ICANN was stupid for promoting.XXX as an official domain.. honestly, what is the point?
1. Existing pornographic sites will not relenguish control of their existing domains.
2. Governments may be enticed to pass legislation to force those sites to the.XXX domain. Of course, said sites could be run remotely and avoid legislation. This leading to (perhaps) more costly hosting for the rest of us (bandwidth, available regional options, etc..)
3. A huge debate happens as to what is "pornography" -- is it showing an ankle? A women not fully clothed? A nude photo (no its art!)? Depending on your culture and beliefs, it varies widely.
Its just stupid. It is not a black and white issue where a site can be deemed ".XXX" worthy or not. A ratings system would be better (quantify the content type) but again, poses many of the same issues.
The fact that ICANN went ahead with.XXX in the first place either indicates its a $$$ maker domain which is unacceptable (IMHO) or the policy makers are complete idiots who do not think about the consequences of their actions, which is again, unacceptable. Either way, ICANN sucks and the Bush admin sucks for thinking porn will go away if.XXX is abolished. hah..
1. Not saving as a Word doc? Perhaps the major problem is Word does not open Star Office format? There are lots of people now using Star Office/OpenOffice.org -- I'm guessing it is the 2nd or 3rd most popular office suite. Besides, its free -- so if people receive an OpenOffice or Star Office document, they can download openoffice.org for free and open it.
2. I agree, teaching how to use an application instead of how different types of software should work is a HUGE problem. Perhaps an Office -> OpenOffice.org tutorial is needed (not sure if it exists) that outline the differences.
3/4. This seems better addressed in OOo2 -- though in defense of OpenOffice, atleast it is the same version available on a wide variety of platforms. MS Office is only available on Windows and Macintosh and even at that, they are not the same version and have different feature sets. In addition to this, Microsoft generally create new interface guidelines (widgets, icons, etc..) with new versions of Office that do not integrate seamlessly with the rest of the desktop.
5. One reason why people go for the font drop down in MS Office is because using style sheets in MS Office is a PITA. Not only is it hidden but awkward to use. Sure, use of style sheets might be foreign to an MS Office user, but it is such a HUGE productivity enhancer (consistency in formating, dynamic creation of table of contents and other indexes, etc..) that it makes it worth learning that workflow.
Besides that, logicly, I believe the OpenOffice.org interface is laid out much more soundly. For example, if I want to format a character or paragraph, I can go under "Format". But if I want to format the page, in OpenOffice.org it is consistent and under "Format" -- MS Office? The file menu! Sure MS Office users are use to this inconsistency (among a huge list of others) but for some reason, I'm guessing a new user would find the OpenOffice.org workflow and layout easier to master.
Out of all of your "major problems" I think #2 is the biggest problem. As classes are tailored for a specific version of a Microsoft product (taking a look at your local community college's class offerings will confirm this) and other classes assume specific knowledge of Microsoft Office, it is no surprise that this is what people are most comfortable with. Unfortunately, as you mentioned the default methods of document creation in Word are, to put it kindly, horrible. The fact it does not promote separation of content and layout via stylesheets or even provide an adequate stylesheet system is a HUGE productivity drain on businesses -- much more so than the difference in acquisition or support cost.
Isn't a BSD distro going to be about the same as a Linux distro?
In the fact that the majority of the same software runs on both platforms and if you know one system you can get up to speed on the other quickly, yes, they are about the same.
The difference (for me) comes in the execution. Here are a few things I really like about FreeBSD:
1. FreeBSD is an operating system, not a kernel. As a result, there is a userland and tools that are apart of a base FreeBSD system that you can expect to be there on ALL FreeBSD systems.
2. Clean separation of the OS and applications. There is a very clear distinction of what is apart of the OS and what is an application. This keeps binaries, configuration files and other elements separated on the file system and easy to find.
3. Consistently followed file hierarchy. This ties into #2 -- the entire ports collection (over 13,000 apps last time I checked) follows the default file hierarchy of FreeBSD. As a result, everything loads and places files on the system in a consistent manner.
4. Package management. The use of ports, portupgrade, portaudit and the pkg_* commands makes it easy to maintain, be notified of security vulnerabilities and update the applications on a given system. If an upgrade is not straight forward, the ports team will include documentation (/usr/ports/UPDATING) on the correct procedure to ensure a working system.
5. ipfw/ipfilter.. I have never really understood iptables. ipfw and ipfilter seem much more logical and straight forward to me and as a result, as an admin, I fully understand how the firewall works and is configured -- which is important.
6. Great documentation. The FreeBSD handbook, Absolute BSD, man pages, etc are comprehensive and straightforward. As a result, I was able to get to a point of feeling empowered with FreeBSD faster than Linux and *generally* can find answers to my problems very quickly (instead of resorting to forums or google searches). This documentation is augmented with sites such as O'Reilly's BSD site and the FreeBSD Diary.
