has been scaled to work on the fastest of the supercomputers. It has proven itself as a capable server system at a variety of levels. Besides perhaps a lack of 3rd party support, it is even a very respectable desktop OS that continues to gain traction and marketshare. Linux and BSD's are able to run on a variety of hardware platforms and CPUs. It is generally one of the first operating systems to be ported to a new platform. Take a look at how long 64-bit Windows took to debut. This is what I am talking about. Staying with Windows limits your choice in hardware and your choice in how to deploy your technical solution. While it may work adequately for tasks Microsoft has deemed within its focus, it has not proven to scale well outside of that realm.
Having said that, it would be discourteous of me not to respond to your lengthy reply. So here it comes..
1. As far as "more abstracted"
2. There is more commonality in different variations of *nix than differnet releases of Windows (Windows 3.x vs Windows Vista for example). While commands may different, underlying philosophies are largely the same and while it does take some time to become acquanted with a different variety of *nix, a large set of core knowledge about the system is applicable from one to the next (ie hopping between FreeBSD and RedHat or Solaris). Given that there is no equivilent in the Windows world, I think this is largely a moot point.
3. I'll agree with Linux being a patchwork... its definitely getting better, but I migrated from Linux to FreeBSD about 3 years ago and while I do like to tinker with Linux on occasion, I definitely enjoy the consistency of FreeBSD. When discussion *nix vs Windows, I tend to overlook the patchwork-esqe qualities of Linux due to my far more familiarity with FreeBSD, and for this, I am wrong.
4. When I originally started using Linux after using Windows for close to a decade, I was put off by the lack of GUI config tools.. while the situation has improved, I have found the use of these GUI interfaces to be lacking when compared to editing text based configs. I find the power to script against the configs, configuring versioning and test different configs very useful. The fact that Microsoft is developing monad and greatly enhancing its CLI interface (along with XML based configs, etc..) tend to lend creedence that perhaps configs *should* remain in the textual relam.
5. "capable of meeting todays computing needs"
Perhaps after typing all of this, looking back, everything is largely "market driven" and perhaps not "technically better" (in a very broad sense). Ultimately the issue en