7. Generally pretty good hardware support. It supports what I need but my hardware requirements are pretty modest.
8. Easy kernel config.. generally speaking, it is pretty easy to configure the kernel and update the system (within the same version (ie 4.x series)). The kernel config is a documented text file where you can comment out items you don't need, add items that you need. If you foobar your kernel (non-bootable system) it is possible to boot from your old kernel (it is auto renamed to "kernel.old") or you can keep as many kernels as you want (ie for testing purposes). If you don't want to mess with the kernel, I believe you can load all optional features as modules.
Manuals? haha.. Lets see.. I did some volume licenses of Office, WinXP and Win2003 Server. What did I get for my $$$? Well umm.. a FedEx'd letter of a webpage printout with a login and code on it.
That was it. No manuals, no software, no nothing. I had to buy the CDs separate ($25 a pop) and even at that, it wasn't the latest versions (it came with service packs and other updates on additional CDs).
Do I have a manual for it? Why sure.. I went to the bookstore and bought a manual, just the same way I did for FreeBSD, Linux and all of my other servers and apps.
The biggest difference is with FreeBSD and Linux, I generally get the latest versions of software (not having to install service packs right off the bat or research "slipstreaming" or other non-sense).
And several development models can and should coexist in healthy competition... but Microsoft does not like competition and kills it at any opportunity by buying out companies, having "creative" budling agreements with OEMs and so forth. Many companies have decided against investing into new ideas because of these anti-competitive practices. How does that create a "healthy competition"?
Microsoft's participation in standards bodies such as IETF, W3C and OASIS
I was unaware of their participation in OASIS.. does this mean we will see the Office suite utilize OASIS in the future or atleast provide the ability to easily import/export OASIS documents?
How about W3C? Seems like MS is very much behind in their "participation" to web standards.
The VCR is a good example of a standards-based product that allowed any video tape* to play on any player - providing a marketplace of competitive VCR implementations, competitive tape media suppliers, and commercial opportunities.
Kinda like how FOSS can *generally* be easily run on any operating system providing a marketplace of competitive OS implementations, hardware architectures and commercial opportunities. Like you said "best tool for the job.."
At the end of the day, we want software to "just work" too. That's what great software is all about.
What better way to make it "just work" than have Microsoft create it all! hehe..
Overall I don't see much value in this interview at all.. there is no doubt that the Linux guy at Microsoft would get this level of treatment. Microsoft needs guys inside the company that fully understands the competition and can provide information so the execs/marketing can maintain success in the marketplace.
The bottom line is this guys reality is not anywhere close to what is happening. Microsoft fostering friendly competition? Microsoft adhering to and promoting open standards (w3c, oasis as examples??)? 'co-opetition'?? Blech.
Needless to say, I wouldn't mind having this guys job.. mess around with FOSS all day, get paid well to do so and every once in a while fire off some BS to management about the state of FOSS (copy and paste from slashdot??:)
Why the heck dont schools use thin clients to servers.
I'll keep you posted. I am working with a school that is changing out their W2k-based network for Sun thin clients -- 20 thin clients per server. The numbers look really interesting. The first phase of the installs should be done by end of September. If all things go well, the entire school (labs, computers in classrooms and administrative computers) will be thin clients within two years.
I agree. I'd like to nominate Microsoft to make this happen. But lets instead of having an arbitrary license, tap this into DRM! That way, our DRM-enabled trustworthy computers will be allowed to install/run certain types of trusted apps depending on the DRM-based license of the operator.
Of course, we won't have un-trustworthy computers *cough*everything non-Windows Vista*cough* on this new world order Internet because well.. that would just screw everything up. And pigs will fly out of my ass.
Not to mention that the last time I checked, IE was given away for free too.
It's not like anyone's paying any money for it.
Hmm.. Internet Explorer is a part of Windows. If you have a legitimate copy of Windows than yes, you did pay for it.
Free versions of Internet Explorer (ie Internet Explorer for the Macintosh, updates for Win95/98/Me) have been long gone.
Microsoft has stated that new versions of Internet Explorer would co-incide with new versions of Windows.. of course, IE7 _seems_ slightly different but I'm guessing given that they are planning on adding a LOT of new features to Beta 2 (as per the weblog) it wouldn't surprise me if it DID coincide with Vista's release.. wait and see.
I wish they would include an auto update feature. I hate having to go to fifty different company's websites to see if they've released a new version of their software or driver. That's one thing MS got right. Firefox did it too.
How you figure Microsoft got it right? If they got it right, their auto-update would be an API that all of your software can tap into and provide you a list of updates so you can choose to apply updates (a la FreeBSD Ports, Gentoo Portage, etc..etc..)
The only thing Microsoft has is a half-baked solution that only updates a subset of their own software.
Bloat is the fact that each is designed around a different framework, so they don't get reused as much as they should be because you have to install both to get all your applications to work.
I agree. Good luck getting the Gnome devs to switch over to KDE.:-) IT would be great to have all apps have a KDE native interface (so you can tap into all the framework benefits of KDE/QT)
No. The point of schools is to create good little conforming consumers/drones. If you want your children to be taught how to think you'll need to do it yourself.
duh. your right. i don't know what i was thinking. having people think for themselves might result in *gasp* conflict of ideas and perhaps *gasp* getting your feelings hurt once in a while. can't have that. individualism bad, conformation good. *Sigh*
I think the general problem is that the populace just doesn't know enough about general functions of the computer, and instead have just familiarized themselves with whatever interface they use
Exactly. The only way your going to get around that is to introduce people to a variety of computer systems.
When I was in school, I had access to a wide variety of systems -- Apple IIe, IIgs, Mac classic, Commodore (various models), DOS, Windows 3.0/3.1, Windows 95.
Today, so many of the schools I visit have standardized on Windows 2000 or Windows XP. Its a shame. Whats worse, is teachers focus on only teaching the interface and not underlying concepts.
In the time I was at school, I learned how to operate many different systems, many different programs and was able to separate concepts from interfaces so when using a new system, it didn't take much to learn (or program).
Isn't that the point of schools? Teach people how to learn for themselves?
I don't know if this will actually work or not, as I've never encountered such a situation myself-- but have you thought of placing those programs in non-standard folders and renaming the executables? I'm guessing such programs just use some kind of getProcByName() type function, as doing anything more complex would be, well, more complex. In my experience, not many things go for the more complex route. Worth a try at least!
Thanks for the suggestion.. I did end up doing that and it still closed it. So I wrote the developer a nasty-gram and found another solution. My thoughts are developers who write crap code to being with then get in my way when I try to make it work has no place on the systems I admin.
Any time you think it requires administrator rights, you can pretty easily figure out what resource it needs and give them access to it explicitly using tools like regmon and filemon.
Sure.. you can use regmon/filemon to see whats going on under the hood..but it takes time.. The obvious problems can be fixed pretty quick.. but then there are hidden problems (ie writing to a registry key and failing silently) or infrequent items (occasionally writing to the file system when certain criteria are met). This can take a LOT of time and a LOT of end-user aggervation.
Even AFTER debugging these apps (which is exactly what is going on) for FREE (atleast relative to the developer) do they include the fixes? So far, nope. Infact, most were not interested. Their recommendation? Run as administrator.
It just blows my mind that these companies write user level applications and do not fully test with regular user level accounts. Whats truly aggervating is most of the time, it is simple stuff -- writing to the wrong spot in the registry, not keeping application data in the user folder. Simple stuff.
And this points at where the problem lies - the users. They're generally lazy and uninformed.
While this might be true.. its not the entire story. The entire story is simple -- there is still a LOT of software out there that simply DOES NOT RUN 100% CORRECTLY OUT OF THE BOX in anything BESIDES an administrative level account.
Even things that SHIP WITH WINDOWS are prone to oversight which tells me one thing (and has been second'ed but not necessarily confirmed on/.) -- Microsoft doesn't believe in restricted access in its development model (read: Microsoft employees all have administrative level access).
So is it any wonder that people DON'T do this? Its one thing to have a slight PITA factor when installing apps (as you can't simply say "hey here is my administrative level password.. install away!") but when you install apps and they may or may not work.. or might load but not work fully (ie write to a restricted part of the registry or file system without checking for success and not providing good error messages on what went wrong).
From my professional experience setting up a "secure" windows environment -- there is a LOT of use of filemon, regmon and other tools to basically guess as to why apps fail and make the environment slightly more insecure so these apps can run (ie provide user write permissions to system registry nodes or certain file system areas)... even then, my success is quite low given the extremely LARGE amount of data that is spewed from these apps (not to mention certain apps that cause the said apps to close so they can't capture the data (piracy checking??))
anyways.. its not even close to a reality. The mindset of programmers, developers, managers and microsoft is still NOT high on restricted user rights security and it is VERY apparent.
Is it better? sure.. but its still not even CLOSE to being as good as on the *nix side even AFTER well over a decade since NT debuted.. fun.
The thing about ClamAV's distribution that is particularly innovative is the use of DNS TXT, not incremental updates. By uses DNS for systems to check for updates, it is refreshed every *15 minutes*. As a result, if a new definition is out, in theory, it could be distributed to all ClamAV systems within a half hour.
The addition of use of small incremental updates, low-cost (from a server/bandwidth POV) technologies for distribution and a push system for mirror updates, it really makes a difference to get virus updates to systems ASAP.
When this is compared to 3hr, 6hr or even 1 week updates (as it pertains to Symantec's LiveUpdate) the response time of the ClamAV team + network is truly innovative. The fact it "Just Works" makes it "insanely great"
ClamAV virus definition distribution model (use of incremental updates, dns txt field checks for new updates, automatic, etc..) -- compare this to the weekly (!) updates of Symantec (or manually updating slightly more frequently) or even some of the "download a big chunk from a centralized location" method of commercial competitors.
BitTorrent
So many things in KDE its insane.. (just check out all the awards, including Software Innovation of the Year - CeBit!)
Plone, Zope, Typo3 - These content management systems lead the way for both commercial and opensource.. so much innovation going on here
CUPS - While not glamerous, I have setup lots of print servers and the flexibility and modularlity of CUPS (in my experience) is unmatched.
The spam fighters: greylisting, spamassassin, amavisd, postfix, dnsrbl, etc.. developed under or made popular due to opensource.. I have yet to come across _any_ non-FOSS solution that comes close to the success and accuracy of the OSS tools for spam filtering
If the current rate to Verisign is $4.25 per domain and there are 5,324,213 registered.net domains as of January 16th.. thats $22,627,905. So exactly what is all this money used for?
A few distributed dns servers and a (should be) highly automated system for managing domains and a handful of support people? That sure doesn't seem like $22m worth of expenses.. what else is it used for?
I recently evaluated SugarCRM for use with an internal system we are building.
Hmm.. its internal? I thought section 3 of the GPL allowed companies to build on GPL software WITHOUT releasing source as long as it wasn't distributed outside the company?
I've had the enjoyment of learning all about LUA about two months ago. A very umm.. textbook example of a small network -- Win2k3 server, WinXP Pro clients.
Needless to say, this was not even CLOSE to what a UNIX user account is like.
Few thoughts..
1. App compatibility - very annoying. While some apps are kind enough to out-right say they suck and are not compatible, there are LOTS of apps that fail in *silent* ways. Mostly writing to folders and registry w/o checking for access rights. There are many apps that attempt to write temporary files outside of user folders (ie the Program Files folder) or even store user prefs in the system registry.
2. Along with #1 -- there are many things INSIDE WinXP that fail. One very annoying example is msconfig.. it throws up a dialog after an admin does some changes but for a user and does not acknowledge the user's response (silently fails when writing to a system registry key). I have no idea why a user is prompted when an admin does a modification. Same thing with user defaults -- the system, even though it prompts to set a browser as default, silently fails when setting registry keys (again, not a user registry key). Apparently there is no way to adjust registry key security from a GPO or script to grant users this access (w/o going to each system manually)..
3. runas.. hehe.. that is so not even close to su/sudo -- while there appears to be lots of little workarounds (ie logging into administrative network shares of drives) its cumbersome and adds so much extra time to troubleshooting.
4. Fonts.. I really don't understand why users don't have their own fonts folder. I had to manually go into each computer, modify the registry to give permission to add fonts, adjust the fonts folder permissions, yada yada.. PITA. A user font folder (that follows them if roaming profiles is enabled) would have been a piece-of-cake while leaving the system font folder small and fast.
I agree. I use to run ClamAV side-by-side with other commercial scanners (mcafee, symantec, sophos) and I found ClamAV to be easier to setup, easier to keep up-to-date and it has kept up with or exceeded the commercial competition.
Needless to say, as subscriptions have come up for renewal, I have not been renewing. ClamAV has been absolutely fantastic as a mail gateway virus scanner.
Yah but then you get into a HUGE question as to what *SHOULD* be the operating system. Some people would argue that the OS only truly needs to be the kernel/drivers to provide a standardized API to the hardware. Everything else -- compilers, shells, graphical interfaces, web browsers, etc are "applications" and not "operating system" and hense should not be included.
This is basically how Linux operates, even though most distros will include many of the same standard software parts that have become known as the "Linux OS" (gnu, xorg, kde/gnome, etc..)
I suppose you could do a piece-meal Windows -- start with the NT kernel and add functions as you want them or as applications you want to use demand them (ideally the apps would be written to a certain open API so you could install a different chunk of code (ie firefox instead of IE) if so desired) but in all honesty, does anyone REALLY want that? You instantly made computing much more complex and for what purpose?
As long as the default back-end apps for a particular API are *good enough* there is not a HUGE demand for a third-party solution. Which kinda sucks -- the competition needs to be FAR better than the OS manufacturers solution and even at that, the OS can simply improve their implimentation to keep control of a given subsystem.
Needless to say, it sucks. Though honestly, I don't know what the solution is. I really don't think it is forcing arbitrary regulations on the OS manufacturer (such as the EU ruling). Ideally, competition would be healthy to provide alternative operating systems to choose from (and in this way, it seems to be getting healthier with OS X and Linux making in-roads) but ultimately the same fate awaits on those platforms as well..
I think ultimately most non-niche software will be integrated to some extent into the base OS. So either you have your own OS or focus on being a niche software team.
No, what Linux need to become a prominent desktop OS is for a major VAR (Mike Dell, I'm talking to you) to start selling Linux-based desktops for a significant cost-savings over Windows XP.
Nah.. sell it for MORE than WinXP.. then people will be like "whoah.. Linux costs MORE than WinXP.. it must be much better!!"
Which of course, you can market it that way.. no spyware issues, virus issues, better default security model, more/better included software, default theme-able interface (KDE), free software updates to new major versions, etc..:)
Of course, we do need to address the "but does XYZ[1] software app run under Linux?" question..
[1] quickbooks, quicken, turbotax, photoshop, particular video editing software, certain game, etc..
Slashdot Mirror
ICANN was stupid for promoting .XXX as an official domain.. honestly, what is the point?
.XXX domain. Of course, said sites could be run remotely and avoid legislation. This leading to (perhaps) more costly hosting for the rest of us (bandwidth, available regional options, etc..)
.XXX in the first place either indicates its a $$$ maker domain which is unacceptable (IMHO) or the policy makers are complete idiots who do not think about the consequences of their actions, which is again, unacceptable. Either way, ICANN sucks and the Bush admin sucks for thinking porn will go away if .XXX is abolished. hah..
1. Existing pornographic sites will not relenguish control of their existing domains.
2. Governments may be enticed to pass legislation to force those sites to the
3. A huge debate happens as to what is "pornography" -- is it showing an ankle? A women not fully clothed? A nude photo (no its art!)? Depending on your culture and beliefs, it varies widely.
Its just stupid. It is not a black and white issue where a site can be deemed ".XXX" worthy or not. A ratings system would be better (quantify the content type) but again, poses many of the same issues.
The fact that ICANN went ahead with
Major problems?
1. Not saving as a Word doc? Perhaps the major problem is Word does not open Star Office format? There are lots of people now using Star Office/OpenOffice.org -- I'm guessing it is the 2nd or 3rd most popular office suite. Besides, its free -- so if people receive an OpenOffice or Star Office document, they can download openoffice.org for free and open it.
2. I agree, teaching how to use an application instead of how different types of software should work is a HUGE problem. Perhaps an Office -> OpenOffice.org tutorial is needed (not sure if it exists) that outline the differences.
3/4. This seems better addressed in OOo2 -- though in defense of OpenOffice, atleast it is the same version available on a wide variety of platforms. MS Office is only available on Windows and Macintosh and even at that, they are not the same version and have different feature sets. In addition to this, Microsoft generally create new interface guidelines (widgets, icons, etc..) with new versions of Office that do not integrate seamlessly with the rest of the desktop.
5. One reason why people go for the font drop down in MS Office is because using style sheets in MS Office is a PITA. Not only is it hidden but awkward to use. Sure, use of style sheets might be foreign to an MS Office user, but it is such a HUGE productivity enhancer (consistency in formating, dynamic creation of table of contents and other indexes, etc..) that it makes it worth learning that workflow.
Besides that, logicly, I believe the OpenOffice.org interface is laid out much more soundly. For example, if I want to format a character or paragraph, I can go under "Format". But if I want to format the page, in OpenOffice.org it is consistent and under "Format" -- MS Office? The file menu! Sure MS Office users are use to this inconsistency (among a huge list of others) but for some reason, I'm guessing a new user would find the OpenOffice.org workflow and layout easier to master.
Out of all of your "major problems" I think #2 is the biggest problem. As classes are tailored for a specific version of a Microsoft product (taking a look at your local community college's class offerings will confirm this) and other classes assume specific knowledge of Microsoft Office, it is no surprise that this is what people are most comfortable with. Unfortunately, as you mentioned the default methods of document creation in Word are, to put it kindly, horrible. The fact it does not promote separation of content and layout via stylesheets or even provide an adequate stylesheet system is a HUGE productivity drain on businesses -- much more so than the difference in acquisition or support cost.
Isn't a BSD distro going to be about the same as a Linux distro?
.. I have never really understood iptables. ipfw and ipfilter seem much more logical and straight forward to me and as a result, as an admin, I fully understand how the firewall works and is configured -- which is important.
.. generally speaking, it is pretty easy to configure the kernel and update the system (within the same version (ie 4.x series)). The kernel config is a documented text file where you can comment out items you don't need, add items that you need. If you foobar your kernel (non-bootable system) it is possible to boot from your old kernel (it is auto renamed to "kernel.old") or you can keep as many kernels as you want (ie for testing purposes). If you don't want to mess with the kernel, I believe you can load all optional features as modules.
In the fact that the majority of the same software runs on both platforms and if you know one system you can get up to speed on the other quickly, yes, they are about the same.
The difference (for me) comes in the execution. Here are a few things I really like about FreeBSD:
1. FreeBSD is an operating system, not a kernel. As a result, there is a userland and tools that are apart of a base FreeBSD system that you can expect to be there on ALL FreeBSD systems.
2. Clean separation of the OS and applications. There is a very clear distinction of what is apart of the OS and what is an application. This keeps binaries, configuration files and other elements separated on the file system and easy to find.
3. Consistently followed file hierarchy. This ties into #2 -- the entire ports collection (over 13,000 apps last time I checked) follows the default file hierarchy of FreeBSD. As a result, everything loads and places files on the system in a consistent manner.
4. Package management. The use of ports, portupgrade, portaudit and the pkg_* commands makes it easy to maintain, be notified of security vulnerabilities and update the applications on a given system. If an upgrade is not straight forward, the ports team will include documentation (/usr/ports/UPDATING) on the correct procedure to ensure a working system.
5. ipfw/ipfilter
6. Great documentation. The FreeBSD handbook, Absolute BSD, man pages, etc are comprehensive and straightforward. As a result, I was able to get to a point of feeling empowered with FreeBSD faster than Linux and *generally* can find answers to my problems very quickly (instead of resorting to forums or google searches). This documentation is augmented with sites such as O'Reilly's BSD site and the FreeBSD Diary.
7. Generally pretty good hardware support. It supports what I need but my hardware requirements are pretty modest.
8. Easy kernel config
Funny moderation? umm.. I'm glad I could humor someone, but they ARE doing this. The initial contract is signed and installation scheduled.
Manuals? haha.. Lets see.. I did some volume licenses of Office, WinXP and Win2003 Server. What did I get for my $$$? Well umm.. a FedEx'd letter of a webpage printout with a login and code on it.
.. I went to the bookstore and bought a manual, just the same way I did for FreeBSD, Linux and all of my other servers and apps.
That was it. No manuals, no software, no nothing. I had to buy the CDs separate ($25 a pop) and even at that, it wasn't the latest versions (it came with service packs and other updates on additional CDs).
Do I have a manual for it? Why sure
The biggest difference is with FreeBSD and Linux, I generally get the latest versions of software (not having to install service packs right off the bat or research "slipstreaming" or other non-sense).
And several development models can and should coexist in healthy competition. .. but Microsoft does not like competition and kills it at any opportunity by buying out companies, having "creative" budling agreements with OEMs and so forth. Many companies have decided against investing into new ideas because of these anti-competitive practices. How does that create a "healthy competition"?
.. does this mean we will see the Office suite utilize OASIS in the future or atleast provide the ability to easily import/export OASIS documents?
.. mess around with FOSS all day, get paid well to do so and every once in a while fire off some BS to management about the state of FOSS (copy and paste from slashdot?? :)
Microsoft's participation in standards bodies such as IETF, W3C and OASIS
I was unaware of their participation in OASIS
How about W3C? Seems like MS is very much behind in their "participation" to web standards.
The VCR is a good example of a standards-based product that allowed any video tape* to play on any player - providing a marketplace of competitive VCR implementations, competitive tape media suppliers, and commercial opportunities.
Kinda like how FOSS can *generally* be easily run on any operating system providing a marketplace of competitive OS implementations, hardware architectures and commercial opportunities. Like you said "best tool for the job.."
At the end of the day, we want software to "just work" too. That's what great software is all about.
What better way to make it "just work" than have Microsoft create it all! hehe..
Overall I don't see much value in this interview at all.. there is no doubt that the Linux guy at Microsoft would get this level of treatment. Microsoft needs guys inside the company that fully understands the competition and can provide information so the execs/marketing can maintain success in the marketplace.
The bottom line is this guys reality is not anywhere close to what is happening. Microsoft fostering friendly competition? Microsoft adhering to and promoting open standards (w3c, oasis as examples??)? 'co-opetition'?? Blech.
Needless to say, I wouldn't mind having this guys job
Why the heck dont schools use thin clients to servers.
I'll keep you posted. I am working with a school that is changing out their W2k-based network for Sun thin clients -- 20 thin clients per server. The numbers look really interesting. The first phase of the installs should be done by end of September. If all things go well, the entire school (labs, computers in classrooms and administrative computers) will be thin clients within two years.
internet user license program
I agree. I'd like to nominate Microsoft to make this happen. But lets instead of having an arbitrary license, tap this into DRM! That way, our DRM-enabled trustworthy computers will be allowed to install/run certain types of trusted apps depending on the DRM-based license of the operator.
Of course, we won't have un-trustworthy computers *cough*everything non-Windows Vista*cough* on this new world order Internet because well.. that would just screw everything up. And pigs will fly out of my ass.
Not to mention that the last time I checked, IE was given away for free too.
.. Internet Explorer is a part of Windows. If you have a legitimate copy of Windows than yes, you did pay for it.
.. of course, IE7 _seems_ slightly different but I'm guessing given that they are planning on adding a LOT of new features to Beta 2 (as per the weblog) it wouldn't surprise me if it DID coincide with Vista's release.. wait and see.
It's not like anyone's paying any money for it.
Hmm
Free versions of Internet Explorer (ie Internet Explorer for the Macintosh, updates for Win95/98/Me) have been long gone.
Microsoft has stated that new versions of Internet Explorer would co-incide with new versions of Windows
I wish they would include an auto update feature. I hate having to go to fifty different company's websites to see if they've released a new version of their software or driver. That's one thing MS got right. Firefox did it too.
How you figure Microsoft got it right? If they got it right, their auto-update would be an API that all of your software can tap into and provide you a list of updates so you can choose to apply updates (a la FreeBSD Ports, Gentoo Portage, etc..etc..)
The only thing Microsoft has is a half-baked solution that only updates a subset of their own software.
Bloat is the fact that each is designed around a different framework, so they don't get reused as much as they should be because you have to install both to get all your applications to work.
:-) IT would be great to have all apps have a KDE native interface (so you can tap into all the framework benefits of KDE/QT)
I agree. Good luck getting the Gnome devs to switch over to KDE.
No. The point of schools is to create good little conforming consumers/drones.
If you want your children to be taught how to think you'll need to do it yourself.
duh. your right. i don't know what i was thinking. having people think for themselves might result in *gasp* conflict of ideas and perhaps *gasp* getting your feelings hurt once in a while. can't have that. individualism bad, conformation good. *Sigh*
I think the general problem is that the populace just doesn't know enough about general functions of the computer, and instead have just familiarized themselves with whatever interface they use
Exactly. The only way your going to get around that is to introduce people to a variety of computer systems.
When I was in school, I had access to a wide variety of systems -- Apple IIe, IIgs, Mac classic, Commodore (various models), DOS, Windows 3.0/3.1, Windows 95.
Today, so many of the schools I visit have standardized on Windows 2000 or Windows XP. Its a shame. Whats worse, is teachers focus on only teaching the interface and not underlying concepts.
In the time I was at school, I learned how to operate many different systems, many different programs and was able to separate concepts from interfaces so when using a new system, it didn't take much to learn (or program).
Isn't that the point of schools? Teach people how to learn for themselves?
I don't know if this will actually work or not, as I've never encountered such a situation myself-- but have you thought of placing those programs in non-standard folders and renaming the executables? I'm guessing such programs just use some kind of getProcByName() type function, as doing anything more complex would be, well, more complex. In my experience, not many things go for the more complex route. Worth a try at least!
Thanks for the suggestion.. I did end up doing that and it still closed it. So I wrote the developer a nasty-gram and found another solution. My thoughts are developers who write crap code to being with then get in my way when I try to make it work has no place on the systems I admin.
Any time you think it requires administrator rights, you can pretty easily figure out what resource it needs and give them access to it explicitly using tools like regmon and filemon.
.. you can use regmon/filemon to see whats going on under the hood. .but it takes time.. The obvious problems can be fixed pretty quick.. but then there are hidden problems (ie writing to a registry key and failing silently) or infrequent items (occasionally writing to the file system when certain criteria are met). This can take a LOT of time and a LOT of end-user aggervation.
Sure
Even AFTER debugging these apps (which is exactly what is going on) for FREE (atleast relative to the developer) do they include the fixes? So far, nope. Infact, most were not interested. Their recommendation? Run as administrator.
It just blows my mind that these companies write user level applications and do not fully test with regular user level accounts. Whats truly aggervating is most of the time, it is simple stuff -- writing to the wrong spot in the registry, not keeping application data in the user folder. Simple stuff.
And this points at where the problem lies - the users. They're generally lazy and uninformed.
.. its not the entire story. The entire story is simple -- there is still a LOT of software out there that simply DOES NOT RUN 100% CORRECTLY OUT OF THE BOX in anything BESIDES an administrative level account.
/.) -- Microsoft doesn't believe in restricted access in its development model (read: Microsoft employees all have administrative level access).
.. install away!") but when you install apps and they may or may not work .. or might load but not work fully (ie write to a restricted part of the registry or file system without checking for success and not providing good error messages on what went wrong).
While this might be true
Even things that SHIP WITH WINDOWS are prone to oversight which tells me one thing (and has been second'ed but not necessarily confirmed on
So is it any wonder that people DON'T do this? Its one thing to have a slight PITA factor when installing apps (as you can't simply say "hey here is my administrative level password
From my professional experience setting up a "secure" windows environment -- there is a LOT of use of filemon, regmon and other tools to basically guess as to why apps fail and make the environment slightly more insecure so these apps can run (ie provide user write permissions to system registry nodes or certain file system areas)... even then, my success is quite low given the extremely LARGE amount of data that is spewed from these apps (not to mention certain apps that cause the said apps to close so they can't capture the data (piracy checking??))
anyways.. its not even close to a reality. The mindset of programmers, developers, managers and microsoft is still NOT high on restricted user rights security and it is VERY apparent.
Is it better? sure.. but its still not even CLOSE to being as good as on the *nix side even AFTER well over a decade since NT debuted.. fun.
The thing about ClamAV's distribution that is particularly innovative is the use of DNS TXT, not incremental updates. By uses DNS for systems to check for updates, it is refreshed every *15 minutes*. As a result, if a new definition is out, in theory, it could be distributed to all ClamAV systems within a half hour.
The addition of use of small incremental updates, low-cost (from a server/bandwidth POV) technologies for distribution and a push system for mirror updates, it really makes a difference to get virus updates to systems ASAP.
When this is compared to 3hr, 6hr or even 1 week updates (as it pertains to Symantec's LiveUpdate) the response time of the ClamAV team + network is truly innovative. The fact it "Just Works" makes it "insanely great"
If the current rate to Verisign is $4.25 per domain and there are 5,324,213 registered .net domains as of January 16th.. thats $22,627,905. So exactly what is all this money used for?
A few distributed dns servers and a (should be) highly automated system for managing domains and a handful of support people? That sure doesn't seem like $22m worth of expenses.. what else is it used for?
I recently evaluated SugarCRM for use with an internal system we are building.
Hmm.. its internal? I thought section 3 of the GPL allowed companies to build on GPL software WITHOUT releasing source as long as it wasn't distributed outside the company?
I've had the enjoyment of learning all about LUA about two months ago. A very umm.. textbook example of a small network -- Win2k3 server, WinXP Pro clients.
.. it throws up a dialog after an admin does some changes but for a user and does not acknowledge the user's response (silently fails when writing to a system registry key). I have no idea why a user is prompted when an admin does a modification. Same thing with user defaults -- the system, even though it prompts to set a browser as default, silently fails when setting registry keys (again, not a user registry key). Apparently there is no way to adjust registry key security from a GPO or script to grant users this access (w/o going to each system manually)..
.. hehe.. that is so not even close to su/sudo -- while there appears to be lots of little workarounds (ie logging into administrative network shares of drives) its cumbersome and adds so much extra time to troubleshooting.
.. I really don't understand why users don't have their own fonts folder. I had to manually go into each computer, modify the registry to give permission to add fonts, adjust the fonts folder permissions, yada yada.. PITA. A user font folder (that follows them if roaming profiles is enabled) would have been a piece-of-cake while leaving the system font folder small and fast.
Needless to say, this was not even CLOSE to what a UNIX user account is like.
Few thoughts..
1. App compatibility - very annoying. While some apps are kind enough to out-right say they suck and are not compatible, there are LOTS of apps that fail in *silent* ways. Mostly writing to folders and registry w/o checking for access rights. There are many apps that attempt to write temporary files outside of user folders (ie the Program Files folder) or even store user prefs in the system registry.
2. Along with #1 -- there are many things INSIDE WinXP that fail. One very annoying example is msconfig
3. runas
4. Fonts
I agree. I use to run ClamAV side-by-side with other commercial scanners (mcafee, symantec, sophos) and I found ClamAV to be easier to setup, easier to keep up-to-date and it has kept up with or exceeded the commercial competition.
Needless to say, as subscriptions have come up for renewal, I have not been renewing. ClamAV has been absolutely fantastic as a mail gateway virus scanner.
It was applied for on February 13th, 1996.
.. Tuesday. :)
Let me guess, it was a Friday?
Nope
Yah but then you get into a HUGE question as to what *SHOULD* be the operating system. Some people would argue that the OS only truly needs to be the kernel/drivers to provide a standardized API to the hardware. Everything else -- compilers, shells, graphical interfaces, web browsers, etc are "applications" and not "operating system" and hense should not be included.
This is basically how Linux operates, even though most distros will include many of the same standard software parts that have become known as the "Linux OS" (gnu, xorg, kde/gnome, etc..)
I suppose you could do a piece-meal Windows -- start with the NT kernel and add functions as you want them or as applications you want to use demand them (ideally the apps would be written to a certain open API so you could install a different chunk of code (ie firefox instead of IE) if so desired) but in all honesty, does anyone REALLY want that? You instantly made computing much more complex and for what purpose?
As long as the default back-end apps for a particular API are *good enough* there is not a HUGE demand for a third-party solution. Which kinda sucks -- the competition needs to be FAR better than the OS manufacturers solution and even at that, the OS can simply improve their implimentation to keep control of a given subsystem.
Needless to say, it sucks. Though honestly, I don't know what the solution is. I really don't think it is forcing arbitrary regulations on the OS manufacturer (such as the EU ruling). Ideally, competition would be healthy to provide alternative operating systems to choose from (and in this way, it seems to be getting healthier with OS X and Linux making in-roads) but ultimately the same fate awaits on those platforms as well..
I think ultimately most non-niche software will be integrated to some extent into the base OS. So either you have your own OS or focus on being a niche software team.
No, what Linux need to become a prominent desktop OS is for a major VAR (Mike Dell, I'm talking to you) to start selling Linux-based desktops for a significant cost-savings over Windows XP.
.. then people will be like "whoah.. Linux costs MORE than WinXP .. it must be much better!!"
:)
Nah.. sell it for MORE than WinXP
Which of course, you can market it that way.. no spyware issues, virus issues, better default security model, more/better included software, default theme-able interface (KDE), free software updates to new major versions, etc..
Of course, we do need to address the "but does XYZ[1] software app run under Linux?" question..
[1] quickbooks, quicken, turbotax, photoshop, particular video editing software, certain game, etc..